Cathodoluminescence

CATHODOLUMINESCENCE
Edited by Naoki Yamamoto

Cathodoluminescence
Edited by Naoki Yamamoto

Published by InTech
Janeza Trdine 9, 51000 Rijeka, Croatia

Copyright 2012 InTech
All chapters are Open Access distributed under the Creative Commons Attribution 3.0
license, which allows users to download, copy and build upon published articles even for
commercial purposes, as long as the author and publisher are properly credited, which
ensures maximum dissemination and a wider impact of our publications. After this work
has been published by InTech, authors have the right to republish it, in whole or part, in
any publication of which they are the author, and to make other personal use of the
work. Any republication, referencing or personal use of the work must explicitly identify
the original source.

As for readers, this license allows users to download, copy and build upon published
chapters even for commercial purposes, as long as the author and publisher are properly
credited, which ensures maximum dissemination and a wider impact of our publications.

Notice
Statements and opinions expressed in the chapters are these of the individual contributors
and not necessarily those of the editors or publisher. No responsibility is accepted for the
accuracy of information contained in the published chapters. The publisher assumes no
responsibility for any damage or injury to persons or property arising out of the use of any
materials, instructions, methods or ideas contained in the book.

Publishing Process Manager Danijela Duric
Technical Editor Teodora Smiljanic
Cover Designer InTech Design Team

First published March, 2012
Printed in Croatia

A free online edition of this book is available at www.intechopen.com
Additional hard copies can be obtained from [email protected]

Cathodoluminescence, Edited by Naoki Yamamoto
p. cm.
ISBN 978-953-51-0362-2

Contents

Preface IX
Part 1 Fundamental Techniques 1
Chapter 1 Cathodoluminescence Studies of Electron
Injection Effects in Wide-Band-Gap Semiconductors 3
Casey Schwarz, Leonid Chernyak and Elena Flitsiyan
Chapter 2 What is the Pulsed Cathodoluminescence? 31
Vladimir Solomonov and Alfiya Spirina
Chapter 3 Pulsed Cathodoluminescence of
Natural and Synthetic Diamonds Excited by
Nanosecond and Subnanosecond Electron Beams 51
E.I. Lipatov, V.M. Lisitsyn, V.I. Oleshko,
E.F. Polisadova, V.F. Tarasenko and E.H. Baksht
Chapter 4 Study of Defects by
Cathodoluminescence Measurements 71
A. Djemel and R-J. Tarento
Part 2 Application to Semiconductors 87
Chapter 5 Cathodoluminescence Properties of Zno Thin Films 89
M. Addou, J. Eboth, A. El Hichou, A. Bougrine, J.L. Bubendorff,
M. Troyon, Z. Sofiani, M. EL Jouad, K. Bahedi and M. Lamrani
Chapter 6 Multicolor Luminescence from Semiconductor
Nanocrystal Composites Tunable in an Electric Field 121
G.N. Panin
Chapter 7 Cathodoluminescence from
Amorphous and Nanocrystalline Nitride Thin Films
Doped with Rare Earth and Transition Metals 161
Muhammad Maqbool,
Wojciech M. Jadwisienczak and Martin E. Kordesch
VI Contents

Part 3 Application to Oxides and Minerals 207
Chapter 8 Peculiarity of the Cathodoluminescence of
lpha- Alumina Prepared by Calcination of Gibbsite Powder or
Generated by Oxidation of a Metallic FeCrAl Alloy 209
Djelloul Abdelkader and Boumaza Abdecharif
Chapter 9 Cathodoluminescence Properties of
SiO
2
: Ce
3+
,Tb
3+
, SiO
2
:Ce
3+
, Pr
3+
and SiO
2
: PbS 233
Odireleng M. Ntwaeaborwa, Gugu H. Mhlongo, Shreyas S. Pitale,
Mokhotswa S. Dhlamini, Robin E. Kroon and Hendrik C. Swart
Chapter 10 Cathodo- and Photo- Luminescence of
Silicon Rich Oxide Films Obtained by LPCVD 253
Rosa Lpez-Estopier, Mariano Aceves-Mijares and Ciro Falcony
Chapter 11 Textural Characterization of
Sedimentary Zircon and Its Implication 273
Yixiong Qian, Wen Su, Zhong Li and Shoutao Peng
Part 4 Related Luminescence 283
Chapter 12 Novel Problems in the Solid State
Cathodoluminescence of Organic Materials 285
Zheng Xu and Suling Zhao
Chapter 13 A Lighting Mechanism for Flat Electron Emission Lamp 305
Jung-Yu Li, Ming-Chung Liu, Yi-Ping Lin,

Shih-Pu Chen,

Tai-Chiung Hsieh, Po-Hung Wang, Chang-Lin Chiang,
Ming-Shan Jeng, Li-Ling Lee, Hui-Kai Zeng and Jenh-Yih Juang

Preface

Cathodoluminescence (CL) is a non-destructive technique to characterize optical and
electronic properties of nanostructures in many kinds of materials. Major subject is to
investigate basic parameters in semiconductors, impurities in oxides and phase
determination of minerals. CL is similar to photoluminescence (PL) technique, but the
excitation by high energy electrons can produce all the transition to the higher energy
excitation states and induce light emission in the wide energy range concerned with
versatile types of electronic transitions in materials. This book involves 13 chapters to
present the basics in the CL technique and applications to particles, thin films and
nanostructures in semiconductors, oxides and minerals.
The first section presents fundamental techniques to measure basic parameters in
semiconductors such as carrier concentration, diffusion length and life time. Prof.
Flitsiyan shows how to get those parameters as a function of temperature and beam
irradiation using the CL related techniques. Pulsed CL is introduced by Prof.
Solomonov to be used for determination of concentration in binary phase. Prof.
Tarasenko applied the pulsed CL to study spectral-kinetic characterization of the
natural and synthetic diamonds. CL becomes a powerful imaging technique to realize
a high spatial resolution when it combines with a scanning electron microscope (SEM)
and transmission electron microscope (TEM). Prof. Djemel demonstrates the CL
imaging of defects in semiconductors by using SEM.
The application of CL is widely spread over many types of materials. Wide gap
semiconductors are recently intensively studied because of its advantage for
optoelectronic and power devices operated at high temperatures. Prof. Addou
presents the study on CL properties of impurity doped ZnO thin films. Prof. Panin
shows CL spectra from ZnO nanorods, nanoparticles and other nano-structures, and
discusses the mechanism of electric field tunable luminescence. Prof. Kordesch
presents CL peaks originating from impurities of rare earth and transition metals
doped in AlN, GaN and other nitride thin films.
Excited electrons are bounded to atomically localized centers such as impurities and
vacancies in insulators. Peculiarity of CL due to point defects in -alumina prepared
by different procedures is reported by Prof. Boubaza. Prof. Ntwaeaborwa reviews CL
properties of phosphor materials prepared by incorporating Ce, Tb and other ions in
X Preface

amorphous SiO2. CL and PL study of silicon rich oxide films are reviewed by Dr.
Lopez-Estopier. Prof. Yixiong presents characterization of internal structures of zircon
in geological rocks.
Some of the light emissions generated in devices using different excitation procedures
are closely related to CL. Prof. Xu presents a new kind of electroluminescence in
organic materials in the layered device structures. Prof. Li reviews several types of
light emission in a new lighting device.
I would like to thank all the contributors for spending their time and energy to
complete their chapters. I hope that this book provides a unique opportunity for a
reader to gain knowledge about various aspects of CL technique.

Naoki Yamamoto
Physics Department, Tokyo Institute of Technology,
Japan

Part 1
Fundamental Techniques

1
Cathodoluminescence Studies of Electron
Injection Effects in Wide-Band-Gap
Semiconductors
Casey Schwarz, Leonid Chernyak and Elena Flitsiyan
*
Physics Department, University of Central Florida, Orlando, FL,
USA
1. Introduction
Recent developments in doping and growth of ZnO stimulated a renewal of interest in this
material from the point of view of its applications in optoelectronic devices. As a direct wide
bandgap semiconductor (E
g
3.35 eV at room temperature) with high exciton binding
energy (60 meV, compared to 25 meV in GaN), ZnO is a superior candidate for minority-
carrier-based devices, such as light emitting diodes, laser diodes, and transparent p-n
junctions.
In the present state of the art, the development of the full potential of ZnO applications
hinges in part on the availability of quality, highly conductive materials of both n- and p-
type. Similar to GaN, achieving n-type conductivity in ZnO does not present a problem,
since even nominally undoped material is generally n-type, due to the electrical activity of
native defects, such as zinc interstitials, zinc antisites, and oxygen vacancies [1,2], as well as
hydrogen impurities [3].
On the other hand, p-type conductivity with sufficiently high carrier concentrations appears
to be much more elusive, mainly due to high ionization energies of potential acceptors, such
as nitrogen, phosphorus, and arsenic. This problem is compounded by high concentrations
of native and unintentional donors, which act as compensating centers, thus further
reducing the free carrier concentration. It has been shown that the background donor
concentration can be lowered if Mg is incorporated into the ZnO lattice, since each percent
of Mg increases the bandgap of ZnO by 0.02 eV, suppressing the ionization of shallow
donors [4]. The conversion to p-type can then be obtained by heavily doping the resultant
material with phosphorus followed by annealing in O
2
atmosphere [5].
Recently, successful p-type doping of ZnO also has been attained using phosphorus [6,7],
nitrogen [8], arsenic [9], and antimony [10] yielding net hole concentrations up to 10
18
cm
-3
.
Even with the advent of new technology enabling the production of viable p-type materials,
the performance of bipolar devices is fundamentally limited by the transport properties of
minority carriers. In direct band gap semiconductors, including ZnO, minority carrier

*
Corresponding author

Cathodoluminescence

4
diffusion length is generally several orders of magnitude lower than in indirect gap
materials such as silicon or germanium. In order to noticeably increase minority carrier
diffusion length by reducing scattering by the dislocation walls, the edge threading
dislocation density must be reduced by at least two orders of magnitude from a typical
value of about 10
9
cm
-2
in epitaxial ZnO layers [11-14].
Investigation of minority carrier diffusion lengths and lifetimes in both n- and p-type ZnO
is an issue of practical importance, since it has direct implications on the performance of
bipolar devices. Moreover, considering possible applications of these devices in high-
temperature electronics, the insight into the temperature dependence of minority carrier
properties is also of great value. Nonetheless, to the best of our knowledge, the reports on
this subject are rather scarce. It is therefore the goal of this work to summarize the
available information on the subject of minority carrier transport in ZnO and related
compounds, focusing on its temperature dependence and the dynamics of non-
equilibrium carrier recombination. This discussion will be preceded by a brief summary
of the role of minority carrier transport in the performance of bipolar devices, as well as
by the review of techniques of choice for measurement of the minority carrier diffusion
length.
2. Role of minority carrier diffusion length in bipolar device performance
In general, when non-equilibrium carriers are generated in a material due to external
excitation in the absence of an electric field, they diffuse over a certain distance before
undergoing recombination. This parameter, namely the average distance traveled in a
particular direction between generation and recombination, is characterized by the
diffusion length, L. The diffusion length is related to the carrier lifetime, t, (i.e., the time
between generation and recombination of non-equilibrium carriers) through carrier
diffusivity, D:
L Dt = (1)
Diffusivity, or diffusion coefficient, is determined in turn by the mobility of the carriers, ,
according to the Einstein relation:

B
k T
D
q
= (2)
where k
B
is the Boltzmanns constant, T is absolute temperature, and q is the fundamental
charge.
The diffusion process is driven by concentration gradients; since external excitation has a
much larger impact on the concentration of minority carriers than that of majority ones
(because generation density is usually much lower than the majority carrier density), it is the
minority carriers that are more susceptible to diffusion. The electron diffusion current
density is proportional to the gradient of the electron density, n, by the following relation;

n n
dn
J eD
dx
= (3)

Cathodoluminescence Studies of Electron Injection Effects in Wide-Band-Gap Semiconductors

5
Where e is the charge of the electron and D
n
is the electron diffusion coefficient. Relating the
diffusion coefficient to temperature we get:

0
exp
a
B
E
D D
k T
| |
=
|
\ .
(4)
With
a
E being the activation energy of the process. Using equation (1) we arrive at the
equation relating the diffusion length to temperature by:

0
exp
2
| |
=
|
\ .
a
n
B
E
L L
k T
(5)
CL peak intensity, I
CL
, of Near Band Edge (NBE) luminescence decays systematically with
decreasing temperature. Since CL intensity is proportional to the rate of recombination, as
the intensity decreases it is shown that the number of recombination events also decreases.
Also, since the intensity is inversely proportional to the lifetime of carriers in the band, the
decay of I
CL
indicates the increase of with temperature. This follows in an exponential
relationship represented in this equation;

,
1 exp
CL
a T
B
A
I
E
B
k T
=
( A | |
+
( |
( \ .
(6)
Where A and B are scaling factors, E
a,T
is the thermal activation energy, k
B
is the
Boltzmann constant and T is temperature.
Diffusion of minority carriers is a process that is fundamental to the operation of bipolar
photovoltaic devices, with minority carrier diffusion length being the central parameter
defining the device performance. In the presence of a p-n junction or a Schottky barrier, the
non-equilibrium minority carriers generated by external excitation (e.g., light incident on a
photodiode) within a few diffusion lengths of the space-charge region can be collected by
the built-in field and thus contribute to the current flow across the device. The greater the
diffusion length of the carriers, the more current can be collected, leading to the higher
efficiency of the device. In photodiodes, it is usually only one side of the p-n junction that
contributes to photocurrent. If the light is absorbed in the p-region of the junction, the
quantum efficiency, q, can be represented as follows:

exp( )
(1 ) 1
1
n
aW
r
aL
q
(
=
(
+

(7)
where r and a are the reflection and absorption coefficients, respectively, W is the width of
the space-charge region, and L
n
is the diffusion length of minority electrons.
Quantum efficiency is directly related to the spectral responsivity, R, of a photodiode:
( )
ph
op
I
q
R E
P E
q
= = (8)

Cathodoluminescence

6
where I
ph
is total photocurrent, P
op
is optical power incident on the device, q is the
fundamental charge, and E is the energy of the incident photons. The relationship
between minority carrier diffusion length and responsivity of Schottky photodiodes has
been examined in great detail in Ref. [15]. Schottky photodiodes are among the simplest
photovoltaic devices, where the non-equilibrium minority carriers, generated in the bulk
of the semiconductor due to light absorption, are collected by the built-in field of the
Schottky barrier deposited on the surface of the semiconductor. In order for the carriers to
contribute to device current, they have to be generated within a few diffusion lengths of
the collector.
For incident energies greater than the bandgap of the absorber material, non-equilibrium
electron-hole pairs are generated only in the thin layer next to the surface of incidence, with the
maximum depth of 1/a. This value is on the order of 100 nm in ZnO [16,17] and is generally
much smaller than the thickness of the absorber layer. Considering a front-illuminated
configuration (in which the incident light passes through the semitransparent Schottky contact),
if L is greater than the generation depth, most of the non-equilibrium minority carriers are
collected by the built-in field of the space-charge region. In this case, the internal quantum
efficiency of the device approaches 100% and the responsivity is independent of the diffusion
length value. Below this threshold, the responsivity decreases with L, provided that the width
of the space-charge region is smaller than the generation depth.
If the energy of incident light is below the bandgap, light penetration depth is large (several
micrometers), and a fair portion of the non-equilibrium carriers is generated in the neutral
region of the semiconductor due to the ionization of the mid-gap levels. Since only the
carriers within a few diffusion lengths of the space-charge region contribute to
photocurrent, the responsivity at below-bandgap energies is limited by the diffusion length
(unless the diffusion length exceeds the thickness of the absorber layer, in which case the
latter is the limiting factor) [15].
3. Methods for determination of minority carrier lifetime and diffusion length
Although as of the date of this writing minority carrier transport in ZnO remains, with a few
exceptions, essentially unexplored, however, the measurement of minority carrier diffusion
length is a well-established subject. This section reviews three of the most widely used
techniques, namely Electron Beam Induced Current (EBIC), Time-Resolved
Photoluminescence (TRPL), and Time-Dependent Cathodoluminescence (CL) Measurements.
3.1 Electron beam induced current technique
Due to a unique combination of convenience and reliability, Electron Beam Induced Current
(EBIC) method is among the most popular techniques for minority carrier diffusion length
measurements. It requires comparatively simple sample preparation and is used in-situ in a
scanning electron microscope (SEM). Fig. 1 shows a typical measurement configuration
known as planar-collector geometry [18,19]. As a charge collection technique, EBIC method
employs a Schottky barrier or a p-n junction to collect the current resulting from the non-
equilibrium minority carriers generated by the beam of the SEM. As the beam is moved
away from the barrier/junction in a line-scan mode, the current decays as fewer and fewer
minority carriers are able to diffuse to the space-charge region.


7
d
Electron
Beam
Current
Meter

Fig. 1. Experimental setup used for EBIC measurements in planar-collector configuration.
The dashed ellipse represents the generation volume; d is the variable distance between
electron beam and the Schottky contact, represented by a solid rectangle.
The mathematical model for EBICurrent [20], shows that the decay of current, I
EBIC
, can be
described by the following expression:
exp
EBIC
d
I Ad
L
o
| |
=
|
\ .
(9)
where A

is a scaling constant, d is beam-to-junction distance, and o is an exponent related to
the surface recombination velocity, v
s.

The diffusion length is usually extracted by rearranging the terms of equation (9):

( )
ln ln( )
d
Id A
L
o
= + (10)
which yields a linear relationship between ln
( )
Id
o
and d with a slope equal to
1
L
. This
approach is taken to be accurate for d > 2L. It should be noted that in Ref. [20], the authors
analyzed only the two asymptotic cases, namely v
s
= 0 and v
s
= , and found that o = -1/2 for
the former and -3/2 for the latter. Later, Chan et. al. [21] demonstrated that this approach can
be applied to materials with arbitrary surface recombination velocity by selecting o such that a
linear relationship between ln(Id
-
o
) and d is obtained. However, even if any value of o is used

Cathodoluminescence

8
(such that -3/2 o -1/2), the result for the diffusion length changes by less than 20% [22] this
is due to the fact that the exponential term dominating the value of I
EBIC
is independent of o.

Fig. 2. Inside the vacuum chamber of the SEM. (Left) The electron gun (1), secondary
electron collector (2), light guide (3) and (4) plastic hoses through which liquid nitrogen is
passed through in order to cool the sample. (Right) This is the sample stage (6) with the
computer controlled heater and platinum resistance thermometer (5) connected to it.
3.2 Time-dependent cathodolumenscence (CL) measurements
Cathodoluminescence is an optical and electrical phenomenon in which a luminescent
material (semiconductor) emits light upon the impact of an electron beam produced from an
electron gun. The high energy electron beam impacted onto a semiconductor will result in
the promotion of electrons from the valence band to the conduction band. This movement
leaves behind a hole, when the electron and hole recombine a photon may be emitted. These
emitted photons can then be collected and analyzed by an optical system.
Minority carrier lifetime is related to the lifetime of the non-equilibrium carriers in the
conduction band. CL measurements were used to detect changes in carrier lifetimes due to
varying sample temperature as well as electron irradiation.
3.3 Experimental setup
CL measurements were also performed in situ inside the Philips XL30 SEM. The
temperatures varies in the sample were 25
o
C to 125
o
C. The Hamamatsu photomultiplier
tube is sensitive to wavelengths ranging from 185 to 850 nm. Slit size was kept at 4.5 mm.
The CL experimental set up is integrated with the SEM. The monocromator is on the right
hand side of the SEM and houses the mirrors and the diffraction grating system. Located on
the right of the monochromator is the Hamamatsu photomultiplier tube. For each specific
wavelength a PC is used to record the intensity of light by counts.
On the light guide there is a parabolic mirror which collects the emitted light from the
samples. A hole is cut in the middle of the mirror to allow for the electron beam to reach the
sample when the mirror is over the sample. The mirror is positioned a specific working
distance, usually a few millimeters, from that sample which is optimized with Back
Scattered Electrons (BSE) for maximum intensity so that its focal length coincides with the
sample. The focus size is normally in tens of microns.


9
The light collected from this focal point comes from a parallel beam through the hollow
waveguide tube and focused onto the entrance slits of the monochromator. The electron
beam may be used in spot or line scan mode. New areas are sampled after each temperature
measurement is made to avoid contamination.

Fig. 3. Diagram of the Cathodoluminescence method including the sample, light guide and
electron beam.
Determining the thermal activation energy by use of the CL measurements is similar to the
way in which the diffusion length is used to calculate the thermal activation energy using
the EBIC technique. Diffusion is due to the existence of a spatial variation of the carrier
concentration inside a semiconductor. The carriers will move from areas of high
concentration to low concentration, this is known as the diffusion current, equation (3).
When non-equilibrium carriers are generated in a material due to an external excitation in
the absence of an electric field, they will diffuse over a certain distance, in a particular
direction, before recombining. The average distance that the carrier has gone before
succumbing to recombination is defined by L, the diffusion length, which is related to
carrier lifetime ,, by equation (1). Carrier lifetime refers to the time between generation and
recombination of non-equilibrium carriers by the Einstein relationship connecting the
mobility and the diffusion current D
n
by equation (2).
3.4 Time-resolved photoluminescence technique
Time-Resolved Photoluminescence (TRPL) technique is an indirect method that relies upon
measuring the minority carrier lifetime in order to determine the diffusion length according
to equations (1) and (2). It provides a time-efficient way for characterizing the transport of
non-equilibrium minority carriers and is most useful for materials with good luminescence
properties. An advantage of this method is that it also provides valuable insight into the
nature of the recombination mechanisms governing minority carrier lifetime. The detailed
description of the TRPL measurements and analysis is provided in Ref. [23].

Cathodoluminescence

10
TRPL measurements are based on recording the transient decay of near-bandgap (NBE)
photoluminescence excited by a short pulse of light, such as from a pulsed laser source. If
the concentration of non-equilibrium minority carriers decreases exponentially from its
initial value
o
, the intensity of the photoluminescence, I
PL
, also follows an exponential
decay:

0
exp
PL
R
t
I

t t
| |
=
|
\ .
(11)
where
R
t is the radiative recombination lifetime. In most cases, however, the
photoluminescence intensity is affected not only by lifetime of minority carriers in the band,
but also by the diffusion rate out of the absorber region, since the escape of the carriers from
the active region is in this case equivalent to the reduction in their lifetime. Therefore, the
lifetime obtained from the TRPL measurements is often less than the true minority carrier
lifetime. In order to obtain an accurate measurement, it is necessary to introduce a
confinement structure to reduce the influence of diffusion. In such a confinement structure
the semiconductor of interest is sandwiched between two layers with wider bandgap and
like doping type, so that the wide bandgap layers act as minority carrier mirrors.
4. Temperature dependence of minority carrier diffusion length and lifetime
4.1 Studies in n-type ZnO
Because of its intrinsic thermal stability, ZnO is a good candidate for high-temperature
optoelectronic devices. However, the subject of the temperature dependence of minority
carrier transport properties has not yet been adequately addressed in the literature. This
section reviews the results obtained from variable-temperature diffusion length
measurements performed on n-type ZnO [24].
The samples under investigation were weakly n-type ZnO substrates with electron
concentration of ~ 10
14
cm
-3
and mobility of ~ 150 cm
2
/Vs. Secondary Ion Mass Spectroscopy
(SIMS) measurements revealed the Li concentration of about 3x10
16
cm
-3
[25]. The Schottky
contacts for EBIC measurements were deposited on the non-polar a-plane of ZnO crystal by
electron beam evaporation of 80 nm-thick Au layer and subsequent lift-off. ZnO:N samples
were grown using molecular beam epitaxy by SVT Associates. Hall effect measurements
revealed hole concentration of 4.5x10
17
cm
-3
and mobility of ~1 cm
2
/Vs.
The studies of L as a function of temperature were carried out using EBIC method in a
planar-collector configuration with a Schottky barrier (Sec. 3.1). At each temperature, several
measurements were taken by scanning the beam of the SEM along a line perpendicular to
the edge of the Schottky contact and recording the exponential decay of current. The
recorded data were fitted with equation (9) using o = -1/2. This value corresponds to zero
surface recombination velocity which, given the excellent luminescence properties of this
sample and a good fit to the experimental results, is a reasonably good approximation. It
should be noted that each EBIC line-scan was recorded on a previously unexposed area in
order to avoid the influence of electron irradiation on the value of diffusion length (cf. Sec. 5
below).


11
Table 1 summarizes the results of EBIC measurements performed on one of the bulk ZnO
samples and shows that the diffusion length of minority holes in n-ZnO increases with
increasing temperature, T. The increase of L with T is not unique to this semiconductor.
Similar trends were previously observed in GaAs [26] and later in GaN epitaxial layers [27].
In all cases, this increase was exponential with temperature and was modeled with the
expression (5).

Temperature (
o
C) Diffusion Length (m) CL Intensity (10
3
counts)
25
o
C 0.438 0.022 72.1 3.7
50
o
C 0.472 0.060 54.4 3.8
75
o
C 0.493 0.028 49.2 2.4
100
o
C 0.520 0.074 44.6 4.7
125
o
C 0.547 0.086 38.5 6.8
E
A
(eV) 0.045 0.002 0.058 0.007
Table 1. Temperature dependence of minority carrier diffusion length and
cathodoluminescence intensity of the near-band-edge peak in n-ZnO. After Ref. [24].
With the fit using equation (5), the experimental results for n-ZnO yields activation energy
of 45 2 meV [24]. This energy represents carrier de-localization energy, since it determines
the increase of the diffusion length due to reduction of recombination efficiency [26]. The
smaller is the activation energy, the more efficient is the de-trapping of captured carriers at a
fixed temperature (see discussion below).
The role of increasing carrier lifetime is also supported by the results of
cathodoluminescence (CL) measurements, carried out in-situ in SEM, which are presented
in Fig. 4. The inset of Fig. 4 shows a cathodoluminescence spectrum in the vicinity of the
NBE transition at 383 nm (3.24 eV). This feature in bulk ZnO has been attributed to the
transition from the conduction band to a deep acceptor level [28]. It was observed that the
peak intensity, I
CL
, of NBE luminescence decays systematically with increasing
temperature, providing direct evidence that the number of recombination events
decreases. Because the intensity of the NBE luminescence is inversely proportional to the
lifetime of carriers in the band, the decay of I
CL
indicates the increase of t with
temperature. The decay proceeds exponentially according to the equation (6) [29]. Based
on the fit shown in Fig. 4, the activation energy was determined to be 58 7 meV. This
value is in excellent agreement with that obtained by photoluminescence measurements
in Li-doped ZnO films [30]. It is also consistent with the results of the variable-
temperature EBIC measurements, which suggests that the same underlying process is
responsible for both the increase in the diffusion length and the CL intensity decay. This
process is outlined below.

Cathodoluminescence

12

Fig. 4. Experimentally obtained values for the peak NBE CL intensity in ZnO:Li as a function
of temperature (open circles) and the fit (solid line; equation (10)), yielding activation energy of
58 7 meV. Inset: CL spectrum on n-ZnO showing the NBE transition at 25
o
C. After Ref. [24].
The increase in minority hole lifetime in the valence band is likely associated with a smaller
recombination capture cross-section for this carrier at elevated temperatures. In GaAs, for
example, detailed analysis for temperature dependence of capture cross-section indicates an
order of magnitude decrease of recombination efficiency, measured in terms of an effective
capture radius, in the temperature range from 100 to 300 K [26]. Non-equilibrium electron-
hole pairs are generated by the beam of the SEM and subsequently annihilate by
recombining with each other. Since the hole capture cross-section is inversely proportional
to temperature [26,31], the frequency of the recombination events (and, hence, the CL
intensity) decreases as the temperature is raised. This means that non-equilibrium holes
exist in the valence band for longer periods of time and, consequently, diffuse longer
distances before undergoing recombination. Note that carrier diffusivity, D, is also a
temperature-dependent quantity and, therefore, can affect the diffusion length (cf. equation
(1)). On the other hand, it has been demonstrated for n-ZnO that the mobility, , of majority
carriers decreases in the temperature range of our experiments by about a factor of 2 [32].
Assuming that the mobility of the minority carriers exhibits the same behavior [27] and
combining equations (1) and (2), it is clear that the value of the diffusion length is
dominated by the growing lifetime of minority holes. From the Einstein relation (equation
(2)), the above-referenced difference in mobility translates to about a 30% decrease in
diffusivity at 125
o
C as compared to 25
o
C. Based on a 30% difference in diffusivity and
using experimentally obtained values of diffusion length, we conclude that the lifetime of
minority holes at 125
o
C is nearly 2.5 times greater than at room temperature.


13
Preliminary results indicate that temperature-induced increase in carrier lifetime also occurs
in epitaxial ZnO. Nitrogen-doped ZnO samples grown using molecular beam epitaxy were
provided by SVT Associates. Hall effect measurements revealed hole concentration of
4.5x10
17
cm
-3
and mobility of ~1 cm
2
/Vs. Fig. 5 shows the decay of NBE CL intensity fitted
with equation (6). The measurements yielded activation energy of 118 12 meV. This value
is comparable to the activation energy of the nitrogen acceptor in ZnO [34,35], which
indicates possible non-equilibrium carrier trapping on nitrogen-related deep levels.

Fig. 5. Maximum CL intensity of the NBE transition in p-ZnO:N as a function of temperature
(open circles). Solid line shows the fit with equation (10), resulting in activation energy of
118 12 meV. After Ref. [33].
4.2 Studies in p-type ZnO doped with antimony
The possibility of p-type doping with larger radii atoms, such as antimony, has been
explored in Refs. [10,36]. The studies demonstrated that despite the large size mismatch,
which in principle should inhibit the substitution of this impurity on the oxygen site,
effective p-type doping with hole concentrations up to 10
20
cm
-3
can be achieved [36]. These
findings prompted the first-principles investigation by Limpijumnong et al., who suggested
that the role of acceptors in size-mismatched impurity doped ZnO is performed by a
complex of the impurity with two zinc vacancies Sb
Zn
-2V
Zn
, the ionization energy of which
is several-fold lower than that of a substitutional configuration and is consistent with the
independent experimental observations [37]. Despite the encouraging predictions, however,
very few attempts at achieving p-type conductivity in antimony-doped ZnO have been
effective. Aoki et al. reported surprisingly high hole concentrations of up to 5 10
20
cm
-3
in
ZnO:Sb films prepared by excimer laser doping [36]. Some of the authors also obtained p-
type ZnO:Sb by molecular beam epitaxy (MBE) [10].

Cathodoluminescence

14
The characteristics of an acceptor level in Sb-doped, p-type ZnO were studied using
cathodoluminescence spectroscopy as a function of hole concentration. Variable-
temperature CL measurements allowed for the estimation of the activation energy of a Sb-
related acceptor from temperature-induced decay of CL intensity. The experiments were
performed on ZnO:Sb layers grown on Si (100) substrates by an electron cyclotron resonance
(ECR)-assisted MBE. The detailed growth procedures are available in Ref. [10]. Hall Effect
measurements revealed strong p-type conductivity, with hole concentrations up to
1.3 10
18
cm
-3
and mobility up to 28.0 cm
2
/V s at room temperature (Table 2).

Sample Number Hole Concentration (cm
-3
)
Carrier Mobility
(cm
2
/V s)
1 1.3 10
17
28.0
2 6.0 10
17
25.9
3 8.2 10
17
23.3
4 1.3 10
18
20.0
Table 2. Room-temperature electronic properties of Sb-doped p-type ZnO films. After Ref.
[39].
CL measurements were conducted in-situ in the Philips XL30 scanning electron microscope
(SEM) integrated with Gatan MonoCL cathodoluminescence system. The SEM is also fitted
with a hot stage and an external temperature controller (Gatan) allowing for temperature-
dependent experiments. The decay of near-band-edge (NBE) luminescence intensity was
monitored as a function of temperature in the range from 25 to 175
o
C. Accelerating voltage
of 10 kV was used. Note that each measurement was taken in a previously unexposed area
to avoid the potential influence of electron irradiation [24,38].
The investigation of the luminescence properties of Sb-doped ZnO was started with the
acquisition of room-temperature cathodoluminescence spectra shown in Fig. 6.
The inset of Fig. 6 reveals that the CL spectra of all three samples are dominated by the
NBE band, which generally contains the band-to-band transition as well as the transition
from the conduction band to a deep, neutral acceptor level (e, A
0
) [28,38]. Since acceptor
levels form a band in the forbidden gap, the red shift of the NBE peak with increasing
carrier concentration (i.e., higher doping levels) is consistent with the (e, A
0
) emission
and may indicate the broadening of the Sb-related acceptor band [40,41]. Another
observation that can be made from Fig. 6 is the systematic decay in intensity of the NBE
luminescence with increasing doping level. This decrease may be attributed to the
reduction in radiative recombination rates as more disorder is introduced into the ZnO
lattice by large-radius Sb atoms. The increasing trend in the values of the full width at
half-maximum (FWHM) of the NBE spectra provide further evidence for the impact of
the size-mismatched dopant - FWHM values were determined to be about 16.1, 19.4,
23.5, and 21.7 nm (corresponding to 136, 163, 196, and 178 meV) for samples 1, 2, 3, and
4, respectively. Note that while FWHM of NBE transitions in CL spectra tends to be
greater than the width of photoluminescence (PL) peaks, the above values are
comparable to those obtained for (e, A
0
) transitions in CL spectra of other ZnO and GaN
materials [24,38,42].


15

Fig. 6. NBE cathodoluminescence spectra of samples 1-4 taken at room temperature. The
peaks are at 382, 384, 385, and 387 nm, respectively. Inset: broad-range CL spectra of the
same samples. After. Ref. [39].
The intensity of NBE luminescence was also monitored as a function of temperature. It was
observed that the intensity decays with sample temperature, T, in agreement with
expression (6) [29]. From equation (6), it can be deduced that the inverse intensity, 1/I,
should exhibit an exponential dependence on 1/kT. This is shown in the inset of Fig. 7 on
the example of sample 1. Note that the intensity in this and subsequent figures was
normalized with respect to its room-temperature value for each of the samples. The
activation energies, E
A
, were obtained from the slopes of Arrhenius plot shown in Fig. 7. In
case of a (e, A
0
) transition, E
A
is related to the ionization energy of acceptors: the lower the
value of the activation energy, the more likely is the ionization of the acceptor by a valence
band electron (A
0
+ e A
-
); since an ionized level does not participate in recombination via
the (e, A
0
) route, the rate of these transitions (i.e., the intensity of the luminescence)
decreases with E
A
at any given temperature. Conversely, for a constant E
A
, the intensity
decays with increasing temperature as more and more acceptors are ionized.

Cathodoluminescence

16

Fig. 7. Arrhenius plot showing the decay of normalized NBE luminescence intensity with
increasing temperature for sample 1 (open squares), sample 2 (open circles), sample 3 (open
diamonds), and sample 4 (open triangles). The linear fits (solid lines) yielded activation
energies of 212 28, 175 20, 158 22 and 135 15 meV for samples 1, 2, 3, and 4,
respectively. The data were vertically offset for clarity. Inset: exponential decrease of CL
intensity for sample 1 (open squares) and the fit (solid line). After Ref. [39].
It is apparent from Fig. 7 that the activation energy shows a systematic dependence on the
carrier concentration. The values of E
A
are 212 28, 175 20, 158 22, and 135 15 meV for
samples 1, 2, 3, and 4, respectively. These values are in reasonable agreement with the
ionization energy of a Sb
Zn
-2V
Zn
complex predicted by Limpijumnong et al. to have a value
of about 160 meV [37]. Furthermore, the decay of activation energy with carrier density, p,
follows a common pattern observed previously in other semiconductors [41, 42, 43] and is
described by an equation of the type;

1/3
( ) (0) ( )
A A A A
E N E N o

= (12)
where
A
N
is the concentration of ionized acceptors, EA(0) is the ionization energy at very

low doping levels, and o is a constant accounting for geometrical factors as well as for the
properties of the material. Fig. 8 demonstrates that equation (12) provides a reasonable fit to
the experimentally obtained activation energies under the approximation that
A
N
-
D
N
+
= p, where
D
N
+
is the density of ionized shallow donors (due to compensation, the p-
type conductivity is determined by the difference between the concentrations of ionized
donors and acceptors).


17

Fig. 8. Decrease of activation energy as a function of ionized acceptor concentration. After
Ref. [39].
The value of o was found to be equal to 6.4 10
-7
, which is comparable to that in p-GaN and
p-Si [41,43].
D
N
+
can be estimated from the electron concentration in undoped, n-type ZnO
samples grown by the same method and is about 5 10
18
cm
-3
[10]. We note that this is a
rough estimate and does not account for the earlier observation that background donor
concentration in Sb-doped samples may be different from that in undoped ZnO films due to
the creation of Zn vacancies induced by Sb doping. The concentration of Zn vacancies was
shown to depend on Sb doping level [10], which would in turn lead to the variations in
shallow donor density among the samples under investigation.
It should be noted that earlier PL measurements performed on sample 4 showed consistent
activation energy of 140 meV [10]. Furthermore, temperature-dependent measurements of
hole concentration in sample 4 shown in Fig. 9 suggest that the temperature dependence of
luminescence intensity is associated with acceptors. The dependence of p on temperature
can be modeled with a charge-balance equation of the following form:

1
A
D
N
p N
p
|
+
+ =
+
(13)
where
D
N
+
and
A
N
were estimated as described above and

3/2
exp( / )
A
AT E kT | = , with A
being a factor accounting for the degeneracy of acceptor states and the density of states in

Cathodoluminescence

18
the valence band [32]. The fit of the data shown in Fig. 9 revealed activation energy of about
100 meV [10], which is in reasonable agreement with activation energy obtained by CL and
is most likely related to the Sb
Zn
-2V
Zn
complex.

Fig. 9. Temperature dependence of hole concentration in sample 4. The fit yielded activation
energy of about 100 meV. After Ref. [39].
Although the existence of other Sb-related acceptors cannot be categorically excluded, their
involvement in the temperature-induced CL intensity decay is highly unlikely. The
substitutional defect (Sb
O
) as well as the single-vacancy complex (Sb
Zn
-V
Zn
) are predicted to
have ionization energies about an order of magnitude greater than those obtained
experimentally, while other defects can also be ruled out based on their electrical behavior
and/or high formation energies [37].
The phenomenon of variation of the dopant activation energy with carrier concentration in
semiconductors has been attributed to a number of causes. Among these are the formation
of the band-tail states that extend into the forbidden gap, the broadening of the acceptor
band in the gap, and the reduction of binding energy due to Coulomb interaction between
the holes in the valence band and the ionized acceptor states [40-42]. The variable
temperature cathodoluminescence studies of Sb-doped p-type ZnO allowed estimating the
activation energy of the Sb-related acceptor in the range of 135-212 meV. The activation
energy was found to be strongly dependent upon the hole concentration. While the nature
of the acceptor cannot be determined conclusively, evidence suggests that it is a Sb
Zn
-2V
Zn

complex proposed by Limpijumnong et al [37].


19
5. Studies of minority carrier recombination
Deep carrier traps have pronounced implications on minority carrier transport and often
give rise to such undesirable phenomena as radiation-induced optical metastability,
persistent photoconductivity and optical quenching of photocurrent [44-46]. On the other
hand, it has been demonstrated that capture of minority carriers by deep metastable traps is
associated with the increase of minority carrier diffusion length and lifetime [24,38,47].
Furthermore, in GaN this increase was shown to result in a significant (several-fold)
improvement of photoresponse in agreement with equation (7) [22,47].
Extensive studies aimed at achieving p-type conductivity in ZnO reveal that most potential
acceptors, such as nitrogen, phosphorus, and arsenic, tend to form acceptor levels far from
the valence band maximum [48]. Since the ionization fraction of such acceptors is low (due
to their high activation energy), there is a large concentration of neutral states that may act
as traps for non-equilibrium electrons. The effects of electron trapping on the diffusion
length and lifetime of minority carriers can be probed by subjecting the material of interest
to the excitation by the electron beam of the scanning electron microscope. The remainder of
this review deals with the influence of electron irradiation on minority carrier diffusion
length (Sec. 5) and lifetime (Sec. 5.2).
5.1 Influence of electron trapping on minority carrier diffusion length
The measurements of diffusion length as a function of beam irradiation duration were
carried out on bulk n-ZnO doped with lithium [24], molecular beam epitaxy (MBE) p-ZnO
doped with nitrogen [49], phosphorus-doped ZnMgO grown by pulsed laser deposition
(PLD) [38], and on Sb-doped epitaxial ZnO layers MBE grown on a Si substrate [39].
As was already mentioned, bulk ZnO samples (Tokyo Denpa Co.) were weakly n-type,
showing electron concentrations of ~ 10
14
cm
-3
and mobility of ~ 150 cm
2
/Vs at room
temperature. Secondary Ion Mass Spectroscopy (SIMS) measurements revealed the Li
concentration of about 3x10
16
cm
-3
[25] (Li is often added to ZnO to increase the resistivity of
initially n-type samples).
Phosphorus-doped Zn
0.9
Mg
0.1
O layers were fabricated using PLD. Capacitance-Voltage
profiling of similar films, grown using the same procedure, resulted in net acceptor
concentration of ~ 2x10
18
cm
-3
after annealing. Pt/Au (200/800 ) layers were deposited on
phosphorus-doped Zn
0.9
Mg
0.1
O films by electron beam evaporation and patterned by lift-off
with contact diameters ranging from 50 to 375 m. Circular electrode pairs with significantly
different surface areas were employed for the EBIC measurements. The electrodes create an
asymmetric rectifying junction, based on back to back Schottky diodes, with the larger area
electrode being pseudo-Ohmic.
EBIC experiments were conducted at room temperature in-situ in a Philips XL30 SEM using
a planar-collector configuration with a Schottky barrier (Sec. 3.1) to monitor the changes in
minority carrier diffusion length as a function of time. The results of EBIC experiments are
presented in Fig. 10. Note that the results shown in the left panel of Fig. 10 and those
discussed in Sec. 4 were obtained from different bulk ZnO samples, which may offer an
explanation for the significant difference in the initial, room temperature values of diffusion
length. Additionally, while the large diffusion length value of bulk ZnO can be attributed to

Cathodoluminescence

20
the higher quality of the bulk material (compared to the epitaxial layers), the diffusion
length of ~ 2 m in PLD-grown ZnMgO is rather surprising because of polycrystalline
nature of the layers. It appears that in the latter sample, the transport of minority carriers is
not significantly influenced by scattering from the grain boundaries.

Fig. 10. Experimental dependence of minority carrier diffusion length on duration of
electron beam irradiation at room temperature (open circles) and the linear fit (solid line).
After Refs. [24, 38, 49].
The experiments revealed that diffusion length follows a general trend in all samples
studied. Fig. 10 shows that irradiation by the electron beam clearly results in a significant
increase of the carrier diffusion length, and that this increase is linear with respect to the
duration of electron irradiation (t). Up to 50% increase of diffusion length was achieved (cf.
Fig. 10 (center)), with the rates of diffusion length increase ranging from ~13% to ~30% per
1000 s. This appears to be a common occurrence in wide bandgap semiconductors doped
with species that create deep acceptor levels, as similar observations were made in (Al)GaN
doped with Mg, Mn, Fe, and C [39,50,51]. It is also noteworthy that similar experiments
conducted on bulk ZnO without any intentional dopants did no show any significant
changes in minority carrier diffusion length [38].
The observed increase of L is ascribed to charging of the deep, neutral acceptor states by the
electrons generated by the SEM beam, since trapping of non-equilibrium electrons prevents
these levels from participating in recombination [38,39]. Therefore, the difference in the rates
of diffusion length increase is likely explained, at least in part, by the difference in the
concentrations of these deep centers [52].
EBIC measurements in Sb-doped ZnO were performed on the samples by moving the
electron beam of the SEM from the edge of the Schottky barrier (created on the top surface
of ZnO:Sb samples by Ti/Au evaporation followed by lift off) outwards (line-scan) and
recording an exponential decay of induced current.


21
After a single EBIC line-scan was completed (12 seconds), the excitation of the sample was
continued by moving the electron beam back and forth along the same line for the total time
of ~2200 seconds. EBIC measurements were periodically repeated to extract the minority
carrier diffusion length [52], L, as a function of the duration of electron beam irradiation, t.
The effects of electron injection for Sb-doped 0.2 m-thick p-type ZnO epitaxial layers
(p = 1.3x10
17
cm
-3
; = 28 cm
2
/Vs at room temperature) grown on Si substrate by MBE are
summarized in Fig. 11. The activation energy for the e-beam injection-induced increase of L,
AE
A,I
= 219 8 meV, was obtained from the graphs in Fig. 11b,c using the following equation:

, ,
0
exp exp
2
A I A T
E E
R R
kT kT
A A | | | |
=
| |
\ . \ .
(14)
where R
0
is a scaling constant, T is temperature, k is the Boltzmanns constant, AE
A,I
is the
activation energy of electron irradiation effect, and AE
A,T
is the activation energy of
thermally-induced increase of L presented in Fig. 11a (AE
A,T
= 184 10 meV). The value of
AE
A,I
is in reasonable agreement with that for a Sb
Zn
-2V
Zn
acceptor complex, predicted by
Limpijumnong et al. [37] (see Ref. [53] for details).

Fig. 11. a) Diffusion length of minority electrons as a function of temperature (open circles)
and the fit (solid line). Inset: Arrhenius plot of the same data yielding activation energy of
184 10 meV. b) Electron beam irradiation-induced increase of minority electron diffusion
length at different temperatures. The values of the diffusion length were vertically offset for
clarity and are not intended to illustrate the temperature dependence. c) Rate of irradiation-
induced increase of diffusion length as a function of temperature (open circles).The fit with
equation 5 (solid line) gives activation energy of 219 8 meV. After Ref. [53].
It can also be seen from Fig. 11 that the rate, R, of the diffusion length increase is reduced
with increasing temperature. The increase of the diffusion length due to trapping is
counteracted by the release of the trapped electrons that occurs if the carriers gain sufficient
energy to escape the trap. As the temperature is raised, the likelihood of de-trapping
increases, which dampens the irradiation-induced growth of the diffusion length.

Cathodoluminescence

22
The saturation and relaxation of irradiation-induced change of diffusion length was studied
at room temperature [53]. It was demonstrated that L reached its maximum value after
about 50 min of continuous exposure to the electron beam. Further monitoring revealed that
irradiation-induced increase persists for at least one week. Annealing the sample at 175 C
for about 30 minutes resulted in a decrease of the diffusion length to about 1 m. This
behavior further supports the involvement of deep electron traps in the phenomenon of
interest, since temperature-induced de- trapping of carriers (cf. Fig. 11,c) re-activates the
original recombination route (cf. Fig. 11,a), thus reducing carrier lifetime and diffusion
length.
5.2 Optical studies of the effects of electron trapping on minority carrier lifetime
According to equation (1), electron irradiation-induced increase in minority carrier diffusion
length discussed above is associated with the increase of lifetime of non-equilibrium
carriers. Experimental evidence for this dependence was obtained from the
cathodoluminescence measurements performed on the same samples. Time-dependent CL
measurements were conducted in-situ in SEM using setup described in section 3.2. This
setup allows combining periodic acquisition of CL spectra with continuous excitation of the
sample by scanning the beam over the same location. For temperature-dependent CL
measurements, the sample temperature was varied in-situ using specially designed hot stage
and an external temperature controller (Gatan). At each temperature, the electron beam
irradiation and CL measurements were conducted at a different location.
5.2.1 Optical studies of Zn
0.9
Mg
0.1
O doped with phosphorus
Fig. 12 shows a series of NBE transitions in p-type Zn
0.9
Mg
0.1
O:P recorded under continuous
electron irradiation and numbered in order of increasing irradiation duration. The edge of
this spectrum at ~ 355 nm (see also inset of Fig. 12) is in agreement with 10% Mg content in
ZnO lattice, since each atomic percent of Mg is known to increase the ZnO band gap by 0.02
eV [4]. The observed CL spectrum is attributed to the band-to-band as well as band-to-
impurity (P-acceptor) optical transitions. The inset of Fig. 12 shows a wider range spectrum
featuring a broad band, which is likely defect related [54].
While no changes were observed in the broad band CL (cf. inset of Fig. 12), the near band-
edge luminescence in Fig. 12 exhibits a continuous decay with increasing duration of
electron beam irradiation. This demonstrates that exposure to the electron beam results in
the increase of carrier lifetime (t), since I
CL
is proportional to 1/t. Similar phenomena were
also observed in GaN, where the decay of NBE CL intensity occurred concomitantly with
increasing diffusion length [38,50,51].
To characterize the intensity decay, we relate it to the diffusion length, L, which is known to
vary linearly with duration of irradiation (cf. Fig. 13). Since L is proportional to t
1/2

(equation (1)), the inverse square root of normalized (with respect to the initial maximum
value) intensity must also be proportional to L, and consequently, would be expected to
change linearly with duration of electron irradiation. Fig. 13 shows that this is indeed the
case, indicating that the observed increase of the diffusion length is attributable to the
growing lifetime of non-equilibrium carriers.


23

Fig. 12. Room temperature CL spectra of Zn
0.9
Mg
0.1
O:P measured in the same location at
different times of electron injection. 1 is a pre-injection spectrum; 2, 3, and 4 correspond to
duration of electron injection of 359, 793, and 1163 s, respectively. Inset: pre-injection
broad-range CL spectrum taken in a different location than measurements in figure 12.
After Ref. [38].

Cathodoluminescence

24

Fig. 13. Variable temperature dependence for the square root of inverse normalized intensity
on duration of electron injection in Zn
0.9
Mg
0.1
O:P. The rate at every temperature is obtained
from the slope of a linear fit. Inset: temperature dependence for the rate of the square root of
inverse normalized intensity (open circles) and the fit. The slope of the graph yields E
A
=
256 20 meV. After Ref. [38].
In CL, the temperature of the samples was varied from 25 C to 125 C. At each temperature,
injection was performed on a site previously not exposed to electron beam. As is apparent
from the inset of Fig. 13, the rate of the decrease in CL intensity (described by the slope, R, of
the linear dependence of I
CL
-1/2
on t) diminishes with growing temperature. This suggests
that while electron irradiation results in an increase of carrier lifetime, there exists another,
thermally activated process that contributes to its decay.
Taking into account both injection-induced effect on R ( exp(AE
A
/kT)) and its temperature
dependence ( exp(AE
A
/-2kT)) [26], and assuming that activation energies are similar in
both cases, the temperature dependence of R can be described as follows:

0
exp( )
2
A
E
R R
kT
A
= (15)
where R
0
is a scaling constant and AE
A
is the activation energy for the overall process. Fitting
the experimental results with this expression (inset of Fig.13) yielded activation energy of
256 20 meV. This activation energy is in good agreement with that for the phosphorus


25
acceptor obtained based on the simple hydrogenic model. The model assumes phosphorus
substitution on the oxygen site and predicts the activation energy of about 250-300 meV [55].
The experimentally obtained value of the activation energy, combined with the fact that no
electron irradiation effects were observed in undoped ZnO [38], suggests that carrier
trapping on phosphorus acceptor levels plays a crucial role in this phenomenon.
5.2.2 Optical studies of bulk ZnO
Similarly to ZnMgO, the electron irradiation-induced increase of minority carrier diffusion
length in ZnO:Li (Sec. 5.1) was found to correlate with the increase of minority carrier
lifetime. Fig. 14 shows a series of room temperature NBE spectra collected under continuous
electron beam excitation, in which the intensity of the NBE transition can be seen to fall
steadily with increasing t. The peak of this emission occurs at 383 nm (3.24 eV) and has been
assigned to the transition from the conduction band to a deep acceptor level, (e,A0) [28]. The
inverse square root of maximum CL intensity (I
CL
-1/2), which is proportional to t and,
therefore, to L, changes linearly with duration of irradiation (cf. left inset of Fig. 14), which is
consistent with the results of EBIC measurements and indicates that the increase in L occurs
due to the irradiation-induced growth of carrier lifetime.

Fig. 14. Room temperature cathodoluminescence spectra of ZnO:Li taken under continuous
excitation by the electron beam. 1 is the pre-irradiation spectrum and 5 is the spectrum after
1450 s of electron irradiation. Left Inset: Variable-temperature dependence of inverse square
root of normalized intensity on duration of electron irradiation and the linear fit with the
rate R. Right Inset: Arrhenius plot of R as a function of temperature yielding an activation
energy AE
A,I
of 2839 meV. After Ref. [24].

Cathodoluminescence

26
CL measurements conducted at elevated temperatures confirmed the same trend for the
irradiation-induced change of luminescence intensity. It can be seen from the left inset of
Fig. 14 that the inverse square root of intensity increases linearly for all temperatures. The
temperature dependence of rate, R, can be used to determine the activation energy of the
irradiation-induced processes according to equation (15). On the other hand, our earlier
studies of the temperature-induced CL intensity decay (Sec. 4) yielded the activation energy
of about 60 meV [38], thus allowing us to separate the two components as followed from
equation (14): the activation energy of electron irradiation effect
, A I
E A ; and, the previously
determined, activation energy of thermally induced intensity decay
, A T
E A . This treatment
yielded the value for
, A I
E A of 283 9 meV. Incidentally, Ref. [25] also reports high
concentration of ~0.3 eV electron traps found in the same material by Deep Level Transient
Spectroscopy (DLTS). The significant difference between
, A I
E A and
, A T
E A observed in bulk
ZnO suggests that, unlike in other materials, temperature- and irradiation-induced changes
of the minority carrier transport characteristics are two distinctly different processes.
Although several theoretical works have predicted a very shallow Li
Zn
level [56,57], these
predictions have not been substantiated experimentally, as most studies find a rather deep
Li-acceptor with activation energies of several hundred meVs [34,58]. In fact, recent first-
principles calculations by Wardle et al., also suggest that the Li
Zn
acceptor state lies at about
0.2 eV above the valence band maximum [58], which is in reasonable agreement with the
, A I
E A of 283 9 meV obtained in this work.
It should be clarified that the weak n-type character of the sample is not necessarily in
contradiction with the dominant behavior of acceptor states observed in electron trapping
phenomena. As was mentioned, the n-type conductivity in nominally undoped ZnO is due
to the shallow donor states, whereas in presence of deep electron traps the Fermi level may
lie far below these states. Although shallow donors may capture non-equilibrium electrons
under excitation, those are quickly released if the temperature is sufficiently high. Therefore,
if the difference in the energetic position between the donor and trap states is large, the
latter dominate the kinetics of electron trapping [59].
5.2.3 Optical studies of ZnO Doped with nitrogen
CL measurements performed on MBE-grown, nitrogen-doped p-ZnO revealed behavior similar
to that of bulk ZnO [40] and PLD-grown ZnMgO [38]. Room temperature ZnO:N spectra are
shown in Fig. 15 and feature a NBE luminescence band with a maximum at about 388 nm (~3.20
eV). This band includes the (e, A
0
) transition as well as the donor-acceptor pair (DAP)
recombination, with nitrogen identified as the acceptor in both processes [60]. Additionally, a
violet band centered on 435 nm has been attributed to the radiative recombination of the
electrons trapped at grain boundaries with the holes in the valence band [61].
As expected, irradiation with electron beam resulted in decay of the intensity of NBE
luminescence, indicating increasing lifetime. One can observe from the inset of Fig. 15 that,
in agreement with the diffusion length measurements (cf. Fig. 10 in Sec. 5.1), the inverse
square root of the peak normalized intensity of the NBE transition changes linearly with
irradiation time, yielding the rate R. Note that the intensity of the violet band is not affected
by electron irradiation, which suggests that electron trapping at the grain boundaries does
not play a significant role in the irradiation-induced increase of carrier lifetime.


27

Fig. 15. Room temperature CL spectra of ZnO:N taken under continuous excitation. Trace 1
corresponds to the pre-irradiation spectrum and trace 5 to the spectrum after 1940 s of
electron irradiation. Inset: Linear dependence of the inverse square root of normalized peak
intensity at room temperature. After Ref. [49].
CL measurements at elevated temperatures (Fig. 16) confirmed that R decreases with
temperature, indicating a thermally activated process that counteracts the effects of electron
injection, similar to what occurs in bulk ZnO and ZnMnO:P. Note that while the values of R
were obtained based on the intensity normalized with respect to its initial value, the data
displayed in Fig. 16 are offset by shifting the normalized results along the y-axis to avoid the
overlap of the data points. The activation energy (
, A I
E A ) of about 134 10 meV was
determined from the Arrhenius plot shown in the inset of Fig. 16, based on equation (15)
and using
, A T
E A = 118 meV obtained from the temperature-dependent CL measurements
(Sec. 4). This value is in reasonable agreement with the ionization energy of the nitrogen
acceptor in ZnO [34,35 ,60,62] and indicates that electron trapping by these levels plays an
important role in the recombination dynamics of minority carriers.

Cathodoluminescence

28

Fig. 16. Variable-temperature dependence of the inverse square root of normalized intensity
in ZnO:N on duration of electron irradiation and the linear fit with a rate R. The data are
offset for clarity. Inset: Arrhenius plot of R as a function of temperature with a fit yielding
AE
A,I
=134 10 meV. After Ref. [49].
6. Summary
Issues affecting minority carrier transport in ZnO have been discussed, with special
attention given to the temperature dependence of minority carrier diffusion length and
lifetime, as well as to the recombination dynamics of non-equilibrium minority carriers.
The mechanisms governing temperature- and irradiation-induced effects have been
presented.
7. References
[1] F.Oba, S.R.Nishitani, S.Isotani, H.Adachi, and I.Tanaka, J. Appl. Phys., 90, 824 (2001)
[2] G.W.Tomlins, J.L.Routbort, and T.O.Mason, J. Appl. Phys., 87, 117 (2000)
[3] C.G.Van de Walle, Phys. Rev. Lett., 85,1012 (2000)
[4] T. Gruber, C. Kirchner, R. Kling, F. Reuss and A. Waag, Appl. Phys. Lett., 84, 5359 (2004).
[5] K. Ip, Y W. Heo, D. P. Norton, S. J. Pearton, J. R. LaRoche and F. Ren, Appl. Phys.Lett., 85,
1169 (2004).
[6] K.G. Chen, Z. Z. Ye, W. Z. Xu, B. H. Zhao, L. P. Zhu and J. G Lv, J. Cryst. Growth, 281,458
(2005).
[7] V. Vaithianathan, B. T. Lee and S. S. Kim, J. Appl Phys., 98, 043519 (2005).


29
[8] E. J. Egerton, A. K. Sood, R. Singh, Y R. Puri, R. F. Davis, J. Pierce, D. C. Look, and T.
Steiner, J. Electron. Mater., 34, 949 (2005).
[9] V. Vaithianathan, B. T. Lee, and S. S. Kim, Appl Phys. Lett., 86, 062101 (2005).
[10] F. X. Xiu, Z. Yang, L. J. Mandalapu, D. T. Zhao, J. L. Liu, and W. P. Beyermann,
Appl.Phys. Lett., 87, 152101 (2005).
[11] S. H. Lim, D. Shindo, H. B. Kang and K. Nakamura, J. Vac. Sci. Technol B, 19, 506 (2001).
[12] K. Miyamoto, M. Sano, H. Kato and T. Yao, J. Cryst. Growth, 265, 34 (2004).
[13] Y Wang, X. L. Du, Z. X. Mei, Z. Q. Zeng, Q. Y Xu, Q. K. Xue and Z. Zhang, J. Cryst.
Growth, 273, 100 (2004).
[14] M. W. Cho, A. Setiawan, H. J. Ko, S. K. Hong and T. Yao, Semicond. Sci. Technol,20, S13
(2005).
[15] E. Monroy, F. Calle, J. L. Pau, F. J. Sanchez, E. Munoz, F. Omnes, B. Beaumont, and P.
Gibart, J. Appl. Phys., 88, 2081 (2000).
[16] J. F. Muth, R. M. Kolbas, A. K. Sharma, S. Oktyabrsky and J. Narayan, J. Appl. Phys.,85,
7884 (1999).
[17] H. Yoshikawa and S. Adachi, Jpn. J. Appl. Phys. Part 1 - Regul. Pap. Short Notes Rev. Pap.,
36, 6237 (1997).
[18] D. E. loannou and S. M. Davidson, J. Phys. D-Appl. Phys., 12, 1339 (1979).
[19] D. E. loannou and C. A. Dimitriadis, IEEE Trans. Electron. Devices, 29, 445 (1982).
[20] J. Boersma, J. J. E.Iindenkleef, and H. K. Kuiken, J. Eng. Math, 18, 315 (1984).
[21] D. S. H. Chan, V. K. S. Ong, and J. C. H. Phang, IEEE Trans. Electron. Devices, 42, 963
(1995).
[22] L. Chernyak, A. Osinsky and A. Schulte, Solid-State Electron., 45, 1687 (2001).
[23] R. K. Ahrenkiel, Solid-State Electron., 35, 239 (1992).
[24] O. Lopatiuk, L. Chernyak, A. Osinsky, J. Q. Xie, and P. P. Chow, Appl. Phys. Lett., 87,
214110 (2005)
[25] A. Y. Polyakov, N. B. Smirnov, A. V. Govorkov, E. A. Kozhukhova, S. J. Pearton, D. P.
Norton, A. Osinsky, and A. Dabiran, J. Electron. Mater, 35, 663 (2006).
[26] M. Eckstein and H. U. Habermeier, J. Phys. IV, 1, 23 (1991).
[27] L. Chernyak, A. Osinsky, H. Temkin, J. W. Yang, Q. Chen and M. A. Khan, Appl Phys.
Lett., 69, 2531 (1996).
[28] K. Thonke, T. Gruber, N. Teofilov, R. Schonfelder, A. Waag, and R. Sauer, Physica B,
308, 945 (2001).
[29] D. S. Jiang, H. Jung and K. Ploog, J. Appl. Phys., 64, 1371 (1988).
[30] A. Ortiz, C. Falcony, J. Hernandez, M. Garcia and J. C. Alonso, Thin Solid Films, 293, 103
(1997).
[31] J. I. Pankove, Optical Processes in Semiconductors, Prentice-Hall, Englewood Cliffs, New
Jersey (1971).
[32] D. C. Look, D. C. Reynolds, J. R. Sizelove, R. L. Jones, C. W. Litton, G. Cantwell and W.
C. Harsch, Solid State Commun., 105, 399 (1998).
[33] O. Lopatiuk-Tirpak, Influence of Electron Trapping on Minority Carrier Transport
Properties of Wide Band Gap Semiconductors, Ph.D. Dissertation, University of
Central Florida, Orlando, 2007.
[34] B. K. Meyer, H. Alves, D. M. Hofmann, W. Kriegseis, D. Forster, F. Bertram, J. Christen,
A. Hoffmann, M. Strassburg, M. Dworzak, U. Haboeck, and A. V. Rodina, Phys.
Status Solidi B - Basic Res., 241, 231 (2004).

Cathodoluminescence

30
[35] S. Yamauchi, Y. Goto, and T. Hariu, J. Cryst. Growth, 260, 1 (2004).
[36] T. Aoki, Y. Shimizu, A. Miyake, A. Nakamura, Y. Nakanishi and Y. Hatanaka, Phys.
Status Solidi B-Basic Res., 229, 911 (2002).
[37] S. Limpijumnong, S. B. Zhang, S. H. Wei and C. H. Park, Phys. Rev. Lett., 92, 155504
(2004).
[38] O. Lopatiuk, W. Burdett, L. Chernyak, K. P. Ip, Y. W. Heo, D. P. Norton, S. J. Pearton, B.
Hertog, P. P. Chow and A. Osinsky, Appl. Phys. Lett., 86, 012105 (2005).
[39] O.Lopatiuk-Tirpak, W.V.Schoenfeld, L.Chernyak, F.X.Xiu, J.L.Liu, S.Jang, F.Ren,
S.J.Pearton, A.Osinsky, P.Chow, Appl. Phys. Lett., 88, 202110 (2006).
[40] M. G. Cheong, K. S. Kim, C. S. Kim, R. J. Choi, H. S. Yoon, N. W. Namgung, E. K. Suh
and H. J. Lee, Appl. Phys. Lett., 80, 1001 (2002).
[41] P. Kozodoy, H. L. Xing, S. P. DenBaars, U. K. Mishra, A. Saxler, R. Perrin, S. Elhamri
and W. C. Mitchel, J. Appl. Phys., 87, 1832 (2000).
[42] W. Gotz, R. S. Kern, C. H. Chen, H. Liu, D. A. Steigerwald and R. M. Fletcher, Mater. Sci.
Eng. B-Solid State Mater. Adv. Technol., 59, 211 (1999).
[43] P. P. Debye and E. M. Conwell, Phys. Rev., 93, 693 (1954).
[44] S. Dhar and S. Ghosh, Appl. Phys. Lett., 80, 4519 (2002).
[45] V. Ursaki, I. M. Tiginyanu, P. C. Ricci, A. Anedda, S. Hubbard and D. Pavlidis, J. Appl.
Phys., 94, 3875 (2003).
[46] J. Ryan, D. P. Lowney, M. O. Henry, P. J. McNally, E. McGlynn, K. Jacobs and L.
Considine, Thin Solid Films, 473, 308 (2005).
[47] L. Chernyak, G Nootz and A. Osinsky, Electron. Lett., 37, 922 (2001).
[48] C. Look and B. Claftin, Phys. Status Solidi B - Basic Res., 241, 624 (2004).
[49] O. Lopatiuk, A.Osinsky, L.Chernyak, in Zinc Oxide Bulk, Thin Films and Nanostructures,
pp.241-265, Edited by C. Jagadish and S. Pearton, Elsevier Ltd. (2006)
[50] L. Chernyak, W. Burdett, M. Klimov and A. Osinsky, Appl. Phys. Lett., 82, 3680 (2003).
[51] W. Burdett, O. Lopatiuk, L. Chernyak, M. Hermann, M. Stutzmann, and M. Eickhoff, J.
Appl. Phys., 96, 3556 (2004).
[52] L. Chernyak, A. Osinsky, V. Fuflyigin and E. F. Schubert, Appl. Phys. Lett., 77, 875 (2000).
[53] O.Lopatiuk-Tirpak, L.Chernyak, F.X.Xiu, J.L.Liu, S.Jang, F.Ren, S.J.Pearton,
K.Gartsman, Y.Feldman, A.Osinsky, P.Chow, J.Appl. Phys.,100, 086101 (2006).
[54] Y. W. Heo, K. Ip, S. J. Pearton and D. P. Norton, Phys. Status Solidi A - Appl. Res., 201,
1500 (2004).
[55] S. J. Pearton, D. P. Norton, K. Ip, Y. W. Heo and T. Steiner, Prog. Mater. Sci, 50, 293
(2005).
[56] C. H. Park, S. B. Zhang and S. H. Wei, Phys. Rev. B, 66, 073202 (2002).
[57] E. C. Lee and K. J. Chang, Phys. Rev. B, 70, 115210 (2004).
[58] M. G. Wardle, J. P. Goss and P. R. Briddon, Phys. Rev. B, 71, 155205 (2005).
[59] M. Salis, A. Anedda, F. Quarati, A. J. Blue and W. Cunningham, J. Appl. Phys., 97,
033709 (2005).
[60] F. Reuss, C. Kirchner, T. Gruber, R. Kling, S. Maschek, W. Limmer, A. Waag, and P.
Ziemann, J. Appl. Phys., 95, 3385 (2004).
[61] R. Ghosh, B. Mallik, S. Fujihara, and D. Basak, Chem. Phys. Lett., 403, 415 (2005).
[62] G. Xiong, K. B. Ucer, R. T. Williams, J. Lee, D. Bhattacharyya, J. Metson, and P. Evans, J.
Appl. Phys., 97, 043528 (2005).
2
What is the Pulsed Cathodoluminescence?
Vladimir Solomonov and Alfiya Spirina
The Institute of Electrophysics of Ural Branch of the Russian Academy of Sciences
Russia
1. Introduction
A cathodoluminescence phenomenon was revealed in the 19
th
century. The first explanation
of this phenomenon was given by Julius Plcker in 1858, and two decades later by Sir
William Crookes (1879). A cathodoluminescence intensity is defined by the concentration of
electrons (holes) n
e(h)
, generated by electrons injected into substance. Generally, this
concentration is described by the kinetic equation

( )
2
( ) ( )
e h
e h e h
dn
G An Bn
dt
= ,
0 e
e i
E j
G
ed c
= . (1)
Here A and B are the coefficients of electrons (holes) recombination processes which obey to
linear and quadratic laws. Coefficient A is a generalized characteristic of the processes
which relate to formation and dissociation of electron and hole centres. These processes are
accompanied with the capture or the release of free electron (hole). The luminescence
intensity of these centres is proportional to concentration, I
e(h)
~An
e(h)
. Coefficient B is
defined by the annihilation of free electron-hole pairs which resulted in origination of
interband, excitonic, and intracentre kinds of luminescence with intensity of I~Bn
2
e(h)
. Thus,
the coefficients A and B are the characteristics of certain substance. Usually for pure crystals
(undoped) the coefficient A has an order of 10
5
-10
6
s
-1
whereas it is of 10
8
s
-1
in case of
crystals doped with donor (acceptor) ions. The value of coefficient B amounts to 10
-10
cm
3
s
-1

for the interband transition in large-band-gap semiconductors [Bogdankevich et al., 1975,
Galkin, 1981]. Coefficient G is the generation rate of electron-hole pairs inside a sample that
is irradiated by primary electrons. The energy and current density of these electrons are
determined as follows

0 0
,
e
e
i
E eU j
S
= = . (2)
In Eqs. (1) and (2) U
0
, e, i
e
, S, d
e
are an accelerating potential, electron charge, a current of
accelerated electrons, an irradiated area of sample and an electron penetration depth,
respectively. An average ionization energy c
i
in Eq. (1) can be approximately estimated as
c
i
~3E
g
, where E
g
is a band-gap energy.
The current of accelerated electrons (i
e
), being injected into the sample, can be determined
with the help of equivalent circuit that is shown in Fig. 1.

Cathodoluminescence

32

Fig. 1. The equivalent circuit of electron current
Here electron accelerator is represented as a source of accelerating potential U
0
with internal
resistance R
0
=U
0
/i
0
=const. The sample with irradiated surface
2
0
S r t = forms a capacitor
with C=2tr
0
cc
0
. Here c
0
is the dielectric constant and c is the permittivity of environment
(e.g. air). Resistor R
t
is introduced as a shunt for capacitor C and it provides the discharge of
the sample surface. For this circuit at R
t
=const and initial condition i
e
(0)=i
o
, the electron
current is described by the equation

0
0
0
( ) 1 exp exp
e e
t c c
U t t
i t i
R R t t
(
| | | |
= + ( | |
| |
+
( \ . \ .

, (3)
where t
c
=CR
0
R
t
/(R
0
+R
t
) is a typical charge time of capacity C.
For example, in the cathodoluminescent microscope [Ramseyer et. al., 1989, Petrov, 1996] the
shunting of capacity C is provided by the emission of secondary electrons over the
irradiated sample surface. This emission results in setting the finite value of shunting
resistance R
t
. Dynamic balance between the primary electrons which are injected into the
sample, and the secondary electrons which leave the sample is equilibrated at t>>t
c
. After
that time the current of injected electrons i
e
tends to achieve the value defined as
i
e
=U
0
/(R
0
+R
t
) and the constant generation rate of electron-hole pairs G is equilibrated. These
conditions are realized for the narrow energy range of the primary electrons 1<E
e
<12 keV
resulted in a small depth of electron penetration d
e
=0.1-1.5 m. Thus, the lower energy of
the primary electrons is limited by the work function of the secondary electrons through the
sample surface. The upper one is limited by the energy loss of secondary electrons, which
appeared on the large depth inside the sample, under diffusion to the sample surface.
The solution of Eq. (1) using G=const reveals that under irradiation by electron beam the
concentration of electron-hole pairs inside the sample volume which is determined by the
beam cross-section and penetration depth of electrons is saturated rapidly with time
according to the following equation:

( )
2
exp( / ) 1 1
2 2 ,
exp( / ) 1
4
t
i i
e h i i
i i
t
n G G
t A
BG A
t t
t t
o t t
= =
+ +
+
. (4)
Here t
i
is an ionization time of substance, o=1+At
i
, =1-At
i
, where At
i
<1. From the Eq. (4) it
can be seen that the concentration of electron-hole pairs increases with increasing the


33
coefficient G. Therefore in order to achieve a high brightness of luminescence, the electron
beam is focused on the sample surface in the spot with a diameter of 1-50 m and current
density j
e
in the range from 10
-2
to 10 A/cm
2
. The coefficient G amounts to 10
23
-10
26
cm
-3
s
-1

and the range of typical time t
i
has an order of 10
-6
-10
-9
s.
Since the 1980-s the electron beams with the energy increased to 20-70 keV and density of
electron current of j
e
=0.1-10 A/cm
2
are applied. These electrons are able to penetrate into the
sample on a depth of 3-30 m [Chukichev et. al., 1990, Yang et. al., 1992]. The secondary
electrons originated on such a depth inside the sample dissipate their own energy while
moved to the outer surface and cant emit outside. In this case the dynamic balance is
provided by the flow of the surplus charge via thin metal film previously deposited on the
sample surface being irradiated and served as a ground. Now the shunt R
t
is determined by
the contact electroconductivity of the irradiated area with the metal film. At these
parameters of electron beam the coefficient G, the ionization time t
i
and the charging time of
capacity t
c
have values of the same order those were mentioned above. However, at
increased energy of electrons the luminescence intensity increases due to the deeper
penetration of electrons and as a result the larger excited volume of substance.
To decrease the thermal load on the irradiated surface at the electron energy of 20-70 keV
the pulsed electron beam with the pulse duration of 1-10 ms [Chukichev et. al., 1990] or
modulated electron beam with the modulation frequency of 100-300 Hz [Yang et. al., 1992]
are applied. The luminescence, excited by such electron beams, is usually called the pulsed
cathodoluminescence (PCL) [Chukichev et. al., 1990]. However, this PCL is the steady-state
one since the pulse duration of injected electrons is much greater than t
i
and t
c
yet.
PCL [Solomonov et. al., 2003], to be talked about in the present chapter, is excited at the
conditions when the dynamic balance between the injected and left electrons is absent, i.e. at
R
t
. The Eq. (3) shows in this case that the current of injected electrons damps
exponentially with characteristic time constant t
v
, as
( ) exp( / )
e eo
i t i t
v
t = ,
0 0
0
0 0
2
e
U
R C
j r
v
cc
t = = . (5)
This time, t
v
, increases with energy of injected electrons. Its value is about 3.5 ns at E
0
=200 keV,
j
e0
=100 A/cm
2
, r
0
=1 mm, and c=1 according to Eq. (5). Therefore the PCL excitation should be
carried out by the electron beam with duration t
e
of the same order as t
v
. The electrons with
energy of 100-200 keV penetrate into dielectric solids on the depth of 100-150 m. Due to the
large penetration depth the coefficient G reaches the value of 10
26
-10
27
cm
-3
s
-1
, which is similar
to that realized at maximum excitation conditions of the steady-state cathodoluminescence.
When t
e
st
v
the concentration of electron-hole pairs comes to n
e(h)
>0.5n
e(h)max
according to Eq.
(4). Here n
e(h)max
is a maximal value of the concentration at t
e
. It means that PCL brightness
is higher than that of steady-state cathodoluminescence excited at the maximal conditions.
PCL spectrum gives the information about the composition and crystal structure of the
sample bulk rather than interface layer. Interface layers are usually characterized with
presence of many absorbed molecules and defect of crystal structure and their properties are
not inherent to the bulk of materials. In PCL the interface layers with a thickness up to 20
m dont have a significant influence on the PCL spectrum quality [Ramseyer et. al., 1990].

Cathodoluminescence

34
It should be emphasized that despite of short time of electron beam impact PCL persistence
occurs and its kinetics is ascribed by complicated laws. This is associated with that the
primary source of luminescence excitation is the electron-hole pairs. Their concentration
according to Eq. (1) over the time of electron beam impact (G=0) is given by

0
( )
0
exp( )
,
1 exp( )
eh
e h
eh
B n A At
n
B At A B n
o
o
o

= =
+
. (6)
Here n
eh0
is the concentration of electron-hole pairs introduced by the electron beam. The
luminescence intensity of the electron and hole centres changes the same law. The intensity
of interband luminescence falls proportionally to n
2
eh
. These kinds of luminescence reach
their maxima at the time moment when the excitation is over. The intensity of intracentre
luminescence changes more difficulty. The first maximum is also reached at the same time
moment but further behaviour depends on the life time of radiative level (t
r
). In the paper
[Solomonov et. al., 1996, 2003] it has been shown that there is the second maximum of the
intensity in the long persistence at t
r
>( 2 -1)/A. After this maximum the intensity falls
according to the exponential law with the characteristic time constant t
r
. Moreover the
second maximum can be more intensive than the first one. If the t
r
<( 2 -1)/A the second
maximum doesnt appear and an exponential decay of luminescence occurs but with
characteristic time constant that is proportional to 1/(2A). It is worthy to note that in case of
using nanosecond exciting electron beams the integral intensity of persistent luminescence is
usually similar or even higher than that during excitation.
2. Apparatus for the PCL registration
The generation of high-current nanosecond electron beam with the energy higher than 100
keV became possible after creation of electron accelerators by G.A. Mesyats in the 1970-s.
These accelerators are founded on the explosive electron emission [Mesyats, 1974]. The
electrons having this energy extend at great distance (more than 10 cm) in air. The
samples excited in air can be used in the form of pieces, powders, and solution. The
irradiation in air furthers also to the partial compensation of injected charge into sample
by the stream of positive air ions, created by the electron beam. The large penetration
depth of these electrons into sample simplifies the sample preparation for analysis
considerably, namely there is no need to undergo the sample to grinding and polishing
procedures. Moreover irradiated surface doesnt require metallization. This is very
important for the analysis of the finished product, in particular, jewels. It should be also
noted that the problem of sample warming, which is typical for the steady-state
cathodoluminescence, is solved due to the introduction of the small energy density (s 3
J/cm
3
).
The investigation of PCL in the different mediums has shown that the portable nanosecond
accelerators of RADAN [Mesyats et. al., 1992] are most applicable for its excitation. These
accelerators include the sealed vacuum electron tube. The biological shielding from X-ray
emission is provided by the design of analytic chamber, which is connected with the output
of the accelerator. The analysed samples are placed into the chamber. The pulsed type of the
luminescence allows using the different methods of its registration.


35
As a first, the traditional method of the registration with the help of optical monochromator,
photoelectronic multiplier, and oscillograph is applied [Vaysburd et.al., 1982, Solomonov et.
al., 1996]. The intensity kinetics of separated luminescence band is measured by this
method. This is necessary for the identification of its nature. The application of the scanning
monochromator allows registering the intensity distribution by wavelengths I (,t).
However, two PCL features have to be kept in mind. The first feature is that the PCL is
characterised by the certain degree of instability of the registered parameters because of the
pulsed regime. Therefore the spectrum measurement has to be performed in the averaging
mode. The second feature is caused by the different kinetics of PCL bands with the various
nature and spectrum registered by such an approach strongly depends on time.
As a second, the time-integral intensity of the luminescent bands can be measured

2
1
( ) ( , )
t
t
I I t dt =
}
.

(7)
Here I (,t) is the current intensity, t
1
is the beginning of registration and t
2
is the ending of
registration. This intensity is registered with the help of multichannel semiconductor
photodetectors based on diode matrix and charge-coupled device [Solomonov et. al., 2003].
In this case optical spectrograph is applied instead of the scanning monochromator and the
wide spectral range for one frame is measured. This method can be used for the PCL
research when intracentre luminescence is dominant. Also the kinetic information about all
registered spectrum can be obtained by means of changing of the integration limits t
1
and t
2
.
Fig. 2 demonstrates the scheme of experimental setup for the receiving of PCL spectra. The
setup consists of the luminescence excitation block (1), multichannel photodetector (2) and
computer (3).

Fig. 2. Scheme of experimental setup

Cathodoluminescence

36
The excitation block (1) represents a combination of RADAN-220 pulsed electron accelerator
and analytical chamber. The operating principle of the accelerator is based on the explosive
emission of electrons from the cold cathode of accelerating tube. The RADAN-220 generates
electron beam with the duration of 2 ns. The voltage that can be applied to the accelerating
tube ranges from 150 to 220 keV. The commercially available IMA3-150E tube is placed in
the analytic chamber. The generated electron beam is extracted to air through the beryllium
foil and directed vertically downwards. The luminescence stream is transferred to the
multichannel photodetector (2) by means of the silica multifiber. Computer (3) is the control
system of the experimental setup. Specad software makes possible to realize various
modes of the photodetector. It provides the calibration, registration, reviewing, processing
and archiving of obtained spectra. The commercially available pulsed cathodoluminescent
spectrograph CLAVI [Michailov et. al., 2001] was created on basis of this experimental
setup.
3. The application of the pulsed cathodoluminescence for the luminescent
analysis of Nd
3+
:Y
3
Al
5
O
12
and Nd
3+
:Y
2
O
3

In the last year the intensive investigations in the field of the optical ceramics creation based
on the metal refractory oxide doped with rare-earth ions, particularly Nd
3+
:Y
3
Al
5
O
12

Nd
3+
:Y
2
O
3
are carried out [Ikesue et. al., 1995, Lu et. al., 2001, Bagaev et. al., 2009]. The
advantages of the laser ceramics against single crystals include the possibility of creating
multilayer elements with sizes greater than those of single crystals, larger concentration of
active ions, and lower manufacturing cost. The fitness of crystals or ceramics for active laser
elements is determined usually by the photoluminescent methods in infrared region by
means of lifetime measurement of upper laser Nd ion level
4
F
3/2
[Hoskins et. al., 1963, Lupei
et. al., 1995]. For this aim the method is effective, however it doesnt display couses of the
lifetime decrease of the laser level. This is necessary to know to correct the conditions the
conditions of crystal and ceramics synthesis of synthesis of crystalls and ceramics. Below the
investigation of the PCL spectra is given. The possibility of realization of qualitative and
quantitative luminescent analyses of Nd
3+
:Y
3
Al
5
O
12
, Nd
3+
:Y
2
O
3
laser materials is developed.
3.1 The luminescence of Nd
3+
:Y
3
Al
5
O
12

The emission lines of neodymium ions in Nd
3+
:Y
3
Al
5
O
12
in visible range correspond to the
transitions from 4f
2
5d
1 2
F2
5/2
level, which has three Stark components v
0
=37775 cm
-1
,
v
1
=37864 cm
-1
, v
2
=38153 cm
-1
, to the levels of 4f
3
configuration of neodymium ion
[Kolomiycev et. al., 1984]. The wavelengths of observed luminescent lines and their
identification for Nd
3+
:Y
3
Al
5
O
12
single crystal are presented in Table 1 in the first and the
second columns, respectively. The numbers of Stark sublevels in according to nomenclature
[Koningstein et. al., 1964] at increasing their energy, starting with zero, are pointed next to
symbol of electron level in brackets; then emission band wavelengths, presented in
[Kolomiycev et. al., 1982], are shown in brackets.
A conspicuous difference appears in PCL spectra of neodymium ions in yttrium aluminates
in case of different crystal structure. The spectrum of orthorhombic Nd
3+
:YAlO
3
single
crystal together with the spectrum of cubic Nd
3+
:Y
3
Al
5
O
12
are presented in Fig. 3 as an
illustration of this difference.


37
Nd
3+
:Y3Al5O12 Nd
3+
:YAlO3
,
nm
Identification of optical trasfer
,
nm
Identification of optical trasfer
389,9
2
F25/2 (2)
2
H9/2 (0), (390,0)
394,6
2
F25/2 (0)
4
F5/2 (2), (394,6)
399,2
2
F25/2 (0)
2
H9/2 (2), (397,5) 398,1
2
F25/2 (0)
2
H9/2 (3), (398,3)
401,6
2
F25/2 (0)
2
H9/2 (4), (401,4)
422,5
2
D5/2 (2)
4
I9/2 (1), (423,0)
426,9
2
F25/2 (2)
4
F9/2 (3), (427,3)
429,9
2
F25/2 (1)
4
F9/2 (0), (430,2) 429,9
2
F25/2 (1)
4
F9/2 (0), (429,9)
435,6
2
F25/2 (0)
4
F9/2 (2), (435,1) 435,2
2
F25/2 (1)
4
F9/2 (1), (434,8)
439,0
2
D5/2 (0)
4
I9/2 (4), (438,8)
440,7
2
P1/2 (0)
4
I9/2 (3), (441,2)
450,4
2
F25/2 (2)
2
H11/2 (3), (450,5) 450,4
2
F25/2 (2)
2
H11/2 (3), (450,4)
455,4
2
F25/2 (0)
2
H11/2 (1), (455,9) 456,0
2
F25/2 (1)
2
H11/2 (3), (456,0)
458,8
2
F25/2 (0)
2
H11/2 (3), (458,3)
461,0
2
F25/2 (0)
2
H11/2 (4), (461,4)
479,3
2
F25/2 (1)
4
G5/2 (1), (479,0) 480,7
2
F25/2 (1)
4
G5/2 (2), (480,5)
487,5
2
F25/2 (0)
2
G7/2 (0), (487,1) 487,8
2
F25/2 (0)
2
G7/2 (2) (488,1)
494,4
4
G11/2 (2)
4
I9/2 (4) (494,2)
525,2
2
F25/2 (0)
4
G7/2 (0), (524,9) 525,4
2
F25/2 (1)
4
G7/2 (1), (525,4)
527,8
2
F25/2 (1)
4
G7/2 (2), (527,8)
538,0
2
F25/2 (1)
4
G9/2 (2), (538,3)
540,6
2
F25/2 (0)
2
K13/2+
2
G9/2 (2), (541,0) 539,5
2
F25/2 (0)
4
G9/2 (0), (539,3)
541,6
2
F25/2 (2)
2
K13/2 (2), (542,0)
545,7
2
F25/2 (2)
2
K13/2 (4), (545,7)
547,5
2
F25/2 (2)
2
K13/2 (5), (547,5)
549,1
2
F25/2 (0)
2
K13/2+
2
G9/2 (7), (549,4) 549,2
2
F25/2 (0)
2
K13/2 (0), (549,0)
554,5
2
F25/2 (0)
2
K13/2 (3), (554,2)
556,3
2
F25/2 (0)
2
K13/2 (4), (556,2)
557,4
2
F25/2 (0)
2
K13/2+
2
G9/2 (10),(557,0)
562,9
2
K15/2 (3)
4
I13/2 (0), (562,6) 563,6
2
K13/2 (4)
4
I11/2(1), (563,5)
576,4
Superposition
2
G7/2 (2,3)
4
I9/2 (0,2), (576,3)

585,4
2
F25/2 (2)
4
G9/2 (4), (585,2)
587,4
2
F25/2 (0)
4
G9/2 (0), (586,8)
596,2
Superposition
2
F25/2 (0)
4
G11/2 (0,1), (596,2)

600,6
2
F25/2 (0)
4
G11/2 (3), (600,1)
602,1
2
F25/2(2)
4
G11/2+
2
K15/2+
2
D3/2(6),(601,8)
610,4
2
F25/2(0)
4
G11/2+
2
K15/2+
2
D3/2(4),(610,2)
612,1
2
F25/2(1)
4
G11/2+
2
K15/2+
2
D3/2(6),(611,8)
615,4
2
F25/2(2)
4
G11/2+
2
K15/2+
2
D3/2(13),(614,7)
620,1
2
F25/2 (0)
2
K15/2 (2), (620,8) 620,1
2
F25/2(0)
4
G11/2+
2
K15/2+
2
D3/2(9),(619,8)
622,1
2
F25/2(1)
4
G11/2+
2
K15/2+
2
D3/2(11),(621,4)
625,4
2
F25/2(1)
4
G11/2+
2
K15/2+
2
D3/2(13),(625,2)
638,3
2
H11/2 (1)
4
I9/2 (2) (637,7)
639,5
2
H11/2 (0)
4
I9/2 (2) (639,1)
646,2
2
H11/2 (3)
4
I9/2 (3) (645,4)
652,4
2
H11/2 (3)
4
I9/2 (4) (652,6)
660,9
2
G7/2 (3)
4
I11/2 (4) (660,8)
665,3
2
G7/2 (0)
4
I11/2 (3) (665,3)
668,6
2
G7/2 (0)
4
I11/2 (4) (667,9)
Table 1. PCL lines and their identification for Nd
3+
:Y
3
Al
5
O
12
and Nd
3+
:YAlO
3
single crystals

Cathodoluminescence

38

Fig. 3. The PCL spectra of Nd
3+
:Y
3
Al
5
O
12
(1) and Nd
3+
:YAlO
3
(2) single crystals
The wavelengths of fundamental neodymium luminescent lines and their identification for
Nd
3+
:YAlO
3
single crystal are presented in Table 1 in the third and the fourth columns,
respectively [Osipov et. al., 2011].
The Fig. 3 and Table 1 show that the principal change is manifested in the considerable
increase of the luminescent band numbers in the yttrium monoaluminate spectrum. This takes
place due to activation of d-f transitions between the different Stark sublevels and the
appearance of f-f transitions. These changes arise from the distortion of crystalline field
symmetry in positions of individual neodymium ions that leads to the modifications of the
oscillator strength and optical transition probability. Thus, the distortion of crystalline field
symmetry appears in the spectrum as the change in intensity and numbers of emission bands.
The differences in the spectra can be used for the determination of the second phase content
in Nd
3+
:Y
3
Al
5
O
12
[Osipov et. al., 2011]. The luminescence lightsum in the spectrum region
from
1
to
2
(
2
1
( ) S I d
=
}
) can be presented by the additive function depended from the
dominant phase content C
g
(cubic Nd
3+
:Y
3
Al
5
O
12
) and the second phase C
im
=1-C
g

(1 )
g g
S C C o | = + , (8)
where o and | are the coefficients of proportionality. They are determined by the integration
range and excitation conditions. To eliminate the influence of the intensity instability it is
necessary to use the ratio of lightsums (S
1
/S
2
) as the analytical parameter calculated for two
ranges of the spectrum. The lightsums in the ranges 350-500 nm (S
1
) and 501-650 nm (S
2
) to
obtain the functional relation between the C
g
and the luminescence intensity of neodymium
ions have been chosen. In that case in accordance with Eq. (8) the content of cubic phase into
Nd
3+
:Y
3
Al
5
O
12
is defined the following equation


39

1
1 2
2
1
2 2 1 2
2
( ) ( )
g
S
S
C
S
S
| |
o | o |
=

(9)
In Fig. 4 the correlation between the C
g
and S
1
/S
2
, calculated for the samples with known
content of cubic phase is shown.

Fig. 4. The correlation between the C
g
and S
1
/S
2
.
This dependence (Fig. 4) is approximated by the following equation with the r
2
>0.99

1
2
1
2
1.071 0.084
0.069
g
S
S
C
S
S

=
. (10)
Moreover the obtained data validity was checked out by the analysis of samples with
electron and optical microscopes.
3.2 The luminescence of Y
2
O
3
, Nd
3+
:Y
2
O
3

The wide band of intrinsic radiation in visible range is a visiting luminescent card of pure
yttria. Earlier the other authors observed this band at different excitation type [Conor, 1964,
Kuznetsov et. al., 1978, Bordun et. al., 1995]. Even at cryogenic temperature of the samples
the unresolved band was registered.
We investigated commercially available yttria powders with a particle sizes of 1-3 m and 5-
10 m. All the powders have a cubic lattice of the o-Y
2
O
3
. From these commercial powders,
nanopowders with the average particle size of 10-12 nm were prepared by the laser
evaporation method. Particles were crystallized in the metastable monoclinic phase -Y
2
O
3
.
After annealing they transformed to the o-Y
2
O
3
.

Cathodoluminescence

40
The spectrum of the powder with particle size of 1-3 m has a broad asymetric band peaked
at 437 nm and long wavelength wing shows local maxima (Fig. 5, curve 1).

Fig. 5. PCL spectra of commercial yttria powders with particle size 1-3 m (1), 5-10 m (2)
and nanopowders with the average particle size of 10-12 nm (3).
In the spectrum of the powder with the particle size of 5-10 m almost all local maxima are
transformed into narrow bands (Fig. 5, curve 2). They are grouped into four series 435 510
nm (the blue series), 515 640 nm (the orange series), 645 700 nm (the red series) and 785
840 nm (the infrared series). The PCL spectra of nanopowders, irrespective of the crystal
phase (either the -Y
2
O
3
or the o-Y
2
O
3
phase) and of the initial coarse powder, have a similar
structures (Fig. 3, curve 3). The broad band with the maximum at 485 nm dominates in these
spectra. The peak range of this band exhibits local maxima of the blue series. Also the lines
of orange series at 573, 583, 612 nm become apparent. The red series is weak, while the
infrared series is hardly seen.
The range of the band series observed in the spectra of pulsed cathodoluminescence
corresponds to the range of intrinsic radiation of yttria, which is identified as the radiation
of associated donor-acceptor pairs Y
3+
- O
2-
[Bordun, 2002].
Since the luminescence wavelengths of narrow bands of commercial powders, nanopowders
coincide, we can assume that these materials contain intrinsic luminescence centers of the
same type.
The series of PCL bands of yttria resemble the radiation of free YO radicals, which is
observed, for example, in laser plume of yttria-containing target [Osipov et. al., 2005]. This
radical has been fairly well studied [Pearse et. al., 1949]. The Table 2 shows the wavelengths
of the bands observed in PCL spectra and their identification. In the second column of this
table the wavelengths of the strongest bands are in boldface.


41
Intrinsic luminescence center
VV , nm
Blue band series, the electronic transition B
2
EX
2
E
00 453.8
22 458.6
33 461.1
01 470.6
23 475.2
02 488.7
Orange band series, the electronic transition A
2
HX
2
E
20+(T
g
+A
g
)=380 542.8
10 551.6
10+(T
g
+A
g
)=380 563.5
00 572.9
33 583.6
01 600.0
34 612.2
02 629.3
Red band series, the electronic transition A
2
HX
2
E
03-(T
g
+A
g
)=162 655.3
03 662.4
03+(T
g
+A
g
)=162 669.6
03+T
g
=469 683.8
Infrared band series, the electronic transition A
2
HX
2
E
05+(T
g
+A
g
)=380 760.8
06 785.0
801.1
818.4
Table 2. Parameters of the PCL lines in the yttria spectrum
Based on these data, we constructed the energy scheme of the intrinsic luminescence center
(Fig. 6). Qualitatively this scheme coincides with that of the YO free radicals. In this scheme
the configuration curves were calculated in the harmonic oscilator approximation as

2
16 2 0
0
2
10 ( )
i
i i i
c m
E E r
h
t v

= +
,
(11)
where i =X, A and B denotes the electronic states X
2
E, A
2
H, and B
2
E, E
0i
,
I
, and v
i
are the
minimal energy, the equilibrium distance, and the wavenumber of the vibration mode of i
th

electronic state, respectively; m
0
is the mass of the oxygen atom; c and h are the light speed
and Plancks constant. The energy E and wavenumbers v
i
are expressed in (11) in reciprocal
centimetres, the amplitudes of the (r-
i
) vibrations are given in nanometres.

Cathodoluminescence

42

Fig. 6. Energy scheme of the intrinsic luminescence center
For the configuration curves of the X
2
E, A
2
H, and B
2
E states (Fig. 6) E
0i
=0, 17510, and 22090
cm
-1
and v
i
=786, 675, and 675 cm
-1
, respectively. With these parameters, up to the
measurement error, the wavelengths of pulsed cathodoluminescence coincide with the
wavelengths of the optical transitions shown in Table 2. The Franck Condon principle for
molecular transitions is most precisely implemented at
A
=
X
+10.870910
-3
nm and
B
=
X
+1.235110
-3
nm, and
X
can be estimated as half of elementary cube edge,
X
=0.1385
nm. For electronic vibration transitions for which one of the vibration quantum numbers V
is large, this principle is implemented only if these transitions involve the most strong
phonons [Schaak et. al., 1970], Table 2. Under these parameters, the configuration curves of
the electronic states A
2
H and B
2
E intersect at the point with E=25256 cm
-1
.
The qualitative coincidence of the emission bands and the energy structure of intrinsic
luminescence centres observed by us and YO free radicals allow us to conclude that intrinsic
luminescence centres in yttria contain bound YO radicals [Osipov et. al., 2008]. Consider the
possibility of formation of such intrinsic luminescence centre. It is known [Schaak et. al.,
1970] that the cubic yttria has unit cell composed of 16 formula units Y
2
O
3
. Twenty four
cations occupy positions with C
2
symmetry and eight cations occupy the positions with C
3i

symmetry (Fig. 7). Every cation is surrounded by six oxygen ions which are positioned on
the corners of deformed cube with the edge size of 0.2702 nm, at that two corners is
unoccupied. Thus in one-third of the cubes (YO
6
) two oxygen vacancies are located at the
cube corners along the face diagonal, while, in the remaining cubes, they are located along
the body diagonal (Fig. 7).
For such packing a structure, presented in Fig. 8. ,can be formed at the outer cube face that
contains two oxygen vacancies and that is located at the crystal boundary. In essence, this
structure is the YO radical bound to the crystal lattice by the yttriun ion. On such surfaces
the fraction of faces with two oxygen vacancies is 1/31/6=1/18 and the average distance


43
between them is about 5 nm. All of this leads to the dependence of the luminescence
spectrum of such bound radicals on the particle size of yttria mainly via their shape.

Fig. 7. The unit cell of yttria. The yttrium positions with C
3i
and C
2
symmetry designated by
red and pink balls respectively. The oxygen and the vacant positions designated by blue and
grey balls. The vacant positions are associated by green dot line.

Fig. 8. The structure of intrinsic luminescence centre.
The considered above intrinsic luminescence centre also presents in Nd
3+
:Y
2
O
3
. However the
presence of neodymium results in decrease of the intrinsic band intensity and distortion of
its profile. We studied the monoclinic and cubic Nd
3+
:Y
2
O
3
nanopowders. The nanopowders
were prepared using a mixture of micropowder o-Y
2
O
3
phase and 1 mol.% Nd
2
O
3
powder.
After evaporation of this mixture by CO
2
laser Nd
3+
:Y
2
O
3
nanoparticles were crystallized
into monoclinic phase. To transfer nanopowders into cubic phase annealing in air was
carried out above 900C [Kotov et. al., 2002].

Cathodoluminescence

44
The PCL spectra of all Nd
3+
:Y
2
O
3
samples contain emission lines of neodymium ions, which
have not been previously observed in photoluminescence. Namely, neodymium doped
nanopowders exhibit a strong band peaking at 825 nm (Fig. 9).

Fig. 9. The PCL spectra of monoclinic (1) and cubic (2) Nd
3+
:Y
2
O
3
nanopowders
Its components correspond well to the
4
F
5/2
4
I
9/2
(825, 811, 834 nm) and
2
H
9/2
4
I
9/2
(818
nm) optical transitions between Nd
3+
Starks sublevels with the energies in Y
2
O
3
cubic lattice
[Chang, 1966]. These components are resolved only in cubic samples (Fig. 9). Therefore the
splitting of Starks components allows us to conclude about the presence of the dominant
phase into Nd
3+
:Y
2
O
3
.
To check this assumption the additional investigations were made. In Fig 10. the PCL
spectra of pressed nanopowders (compacts) are presented. The compacts were annealed at
530, 750, 950, 1100, and 1300C. The X-ray analysis for this samples showed that the
unannealed compact and annealed compact at 530C have monoclinic phase, all remaining
compacts are cubic samples. It is shown that the splitting of neodymium band at region of
800-840 nm only takes place in cubic samples and one component appears at 825 nm in
monoclinic samples.
In addition to the band in the region of 800-840 nm two emission bands of neodymium ions
arise in the Nd
3+
:Y
2
O
3
. These are a weak band at 720 nm due to
4
F
9/2
4
I
9/2
transition and
stronger band at 750 nm with the components due to the transitions between the Stark
sublevels:
4
F
7/2
4
I
9/2
and
4
S
3/2

4
I
9/2
.
The intensity weaking of intrinsic band into Nd
3+
:Y
2
O
3
is associated with the quantitative
decrease of this centres, since the part of yttrium ions are replaced by the neodymium ions.
The distortion of the intrinsic band is determined by the neodymium absorption of it. The
most absorption is observed in region at 560-613 nm [Osipov et. al., 2009].


45
In addition to the present bands into Nd
3+
:Y
2
O
3
the appearance of four well-resolved
components in the range of 610-660 nm can be seen. The specta of this band for compacts
annealed at 950 and 1300C are presented in Fig. 11. The band contains the following four
narrow lines at 620.6, 630.6, 645.3, 655.6 nm which we identify using Raman spectra, as
luminescence of oxygen molecular ion
2
O
[Solomonov et. al., 2011].

Fig. 10. The PCL spectra of unannealed compact (1) and compacts annealed at 530C (2),
750C (3), 950C (4), 1100C (5), 1300C (6).

Cathodoluminescence

46
The bands with the frequencies of 1615 and 1702 cm
-1
correspond to vibrations of the
molecular ion in the ground state, while the frequencies at 966 and 993 cm
-1
correspond to
the excited state for the two sites of the oxygen molecular ion in the yttria lattice.
Fig. 12 demonstrates two sites of defects
2
O
taking into account the occurrence of two types

of natural vacancies about we talked earlier.

Fig. 11. The PCL of oxygen molecular ions.

Fig. 12. Two sites of oxygen molecular ions defects in lattice of cubic Y
2
O
3

Fig. 13 presents two potentional curves and formation of luminescence bands. The lowest
excited state A
2
H
u
of the oxygen molecular ion can be stabilized.
The observed luminescence band results of the transition from the vibration level V=0 of
the excited electronic state A
2
H
u
to one vibration level V=3 of the ground state X
2
H
g
. The


47
bands at 630.6 and 655.6 nm are formed due to the transition to the vibration level of the
groun state with the participation of lattice phonons. For the first curve the phonon energy
is 255 cm, while, for the second curve it is 244 cm
-1
. These phonons are observed in the
Raman spectrum. The excited electronic state of the molecular ion is spaced from the ground
state by 20660-21580 cm
-1
[Solomonov et. al., 2011].

Fig. 13. Configuration curves for two sites of oxygen molecular ions
On the basis of the qualitative luminescent analysis of Nd
3+
:Y
2
O
3
the determination of
neodymium concentration by means of the calibration curve construction is possible
because of intensity of neodymium lines is proportional to it concentration I
Nd
(
i
)=a
i
C
Nd

(for example 750 or 825 nm). However we cant use this equation because PCL spectrum is
characterized by the instability. Therefore we also chose lightsums ratio as analytic
parameter in regions of 730-840 nm and 350-840 nm. The first region of Nd
3+
:Y
2
O
3
spectrum
involves only neodymium bands, but the second one includes in addition the intrinsic band
which is distorted by the neodymium absorption and can be ascribed as follows

Cathodoluminescence

48

0
1 exp( )
( )
i Nd
YO i i
i Nd
k C l
I I
k C

(12)
Here I
0i
, l, k
i
are the intensity of i intrinsic band without absorption, the thickness of
samples, coefficient of absorption of i band, respectively. Hence relation of lightsums ratio
(S(350-840)/S(730-840)) with C
Nd
has to include the equation (12). Really this relation is
decribed by the following equation

2
1 exp( ) (350 840)
(730 840)
Nd
Nd
B C S
A D
S C

= +
.
(13)
Fig. 14 demonstrates the calibration curve (r
2
>0.99) for the determination of neodymium
concentration in region of 0.11 1.07 at. %.

Fig. 14. The calibration curve for the determination of neodymium concentration in region of
0.11 1.07 at. %.
4. Conclusion
Thus, the possibility of realization of rapid, nondestructive, qualitative and quantitative
luminescent analyses of laser materials, in particular Nd
3+
:Y
3
Al
5
O
12
, Nd
3+
:Y
2
O
3
, with the
help of pulsed cathodoluminescence was shown.
5. References
Bagaev, S.; Osipov, V.; Ivanov, M.; Solomonov, V.; Platonov, V.; Orlov, A.; Rasuleva, A. &
Vatnik, S. (2009).Fabrication and characteristics of neodymium activated yttrium
oxide optical ceramics. Optical Materials, Vol.31, pp. 740-743, ISSN 0925-3467


49
Bogdankevich, O.; Darznik, S. & Eliseev, P. (1975). Semiconductor lasers, Nauka, Moscow,
Russia
Bordun, O.; Bordun, I. & Novosad, S. (1995). Luminescence centres into Y
2
O
3
. Journal of
applied spectroscopy, Vol.62, No.6, pp. 91-95, ISSN 0021-9037
Bordun, O. (2002). Influence of oxygen vacancies on the luminescence spectra of thin Y
2
O
3

films. Journal of applied spectroscopy, Vol.69, No.3, pp. 371-374, ISSN 0021-9037
Chang, N. (1966). Energy levels and crystal field splittings of Nd
3+
in yttrium oxide. J.
Chem. Phys., Vol.44, pp. 4044-4050, ISSN 0021-9606
Chukichev, M.; Sabri, D.; Sokolov, V. & Surcova T. (1990). Cathodoluminescence of solid
solution Zn
1-x
Mn
x
Se. Optics and Spectroscopy, Vol.68, pp. 2000-2021, ISSN 0030-4034
Connor, O. (1964). A theory of thermoluminescence of fluorite (CaF
2
:Y). Radiative
recombination from highly associated electron hole pairs. Appl. Phys. Lett., Vol.4,
No.1, pp. 126-129, ISSN 0003-6951
Galkin, G. (1981). The interband processes of the recombination into semiconductors at high
excitation level. Proceedings of the P.N. Lebedev Physical Institute (Trudi FIAN),
Vol.128, pp. 3-64
Hoskins, H. & Soffer, B. (1963). Stimulated emission from Y
2
O
3
:Nd
3+
. Appl. Phys. Lett., Vol.4,
pp. 22-23, ISSN 0003-6951
Ikesue, A.; Kinoshita, T.; Kamata, K. & Yoshida, K. (1995). Fabrication and optical properties
of high-performance polycrystalline Nd:YAG ceramics for solid-state lasers. J.
Amer. Ceram. Soc., Vol.78, pp. 1033-1040, ISSN 1551-2916
Kolomiycev, A.; Meylman, M.; Volodina, I.; Chukichev, M.; Smagin, A. & Bagdasarov H.
(1982). Luminescence of neodymium activated yttrium aluminum garnet
crystals in ultra-violet and visible ranges at high energy excitation. Deposition in
VINITI, No. 5995-82, pp. 1-15.
Kolomiycev, A.; Meylman, M.; Volodina, I.; Chukichev, M.; Smagin, A. & Bagdasarov H.
(1984). Luminescence of neodymium activated yttrium aluminum garnet
crystals in ultra-violet and visible ranges at high energy excitation. Optics and
Spectroscopy, Vol.56, pp. 365-367, ISSN 0030-4034
Koningstein, J. & Geusic, J. (1964).Energy levels and crystal field calculations of
neodymium in yttrium aluminum garnet. Phys. Rev., Vol.136, No. 3A pp. 711-716,
ISSN 1893-1912
Kotov, Yu.; Osipov, V.; Ivanov, M.; Samatov, O.; Platonov, V., Azarkevich, E.; Murzakayev,
A. & Medvedev. A. (2002). Properties of oxide nanopowders prepared by target
evaporation with a pulse periodic CO
2
laser. Technical Physics, Vol.47, No.11 pp.
1420-1426, ISSN 1063-7842
Kuznetsov, A.; Abramov, V.; Rooze, N. & Savikhina, T. (1978). Autolocalized excitons into
Y
2
O
3
. JETP Letters, Vol.28, No.10, pp. 602-605, ISSN 0021-3640
Lu, J.; Murai, T.; Takaichi, K.; Ueda, K.; Yagi, H.; Yanagitani, T. & Kaminskii, A. (2001)
Nd
3+
:Y
2
O
3
ceramic laser. Jpn. J. Appl. Phys., Vol.40, pp. 1277-1279, ISSN 0021-4922
Lupei, V.; Lupei, A.; Tiseanu, C.; Georgescu, S.; Stoicescu, C. & Nanau, P. (1995) High
resolution optical spectroscopy of YAG: Nd: A test for structural and distribution
models. Phys. Rev. B, Vol.51, No.1 pp. 8-17, ISSN 0003-6951
Mesyats, G. (1974). Generating of highpower nanosecond pulses, The Soviet radio, Moscow,
Russia

Cathodoluminescence

50
Mesyats, G.; Shpak, V.; Yalandin, M. & Shunailov S. (1992). Compact RADAN electron
accelerators for testing new radiation technologies and sterilization. Radiat. Phys.
Chem., Vol.46, pp. 489-492, ISSN 0969-806X
Michailov, S.; Osipov, V. & Solomonov V. (2001). Pulsed cathodoluminescent analyzer of
materials. Pribory i tekhnika eksperimenta, No.3, pp. 164-165, ISSN 0032-8162
Osipov, V.; Solomonov, V.; Platonov, V.; Snigireva, O.; Ivanov, V. & Lisenkov, V. (2005).
Laser plume spectroscopy. 2. Graphite yttrium stabilized and zirconium oxide
targets. Quantum Electron., Vol.35, No.7 pp. 633-637, ISSN 0368-7147
Osipov, V.; Rasuleva, A. & Solomonov, V. (2008). Luminescence of pure yttria. Optics and
Spectroscopy, Vol.105, No.4 pp. 524-530, ISSN 0030-400X
Osipov, V.; Solomonov, V.; Spirina, A.; Ivanov, M. & Orlov, A. (2009). Luminescence of
yttrium oxide doped with neodymium. Optics and Spectroscopy, Vol.106, No.1 pp.
78-83, ISSN 0030-400X
Osipov, V.; Solomonov, V. & Spirina, A. (2011). Luminescent investigation of neodymium
doped yttrium aluminates. J. of Optical Technology, Vol.78, No.6 pp. 81-87, ISSN
0030-4042
Pearse, R. & Gaydon A. (1949). The identification of molecular spectra, Inostrannaya Literatura,
Moscow, Russia
Petrov, V. (1996). Cathodoluminescent microscopy. Physics - Uspekhi, Vol.39, pp. 807-818,
ISSN 0042-1294
Ramseyer, K.; Fisher, J.; Matter, A.; Eberhardt, P. & Geiss J. (1989). A cathodoluminescence
microscope for low intensity luminescence. Journal of Sedimentary Research, Vol.59,
pp. 619-622, ISSN 0022-4472
Ramseyer, K. & Millis J. (1990). Factors influencing short-lived blue cathodoluminescence in
o - quartz. Journal of Sedimentary Research, Vol.59, pp. 619-622, ISSN 0003-004X
Schaak, G. & Koningstein, J. (1970). Phonon and electronic raman spectra of cubic rare-earth
oxides and isomorphous ittrium oxide. J. Opt. Soc. Amer. Vol.60, No.8, pp. 1110-
1115, ISSN 1551-2916
Solomonov, V. & Michailov, S. (2003). Pulsed cathodoluminescence and its application for the
analysis of condensed substances, UB RAS, ISBN 5-7691-1357-X, Yekaterinburg, Russia
Solomonv, V. (2003). Kinetics of pulsed cathodoluminescence. Optics and Spectroscopy,
Vol.95, pp. 248-254, ISSN 0030-4034
Solomonv, V.; Michailov, S; & Deykun, A. (1996). About the mecanism of excitation and the
structure of the pulsed cathodoluminescence bands of Cr
3+
, Mn
2+
impurity ions into
minerals. Optics and Spectroscopy, Vol.80, pp. 447-458, ISSN 0030-4034
Vaysburd, D.; Semin, B.; Tavanov, E.; Matlis, S.; Balichev, I. & Gering G. (1982) Highenergy
electronics of solid, Nauka, Novosibirsk, Russia
Yang, B.; Luff, B. & Townsend F. (1992). Chatodoluminescence of natural zircons. J. Phys.:
Condens. Matter., Vol.4, pp. 5617-5624, ISSN 0953-8984
3
Pulsed Cathodoluminescence of Natural and
Synthetic Diamonds Excited by Nanosecond
and Subnanosecond Electron Beams
E.I. Lipatov
1
, V.M. Lisitsyn
2
, V.I. Oleshko
2
,
E.F. Polisadova
2
, V.F. Tarasenko
1
and E.H. Baksht
1

1
High Current Electronics Institute, Tomsk,
2
Tomsk Polytechnic University, Tomsk,
Russian Federation
1. Introduction
Research on the luminescence of crystals due to ionizing radiation is an important field in
spectral analysis of solids that allows determination of the chemical composition of material,
its structure (the presence of defects), state (internal stress), etc. [Marfunin; Gritsenko et al.;
Kravchenko et al.; Solomonov&Mikhailov]. The luminescence spectra and kinetics of various
crystals are most commonly studied using pulsed accelerators with vacuum diodes
[Gritsenko et al; Korepanov et al., 2000; Korepanov et al., 2005; Kravchenko et al.; Lipatov et
al., 2007 (1); Lisitsyna et al., 2002; Lisitsyna et al., 2008; Lisitsyna et al., 2011;
Solomonov&Mikhailov]. These accelerators produce electron beams of varying energy,
current density, and pulse duration: the beam current pulse duration is varied by varying
the voltage pulse parameters; the electron energy in a vacuum diode is determined by the
voltage across the interelectrode gap. However with short beam current pulses (~ 100 ps),
the design of accelerators with vacuum diodes is added much complexity [Zheltov]. In this
case, electron accelerators with gas diodes are more preferable to use [Tarasenko et al., 2008;
Tarasenko et al., 2009; Tarasenko et al., 2010]; they are simpler in design and make it possible
to produce an electron beam of current density ~ 100 /cm
2
and to control the pulse
duration in the range from 100 to 500 ps [Baksht]. In this work, experiments were performed
on two electron accelerators with vacuum and gas diodes. Compact electron accelerators
based on vacuum and gas-filled diodes ensure high-intensity pulsed cathodoluminescence
(PCL) typical of high-current electron beams [Solomonov&Mikhailov].
In our previous works, we demonstrated intense PCL of spodumene [Lipatov et al., 2007
(1)], ruby [Lipatov et al., 2005 (1)], calcite [Baksht et at], fluorite [Baksht et al], and diamond
[Baksht et al.; Lipatov et al., 2005 (2)] excited by a subnanosecond avalanche electron beam
and measured the integrated PCL spectra [Baksht et al.; Lipatov et al., 2005 (1); Lipatov et al.,
2005 (2)] and the luminescence decay kinetics [Baksht et al.; Lipatov et al., 2007 (1)].
The objective of the present paper was to study the luminescence of two specimens of
diamond (natural and synthetic) excited by nanosecond and subnanosecond electron beams

Cathodoluminescence

52
with a current density of 10200 A/cm
2
in the spectral range from 200 to 850 nm. The use of
compact nanosecond and subnanosecond high-current electron accelerators as excitation
sources allowed pulsed luminescence spectrometry with high temporal resolution, high
light signal intensity, and hence high sensitivity and accuracy of measuring the decay
characteristics over a wide time interval (10
10
10
1
s).
The characteristics under study were the PCL spectra and decay kinetics of the diamond
specimens, including the time evolution of the spectra. The both specimens revealed a PCL
band due to intrinsic structural defects (Band-A). In the PCL spectra of the synthetic
diamond, because of the lower impurity content, a radiative recombination band of free
excitons was found, whereas in the PCL spectra of the natural diamond this band escaped
detection because of the high N impurity content. In the spectrum of the natural diamond,
the N3 system due to nitrogen-containing centers was observed.
2. Experimental setups, methods, and specimens
The integrated PCL spectra of the specimens were measured as shown in Fig. 1a. The
electron beam produced by accelerator 1 (the characteristics of the accelerators are presented
in Table 1) excited pulsed cathodoluminescence in specimen 2. The luminescence was
transferred through optical fiber 3 to spectrometer 5 (EPP-2000C Stellar-Net Inc.) and
converted to an electrical signal which was transmitted to computer 9 and represented as a
spectrum. In PCL spectrum processing, the optical absorption spectra of the specimen 2 and
fiber 3 and the spectral sensitivity of the spectrometer 5 were taken into account.

Accelerator E
m
, keV t
0.5
, ns j
e
, /cm
2
, mJ/cm
2

SLEP-150 100 0.10.65 10100 1.58.0
RADAN-EXPERT IMA3-150E 150 1.5 100 20
RADAN-220 IMA3-150E 220 2 200 60
NORA 240 4 200 120
Vacuum diode with GIN-600 360 10 30 100
Table 1. Characteristics of the accelerators and electron beams that excited PCL in the
specimens.
The optical absorption spectra were also measured as shown in Fig. 1a, except that a KrCl-
lamp [Lipatov et al., 2010] was used instead of the accelerator 1. The optical radiation was
delivered to the spectrometer 5 through the optical waveguide 3. Once the reference
spectrum was measured, the entrance aperture of the waveguide 3 was covered with the
specimen 2. The plane-parallel specimen 2 was arranged on a special cooled holder with
forevacuum pumping such that the output facet was parallel to the entrance aperture of the
waveguide 3. Because the radiation intensity of the KrCl-excilamp at 200260 nm was much
higher than that at 260800 nm, the foregoing procedure was followed for each range
separately. Two parts of the optical absorption spectrum was joined through multiplying
the short-wave part by an appropriate factor.
Synthetic Diamonds Excited by Nanosecond and Subnanosecond Electron Beams

53

Fig. 1. Measurement scheme a) the integral spectra of OA and PCL samples and b) PCL
spectra with a time resolution. 1) light source or an electron accelerator, 2) the test specimen
of diamond, 3) fiber, 4) the collecting lens, 5) spectrometer, 6) monochromator, 7)
photomultiplier or photodiode, 8) digital oscilloscope, 9) PC.
The PCL spectra and decay kinetics of the specimens were studied using a MDR-23
monochromator, a FEU-100 (FEU-97) photomultiplier tube, and a FEK-22 photodiode. The
measuring procedure is shown in Fig. 1b. The electron beam produced by the accelerator 1
excited luminescence in the specimen 2, which was fixed on a special cooled holder with
forevacuum pumping. The irradiated part of the specimen surface was in the focus of
collecting lens 4. The collimated light beam was delivered to the entrance slit of
monochromator 6. The entrance and exit slits were normally 10 m wide. Spectrum scanning
was provided by rotation of a diffraction grating of 600 and 1200 groove/mm with a stepping
motor. The spectral resolution was about 1 nm. The optical signal was converted to an
electrical signal by photomultiplier tube (or photodiode) 7, was recorded as a pulse by TDS-
3032 and TDS-6604 Tektronix digital oscilloscopes 8, and was transmitted to the computer 9.
The luminescence pulse was recorded in a separate file for each wavelength. The luminescence
spectrum at an arbitrary point in time was reconstructed by software processing. The resulting
spectrum was normalized to the spectral sensitivity of the photomultiplier tube or photodiode.
The time interval of measurement of the luminescence decay kinetics (nano- or milliseconds)
was determined by choosing the load on the oscilloscope 50 Ohm or 1 MOhm.
The luminescence relaxation kinetics was determined from approximation of the light pulse
fall time by a simple exponential function for each wavelength. The thus obtained values
were used to reconstruct the luminescence relaxation time spectrum of the specimens.
We had ten diamond specimens at our disposal. The main experiments were performed
with one natural diamond specimen and one synthetic diamond specimen whose
description and characteristics are given in Table 2.

Cathodoluminescence

54
No Description Dimensions, mm Type
*
Synthesis method
1 Disk C5 0.25 2 natural
2 Square plate 10 10 0.1 2 CVD
**

*
standard physical classification [Zaitsev];
**
chemical vapor deposition.
Table 2. Number, description, dimensions, type, and method of synthesis of the diamond
specimens.

Fig. 2. The left axis: the transmission spectra of natural (continuous curve) and synthetic
(broken curve) samples of diamond, measured at room temperature, and the transmission
spectrum, calculated according to expression (1), for an ideal (non-absorbing) plane-parallel
specimen of the diamond. The right axis: absorption spectra, calculated according to
expression (1) and (2), of natural (continuous curve) and synthetic (broken curve) samples of
the diamond.
Figure 2 shows the optical absorption spectra of the specimens measured by the foregoing
procedure. The absorption spectra of the specimens was calculated by the dichotomy
technique with resort to the available data [Zaitsev] on dispersion of the refractive index of
diamond and to the expressions:
( )
( ) ( )
( )
( )
( )
( )
2
2 2
1
1
d
d
r e
T
r e
o
o

=

, (1)
( )
( ) ( )
( ) ( )
2
2
1
1
n
r
n

=
+
, (2)
where n() is the refraction index, r() is the reflection coefficient, o() is the absorption
coefficient [cm
1
], and d is the optical thickness of a specimen. The calculated transmission
spectrum of a perfect plane-parallel diamond shown in Fig. 2. Noteworthy is the absence of
narrow-band peculiarities in the absorption spectra of the diamond specimens, the presence

55
of a clearly defined fundamental absorption edge at 225 nm, and high imperfection of the
synthetic diamond. According to the standard physical classification of diamonds [Zaitsev],
the specimens are referred to type 2a.
3. Results and discussion
Figure 3 shows integrated PCL spectra of the natural (continuous spectrum) and synthetic
(dotted spectrum) diamonds cooled with liquid nitrogen.
The integrated PCL spectrum of the natural diamond is dominated by the N3 vibronic
system with a zero-phonon line (ZPL) at 415.2 nm and its phonon replicas at 420450 nm
due to the presence of N
3
V defects [Zaitsev]. The N3 system is thus superimposed on a
structureless luminescence band-A in the spectral range 300700 nm. According to
[Takeuchi], the luminescence band-A owes to intrinsic defects of the diamond lattice sp
2
-
hybridized carbon bonds.

Fig. 3. Integrated PCL spectra of natural (continuous line) and synthetic (broken curve) of
diamond samples, measured at liquid nitrogen temperatures. Inset: the band of the
radiative recombination of free excitons (FE
TO
at 235 nm), the zero-phonon line of N
3
V-
defects (ZPL 415.2 nm), the emission bands of the second positive system of molecular
nitrogen in the air.
On the contrary, the integrated PCL spectrum of the synthetic diamond revealed a very
weak band-A and was dominated by a radiative recombination band of free excitons FE
TO
at
235 nm. Because diamond is an indirect semiconductor, radiative recombination of free
excitons occurs during the generation of phonons or a cascade of phonons [Fujii]. In the
radiative recombination band of free excitons in the diamond, local maxima are observed
according to the following expression:

h g b p
E E E E
v
= , (3)

Cathodoluminescence

56
where E
hv
is the maximum energy in the band, E
g
is the width of the band gap of diamond
(5.49 eV), E
b
is the binding energy of a free exciton (80 meV), and E
p
is the energy of the
corresponding phonon. The radiative recombination spectrum of free excitons in pure
diamond normally reveals local maxima corresponding to the following phonons of the
diamond lattice [Fujii]: transverse acoustic phonon (TA, 87 meV) at 232.8 nm, transverse
optical phonon (TO, 141 meV) at 235.2 nm, longitudinal optical phonon (LO, 163 meV) at
236.2 nm, and their cascades TO+LO at 242.7 nm and TO+LO+LO at 250.7 nm. The
dominant component is thus the transverse acoustic phonon FE
TO
at 235.2 nm against the
background of which the longitudinal optical phonon at 236.2 nm is not distinguished.
In the PCL spectrum of the synthetic diamond in Fig. 3, the classical pattern of radiative
recombination of free excitons is observed.
In the PCL spectrums of the both specimens in Fig. 3, luminescence of the second positive
system of molecular nitrogen is detected in the form of a series of bands at 305395 nm. The
luminescence of the second positive system of molecular nitrogen was observed earlier in
studies of the luminescence properties of natural spodumene (LiAlSi
2
O
6
with Mn
2+

impurity) in [Lipatov et al., 2007 (1)].
The considerable difference in the spectra of the natural and synthetic diamonds owes to the
difference in their impurity contents. The ratio between radiative recombination bands of free
excitons and luminescence band-A in a diamond allows one to judge the specimen perfection
[Kawarada], i.e., as the defect or impurity concentration increases, the exciton luminescence
intensity decreases and the band-A intensity increases due to amplification of radiationless
recombination of free excitons and amount of sp
2
- hybridized carbon bonds. According to
[Kawarada], the ratio between the exciton and band-A intensities (FE
TO
/A) is close to unity at
a defect concentration of ~ 10
17
cm
3
.
The natural diamond contains nitrogen in sufficient concentration (~ 10
18
cm
3
) to
demonstrate the vibronic luminescence band of N
3
V defects. It was assumed that the
excitation from N
3
V defects is transferred to the band-A centers thus increasing the band-A
contribution to the integrated luminescence spectrum [Lipatov et al., 2007 (2)].
In the natural diamond, exciton luminescence bands were observed, but at FE
TO
/A < 0.01
[Dean], i.e., the exciton luminescence under our conditions is impossible to distinguish
against the background of the noise component.
3.1 Integrated PCL spectra of natural diamond
Pulsed cathodoluminescence of the natural diamond was excited by electron beams of
duration 0.110 ns. Figure 4 shows integrated PCL spectra of the natural diamond excited by
electron beams of duration 0.1, 2, and 4 ns.
The general PCL characteristics depend almost not at all on the duration of electron beam
pulses. The band-A was found to dominate at 300650 nm and was superimposed by the
vibronic N3 system with a ZPL at 415.2 nm and phonon replicas at 420480 nm. The
maximum of the integrated spectra was between 430 and 440 nm.
It is found that the longer the beam duration and the higher the beam current density, the
larger the contribution of the N3 system to the integrated PCL spectrum. Presumably,
increasing the beam current density increased the duration of luminescence of the N3 system.

57

Fig. 4. Integrated luminescence spectra of natural diamond sample at various durations of
the electron beam (0.1 ns, 2 ns, 4 ns) at room temperature (continuous lines) and at cooling
with liquid nitrogen (broken lines). Zero-phonon line of N
3
V-defects (ZPL 415.2 nm) is
marked. Inset: the spectral range of luminescence bands of the second positive system of
molecular nitrogen in the air, and integral luminescence spectra of natural diamond sample
under excitation by the electron beam with a duration 0.1 ns on a larger scale.
Cooling of the natural specimen with liquid nitrogen increased the ZPL intensity at
415.2 nm by a factor of ~ 2.5, ~ 8.8, and ~ 6.8 with a beam of duration 0.1, 2, and 4 ns,
respectively. The intensity of phonon replicas of the ZPL of the N3 system increased on
cooling with liquid nitrogen by a factor of ~ 2.1, 6.2, and 6.3, respectively. The band-A
intensity was nearly the same and increased ~ 1.8 times at all three beam durations.
Thus, the temperature quenching of luminescence was weaker for recombination radiation
(the band-A) than for intracenter transitions (the N3 system) [Solomonov].
On cooling of the natural specimen, the intensity of lines of the second positive system of
molecular nitrogen also increased. The causes for this phenomenon were discussed for mineral
spodumene earlier in [Lipatov et al., 2007 (1)] and are beyond the scope of the present study.
3.2 Time-resolved PCL spectra of natural diamond
The PCL spectra of the natural diamond were measured at room temperature by the
procedure illustrated in Fig. 1b. Figure 5 shows PCL spectra of the natural diamond excited
by an electron beam of duration 0.1 and 10 ns; the spectra were reconstructed at the maxima
of light pulses and within 2 ms after the beginning of the pulse.

Cathodoluminescence

58

Fig. 5. PCL spectra of natural diamond sample recovered from the amplitudes of the light
signal at the maximums, and 2 ms after the onset of a light signal when excited by an
electron beam of duration 0.1 ns (continuous lines) and 10 ns (broken line). For ease of
comparison, the spectra measured at 2 ms after the start of the pulse, are enlarged in 1000
times. Inset: the spectral range of zero-phonon line 415.2 nm in an enlarged scale. The
measurements were performed at room temperature.
With time, the PCL spectrum of the natural diamond underwent considerable
transformations. At the maxima of the light pulse (within 1015 ns after the beginning of the
pulse), the PCL spectrum contained an intense band of the N3 system with a ZPL at 415.2
nm and phonon replicas at 420480 nm.
Within 2 ms after the beginning of the excitation pulse, the PCL band intensity of the
natural diamond decreased three orders of magnitude (for convenient comparison, the
spectra are magnified by the corresponding factor). In this case, the N3 system escaped
detection, whereas the band-A was detected in its undistorted form with a maximum at
450470 nm.
For two excitation pulse (electron beam) durations of 0.1 and 10 ns at comparable current
densities, the reconstructed spectra were almost the same. With a duration of 10 ns, the
phonon replicas of the ZPL at 415.2 nm in the spectra of maximum light signals were much
more pronounced. With a duration of 10 ns, the ZPL intensity at 415.2 nm was thus 1.6 times
higher than that found with a duration of 0.1 ns. These differences were likely to be due to
transient processes.
Within 2 ms after the beginning of the excitation pulse, the band-A was the same for both
excitation pulse durations.

59
3.3 Integrated PCL spectra of synthetic diamond
With an excitation pulse duration of 2 and 1.5 ns, the integrated PCL spectra of the synthetic
diamond contained a weak band-A, an intense recombination band of free excitons, and
lines of the second positive system of atmospheric nitrogen (see Fig. 6).

Fig. 6. Integrated luminescence spectra of synthetic diamond at room temperature (RT,
continuous lines) and measured at temperature of liquid nitrogen (80 K, broken lines) under
excitation by electron beam with a duration 2 ns and 1.5 ns. Spectra of PCL for the duration
of the electron beam of 1.5 ns is shown in an enlarged scale. The second positive band
system of molecular nitrogen of air is marked. Inset: the spectral range of the band radiative
recombination of free excitons (FE
TO
at 235 nm).
Cooling of the specimen with liquid nitrogen increased the exciton luminescence intensity ~
2.5 times with a modest increase in band-A intensity.
For observation of the band-A of the synthetic diamond, a setup incorporated an image
converter tube with a spectral range of 350750 nm and excitation pulse duration of 1.5 ns
was used [see Lipatov et al., 2007 (1) and references herein]. Note that the thus obtained
band-A had a spectral maximum at 430450 nm, i.e., was shifted by 20 nm toward the short-
wave region with respect to the band-A of the natural diamond (see Fig. 5). The short-wave
wing of the band-A of the synthetic diamond terminated steeply at 400 nm, whereas that of
the natural diamond decreased gradually down to 350 nm.
The long-wave wing of the band-A was observed to 650 nm for both specimens; however,
the intensity for the synthetic diamond decreased rapidly from 450 to 480 nm, whereupon it
decreased gradually to 650 nm.

Cathodoluminescence

60
It is conceivable that the described distinctions of the band-A for the synthetic diamond owe
to the peculiar spectral sensitivity of the multichannel setup with an image converter tube.
The recombination band of free excitons at room temperature (RT) and on cooling with
liquid nitrogen (80 K) was similar. All phonon components inherent in the luminescence of
free excitons in diamond were observed (FE
TA
at 232.8 nm, FE
TO
at 235.2 nm, FE
LO
at
236.2 nm, FE
TO+LO
at 242.7 nm, and FE
TO+LO+LO
at 250.7 nm) [Fujii]. Cooling with liquid
nitrogen increased the intensity (2.5 times for the FE
TO
component) with a decrease in the
contribution of the FE
TA
component.
At 80 K, the intensity of lines of the second positive nitrogen system also increased, as was
the case for the natural diamond (see Fig. 4). However at an excitation pulse duration of
0.1 ns, the measured spectrum was free of lines of the second positive nitrogen system
(see Fig. 7).

Fig. 7. Integrated luminescence spectra of synthetic diamond sample, measured at room
temperature, under excitation by electron beam with durations 0.1 ns and 2 ns. The second
positive band system of molecular nitrogen of the air is marked. Inset: the spectral range of
the band radiative recombination of free excitons (FE
TO
at 235 nm).
The recombination luminescence bands of free excitons with excitation pulse durations of
0.1 and 2 ns were distinct (Fig. 7). So in both cases, the FE
TO
component of exciton
luminescence of the diamond at 235.2 nm dominates; however, with an excitation pulse
duration of 0.1 ns, the FE
TA
component at 232.8 nm is not distinguished and mixing of
multiphonon components with a common maximum at 242246 nm in the long-wave region
of the band is found. The difference of the radiative recombination band of free excitons in
the diamond from the classical form at an excitation pulse duration of 0.1 ns presumably
owes to transient processes.

61
At both excitation pulse durations, the band-A was weak and structureless.
The intensity of the FE
TO
component of exciton luminescence of the synthetic specimen
increased nonlinearly with increasing the beam current density, as evidenced in Fig. 8.
Approximation of the dependence by a power function gave an exponent of 1.4 (from 1.2 to
1.6 within confidence intervals). The nonlinear dependence of the output PCL on the
electron beam current density made the radiative recombination band of free excitons
undetectable against the background of the noise component in the low-impurity diamond
at low excitation intensities. Apparently, this effect impeded the observation of exciton
luminescence in the synthetic diamond excited by spontaneous UV sources (excilamps),
whereas the excitation of photoluminescence by a pulsed KrCl-laser allowed observation of
an intense FE
TO
component of exciton luminescence [Lipatov et al., 2010].

Fig. 8. The intensity of the band 235 nm (radiative recombination of free excitons FE
TO
) in
the integral PCL spectra from electron beam current density for a sample of synthetic
diamond.
3.4 Time-resolved PCL spectra of synthetic diamond
Figure 9 shows PCL spectra (at 232238 nm) of the synthetic diamond excited by an electron
beam of duration of 2 ns; the spectra were reconstructed at the maxima of light pulses
(within 0.60.9 ns after the beginning of the excitation pulse); the measurements were taken
at room temperature and at 80 K.
The spectra reveal only the FE
TO
component at 235.2 nm and the other phonon components
present in the integrated luminescence spectra of this diamond (see Figs. 6 and 7) escape
detection at an excitation pulse duration of 2 ns. The absence of phonon components of
exciton luminescence, except for the FE
TO
component, in the time-resolved PCL spectrum of
[mJ/cm
2
]

Cathodoluminescence

62
the synthetic diamond confirms the effect of transient processes on the luminescence of
phonon components the assumption of which was made in analyzing the integrated PCL
spectrum of the diamond excited by an electron beam of duration 0.1 ns (Fig. 7). So the
integrated PCL spectrum was free of the FE
TA
component at 232.8 nm and mixing of
multiphonon components with a common maximum at 242246 nm was observed in the
spectrum.

Fig. 9. Luminescence spectra and the decay time PCL of synthetic diamond sample in the
spectral range of the band radiative recombination of free excitons (FE
TO
at 235 nm). The
intensity of luminescence is the maximum amplitude of the light pulse, which corresponds
to the time 1 ns after the beginning of the excitation pulse. Duration of the electron beam
was 2 ns. The decay time was calculated from the approximation of relaxation curve by an
exponential function. The measurements were performed at room temperature (RT, full
symbols) and at cooling with liquid nitrogen (80 K, empty symbols).
Cooling of the diamond to 80 K increased the intensity of the FE
TO
components two times
with a decrease in the contribution of the short-wave wing of the exciton band, as is the case
in the integrated spectrum in Fig. 6.
The PCL spectra of synthetic diamond sample excited by the electron beam of duration of 10
ns are shown in Figure 10. The spectra were reconstructed from the amplitudes of the light
pulse, measured at 0.5, 1 and 2 ms after the start of the excitation pulse. As we can see, the
band-A with maximum about 450 nm in this time range is dominant in the spectrum of
synthetic diamond; in addition, is registered a weak peak at 415 nm.

63

Fig. 10. PCL spectra of synthetic diamond sample under excitation by electron beam with a
duration 10 ns, shown at 0.5, 1, 2 ms after the beginning of the excitation pulse.
3.5 PCL decay kinetics in natural and synthetic diamonds
The PCL decay kinetics in the diamonds was studied at an electron beam duration of 0.110
ns by the procedure shown in Fig. 1b. The electron accelerator was operated in two modes: a
single-pulse mode at a FWHM of 0.110 ns and a two-pulse mode at a FWHM of 0.65 ns
with a 3-ns interval. Figure 11 shows oscillograms of the electron beam current at a beam
duration of 0.2 and 0.65 ns (two-spike mode). The same figure presents oscillograms of PCL
at 300650 nm (the region of the band-A and N3 system) with no spectral resolution (for
which we removed the monochromator 6 from the circuit shown in Fig. 1b and replaced the
photomultiplier tube 7 by the photodiode for the natural (atop) and synthetic (at the middle)
diamonds.
For the natural diamond, the luminescence duration at 300650 nm was an order of
magnitude longer than that for the synthetic diamond. In the two-pulse excitation mode, the
luminescence duration was also much longer than that in the single-pulse mode (Fig.11).
For the synthetic diamond, the light pulse decay time is nearly the same in the single- and
two-pulse excitation modes, i.e., in PCL excitation by an electron beam of duration 0.250.65
ns, the light pulse duration was constant and was ~ 2 ns (Fig.11, middle).
Figure 9 shows PCL decay spectra of free excitons in the synthetic diamond excited by a
pulse of duration 2 ns at room temperature and at 80 K; the spectra were obtained through
approximation of the light pulse decay time in the single-pulse excitation mode by a first-
order exponential function. At room temperature, the decay time was nearly constant at
233237.5 nm and was 1013 ns.
Cooling to 80 K caused an increase of the PCL decay time of the excitons to 13-14 ns in
spectral range 234-236 nm. Outside this range the PCL decay time decreased to 8-11 ns.

Cathodoluminescence

64

Fig. 11. Oscillograms of the luminescence of natural (atop) and synthetic (middle) diamond
samples under electron beam excitation at single-pulse operation (duration of pulse 0.25 ns,
broken curves) and two-pulse operation (duration of pulse 0.65 ns, continuous curves).
Oscillograms of the beam current duration 0.25 ns (broken curve) and 0.65 ns (continuous
curves) are shown down.
The PCL decay in the natural diamond was studied in the single-pulse mode at three
excitation pulse durations: 0.1, 2, and 10 ns. The PCL decay spectra of the natural diamond
are shown in Fig. 12.
In the nanosecond range, the decay time was 1560 ns. With all excitation pulse durations,
an increase in PCL decay time to 4045 ns was observed in the vicinity of the ZPL at 415.2
nm. In the region of phonon replicas of the ZPL at 415.2 nm (420-480 nm), the PCL decay
time increased to 3545 ns with an excitation pulse duration of 0.1 and 10 ns and to 4560 ns
with an excitation pulse duration of 2 ns.
Thus, in the nanosecond range the N3 system dominates, and this is evidenced by the PCL
spectrum (Fig. 5) and the spectra of decay time (Fig. 12).
The electron beam current densities at a beam duration of 0.1 and 10 ns were
comparable, as can be seen from Table 1. However with an excitation pulse duration of 2
ns, the current density increased several-fold and the PCL decay time in the nanosecond
range also increased compared to the decay times with excitation pulse durations of 0.1
and 10 ns.

65

Fig. 12. The spectra of decay time of PCL of natural diamond sample under excitation by
electron beams of duration 0.1, 2 and 10 ns at room temperature in nano- and millisecond
ranges.

Fig. 13. The spectra of decay time of PCL of natural diamond sample under excitation by
electron beams of duration 2 ns at room temperature (continuous curves) and at cooling
with liquid nitrogen (80 K, broken curves) in nano- and millisecond ranges.

Cathodoluminescence

66
Note that increasing the excitation intensity (the beam current) increases the luminescence
duration of the N3 system, and this is confirmed by the increase in PCL duration in the two-
pulse mode (Fig. 11) compared to that in the single-pulse mode.
In the millisecond range (Fig. 13), the PCL decay time in the natural diamond was 8.59 ms
in the spectral luminescence region of the band-A and decreased steeply beyond this region.
The PCL decay time thus did not depend on the excitation pulse duration. Actually in the
millisecond range, only the luminescence of the band-A was observed in the natural
diamond and that of the N3 system was entirely absent (see Figs. 5 and 12 see).
Cooling of the natural diamond to 80 K increased the decay time 1.8-2 times both for the N3
systems (nanosecond range) and for the band-A (millisecond range).
Figure 14 shows the decay curves of PCL at a wavelength of 235 nm for samples of natural
and synthetic diamond under excitation by electron-beam duration of 10 ns. Rise time of
light pulse is the same for both samples. Decay times are slightly different: about 25 ns for
the synthetic sample and 45 ns for the natural sample.
-0,2
0
0,2
0,4
0,6
0,8
1
1,2
-100 0 100 200 300 400
I [arb.un]
Time [ns]
natural diamond
synthetic diamond

Fig. 14. Decay curve of PCL of natural and synthetic diamond sample at 235 nm under
excitation by electron beams of duration 10 ns at room temperature.
Studies of the decay kinetics of 415 nm (Fig. 15) under excitation of samples by electron
pulse with a duration of 10 ns showed that in this spectral range the decay time of PCL for
synthetic sample is also smaller than for natural sample (16 and 50 ns, respectively) under
identical conditions of excitation. That difference of the decay time may be due to the
presence the channel of nonradiative energy transfer from the excited radiative levels in
crystals of synthetic diamond.
Luminescence decay times of band-A in natural and synthetic diamond samples are also
different (Fig. 16) under excitation by electron pulse of duration 10 ns. PCL decay in the
natural sample has been slower, t is about 50 ns, whereas in the synthetic one is about 10 ns.
Decay kinetics of PCL for emission 235 nm, 415 nm, 450 nm when changing the excitation
density from 6 to 300 mJ/cm
2
, were investigated. It is established that the decay kinetic of
luminescence at 235 and 415 nm has changes weak with increasing density of excitation for
both natural and synthetic diamond samples. The decay time is independent of excitation
density.

67


Investigation of the decay of luminescence at band 450 nm showed that the decay time of
PCL in nanosecond range for synthetic diamond does not depend on the excitation density
(Fig. 17). It was found that the decay time of PCL for natural sample increases from 12 to 50
ns with increasing excitation density. Dependence is shown in Figure 17. This feature is
characteristic for the recombination luminescence.

Cathodoluminescence

68

Fig. 17. The dependence of the decay time of PCL in the band 450 nm from excitation
density for a sample of natural and synthetic diamond. The sample was excited by the
electron beam of duration 10 ns.
4. Conclusion
In the work, we studied the integrated and time-resolved spectral-kinetic characteristics of
pulsed cathodoluminescence (PCL) of natural and synthetic diamonds. The PCL was excited
by electron beams of duration from 0.1 to 10 ns and current density from 10 to 200 /cm
2
.
According to the physical classification of diamonds, the diamonds under study were of
type 2, i.e., their optical transmission spectra revealed a clearly defined fundamental
absorption edge at 225 nm, transparency from the UV to IR range, and absence of resolvable
features.
The PCL of the natural diamond displayed two luminescence bands differing in nature,
spectral composition, and time response. In the nanosecond range, the vibronic N3 system
due to N
3
V defects dominated. The N3 system displayed a ZPL at 415.2 nm and phonon
replicas at 420480 nm. The characteristics decay time for the N3 system was 1565 ns. In the
vicinity of the ZPL at 415.2 nm, an increase in decay time to 45 ns was observed. In the
region of phonon replicas of the ZPL, an increase in decay time to 4065 ns was also found.
Increasing the electron beam current density increased the intensity of the ZPL at 415.2 nm
and its phonon replicas. The N3 system was reliably detected in the integrated and time-
resolved PCL spectra, though it was not resolved in the absorption spectra of this diamond.
In the millisecond range, the N3 system escaped detection and the PCL spectra of the
natural diamond revealed a broad structureless luminescence band-A at 350650 nm with a
decay time of 8.59 ms being dependent neither on the excitation pulse duration, nor on the
electron beam current density.

69
Cooling of the diamond increased the intensity and the decay time two-fold at all excitation
pulse durations and beam current densities. The N3 system thus revealed a double increase
in decay time. The intensity of the ZPL at 415.2 nm and its phonon replicas also increased on
cooling. The higher the current densities, the more considerable the increase in intensity
was.
The synthetic diamond also revealed the luminescence band-A in the PCL spectra.
However, its intensity was too weak to study its time characteristics. The band-A was
structureless in its spectral composition and was in the same spectral range as the band-A in
the natural diamond.
The spectra of the synthetic diamond were dominated by a radiative recombination band of
free excitons with several single-phonon and multiphonon components in which the
dominating component was the FE
TO
component at 235.2 nm. At excitation pulse durations
of 0.12 ns, this component was the main one, and the FE
TA
, FE
LO
, FE
TO+LO
, and FE
TO+LO+LO

components (at 232.8, 236.2, 242.7, and 250.7 nm, respectively) were affected by transient
processes.
The intensity of the exciton FE
TO
component depended on the beam current density
according to the power law with an exponent of ~ 1.4.
Cooling of the synthetic diamond increased the PCL intensity and decay time by a factor of
1.31.6 with a modest increase in band-A intensity. Cooling of the diamond also froze out
the phonon FE
TA
component of exciton luminescence.
It is shown that the decay time of PCL at band-A depends on the excitation density and has
changes from 12 ns (at 6 mJ/cm
2
) to 50 ns (at 300 mJ/cm
2
) for a sample of natural diamond.
5. References
Baksht E. K.; Burachenko A.G. and Tarasenko V.F. (2010) Tech. Phys. Lett., Vol.36, pp. 1020-
1023, ISSN 0320-0116.
Dean P.J., Jones I.H. Recombination radiation from diamond (1964) Physical review, Vol.133,
6A, pp.A1698-A1705.
Fujii A., Takiyama K., Maki R., Fujita T. (2001) Lifetime and quantum efficiency of
luminescence due to indirect excitons in a diamond. Journal of luminescence, 94-95,
pp.355-357. ISSN 0953-4075
Gritsenko B.P.; Lisitsyn V.M., Stepanchuk V.N. (1981) Phys. of the Solid State, Vol.23, pp. 393-
396. ISSN 0367-3294
Kawarada H.; Tsutsumi T.; Hirayama H.; et al. (1994) Appl. Phys. Lett., Vol.64, pp. 451-453.
ISSN 0003-6951.
Korepanov V. I.; Lisitsyn V. M. and. Oleshko V. I. (2000). Russian Physics Journal, Vol.43, pp.
185-192. ISSN 0021-3411.
Korepanov V.I.; Vil'chinskaya S.S.; Lisitsyn V.M.; Kuznetsov M.F. (2005) Optics and Spectr.,
Vol.98, pp. 401-404. ISSN 0030-4034.
Kravchenko V.D.; Lisitsyn V.M.; Yakovlev V.Yu. (1985) Phys. of the Solid State., Vol.27, pp.
2181-2183. ISSN 0367-3294
Marfunin A.S.(1975) Spectroscopy, Luminescence and Radiation Centers in Minerals; Nedra:
Moscow, 327 p. (in Russian).

Cathodoluminescence

70
Lipatov E.I.; Tarasenko V.F.; Orlovskii V.M.; Alekseev S.B. and Rybka D.V. (2005) Tech. Phys.
Lett. Vol.31, pp. 231232. ISSN 0320-0116.
Lipatov E.I., Tarasenko V.F., Orlovskii V.M., Alekseev S.B. (2005) Luminescence of crystals
excited by a KrCl laser and a subnanosecond electron beam. Quantum electronics,
Vol.35, 8, pp. 745-748. ISSN 0368-7147
Lipatov E.I., Orlovskii V.M., Tarasenko V.F., Solomonov V.I. (2007) Comparison of
luminescence spectra of natural spodumene under KrCl laser and e-beam
excitation. Journal of luminescence, Vol.126, 2, pp.817821. ISSN 0953-4075.
Lipatov E.I.; Lisitsyn V.M.; Oleshko V.I.; and Tarasenko V. F. (2007) Russian Phys. J., Vol.50,
pp. 52-57. ISSN 0021-3411.
Lipatov E.I., Avdeev S.M., Tarasenko V.F. (2010) Photoluminescence and optical
transmission of diamond and its imitators. Journal of luminescence, 130, pp.2106-
2112. ISSN 0953-4075
Lisitsyna L.A.; Korepanov V.I.; Lisitsyn V.M. (2002). Phys. of the Solid State. Vol. 44, pp. 2235-
2239. ISSN 0367-3294
Lisitsyna L.A.; Oleshko V.I.; Putintseva S.N.; Lisitsyn V.M. (2008) Optics and Spectr. Vol. 105,
pp. 531-537. ISSN 0030-4034
Lisitsyna L.A.; Korepanov V.I.; Lisitsyn V.M. Eliseev .. at al. (2011) Optics and Spectr.
Vol.110, pp. 568-573. ISSN 0030-4034
Solomonov V. I. and Mikhailov S. G. (2003) Pulsed Cathodoluminescence and Its Application to
Analysis of Condensed Substances; Publisher: UrO RAN, Yekaterinburg, p. 1-181.
Takeuchi D.; Watanabe H.; Yamanaka S.; et al. (2001) Physical review B. Vol. 63, 245328. ISSN
0163-1829
Tarasenko V.F.; Baksht E.K.; Burachenko A.G.; Kostyrya I.D.; Lomaev M.I. and. Rybka D.V.
(2008) Plasma Devises and Operation. Vol.16, pp.267-298. ISSN 1051-9998
Tarasenko V.F.; Baksht E.K.; Burachenko A.G.; Kostyrya I.D.; Lomaev M.I. and Rybka D.V.
(2009) IEEE Trans. of Plasma Science. Vol.37, pp.832-838. ISSN 0093-3813.
Tarasenko V. F.; Baksht E. K.; Burachenko A.G.; Kostyrya I. D., Lomaev M. I. and Rybka D.
V. (2010) IEEE Transactions on Plasma Science. Vol.38, 741-750. ISSN 0093-3813.
Zaitsev A.M. (2001) Optical properties of diamond; Springer:Berlin; 502 p. ISBN 354066582X
Zheltov K.A. (1991) Pikosekundnye silnotochnye elektronnye uskoriteli (Picosecond High-
Current Electron Accelerators). Energoatomizdat. Moscow, Russia, 1991. 114 p.
ISBN: 5-283-03978-1
4
Study of Defects by Cathodoluminescence
Measurements
A. Djemel
1
and R-J. Tarento
2

1
LPCS, Universit. Mentouri Constantine,
2
LPS, Universit. Paris-Sud, Orsay,
1
Algeria
2
France
1. Introduction
Scanning electron microscopy (SEM) can be used to obtain images of a large variety of
materials resulting from secondary electron. The SEM can also be used to detect different
signals that provide composition information , surface morphology and characteristics of the
local electronic structure. The signals provide different processes in the electron-matter
interaction. The combination of different signals produces an detailled physical image and
qualitative and quantitative analysis of the electronic properties of studied samples.
The cathodoluminescence (CL) is one of different signals that has frequently been used
within SEM to study the semiconductor materials. This technique avoids destruction of the
sample ,has a high resolution and an image depth field which are determined by the beam
current , the beam energy and the beam size.
This method has been used to investigate and to identify the particular features of the
crystal defects(dislocation,precipitate,boundaries (Djemel , Castaing et al ,1990 ); Djemel ,
Castaing et al ,1992 ).
The CL method allows the determination of quantitative information on local electronic and
optical parameters of materials such as the diffusion length (L) ,absorption coefficient ( ) , the
dopping levels and the defects parameters such as the recombination velocity (V
s
) ,the defects
density (N
t
) ,the capture cross section ( ) and the energy level (E
t
) associated to defects.
The cathodoluminescence is based on the study of the interaction of the electron beam with
the semiconductor. This interaction gives rise to electron-hole (e-h) pairs generation within
the sample . The density of e-h pairs generated is limited by the scattering process of the
electron beam within the sample. The distribution of the e-h pairs created depends on the
diffusion length and the recombination behaviour at the surface and in the bulk. The
recombination can be either radiative in which case a photon is emitted or non-radiative
generating phonons , Auger electron ....At the surface ,the recombination process is
generally non-radiative. The generated photon submits to an absoption in the escape from
the sample and those that finally emerge can be collected and subsequently detected to
provide the cathodoluminescence signal . The radiative recombination results from band to

Cathodoluminescence

72
band transition and band to energy level transition. The energy level is associated with
impurities or with crystal defects.
2. Quantitative study
The quatitative determination of CL requires an accurate simulation of the CL signal as a
function of the electron beam paramaters , the beam current (I
p
) and energy beam (E
0
).
However, this depends on various stages :
- a better description of the electron-semiconductor interaction
- the diffusion and the distribution of the generated (e-h) pairs within the sample
- the boundary conditions and the recombination at the surface and in the bulk.
2.1 Generation function
Many models can describe the electron-semiconductor interaction : the polynomial form
(Everhoff&Hoff,1971) , the point or spherical models , the modified Gaussian approximation
(Wu & Wittry,1978 ; Hergert , Reck et al,1987 ) and the Monte Carlo method (Phang, Pey et
al ,1992).
For GaAs, the electronsemiconductor interaction is approximeted by the modified
Gaussian function which gives the local generation rate G(z)
G(z) = (/R) (u) (1)
(u) = A exp[- (u-u
0
)
2
/u] Bexp - (bu/u
0
) (2)
where u = z/R , is the density of semiconductor ( g/cm
3
) ,z is the depth ( cm) and R is the
maximum electron range (g/cm
2
).R has been deduced for GaAs as e function of the beam
energy E
0
(keV):
R = 2.56.10
-3
(E
0
/30)
1.7
(3)
For GaAs , u
0
= 0.125 , u = 0.350 , b = 4.0 , B/A = 0.4.The constant A and B are determined
using the normalization condition.( Wu & Wittry,1978 ).
2.2 Diffusion and distribution
Consider a n-type semiconductor doped with concentration N
d
.
The transport of (e-h) pairs generated has been controlled by the continuity equation which
has the following classic form :
divJ = G(z) - r(z) (4)
where J is the flux of the carrier excess, r(z) is the recombination rate in the bulk and at the
semiconductor surface.
The recombination rate in the bulk (neutral region ) is expressed by:
r(z) = p(z)/
p
(5)

Study of Defects by Cathodoluminescence Measurements

73
where p(z) is the excess hole carrier and
p
is the hole lifetime .
In the litterature there are two definitions of the semiconductor surface . In the first
definition, the surface is defined as a dead-layer ( non-radiative region ) with a thickness Z
T

( Hergert , Reck et al 1987 ; Wittry & Kyser 1967 ).
The distribution of minority excess carriers is determined by the continuity equation in the
bulk or neutral region:
D
p)
d
2
p(z)/dz
2
= G(z) p(z)/
p
(6)
In the second definition , the surface is described by a defects density N
t
and an energy level
E
t
in the band gap ( Djemel , Tarento et al , 1998 ).The consequence of electronic surface
states associated to surface defects consists in the existence of a depletion region that is
linked to a potential barrier E
b
between the free semiconductor surface and the bulk.This
second definition of surface is used in cases of dislocation ( Tarento & Marfaing , 1992 ) and
grain boundary ( Oualid , Singal et al , 1984 ).
The distribution of excess carrier is governed by :
- the continuity equations of both excess majority and minority carrier in the depletion
region ( the recombination is neglected in the region r(z)=0 ) ( Djemel , Tarento et al
,1998 ) :
-d
2
n(z)/dz
2
+ 2(z-Z
d
) dn(z)/dz + 2n(z) = G(z)/D
n
(7)
-d
2
p(z)/dz
2
- 2(z-Z
d
) dp(z)/dz - 2p(z) = G(z)/D
p
(8)
Using the Einstein relation (D/ = kT/e), becomes equal to N
d
e
2
/2kT , D and are the
diffusion coefficient and mobility of carriers and is the electric permitivity.
- and by the continuity equation of excess minority carrier in the neutral region (in the
bulk)
D
p
d
2
p(z)/dz
2
= G(z) - p(z)/
p
(9)
2.3 Surface recombination
The recombination rate at the semiconductor surface is treated by two methods.The free
surface of semiconductor is defined as a dead-layer (non-radiative region) with a thickness
Z
T
and a surface recombination velocity V
S
( Hergert , Reck et al , 1987 ; Wittry & Kyser ,
1967 ). In this case V
S
is found from the condition:
D
p
[dp(z)/dz]
z=0
= V
S
p(0) (10)
In the second method ,the free surface of semiconductor is described by defects density N
t

and an energy level E
t
in the band gap with an occupation probability f given by the
Shockley-Read-Hall theory.
f = [n(0) +n
0
+n
i
exp((E
i
E
t
)
/
kT ) ]/[n(0) +n
0
+ p(0) +p
0
+2n
i
cosh(( E
t
E
i
)/kT )] (11)
n
0
, n
i
, p
0
, E
i
are defined in ( Djemel , Nouiri et al ,2002 ).

Cathodoluminescence

74
n(0) , p(0) are the excess carriers at the surface (z=0) of the electron and hole
respectively.
The recombination rate U(0) at the surface (z=0) is given by:
U(0) = C N
t
(n(0) p(0) + p
0
n(0) + n
0
p(0)) / (n
t
+n
0
+ n(0) + p
t
+ p
0
+ p(0)) (12)
Where C = V
th
is the capture coefficient of electron and hole ( is the capture cross section
which is generally linked to both defect and the environment (Bourgoin & Corbett ,1972),
and V
th
is the thermal velocity ) .
n
t
, p
t
, are given in Ref ( Djemel , Nouiri et al , 2004 ).
The relation between V
S
(first definition of surface) and ( N
t
, E
t
) (second definition of
surface) can be established ( Djemel , Nouiri et al , 2002 ).
2.4 Signal of Cathodoluminescence (CL)
The CL signal is calculated from the excess minority carrier in the neutral region (in the
bulk).The photons are generated within the sample when the carriers recombine radiatively.
It is assumed that the CL signal is proportional to the integral of the carriers excess over the
generation volume. The main optical loss mechanism taken into account is the absorption
within the sample. The attenuation of photons propagation towards the surface is given by
an exponential law.
In the case where the suface is considered as a dead-layer with a thickness Z
T
, the CL signal
is given by: ( Hergert , Hildebrandt et al , 1987 ; Hildebrandt , Schreiber et al ,1988 ) .
I
CL
(1/
r
) sin
c
F(
c
)d ;
c
= /cos (13)
F(x) = G
0
exp(-Z
T
) [ (x,Z
T
) - (Lx+S) (1/L , Z
T
)/(1 + S)]/( 1 - x
2
L
2
) (14)
(x , Z
T
) = exp( -z ) G(z + Z
T
)dz (15)
where
c
denote the critical angle of the total reflexion at the surface , G
0
is the total generation
rate, and
r
are the total and radiative lifetimes , respectively , L is the diffusion length , is
the optical absorption coefficient , S is the reduced surface recombination velocity.
In the second definition where the surface is described by defects density N
t
and an energy
level E
t
in the band gap ,the CL signal is written as follow ( Djemel , Kouissa et al , 2008 )
I
CL
[ p(z)/
p
] exp( -
b
z ) dz (16)
p
is the lifetime of excess minority carriers ,
b
is the optical absorption coefficient that
depends on the wavelength , is the efficiency of radiative recombination and written as
(Yacobi & Holt , 1990 ) :
=
nr
/(
r
+
nr
) (17)
r
and
nr
are the lifetimes of radiative and non-radiative recombination, respectively .
p(z) is the minority carriers excess solution of equations (7) , (8) and (9).


75

( ) ( )
( )
( ) ( )
{ }
p d p
p p d p d p
p z B exp - z -Z /L
L /2D G z exp - z-Z /L - exp - z z - 2Z /L dz
(
A = +

(
+ } +
(

(18)
B
p
= p( z = Z
d
) is a constant and represents the excess carrier at the limit between neutral
region and depletion region .
The influence of parameters of electron beam (beam intensity I
p
and energy beam E
0
) , of
parameters of semiconductor ( N
d
, L
p
,
b
) and of parameters of surface ( V
s
, Z
T
) or ( N
t
,
Et, ) on the CL signal has been intensively studied ( Wittry & Kyser , 1967 ; Djemel , Nouiri
et al ; 2002;Ben Nacer,Matoussi et al ,2009).
On the other hand,there are correlations between the theoretical CL signal and CL
measurement to determine the parameters of surface ( V
s
, Z
T
) ( Hergert , Reck et al , 1987 )
or ( N
t
,Et , ) ( Djemel , Nouiri et al , 2000 ) .
To illustrate the CL calculation where the surface is defined by N
t
,E
t
and , one sample
of p-type GaAs is investigated experimentally. The p-type conduction is obtained by
thermal treatment of semi-insulating GaAs (heated at 1019 C, 24 h), the acceptors
concentration Na=10
16
cm
-3
has been measured by Hall effect (Djemel, Castaing et al ,
1992 ).
Two surfaces are compared: one is non-treated and the other is chemically polished with
H
2
SO
4
:H
2
O
2
:H
2
O (9:1:1) . CL measurements have been performed at room temperature. The
primary electron beam intensity is 15 nA, with an electron energy ranging from 5 to 40 KeV.
To understand the influence of E
t
on the CL intensity, we have reported in figure 1, the
variation of the occupation probability ] (equation 11) as a function of E
t
for different excess
carrier at the surface An (0). This figure shows that ] increases when the energy level E
t

moves away from the conduction band E
c
; it means that the number of the occupied centres
is increased, consequently, the surface recombination velocity increases. Therefore, the
change in energy level E
t
explains the surface recombination. On the other hand, the
variation of ] is important when the excess carrier at the surface An(0) increases.
Figure 2 shows the CL curves (equation 16) for different E
t
. The influence of E
t
on the CL
curves is important for the low electron beam energy. The CL intensity increases and the
maximum moves to low acceleration energy when E
t
is near to conduction band (weak
surface recombination). This behaviour is similar to that of the surface recombination
velocity (Hergert, Reck et al, 1987) .
In figure 3, the CL intensity is measured for two surface states. After chemical treatment of
the surface (open circles), the CL intensity increases in the low electron energy region .In the
high electron energy region, the difference of CL signal between both surface states (treated
and non-treated chemically) is reduced...
The adjustment between the numerical results and experimental data has been obtained by
a change in energy level E
t
. Thus, the two surface states can be characterised by a same
defect density N
t
= 2.10
9
cm
-2,
and an energy level E
t
=1.30 eV (Fig.3, 1), and by an energy
level E
t
=1.33 eV (Fig.3, 2). The similar surface states are obtained in p-GaAs deformed and
non-deformed (Djemel, Nouiri et al, 2000).

Cathodoluminescence

76

0,0 0,2 0,4 0,6 0,8 1,0 1,2 1,4 1,6
0,0
0,2
0,4
0,6
0,8
1,0
1: Fermi-Dirac function
2: Occupation probability
Ev Ec
c
b
a
2
1
Ef
i
f
Et (eV)

Fig. 1. The occupation probability ] as a function of E
t
for different excess carrier at the
surface (An(0)=10
12
cm
-3
(a), 10
13
cm
-3
(b), 10
14
cm
-3
(c))

0 10 20 30 40
0,0
0,1
0,2
0,3
0,4
0,5
0,6
0,7
0,8
0,9
1,0
1,1
3
2
1
1: Et=1.2 eV
2: Et=1.3 eV
3: Et=1.4 eV
I

c
l

(
a
.
u
.
)
Eo (KeV)

Fig. 2. CL intensity as a function of electron beam energy for different E
t
.


77

0 1 0 2 0 3 0 4 0
0
1 0
2 0
3 0
4 0
1
2
I
p
= 1 5 n A
1 : s u r f a c e n o n t r e a t e d
2 : s u r f a c e t r e a t e d
C
L

i
n
t
e
n
s
i
t
y

(
a
.
u
)
E o ( K e V )

Fig. 3. CL intensity as a function of electron beam energy E
0
before and after surface
treatment for p-GaAs. Full lines are the results of proposed model (Djemel, Tarento et al,
1998).
3. Electrical activity of defects
The electrical activity of defects can be imaged by cathodoluminescence .It reveals the
regions where the recombination is radiative or non radiative. This electrical activity can
be observed by the measurement of the contrast (C
NR
) which is defined by the relative
difference between the measured CL intensity over the defect and the measured CL
intensity in the region without defect. The contrast is changed by the modification of the
electronic properties of defects which are induced by the external treatment
(hydrogenation, thermal treatment.. (Djemel , Castaing et al ,1990 ;Djemel , Castaing et al
,1992 ).
Figure 4 illustreates the modifications of the thermal treatment on the electrical activity of
defects in GaAs.

Cathodoluminescence

78

Fig. 4. CL observation after a thermal treatment of 15 hours at 850C followed by quenching,
with Si
3
N
4
encapsulation ( Djemel , Castaing et al ,1990 )(x 1000)
Two kinds of contrast are essentially observed : dark contrast or halo contrast such as
illustrated in fig.5 and white contrast. The two kinds of contrast are explained (i) by the
existence of non radiative recombination centres or energy elimination by any other process
( phonons , Auger ...) ( or ones emitting out of the spectral band of detector ) for the dark
contrast and (ii) by the existence of radiative recombination centres or by an increase of
carrier concentration for the white contrast .

Fig. 5. Typical variation of CL Intensity (I
CL
) for a line scan through a defect and variation of
contrast (C
NR
) : a) dark contrast , b) halo contrast. ( Djemel , Castaing et al ,1990 )
Figure 6 illustreates the modifications of the hydrogenation on the electrical activity of
subgrain boundaries in GaAs.


79

Cathodoluminescence

80

Fig. 6. CL migrographs of deformed GaAs covered with a layer of GaAlAs and exposed to a
hydrogen plasma. Observations for various beam accelerating voltages: a) 15 KeV , b) 20
KeV , c) 25 KeV.( Djemel , Castaing et al ,1992)
Deuterium is introduced in deformed GaAs by RF plasma and the Deuterium profil is
determined by secondary ion mass analysis (SIMS).We have checked that the CL emission
originates from the subsurface region containing Deuterium, by changing the range
penetration of electrons. The values are 1.6 m , 2.6 m, 3.7 m for 15 KeV , 20 KeV , and
25 KeV beam voltages respectively .The subgrain boudaries are invisible at the low beam
voltage (Fig.6c) when electron-hole pairs are generated in the Hydrogenated region of the
specimen (1-2 m obtained by the Deuteurium profil SIMS). For high voltages , the
contrast is strong (Fig.6c) because the hydrogen penetration is small compared to the
electrons one.
4. Spectroscopic study
Defects and Complex defects in semiconductors are still attractive topics.They influence
the electronic and optical properties of the materials. Complex defects are formed
between the defects impurities(doping,already present in the starting material) and the
created intrinsic defects (vacancy,interstitial and substitution position ,anti-site).These
defects include nearest- and second-neighbor point-defect pairs.These defects and
complex defects introduce generally the energy levels in the band gap. They can be
radiative or nonradiative. In the case of radiative defects , the cathodoluminescence
(Yacobi,Holt,1986;Lei,Leipner et al ,2004) and the photoluminescence (Sauncy,Palsule et
al,1996 ;William,1968) allow to obtain a quantitative spectroscopy of the energy levels
associated with defects.


81
Figure 7 shows the liquid-He spectra in semi-insulating GaAs. (Yacobi, Holt,1986)

Fig. 7. Liquid-He CL spectra from (a) bright area and (b) an adjacent dark area (Fig.8) in
semi-insulating GaAs (Yacobi, Holt, 1986)
Using the monochromatic CL micrograph, we can determine the spatial distribution of the
defect corresponding to the emission .Fig.8.

Fig. 8. Liquid-He monochromatic CL micrograph of a cell boundary in semi-insulating GaAs
using the 1.494 eV emission due to residual Carbon (Yacobi, Holt, 1986)

Cathodoluminescence

82
5. Impact of electron beam irradiation
The cathodoluminescence is an effective technique for the analysis of the electronic
properties of semiconductors. However, some materials ( GaAs , ZnO , GaN ..) are sensitive
to the irradiation electron beam (Djemel, 1988 ; Dierre,Yuan et al ,2008) .This sensibility is
shown by a decrease of the CL intensity (increase of contrast) or an increase of CL intensity.
The variation of CL intensity depends on the chemical nature of surface (Ga face, As face, Zn
face ,O face ,N face )(Dierre,Yuan et al ,2008) , on the emission band ( energy level in the gap
band) (Dierre,Yuan et al ,2008) and on the external treatment supported by the materials
(Djemel, 1988) .
Figures 9 and 10 illustrate the impact of electron beam irradiation before and after the
treatment by hydrogen (Djemel, 1988).
Different mechanisms are induced by the electron beam. The electron beam irradiation
induces an Electron Stimulated Reaction (Bourgoin, Corbett, 1972).The incident electrons
depose locally a large energy which causes the bond breaking in the materials and the
formation of the reactive sites. These reactive sites can react with the defects from surface or
in the bulk .The electron beam, across the locally deposited energy, can enhance the
migration of mobile defects and impurities. The interaction , between all defects and the
complex defect responsible of luminescence , changes the electronic environment of the last
defect .Thus ,these defects can turn into radiative and nonradioactive recombination centres
.

Fig. 9. Impact of electron beam irradiation observed by CL in GaAs before hydrogenation
(dark dots) (Djemel, 1988)


83

Fig. 10. Impact of electron beam irradiation observed by CL in GaAs after hydrogenation
(white dots ) (Djemel,1988)
Different mechanisms are induced by the electron beam. The electron beam irradiation
induces an Electron Stimulated Reaction (Bourgoin, Corbett, 1972).The incident electrons
depose locally a large energy which causes the bond breaking in the materials and the
formation of the reactive sites. These reactive sites can react with the defects from surface or
in the bulk .The electron beam, across the locally deposited energy, can enhance the
migration of mobile defects and impurities. The interaction , between all defects and the
complex defect responsible of luminescence , changes the electronic environment of the last
defect .Thus ,these defects can turn into radiative and nonradioactive recombination centres.
6. Effect of temperature on CL intensity
Few theoretical and experimental works on the variation of the CL intensity as a function of
the temperature have been realized (Jones, Nag et al, 1973; Lei, Leipner et al, 2004; Djemel,
Kouissa et al ,2009). These studies concern the temperature variation of CL intensity for the
transition conduction band and valence band (Jones, Nag et al, 1973 ). In this work a large
discrepancy exists between the theoretical calculation and the experimntal data. Lei,Leipner
et al ,2004 have studied the temperature variation of the luminescence bands for n-GaAs in
the temperature range 20K-100K (Lei,Leipner et al , 2004). A new theoretical study on the CL
intensity as a function of temperature and taking into account the influence of temperature
on all physical parameters is done by Djemel,Kouissa et al ,2009 . Using this model, an
improvement in the fitting of experimental data(Lei,Leipner et al , 2004 ) is shown in (Fig.11)
allowing an estimation of capture cross section and the determination of the parameter of
the radiative recombination centre.

Cathodoluminescence

84
0 20 40 60 80 100 120
0,0
0,2
0,4
0,6
0,8
1,0
1,2
C
L

I
n
t
e
n
s
i
t
y
Temperature(K)
D Band

Fig. 11. CL Intensity versus temperature for D band( Et=0.5eV) in GaAs:Si (Lei,Leipner et al ,
2004) Numerical results () Experimental data () ( Djemel,Kouissa et al ,2009)
7. Conclusion
In this article we have shown the ability of the cathodoluminescence technique to study
defects. We have demonstrated how this technique provides much data which are
complementary in nature.
The quantitative study allows the knowledge of the interaction electron-matter in all
aspect: energy dissipation of electron beam , generation function , penetration range ,
creation e-h pair energy ,distribution of generated carriers and their concentration ,
comparison between the calculated CL intensity and the measured CL intensity
,determination of physics parameters of defects (defects density, energy levels in the band
gap, the capture cross section ,recombination velocity ) and the parameters of carriers (
diffusion length).
The CL micrograph visualisation gives us a general cartography of the active defects and
non-active defects .The measurements of luminescence band give valuable information on
the energy levels associated to defects and their place in the band gap. The impact of
electron beam irradiation allows to determine the thermal stability of the electronic
properties of defects ,to study the reactivity between defects and can identify the
crystalline face ( impact different on Zn face and O face ).The measurements of CL
intensity as a function of the temperature predict the evolution of many physical
parameters with temperature. The comparison between the numerical results and the
experimental data of CL intensity allows to extract the data on the radiative
recombination centre.


85
8. References
Bourgoin .J.C, Corbett .J.W, A new mechanism for interstitial migration, Phys. Letter A 38 ,
1972, 135-137
Ben Naser.F, Matoussi.A ,Guermazi.S ,Fakhfakh.Z. Cathodoluminescence Investigation of
GaAs thin layers , Physics Procedia 2,2009,827-833
Dierre.B,Yuan.X.L,Yao.Y.Z,Yokoyama.M ,Sekiguchi.T.J. Impact of electron beam irradiation
on the cathodoluminescence intensity for ZnO and GaAs. Mater.Sci:Mater.Electron
,19 ,2008,S307-S310
Djemel.A . These de Doctorat dEtat , Universit Paris XI Orsay ,Dislocation dans GaAs et
Proprits lectroniques par luminescence :Influence des impurets, 1988
Djemel.A,Castaing.J,Visentin.N,Bonnet.M,Influence of thermal treatments on the electronic
activity of dislocations in GaAs observed by cathodoluminescence
Semicond.Sci.Technol.5,1990,1221-1224
Djemel.A,.Castaing.J,Chevallier.J,Henoc.P.Hydrogenation of GaAs covered by GaAlAs and
subgrain boundary passivation,J.Phys.III,France 2,1992,2301-2307
Djemel.A, Tarento.R.J,Castaing.J,Marfaing.Y,Nouiri A, Electronic Surface Properties of GaAs
in CL Experiments ,Phys.Stat.Sol.(a),1998,168,425-432
Djemel.A,Nouiri.A,,Tarento.R.J , Study of suface defects in GaAs by cathodoluminescence:
calculation and experiment J.Phys.Condens.Mater 12,2000,10343-10347
Djemel.A,Nouiri.A,Kouissa.S,Tarento.R.J.Cathodoluminescence Calculation of n-GaAs.
Surface Analysis and comparaison,Phys.Stat.Sol(a),2002,191,N1,223-229
Djemel.A, Nouiri.A,Kouissa.S, Tarento.R.J. Characterisation of n-GaAs by
cathodoluminescence : Quantitative study and comparison,Current Issues on
Multidisciplinary Microscopy Research and Education ,FORMATEX Microscopy
Series N2 Printed in Spain,2004 ,65-71
Djemel.A,Kouissa.S,Tarento.R.J,Temperature dependence of luminescence centre in
cathodoluminescence ,Physics Procedia 2,2009,845-851
Everhart.T.E,Hoff.P.H, Determination of kilovolt electron energy dissipation vs penetration
Distance in solid materials J.Appl.Phys.Vol.42,1,1971,5837-5846
Hergert.W,Reck.P,Passemann.L,Schreiber.J, Cathodoluminescence measurements using
scanning electron microscope for the determination of semiconductors parameters
Phys.Stat.Sol,(a),101,1987,661-618
Hergert.W, Hildebrandt.S, Pasemann.L Theoretical Investigations of Combined EBIC,
LBIC, CL, and PL Experiments. The Information Depth of the PL Signal
Phys.Stat.Sol,(a),102,1987,819-828
Hildebrandt.S, Schreiber.J, Hergert.W,I. Petrov.V.I Determination of the absorption
coefficient and the internal luminescence spectrum of GaAs and GaAs
1x
P
x
(x
=0.375, 0.78) from beam voltage dependent measurements of cathodoluminescence
spectra in the scanning electron microscope. Phys.Stat.Sol,(a),110,1988,283-291
Jones.G.A.C,Nag.B.R,Gopinath.A,.Temperature variation of cathodoluminescence in direct
gap semiconductors,Scanning Electron Microscopy (Part II)Proceeding of the
worshop on electron specimen interaction :Theory for SEM IIT Research Institute
,1973,309-316
Lei.H,Leipner.H.S,Bondarenko.V,Schreiber.J.Identification of the 0.95 eV luminescence band
in n-GaAs:Si, J.Phys.Condens.Matter,16 ,2004,S279-S285

Cathodoluminescence

86
Oualid.T,Singal.C.H,Dingas.J,Crest.J.P,Amzil.H,. Influence of illumination on the grain
boundary recombination velocity in silicon J.Apl.Phys.55,1984,1195-1205
Phang.J.C.H,Pey.K.L,Chan.D.S.H. A simulation model for cathodoluminescence in the
scanning electron microscope IEEE Trans.Electron.Devices,39,1992,782-791
Sauncy.T, Palsule.C.P, Holts.H, Gangopadhyay.S,. Lifetime studies of self-actived
photoluminescence in heavily silicon-doped GaAs , Phys.Rev.B53,4,1996,1900-1906
Tarento.R.J,Marfaing.Y, Analysis of the recombination velocity and the electron beam
induced current and cathodoluminescence contrasts at a dislocation,
J.Appl.Phys.71,1992,4997-5003
William.E.W.Evidence for self-actived luminescence in GaAs:The Gallium Vacancy-Donor
Centre ,Phys.Rev.168,3,1968,922-928
Wittry.D.B,Kyser.D.K. Measurement of diffusion lengths in direct Gap semiconductors by
electron beam excitation ,J.Appl.Phys.Vol.38,1,1967,375-382
Wu.C.J,Wittry.D.B, Investigation of minority-carrier diffusion lengths by electron
Bombardment of Schottky barriers.J.Appl.Phys.49,1978,2827-2836
Yacobi.B.G,Holt.D.B,Cathodoluminescence Microscopy of Inorganic Solids Plenium Press,
1990
Yacobi.B.G,Holt.D.B,Cathodoluminescence Scanning Electron Microscopy of semiconductors,
J. Appl.Phys.59,4,,1986, R1-R24
Part 2
Application to Semiconductors

5
Cathodoluminescence Properties of
Zno Thin Films
M. Addou et al.
*

University of Ibn Tofail
Morocco
1. Introduction
The industry has a great need for high performance materials to feature well defined. These
needs have prompted the development of methods of study and control of gradually more
sophisticated based on the radiation-matter interaction. To identify the properties of
materials, we must make a spectral analysis on the emitted photon using a spectrometer
combined with a detection system. The analysis of the photon can take place directly in the
form of an electrical current, as well as the spectrum received by the measurement system is
represented by a function I ().
The best radiations are the electrons because they are easy to produce, accelerate and to
focus. In this context we speak of cathodoluminescence, which was long used in devices
including fluorescent screens. This phenomenon, resulting from the excitation of
luminescent materials by electron beam, is leading to photon emission, which is subject to
the laws of the transitions. We usually distinguish the intrinsic luminescence (transition
band to band, free exciton) and extrinsic luminescence from impurities and defects. The
most common application is the television screen (when it is a cathode ray tube). In geology,
the cathodeluminescence microscope is used to examine internal structures of geological
samples in order to determine, for example, the history of rock formation. Another
important application is Cathodoluminescence in image mode. In this mode the
cathodoluminescence can view the spatial distribution of responsible levels for a radiative
transition observed or locate the non radiative defects such as dislocations, grain boundaries
and precipitates. These defects give rise to a generally strong contrast. It is possible to
determine the energy levels that are at the root of this contrast. So we used this technique to
study the nature and distribution of defects and impurities in materials [1].
Semiconductors exhibit energy gap between the valence band and the conduction band, the
order of the electron volt. An incident electron beam of sufficient energy can move electrons
from the valence band to the conduction band, thus producing electron-hole pairs; return to
the ground state can be achieved through nonradiative transitions or by radiative transitions
with photon emission of light. This return is via two mechanisms (direct recombination and

*
J. Eboth, A. El Hichou, A. Bougrine, J.L. Bubendorff, M. Troyon, Z. Sofiani, M. EL Jouad,
K. Bahedi and M. Lamrani
University of Ibn Tofail, Morocco

Cathodoluminescence

90
recombination indirect). The CL signal is formed by detecting the photons in the UV, visible
and IR.

Fig. 1. Cathodoluminescence schema
1. The region of high energy loss (DE> 50eV) corresponding to the excitation of deep
levels K, L, creates the X and Auger radiations.
2. The region of low energy loss (DE <50eV) corresponds to the excitation of the valence
and conduction (photon CL).
The most important mechanisms in the analysis of CL in semi-conductors are:
- Electron-solid interaction.
- Dissipation of energy.
- Generation of carriers (e-h).
From the experimental point of view, a study of CL can be simplified to an electron gun, a
sample and a detector.
Many parameters are involved and contribute to enhance the cathodoluminecence intensity
during the growth process. Temperature effect, doping effect, flow rate deposition effect and
the excitation energy effect are strongly among many other parameters, which influence the
optical properties of thin films. The investigation on their luminescence remains very
limited, especially the cathodoluminescence. However, band-edge photoluminescence in
polycrystalline ZnO has not received as much attention as the bulk material because the
band-edge photoluminescence is usually weaker and lacks the fine structure of the bulk
crystals. Nevertheless, a study of the photoluminescence structure of this material is
interesting, because it can provide valuable information on the quality and purity of the
material. Polycrystalline films may also have different photoluminescence mechanisms as
compared to the bulk material.

Cathodoluminescence Properties of Zno Thin Films

91
A
II
B
VI
compounds, such as transparent and conducting oxides films, have been
attracting ample attention as starting material for electro-luminescent devices because of
their high visible transmittance and low electrical resistivity in the visible region. Zinc
oxide (ZnO) is one of the few metal oxides, which can be used as a transparent
conducting material. It has some advantages over other possible materials such as In
2
O
3
Sn, CdSnO
4
, or SnO
2
, due to its unique combination of interesting properties: non-
toxicity, good electrical, optical and piezoelectric behaviour, high stability in a hydrogen
plasma atmosphere and its low price [2,3]. So, ZnO is a good candidate to substitute
indium tin oxide (In
2
O
3
:Sn) and tin oxide (SnO
2
), in conductive electrodes of amorphous
silicon solar cells.
Furthermore, ZnO has the same crystal structure as GaN: wurtzite crystal structure and
direct wide band gap. It is closely lattice matched to GaN. ZnO therefore offers potential as a
substrate material on which high quality GaN may be grown. The room temperature band
gap of ZnO is 3.3 eV with emission in ultra-violet (UV) region. An outstanding feature of
ZnO is its large excitonic binding energy of 60 meV leading to the existence and extreme
stability of excitons at room temperature and/or even higher temperatures [4-7]. These
characteristics have generated a wide series of applications for example, as gas sensors [8],
surface acoustic devices [9], transparent electrodes [10] and solar cells [3] among others. The
preparation of zinc oxide thin films has been the subject of continuous research. Many
techniques are used for preparing this transparent conductive ZnO such as: RF sputtering
[11], evaporation [12], chemical vapour deposition [13], ion beam sputtering [14] and spray
pyrolysis [1518]. Among these methods spray pyrolysis has attracted considerable
attention due to its simplicity and large scale with low-cost fabrication. Additionally, by
using this technique one can produce large area coatings without the need of ultra high
vacuum.
Different process parameters of the intrinsic and doped ZnO films by spray technique have
been published [1922]. The investigation on their luminescence remains very limited,
especially the cathodoluminescence. However, band-edge photoluminescence in
polycrystalline ZnO has not received as much attention as the bulk material because the
band-edge photoluminescence is usually weaker and lacks the fine structure of the bulk
crystals [23,24]. Nevertheless, a study of the photoluminescence structure of this material is
interesting, because it can provide valuable information on the quality and purity of the
material [25]. Polycrystalline films may also have different photoluminescence mechanisms
as compared to the bulk material.
This Chapter will be divided on three parts. The first one describe the effects of different
process parameters such as substrate temperature, air flow rate and precursors of undoped
ZnO films prepared by spray pyrolysis on glass substrate. Several cathodoluminescence
bands and the corresponding emission processes are identified. It is also shown that SP is an
adapted technique to achieve ZnO films with a quality comparable with that of transparent
conducting oxide thin films prepared by other techniques.
The second one concerns some results of a systematic investigation of the crystallinity, the
surface morphology, optical properties and the cathodoluminescence measurements of the
undoped and Sn-doped ZnO thin films deposited by SP. Several CL bands and the
corresponding emission process are identified.

Cathodoluminescence

92
The last section of this chapter is dealing with, the luminescence of the investigated ZnO
samples is examined accounting for the effects of two main parameters: the Er dopant
concentration and the incident electron beam energy. A general sight of the CL material
behavior is first reported, which precedes a more refined study including the spectral peaks
integration and the near-field imaging aspects.
2. Effect of deposition temperature (Ts), air flow rate (f) and precursors on
cathodoluminescence properties of ZnO thin films
Earlier work indicated that ZnO exhibited three bands of luminescence are centered around
382, 510 and 640 nm, labeled near UV, blue-green and red bands, respectively [2628]. Our
undoped ZnO films displayed the same three cathodoluminescence bands. However, the
relative intensities and the disappearance of the blue-green and red emissions depend
strongly on the precursor and deposition conditions such as flow rate and/or substrate
temperature and doping nature.
2.1 Variation of substrate temperature
Fig. 2 shows the cathodoluminescence of ZnO thin films, at E=5 keV electron beam energy
with a beam current of about 1 nA, deposited at different substrate temperatures and
different flow rates. The substrate temperatures (Ts) were varied from 350 to 500Cwith an
interval of 50C, and with a flow off=5 ml/min. When the substrate temperature increases,
the surface of our films is entirely covered by grains and condensed.

Fig. 2. CL Spectra of ZnO sprayed at flow rate f=5ml/min deposited on different
temperature: (a) : T= 350C, (b): T= 400C, (c) : T= 450C, and (d) : T= 500C.


93
As indicated, the luminescence intensity depends strongly on the deposition temperature.
Extinction of the blue-green emission (centered around 510 nm) is observed at substrate
temperature of 350 and 400C, whereas the near UV emission at 382 nm becomes more
dominant than other transitions (blue-green and red emissions) at 450C. The blue-green
emission (510 nm) appears above substrate temperature 450C but the red emission (640
nm) appears at different substrate temperature. At Ts=500C, the UV transition shifts to
higher wavelength and becomes comparable in cathodoluminescence intensity with blue-
green emission. The maximum value of cathodoluminescence intensity for three bands is
obtained at T=450C.

Fig. 3. XRD of ZnO sprayed at flow rate f=5ml/min deposited on different temperature: (a) :
T= 350C, (b): T= 400C, (c) : T= 450C, and (d) : T= 500C.

Cathodoluminescence

94
Before discussing the cathodoluminescence observations of the films in more detail, it is
useful to consider their crystal structure and morphology. The XRD of ZnO films,
deposited at different substrate temperatures and at flow rate 5 ml/min, indicated that
they possess a hexagonal close packed structure (Fig. 3). Moreover, it is the only main
peak obtained with all films indicating that the increasing of substrate temperature does
not change the preferred textural growth orientation. The [0 0 2] direction corresponds to
the c-axis of the crystal lattice that its normal to the deposition substrate plane. The XRD
intensity depends strongly on the deposition temperature. The narrow range of
deposition temperature permitted for maximum XRD intensity is illustrated by the
pronounced peak at Ts=450C.
The mean crystallite size D was calculated from the (0 0 2) diffraction peak using
Scherrers formula [29]. Values of D are listed in Table 1 for films prepared under various
substrate temperatures. The mean crystal diameter was around 30 nm for the sample
deposited at Ts=450 C which was larger value as compared to all samples. We conclude
that the films, which present strong cathodoluminescence intensity in UV emission, have
a strong XRD intensity and large grain size, therefore a good crystallinity. We confirmed
our results by using SEM images. Fig.4 clearly shows that there is a change in the surface
morphology of ZnO films due to a change in the substrate temperature. It is evident that
the porous structure occurred throughout the films deposited at Ts=350 C (Fig. 4(a)). All
films deposited at Ts>350 C attain a microstructure and had a close-packed morphology.
The films deposited at Ts=450 C consists of hexagonal-like grains of approximately 300
nm size. The mean crystal diameters obtained using Scherrers formula are all case
substantially smaller than the dimension of grains observed by SEM image, indicating
these grains are probably an aggregation of crystallites. The enhancement of
cathodoluminescence intensity at substrate temperature 450C could be due to a large
grain size therefore a better crystallinity. The best luminescence was achieved with
samples grown at T=450C.

Table 1. Grain size values at different substrate temperatures calculated using Sherrers
formula (f=5 ml/min)


95

Fig. 4. Morphology of ZnO sprayed at flow rate f=5ml/min deposited on different
temperature: (a): T=350C, (b): T= 400C, (c) : T= 450C, and (d) : T= 500C.
2.2 Variation of airflow rate
The structural and luminescence properties of ZnO thin films are also investigated by
varying the flow rate in the region 2.5pfp7.5 ml/min, which here is equivalent to the
varying growth rate range of 0.2pro1 mm/s. This formation occurs on a deposited substrate
made of soda glass whose temperature is fixed at Ts=450C. Note that no cracks are
observed on large scan area for all samples (SEM images Fig. 6). The films are continuous
and in fact consist of grains. The cathodoluminescence spectra of these films (E=5 keV
electron beam energy with a beam current of about 1 nA) at different airflow rate and at
Ts=450C are shown in Fig.5. For f=2.5 ml/min, one large band centered at 400 nm have
been observed. When the spray rate enhanced, three emissions have been appeared with a
dominance of the blue-green emission (510 nm) for f=3.75 ml/min and a strong intensity of
the UV transition (382 nm) for f=5 ml/min. Furthermore, the appearance of the UV emission
corresponds to the improved crystal quality. At high flow rate f46.25 ml/min, a degradation
of cristallinity leads to decrease of the intensity related to UV transition and disappearance
of both emissions situated in blue-green and red ranges.

Cathodoluminescence

96

Fig. 5. CL Spectra of ZnO deposited at 450C for different flow rate : (a) : f= 2,5ml/min, (b) :
f= 3.75ml/min, (c) : f= 5ml/min, (d) : f= 6.25ml/min and (e) : f= 7.5ml/min
In Table 2, we have reported the mean grain size of ZnO films at different flow rate
(statistical analysis of more than 100 grains on SEM images, Fig. 6). The some behavior of
our samples at different substrate temperatures has been observed. Indeed, the low
cathodoluminescence intensity of the UV transition is observed for films indicate small grain
sizes and porous structure. We can conclude, that good crystallinity and the best
luminescence are achieved with samples grown at f=5 ml/min. The decrease in surface
grain size with the flow rate beyond f=5 ml/min appears to be mainly the effect of the non-
incorporation of Zn particles on the material surface. The optimum flow rate and substrate
temperature are f=5 ml/ min and Ts=450 C, respectively. The films produced exhibit
optical characteristics comparable to films grown by more sophisticated techniques.

Table 2. Grain size calculated from SEM images at different air flow rates (T=450 C)


97

Fig. 6. SEM micrographs of ZnO films sprayed at substrate temperature Ts=450 C and
different air flow rate: (a) f=2.5ml/min, (b) f=5 ml/min, (c) f=7.5 ml/min and substrate
temperature: Ts=450 C.
2.3 Cathodoluminescence properties and variation of precursor
Cathodoluminescence spectra obtained from samples deposited at optimum condition
Ts=450 C and f=5ml/min is shown in Fig. 2(c). The three main emissions had peaks at 382,
520 and 672 nm. The luminescent at 382 nm corresponds to the band gap transition of ZnO.
The presence of this band is an indicator of the good crystallinity.
The second peak centered at 520 nm is the characteristic of blue-green emission, which is also
typical for ZnO material. Despite the numerous reports on the photoluminescence and
cathodoluminescence of undoped ZnO, the luminescent centre responsible for this emission is
not yet clearly identified. Several assumptions are proposed. Dingle [30] ascribes this emission
peak to a substitution of Zn
2+
by Cu
2+
in the crystal lattice. According to Vanheusden et al. [31]
and Egelhaaf et al. [23] this peak corresponds to a defect-related luminescence (deep-level
luminescence), due to the oxygen vacancies in ZnO. This defect-related luminescence is
explained by radiative transitions between shallow donors (oxygen vacancies) and deep
acceptors (Zn vacancies). In this case, the acceptor level is located 2.5 eV below the conduction
band edge [23,32], while the donor level is known as a shallow level at 0.050.19 eV, leading to
an emission band centered around 508540 nm. Furthermore, Minami et al. [33] proposed that
the blue-green emission in this material might be associated to a transition within a self-
activated centre formed by a double-ionized zinc vacancy (V
Zn
)
-2
and a single-ionized
interstitial Zn
+
at the one and /or two nearest-neighbor interstitial sites. Nevertheless, the
composition studies of our films by XPS and EDX analysis do not reveal presence of copper
and thus the first explanation must be rejected. Therefore, one of the two other hypotheses
may explain the blue-green emission that we observe.

Cathodoluminescence

98

Fig. 7. Cathodoluminescence spectra of ZnO for both precursors (Ts=450 C and ow rate
f=5 ml/min): (a) ZnCl
2
and (b) AcZn

Fig. 8. SEM micrographs of ZnO at different precursors (Ts=450 C and flow rate f=5
ml/min: (a) ZnCl2 and (b) AcZn.


99
Previous works have also attributed the blue-green emission to a luminescent centre formed
by the association of a doubly ionized zinc acceptor vacancy defect and of halogen
impurities such as Cl atom of precursor. Fig.7 shows the cathodoluminescence spectra of
undoped ZnO thin films preparing using zinc chloride (ZnCl
2
) and zinc acetate (AcZn) as
precursors. The presence of blue-green emission for both precursors is not attributed to the
presence of chlorine atom, but confirms the attribution of blue-green emission to a self-
activated centre. Furthermore, the SEM image of both samples shows there is a change in
the surface morphology due to the presence of Ac or Cl atoms (Fig.8). The films prepared
from precursor ZnCl
2
presents a better cathodoluminescence intensity, which confirms our
choice for precursor, and the good cristallinity related to a dense structure and crystallites
size relatively important. Cathodoluminescence intensity is directly correlated to an
improvement of the films crystallinity.
The last peak at 670 nm, which appears markedly broad for this low beam energy (5 keV),
may be due to the defect related deep emission, as shown by photoluminescence [34,35].
Lohnert et al. [36] pointed out that the intensity of broad luminescence observed from about
450 to 700 nm depends on the oxygen partial pressure in the sintering process of ZnO
ceramics. This also suggests the hypothesis of a transition within a V
o
x
neutral and single-
ionized oxygen V
o
-
.
2.4 Summary
The dependence of cathodoluminescence properties of spray-deposited ZnO films on
process parameters such as substrate temperature, airflow rate and precursors, has been
studied in detail. We have found that the luminescence intensity depend strongly on process
parameters. The optimum values of substrate temperature and airflow rate have been
determined to give the best luminescent ZnO films.
In the process of this investigation, it has been observed that the films are polycrystalline
and have a preferred orientation in the [0 0 2] direction. The best films in terms of
cathodoluminescence consisted of close-packed grains. These grains were an aggregation of
many crystallites having a mean diameter around 30 nm.
The cathodoluminescence analysis of the sample deposited at optimum conditions (Ts=450
C; f=5 ml/min) shows three bands: a near ultra-violet emission at 382 nm, a blue-green
emission at 520 nm and a weak red one at 672nm.
3. Cathodoluminescence of undoped and tin-doped ZnO thin films
Cathodoluminescence (CL) spectroscopy, X-ray diffraction and spectrophotometery have
been studied of undoped and tin (Sn)-doped ZnO films prepared by spray pyrolysis (SP)
technique of zinc chloride and tin chloride. The luminescence films had a polycrystalline
hexagonal wurtzite type structure. At room temperature, the cathodoluminescence (CL)
spectra of the undoped and doped ZnO films exhibit the common near ultra-violet (UV)
band-gap peak at = 382 nm but they differ as regards their visible emissions. The undoped
ZnO emits an intensive bluegreen light ( = 520 nm) and a red emission ( = 672 nm). The
presence of tin gives rise to a new light emission corresponding to = 463 nm and the
extension of bluegreen light typical of the intrinsic behavior of the material. CL imaging of

Cathodoluminescence

100
undoped ZnO films show that the luminescence is located at defined sites giving rise to a
grain-like structure inherent to the surface morphology. The presence of tin in the material
leads to great luminescent spots, attributed to large grain sizes.
3.1 Experimental procedure
The luminescent films were prepared by spray pyrolysis using air as the carrier gas
atmospheric pressure. The apparatus used for deposition is described and schematized
elsewhere [37/21]. The spraying solution was prepared from a mixture of 0.05 M zinc
chloride (ZnCl
2
) and deionised water. A small amount of chloridric acid HCl was added in
the solution to prevent the formation of zinc hydroxide (Zn(OH)
2
). The nozzle is directed
towards a substrate with a spraying ow rate of 5 ml/min. This substrate is made of bare
glass sheets whose temperature is maintained constant at T = 450C, that is known to be the
optimal condition for this material formation. The same film thickness of 0.4 mm, estimated
from the deposition time, is retained. A good agreement is obtained with the value deduced
from the cross sectional images obtained by transmission electron microscopy. Tin doping
was achieved by adding SnCl
2
to spraying in a concentration of 6 at.%. The compositionnal
analysis of the films as performed with an EDAX 9100 analyzer indicates that the 6% tin
percentage of the spraying bath is quite recovered in the Sn-doped ZnO samples.
The crystalline structure was investigated by X-ray diffraction using CuK radiation. Optical
transmittance measurements were performed with a Shimadzu 3101 PC UV-VIS-NIR
spectrophotometer. The cathodoluminescent (CL) spectra and images were measured at
room temperature, using a LEO-GEMINI 982 scanning electron microscopy (SEM), operated
from 1 to 30 kV and equipped with a field emission gun. Our CL system, in association with
a spectrometer (Triax 190 from JOBIN YVON), allows monochromatic CL imaging as well as
acquisition of CL spectra on very localized spots of a sample. The mechanical details of our
homemade SFM microscope, implemented inside the SEM, are reported in [37]. In this
paper, the photons enter directly a multimode optical fiber (numerical aperture = 0.48)
leading to far-field images or spectra. The CL images are acquired with a fixed electron
beam, the sample being scanned under the fiber. The spectrometer has a spectral resolution
of 0.5 nm and is equipped with an N2-coolled CCD detector. Monochromatic CL images can
be obtained by selecting a defined wavelength and by sending the light to a photomultiplier
(PM, R4220P from Hamamatsu). The signal is then amplified with an electrometer and
images of 256 256 pixels at a line frequency of about 0.1 Hz are generated. Panchromatic
images are also obtained by directing the light towards the PM without passing through the
spectrometer.
3.2 Results and discussion
X-ray diffraction patterns for undoped and Sn-doped ZnO thin films prepared at optimum
parameters are shown in Fig.9. The diffraction (0 0 2) peaks located at the angular position
2 = 34.4 (undoped ZnO) and 2 = 34.38 (Sn-doped ZnO) are characteristics of the
hexagonal close packed structure of this material. Moreover, it is the only main peak
obtained with both films indicating that the presence of tin in ZnO keeps unchanged the
preferred textural growth orientation. The (0 0 2) direction corresponds to the crystal lattice
that is normal to the deposition substrate plane. It should be pointed out that the peaks


101
intensity of the Sn-doped ZnO films in Fig.9 (b) is higher that the one of the undoped
sample of Fig.9 (a), it is full width at half maximum (FWHM) being broaded. One deduced
that the doped sample has a better crystallinity. Probably, there is the substitution of oxygen
by tin anions in the material sites, leading to a partial healing of the crystal network.

Fig. 9. X-ray diffractograms of: (a) undoped and (b) Sn-doped ZnO lms.
Fig.10 consists of SEM showing the effect of doping on the surface morphology. We can see
that the morphology evolves considerably after doping our samples. The undoped film in
Fig.10 (a), depicts a microstructure consisted of hexagonal-like grain of approximately 200
nm size, the substrate surface being entirely covered for the film thickness retained. The last
remark remains valid for the Sn-doped sample image of Fig.10 (b). Where as Sn-doped film
presents a plate-like grains with polycristalline-like irregular grains. The size of the grains is
larger (350 nm) and not uniform. Therefore, the crystallinity of Sn-doped ZnO films was
significantly improved with Sn-doping. This is in a good agreement with X-ray diffraction
results.

Cathodoluminescence

102

Fig. 10. SEM micrographs of: (a) undoped and (b) Sn-doped ZnO lms.

Fig. 11. Optical transmission spectra of: (a) undoped and (b) Sn-dopedZnO films.
The optical properties of undoped and Sn-doped films are restricted to their transmission
spectra (Fig.11). The films were highly transparent above 85% in the visible wavelength
range. For wavelengths in the absorption edge region, it is observed that it is shifted
towards lower wavelengths. This effect has been already observed and discussed by other
authors [38], significant an increasing in band-gap value. The results from the previous
figures show that the ZnO film with good crystallinity and purity can be obtained by SP.


103

Fig. 12. Cathodoluminescence spectra of: (a) undoped ZnO and (b) Sn-doped ZnO lms at
E=5 keV electron beam energy with a beam current of about 1 nA.
Cathodoluminescence studies exhibit interesting features. In Figure 12, we present the
cathodoluminescence emission spectra of undoped (Fig.12 (a) and Sn-doped ZnO thin film
(Fig.12 (b)) at the electron beam energy of 5 keV. Early work indicated that ZnO exhibited
three photoluminescence bands centered around 380, 520, and 650 nm, labeled near-
ultraviolet (UV), bluegreen and red bands [39,27,28]. Our undoped ZnO films displayed
the same three cathodoluminescence emission, showing in Fig.12 (a). The first peak localized
at = 382 nm (E = 3.25 eV) corresponds to the band-gap transition of this material. It is
clear that the appearance of the band-edge correspond to the good quality and crystallinity
of our films and confirms the previous X-ray diffraction observations. The second peak of
the undoped sample is the bluegreen emission located at = 520 nm (E = 2.38 eV) that is
here typical of ZnO material. The luminescent center responsible for this emission is not
clearly identified. It is ascribed by Dingle [30] to a substitution of Zn
2+
by Cu
2+
in the crystal
lattice. Recently, it has been proposed that the blue green emission in this material might be
associated to a transition within a self-activated center formed by a double-ionized zinc
vacancy V
zn
2
and the single-ionized interstitial Zn
+
at the one and/or two nearest-neighbor
interstitial sites [39,33]. According to Vanheusden et al. [31], the greenyellow luminescence
(deep-level luminescence) is related to the oxygen vacancies in the grown ZnO crystals.

Cathodoluminescence

104
Egelaaf and Oelkrug [45] reported that those defect-related luminescence are caused by
radiative transitions between shallow donors (related to oxygen vacancies) and deep
acceptors (Zn vacancies). The acceptor level (Zn vacancies) is located 2.5 eV below the
conduction band-edge [40,32], while the donor level is known as shallow as 0.050.19 eV. In
our undoped ZnO samples, no copper impurities were detected using electron probe
microanalysis (EPMA) and still we observe the bluegreen emission. Therefore, it is likely
that the bluegreen emission observed in our samples might be better explained by the
second hypothesis. The third peak at = 672 nm (E = 1.84 eV), which appears markedly
broad for this low beam energy, may due to the defect related deep emissions. Lohnert and
Kubalek [41] pointed out that the intensity of broad luminescence observed from about 450
to 700 nm depend on the oxygen partial pressure in the sintering process of ZnO ceramics.
This also suggests the hypothesis of a transition within Vo x neutral and single-ionized
oxygen V
o
.
The emission spectrum for the Sn-doped films, using electron beam energy of 5 keV, is a
broad band composed of two peaks at 372 and 672 nm (Fig.12 (b)). The incorporation of Sn
apparently results in a competitive phenomenon that overshadows the bluegreen emission.
This behavior could be originated by a more favorable transmission from the localized state
generated by the Sn or could be due to a reduction of the self-activated centers responsible
for the bluegreen emission by occupation of the Zn vacancies by Sn ions. This figure
confirms the shift of absorption edge observed by optical results.

Fig. 13. Cathodoluminescence emission of Sn-doped ZnO lms at different electron beam
energies: (a) E = 7 keV, (b) E = 10 keV and (c) E = 15 keV.


105

Fig. 14. Cathodoluminescence images of undoped ZnO thin film taken at E = 3 keV: (a)
polychromatic CL image, (b) monochromatic CL image taken at = 382 8 nm, (c)
monochromatic CL image taken at = 520 8 nm. The full arrow in each image is a marker
for the investigated region of the film surface area.
Fig.13 represents the CL study of the Sn-doped sample at different electron beam energies (E
= 7, 10, and 15 keV) proposed for investigating the material homogeneity versus the
penetration depth of the electrons. It is clearly appeared that the predominant emission
depends strongly by excitation. Indeed, at low excitation alone the tow peaks situated at 372
and 672 nm are observed. The peak localized at 372 nm is due to a transition band-gap with
an increasing in band-gap value witch is due to tin doping. From 10 keV a new peak
centered at 465 (E = 2.66 eV) began to appear. This emission, that seems typical of tin
element, may be associated with a complex luminescent center like (VZn SnZn ). At 15
keV, the peak at 465 nm becomes comparable in intensity with the peak situated at 672 nm.
Therefore, the increase of this peaks intensity with E means that the tin concentration is
higher in the bulk. This result has been verified by X-ray photoelectron spectroscopy (XPS)
measurements after successive etching treatment of the sample surface. A similar
investigation of the undoped ZnO film has shown that the intensity of the three peaks
proportionally increases with the increase of E, implying that the SP deposition method can
lead to the formation of homogeneous samples.

Cathodoluminescence

106
Figs. 14 and 15 show polychromatic and monochromatic images of the undoped and Sn-
doped ZnO films obtained at the exciting beam energy E = 3 and 5 keV, respectively. The
monochromatic images were recorded by selecting the appropriate wavelength with a
spectral bandwidth of 16 nm. The CL images of undoped ZnO films (Fig.14) show that the
luminescence is located at defined sites giving rise to a grain-like structure inherent to the
sample morphology. In view of the relatively low exciting beam energy E = 3 keV, the
lateral resolution of the CL images is noticeable. It is smaller than the mean grain size and
approximates 100 nm, indicating that the diffusion length of the undoped ZnO film is of this
order of magnitude or even smaller. Both monochromatic images at = 382 and 520 nm are
not much different indicating that the luminescent centers responsible for the band-edge
and the bluegreen emissions are approximately located at the same places. Nevertheless, in
Fig. 6(b), the monochromatic image corresponding to the band-edge emission is presented.
A higher density of emitting sites, with smaller luminescent spots than those of Fig. 6(c).

Fig. 15. Cathodoluminescence image of F-doped ZnO thin film taken at 4 keV: (a)
polychromatic CL image, (b) monochromatic CL image taken at =3728 nm. The full arrow
in each image is a marker for the investigated region of the film surface area.
The CL images of Sn-doped ZnO films (Fig.15) show that the emission is less uniform,
appearing more as a granular structure, with a greater luminescent spots than that of the
intrinsic ZnO. The grain-like structure cannot be resolved by CL due to the finite carrier-
diffusion length. Such a modulation of emission intensity would be attributed to areas with
varying grain density, in which the bright regions corresponding to dense grains. Only a
few luminescent defects, indicated by arrows on the images, are present. Notes that it is very
difficult to take the monochromatic image at = 465 nm, because this emission appears at
higher exciting beam energy.


107
In order to study, the effect of Sn doping on the properties of ZnO, we have plotted in
Fig.16, the luminescent evolution versus different level doping at a single electron beam
energy E = 5 keV. We have remarked a disappearance of bleu green emission observed in
undoped ZnO even if level doping and peak intensity becomes more important when level
doping increases.

Fig. 16. Cathodoluminescence emission of Sn-doped ZnO films at different level doping: (a)
2 at.%, (b) 4 at.% and (c) 6 at.%.
3.3 Conclusion
Structural, optical and cathodoluminescence characteristics of sprayed ZnO films have been
studied here. The effects of tin dopant at the atomic percentage of 6% on these film
properties have been analyzed. The hexagonal wurtzite structure of the material is not
modified by the presence of the Sn-dopant. The preferred (0 0 2) growth orientation of the
films is not affected while an improvement of the material crystallinity is observed with this
dopant. The cathodoluminescence analysis of the samples shows that the undoped film
presents three bands: a near UV emission at = 382 nm in the near UV region, a bluegreen
emission at = 520 nm and a weak red one at = 672 nm. Incorporation of tin extinguishes
the bluegreen band while appears a blue light at = 465 nm and increases the value of the
band-gap transition. CL imaging analysis shows that the repartition of the emitting centers
in the material is intimately connected to the film morphology. The presence of tin in the
material leads to great luminescent spots, due to large grain sizes.
4. Cathodoluminescence of undoped and Erbium-doped ZnO thin films
4.1 Introduction
The present work proposes an examination of erbium species influence on the visible and
near UV luminescence of thermally sprayed ZnO-Er films. As shown elsewhere, the
conditions used for this large scale and cheap deposition technique greatly affect the
emitting characteristics of the undoped specimen. Hence, our first objective is the right

Cathodoluminescence

108
monitoring of the formation process for this matrix material from which the doping effect
can be safely investigated. A particular attention is paid to the structure and composition
evolution that mostly remain the basic origin of changes in materials physical properties.
The use of rare earth elements as lanthanides engenders more complex phenomena
resulting from a combined optical effect of both ZnO and the dopant shell as shown for
embedded Er
3+
species investigated nowadays for planar wave guides in telecommunication
applications. In that case, ZnO host appears an excellent amplifier of the Er3+ intra-4f
transition signal engendering an infrared line of 1.55 m wavelength. This somewhat
relegates from the time being to minor importance the role of these species in the near UV
and visible emissions of ZnO thin films. Despite of it, some works are done here and there in
this direction. The study of the lifetime decay in the green fluorescence observed with
thermally annealed (Er
3+
/Si
4+
)-ZnO nanocrystalline films was proposed by Kohls et al. [42]
The room temperature blue luminescence of Er-doped ZnO thin films was also investigated
by Zhang et al. [43] from samples prepared by simultaneous evaporation of ZnO and Er
materials. It clearly appears in most of these reports that only expensive doped Er-ZnO films
are involved in this research area. The cathodoluminescent characteristics of the samples in
the near ultraviolet and visible region depict a complete extinction of the visible emitted
bands ( =445, 526, and 665 nm) at 1 at. % Er content. Their deactivation below this
concentration is explained by a compensation of oxygen defects in the material due to the
oxygen-rich medium of the deposition bath. Their reactivation beyond this particular
concentration is ascribed to the increase of the Er
3+
ion shells whose internal radiative
transitions lead to a recovering of these visible emitted bands. The radiative mechanism of
the transitions from the
4
F
9/2
excited states to then
4
I
15/2
ground state, responsible for the
=665 nm emission, is predominant in that case. The respective normalized intensity of the
violet =445 nm and green =526 nm emitted bands exhibits a maximal value for 3 at. % Er
content, reaching a stabilized regime from about 5 at. % Er.
4.2 Experimental and characterization
Varying Er concentration in ZnO matrix, as proposed in the present work, requires some
precautions due to the possible formation of a ternary compound. In actual fact, the physical
solubility of Er in ZnO has not yet been clearly evaluated. Due to the high atomic weight of
this dopant, this solubility is expected to be somewhat reduced. This necessitates a chemical
and structural examination of our investigated samples in the limits of the selected doping
concentration interval going till 10 at. %. In these conditions, the proposal of
cathodoluminescent study could be done in the light of any change in the material
composition and crystal phase.
4.3 Microstructure and composition of ZnO samples
Figure 17 reports the results of x-ray analysis obtained from three representative
investigated specimens. As shown, one sees that the material depicts in every case a (002)
diffraction peak corresponding to 2 = 34.5 angular position, which is a characteristic of the
hexagonal close packed structure of ZnO c
0
/ a
0
= 1.6024. This result implies that the related
(002) direction, normal to the substrate plane, is the preferential growth orientation in every
case. Figure 17(a) clearly reveals that this is the only main orientation of the undoped
sample. One sees in Figs.17 (b) and 17(c) that the increase of erbium concentration


109
progressively modifies the films textural feature with the appearance of two supplementary
peaks of the same lattice at 2 = 31.8 and 2 = 36.3 that are, respectively, related to the 100
and 101 directions. It should be pointed out that this textural change occurs without any
shift in the angular position of 002 diffraction peak and hence with no strain in ZnO matrix
structure. A refined analysis of these results reveals that the dependence of the individual
intensity of both peaks on the Er content quite differs.
Figure 18 shows that the intensity of the (100) peak always remains higher than the (101)
one, regardless of the Er content. However, the (100) peak is more sensitive to the dopant
presence particularly for the concentration higher than 5 at. % Er from which the 100 peak
becomes markedly enhanced while only a slight increase of the (101) peaks intensity is
observed. It should be noted that no typical peak of erbium or Er
2
O
3
materials appears in the
x-ray diffraction pattern of the films even for 10 at. % Er content. This ultimately gives an
indication on the chemical nature of the investigated ZnO films and hence ensures on the
real role of Er species in the material CL characteristics.

Fig. 17. X-ray diffractograms of ZnO Films: (a) Undoped ZnO (b) Er-doped ZnO 5 at. %,
(c) 10 at. %.

Cathodoluminescence

110
The study of the samples composition leading to the results of Table 3 denotes the almost
entire recovering of erbium proportion of the spraying bath in ErZnO samples, which
raises the problem of the dopant insertion mode. Due to the difference in the valence charge
[Zn(II) and Er(III)] and the ionic radius, [R
i
(Er
3+
)0.09 nm and R
i
(Zn
+2
)0.074 nm], Er is more
likely to be interstitial than substitutional in ZnO matrix. On the other hand, the increase of
Er content in the doped samples goes together with a decrease of Zn /O ratio. That indicates
an increase of oxygen defects in ZnO lattice that is expected to favor the building of ErO
complexes of some interest in the study of the IR emissions of the material.

Fig. 18. Relative integrated intensity I/Io of the 100 and 101 x-ray peaks measured from the
intensity Io of the 002 peak with respect to Er content in the investigated samples.

Table 3. Compositional feature of the investigated Er doped ZnO films deposited a constant
flow rate value f = 5 ml min
1
.
The presence of dopant greatly affects the films surface morphology as depicted in the
SEM images of Fig.19. If the undoped sample image in the inset of Fig.19 (a) exhibits
regular hexagonal-like grains; the insertion of Er species in ZnO matrix engenders
variously shaped crystallites, as appearing in Figs. 19a19c. Besides, one sees that the
samples mean grain size D determined by the Debye-Scherrer broadering method from
the main (002) diffraction peak 21 is greatly affected by the Er content. The corresponding
D =f (Er at. %) curve of Fig. 4 shows that D tremendously increases from 1% concentration
onwards, the D value of the undoped sample being slightly higher than that obtained at
this particular concentration.


111

Fig. 19. SEM images of Er-doped ZnO films collected under the incident electron beam
energy Eo = 4 keV: (a) for 1 at. %, (b) for 5 at. %, and (c) for 10 at. % Er concentrations. The
inset of image (a) depicts the undoped ZnO films micrograph at the same scale, as proposed
in Ref. [44].

Fig. 20. Dependence of the mean grain size (D) extracted from the SEM images on the
sample Er concentration.

Cathodoluminescence

112
4.4 CL Spectral study of the investigated films
4.4.1 Qualitative CL Spectral study of the investigated films
The changes observed in the general profile of the CL spectrum linked to the investigated
samples give the first indications on the effect of the selected parameters. Referring to the
undoped sample result and the common electron beam energy E
0
=5 KeV, one can see in
Fig.21 that the evolution of the sample CL-spectral profile somewhat deviates from the
regular insertion of Er in ZnO matrix. In actual fact, it clearly appears that the spectrum of
the Er 1 at. %_-doped ZnO specimen exhibits only the near UV emission peak of 382 nm
wavelength. This greatly differs from the result of the other samples for which at least one
supplementary peak appears in the visible region. Since these supplementary emission
peaks exist for both the undoped and the Er 2 at. % ZnO samples, one infers the occurrence
of a specific phenomenon in the material for 1 at. % Er. This phenomenon reflects a
deactivation and a reactivation of the peaks, respectively, centered at 440, 520, and 650 nm,
below and beyond this particular Er concentration. These peaks are frequently known to be
characteristic of the undoped ZnO CL spectrum [45]. All of them are commonly assigned to
structural defects, their individual mechanism being not yet clearly identified. The green
emission of 520 nm wavelength is currently associated with either Zn or O defect or excess
[31].
The result in Table 3 regarding this sample confirms the as expected non stoichiometry
feature since it corresponds to the ratio Zn/O=1.006. The observed oxygen defect of the
material can be presumably compensated during the film formation by the oxygen-rich
phase of the Er
2
O
3
deposition solution, the related Er content 1 at. % being marginal. This
compensation, directly depending on the defects rate of the undoped sample, is expected to
engender deactivation of the green emission as observed here before Er 1 at. %
concentration. The increase of Er in ZnO matrix favors the selective excitations of Er
+3
ions
indirectly due to the electron-hole _e-h_ intermediate process via ZnO host, as explained in
Ref [46]. This appears as the main cause of the spectral results for Er content higher than 1
at. % for which the inner shell transitions of the Er
3+
ions engender the emitted peaks. In that
case, practically the same visible bands than the undoped ZnO sample are obtained due to
the radiative mechanisms from different excitated states to the ground state I
15/2
. 4F5/2
excitated state originates the peak centered at 455 nm, while
2
H
11/2
and
4
F
9/2
, respectively,
lead to the 526 and 665 nm emissions.
Assuming that the effect of the electron beam energy on the investigated sample is similar
regardless of the doping concentration, the study is here proposed from a Er 5 at. % doped
specimen. In that case, Fig.22 reveals two main observations: (i) some of the typical CL
peaks appear only with the increase of Eo value, which is clearly depicted comparing
spectra (a) and (f). The modification obtained can be explained by a deeper penetration of
the incident electrons in the material that increases the rate of all the shell electronic
transitions of Er
+3
described above as being responsible for the main peaks obtained in the
visible region. We point out that the near UV emission of 385 nm characteristic of the
excitonic band edge of pure ZnO material is free of doping, and hence is only affected by
this general beam electron penetration that should be the origin of the increase of its peaks
intensity. (ii) The results obtained shows that E
o
differently affects the as-obtained CL peaks.
In actual fact, it clearly appears that the transition from the (H
11/2
) excited states linked to


113
the 526 nm emission peak is predominantly involved in the E
o
increase even though those
from the (F
5/2
) and (F
9/2
) states, respectively, related to 455 and 665 nm, also remain
important. The regular increase of all these peaks intensity contrasts with the appearance of
specific emissions linked to particular excitation energy values observed in the
photoluminescent results of Er doped ZnO thin films reported by Komuro et al [47].

Fig. 21. Dependence of the CL-spectral characteristics of the undoped and Er-doped ZnO films
on the samples Er concentration obtained at the incident electron beam energy E= 10 keV.

Fig. 22. Dependence of the CL spectral characteristics of Er(5 at. %)-doped ZnO sample on
the incident electron beam energies: (a) for Eo = 4 keV, (b) for Eo = 5 keV, (c) for Eo = 7 keV,
(d) for Eo = 10 keV, (e) for Eo = 15 keV, and (f) for Eo = 20 keV.
We point out that the broadening of the CL spectral peaks in Figs. 21 and 22 is likely to be
connected with electronphonon interactions. This is a characteristic of disordered media in
the range of the ZnO band gap (1.53.5 eV), particularly under the Er doping effect. [48]

Cathodoluminescence

114
4.4.2 Integration and normalized intensity of the CL Spectral peaks
An accurate study of a luminescent spectrum requires taking into account the possible
peaks overlap phenomenon. This is currently overcame by a spectral modeling leading to a
better peaks integration that remains necessary for the assessment of their relative intensity.
The application of Gaussian model, as shown in Fig.23, appears to be more appropriate
here. The reported result shows that the typical four CL peak feature of the Er doped ZnO
material obtained with spectra Figs. 21 and 22 remains unchanged in the modelling despite
the shape modification caused by the incident electron beam energy. With the near UV
emission being free of the doping process, only the evolution of the three visible CL peaks
mentioned above as centered at 445, 526, and 665 nm wavelengths, respectively, can be
studied with respect to Er content in ZnO material. Besides, their intensity can lead to
examine the effect of the incident electron beam energy for every Er-doped ZnO sample in
the region higher than 1 at. %. The normalization of their respective intensity from the one
of the near UV value leads to a better assessment of the doping and the incident energy
influences.

Fig. 23. Gaussian decomposition of CL spectra related to the Er (5 at. %)-doped ZnO sample
for two different incident electron beam energies: (a) for E= 5 keV and (b) for E= 15 keV.


115
The results of Fig.24 show that, for a constant incident electron beam E
o
=10 KeV, the
increase of Er content in ZnO matrix leads to different peaks intensity profiles. However, a
common increasing trend is observed till 3 at. % Er content. This concentration corresponds
to a maximum for the intensities of the violet 445 nm and green 526 nm emissions. In both
cases, a decrease is observed from this Er percentage towards a plateau neatly appearing at
nearly 5 at. %. No decrease is obtained for the red emission 665 nm whose plateau onset is 3
at. % with a higher intensity value. One infers from these results that the radiative
mechanism from the excited state F
9/2
to the ground state I
15/2
is quite predominant in the
investigated Er concentration region. The lowering in the rate of the radiative mechanisms
from the F
5/2
and H
11/2
excited states to I
15/2
linked to the 445 and 526 nm peaks intensities
might have a link to be clarify with either the reported textural change of Fig.17 occurring at
Er concentration higher than 3 at. % or the material composition. Since this material texture
remains unchanged below this concentration, the regular increase of the three CL peaks
intensity in this region is likely to be the fact of a regular increase of all the radiative
mechanisms involved going together with the increase of the active Er species in the
material.
The similarity of the electron beam energy effect on the investigated samples assumed
above is here confirmed in Fig.25 from the study of samples having three different Er
contents. It appears in every case that the increase of E
o
leads to that of the normalized
intensity of the green emission peak of 526 nm. We point out that the same increasing
trend is observed with the violet and red emissions of 445 and 665 nm. As already
mentioned, this result has a direct connection with the irradiated volume of the material
due to further Er ions involved. A saturation regime is expected at higher E
o
value, the
related current being mostly affected by the sample Er content and possibly by the
textural change observed between 2 and 10 at. %.

Fig. 24. Normalization (from the UV band) of the integrated intensity related to the main
emission peaks and its dependence on the sample Er concentration for E= 10 keV.

Cathodoluminescence

116

Fig. 25. Normalization (from the UV band) of the integrated intensity related to the green
emission peak ( = 526 nm) and its dependence on the incident electron beam energy for
three different samples Er concentration.
4.4.3 CL imaging and repartition of the emitting centers
The study is here performed with a Er 5 at. % doped ZnO sample under an incident beam
energy E
o
=5 KeV being agreed that the results are similar regardless of the doping
concentration and E
o
value. The CL images are collected at a required wavelength having a
16 nm spectral bandwidth. Figure 10 depicts those obtained from each of the main emitted
wavelengths of the material, the dark mark surrounded by a dashed line in some pictures
just serves for the positioning of the investigated sample zone. Figure 10-a related to the UV
band edge =382 nm clearly shows that the luminescent emitting centers of this particular
wavelength are regularly dispersed all over the material, merging with the material surface
grains. Being free of doping, such a repartition of emitting centers is easily explained since
only ZnO matrix is here concerned. It could be noted that the CL image as obtained is the
neatest one due to the predominant intensity of the related peak as it appears in Fig.21. The
emitting centers image of the green wavelength =526 nm proposed in Fig.26-b shows that
the grain like feature is once again recovered. One infers that Er species are well localized in
the surface grains. The result in Fig. 26-c obtained for the red line =665 nm greatly
contrasts with the two precedent investigated wavelengths since the emitting centers are not
clearly distinguished, the intensity signal being quite negligible in Fig. 21. The relationship
between the Er species repartition and the emitting centers is thereby difficult to be
established. Probably, a better result could be expected for this wavelength at higher
incident energy values.


117

Fig. 26. Monochromatic CL images of the Er (5 at. %)-doped ZnO films collected at E
0
= 5
keV, with a focalized spot of 5 nm diameter: (a) for =382 8 nm, (b) for =526 8 nm, and
(c) for =665 8 nm wavelengths.
5. Summary and conclusion
It comes out of the present work that the role of heavy elements as lanthanides in the
doping of ZnO material has not yet been quite explored. Dealing here with Er species
inserted in this material matrix, we find that the increase of erbium concentration leads to
some textural change in the growth of ErZnO film. A single
002
preferred growth
orientation is obtained at lower Er contents, while the
100
and
101 additional ones

become neatly perceptible from about 5 at. % Er concentration. Restricting our
luminescence study in the near UV and visible regions, we show that Er concentration
greatly affects the CL-spectral feature of the material. All the visible emitted bands
entirely disappear for 1 at. % Er concentration. In the investigated wavelength region, the
normalization of the main visible peaks intensity from that of the UV peak shows that the
radiative transitions in the Er
+3
ion shell from the

F9/2

excited states to I15/2

ground state,

Cathodoluminescence

118
responsible for the 665 nm emitted band, are always predominant in the material. Both,
the green =526 nm and the violet =445 nm emitted bands exhibit a maximal normalized
intensity for 3 at. % Er concentration that is followed by a lowering and stabilization
reached beyond 5 at. % Er content.
6. Conclusion
The dependence of cathodoluminescence properties of spray-deposited ZnO films on
process parameters such as substrate temperature, airflow rate and precursors, has been
studied in detail. We have found that the luminescence intensity depend strongly on process
parameters. The optimum values of substrate temperature and airflow rate have been
determined to give the best luminescent ZnO films.
The produced films at optimum condition exhibit luminescence characteristics comparable
to films grown by more sophisticated techniques. These films may be useful for specific
applications as transparent n-type windows in solar cells or for sensors devices where large
surface areas are needed and may also be an important candidate for UV diode lasers
applications.
It seems here evident that the particular spectral feature of the single emitted near UV band
obtained for 1 at. % Er content in ZnO matrix can be of prime interest in the technology of
low cost light emitting diodes LEDs.
7. References
[1] Cowley. J.M. (1969)- Image contrast in a transmission scanning electron microscope.
Appl. Phys. Lett., 15, 58.
[2] Z.C. Jin, J. Hamberg, C.G. Granqvist, J. Appl. Phys. 64(1988) 5117.
[3] J.B. Yoo, A.L. Fahrenbruch, R.H. Bube, J. Appl. Phys. 68(1990) 4694.
[4] Y. Chen, D.M. Bagnall, H. Foh, K. Park, K. Hiraga,Z. Zhu, T. Yao, J. Appl. Phys. 84 (1998)
3912.
[5] E. Jeff, Nause, III-Vs Review 12 (1999) 28.
[6] Y. Chen, D.M. Bagnall, T. Yao, Mater. Sci. Eng. B 75(2000) 190.
[7] Y.C. Kong, D.P. Yu, B. Zhang, W. Fang, S.Q. Feng,Appl. Phys. Lett. 78 (2001) 407.
[8] R. Menner, B. Dimmler, R.H. Maunch, H.W. Shock,J. Crystal Growth 86 (1988) 906.
[9] C.Y. Lee, Y.K. Su, S.L. Chen, J. Crystal Growth 96 (1989)785.
[10] A. Sarkar, S. Chaudhuri, A.K. Pal, Phys. Stat. Sol. (A) 119(1990) K21.
[11] T. Minamai, T. Yamamoto, T. Miyata, Thin Solid Films366 (2000) 63.
[12] K. Haga, F. Katahira, H. Watanabe, Thin Solid Films 344(1999) 145.
[13] Y. Natsume, H. Sakata, T. Hirayama, H. Yanigida,J. Appl. Phys. 72 (1992) 4203.
[14] M. Ruth, J. Tuttle, J. Goral, R. Nou, J. Crystal Growth96 (1989) 363.
[15] A. Ortiz, C. Falcony, J. Hernandez, M. Garcia,J.C. Alonso, Thin Solid Films 293 (1997)
103.
[16] A. Maldonado, R. Asomoza, J. Canetas-Ortega,E.P. Zironi, R. Hernandez, O. Solorza-
Feria,Sol. Energ. Mater. Sol. Cells 57 (1999) 331.
[17] S.A. Studenikin, M. Cocivera, W. Kelner, H. Pascher,J. Lumin. 91 (2000) 223.


119
[18] M. de la, L. Olvera, A. Maldonado, R. Asomoza,O. Solorza, D.R. Acosta, Thin Solid
Films 394 (2001) 242.
[19] N. Benhamdane, W.A. Murad, R.H. Misho, M. Ziane,Z. Kebbab, Mater. Chem. Phys. 48
(1997) 119.
[20] S.A. Studenikin, N. Golego, M. Cocivera, J. Appl. Phys.83 (1998) 2104.
[21] M. Addou, A. Moumin, B. El idrissi, M. Regragui, A.Bougrine, A. Kachouane, J. Chem.
Phys. 96 (2) (1999) 232.
[22] F. Paraguay, D.J. Morales, W. Estrada, L.E. Andrade,M. Miki-Yoshida, Thin Solid Films
366 (2000) 16.
[23] H.J. Egelhaaf, D. Oelkrug, J. Crystal Growth 161 (1996)190.
[24] T. Sekiguchi, N. Ohashi, Y. Terada, Jpn. J. Appl. Phys.Part 2 Lett. 36 (1997).
[25] P. Fons, K. Iwata, S. Niki, A. Yamada, K. Matsubara,J. Crystal Growth 201/202 (1999)
627.
[26] H. Nanto, T. Minami, S. Takata, Phys. Status Solidi A 65(1981) K131.
[27] J. Zhong, A.H. Kitai, P. Mascher. J. Electrochem. Soc. 140(1993) 3644.
[28] K. Vanheusden, W.L. Warren, C.H. Seager, D.R. Tallant,J.A. Voigt, B.E. Gnade, J. Appl.
Phys. 79 (1996) 7983.
[29] P. Scherrer, Goettinger Nachr 2 (1918) 98.
[30] R. Dingle, Phys. Rev. Lett. 23 (1969) 579.
[31] K. Vavheusden, C.H. Seager, W.L. Warren, D.R. Tallant,J.A. Voigt, Appl. Phys. Lett. 68
(1996) 403.
[32] E.G. Bylander, J. Appl. Phys. 49 (1978) 1188.
[33] T. Minami, H. Nanto, S. Takata, J. Lumin. 24/25 (1981)63.
[34] C. Falcony, A. Ortiz, M. Garcia, J.S. Helman, J. Appl.Phys. 63 (1998) 7.
[35] A. Mitra, R.K. Thareja, V. Ganesan, A. Gupa, P.K.Sahoo, V.N. Kulkarni, Appl. Surf. Sci.
174 (2001) 232.
[36] K. Lohnert, E. Kubalek, Microscopy of SemiconductingMaterials, Institute Of physics
Publishing, Bristol, 1983,
[37] M. Troyon, D. Pastre, J.P. Jouart, J.L. Beaudoin, Ultramicroscopy75 (1) (1998) 15.
[38] I. Shin, C.V. Qin, J. Appl. Phys. 58 (1985) 2400.
[39] S.A. Studenikin, N. Golego, M. Cocivera, J. Appl. Phys. 84 (1998) 4.
[40] H.J. Egelaaf, D. Oelkrug, J. Cryst. Growth 161 (1996) 190.
[41] K. Lohnert, E. Kubalek, Microscopy of Semiconducting Materials,IOP, Bristol, (1983)
303.
[42] M. Kolhs, M. Bonanni, L. Spanhel, D. Su, and M. Giersig, Appl. Phys. Lett. 81, (2002)
3858
[43] X. T. Zhang, Y. C. Liu, J. G. Ma, Y. M. Lu, D. Z. Shen, W. Xu, G. Z. Zhong, and X. W.
Fan, Thin Solid Films 413, (2002) 257 .
[44] A. El Hichou, A. Bougrine, J.-L. Bubendorff, J. Eboth, M. Addou, and M. Troyon,
Semicond. Sci. Technol. 17, (2002) 607
[45] C. Shi, Z. Fu, C. Guo, X. Ye, Y. Wei, J. Deng, J. Shi, and G. Zhang, J. Electron Spectrosc.
Relat. Phenom. 101103, (1999) 629.
[46] X. T. Zhang, Y. C. Liu, J. G. Ma, Y. M. Lu, D. Z. Shen, W. Xu, G. Z. Zhong, and X. W.
Fan, Thin Solid Films 413, (2002) 257.

Cathodoluminescence

120
[47] S. Komuro, T. Katsumata, T. Morikawa, X. Zhao, H. Isshiki, and Y. Aoyagi, J. Appl.
Phys. 88, (2000) 7129.
[48] I. V. Kityk, J. Wasylak, D. Dorosz, J. Kucharski, S. Benet, and H. Kaddouri, Opt. Laser
Technol. 33, (2001) 511.
6
Multicolor Luminescence from Semiconductor
Nanocrystal Composites Tunable in an
Electric Field
G.N. Panin
1
Department of Physics, Quantum-Functional Semiconductor Research Center
Dongguk University, Seoul,
2
Institute of Microelectronics Technology & High Purity Materials,
Russian Academy of Sciences, Chernogolovka,
1
Korea
2
Russia
1. Introduction
Semiconductor nanostructures have interesting electro-optic properties due to quantum
confinement of electrons, strong exciton binding energies, and the possibility to tailor the
bandgap and radiative emission rates. An electric field normal to the quantum well, for
example, shifts the absorption edge to lower energies (red-shift) and increases the
refractive index below the absorption edge allowing to control the intensity of a light
beam through electroabsorption. Such nanostructures can be conveniently used for the
direct modulation of light, since they show much larger electro-optic effects than bulk
semiconductors. Moreover, the optical properties of semiconductor nanostructures can be
controlled by an electric field due to different emission rates from radiative centers in
different charge states.
In the Chapter, the effect of an external electric field on cathodoluminescence from
semiconductor nanocrystals and nanocomposites with different radiative emission rates is
described, giving special emphasis to ZnO nanocrystals in MgO and polymer matrix.
In the next Section we shortly review the preparation of ZnO nanocrystals and
nanocomposites, including a chemical solution deposition (CSD) of core/shell nanocrystals,
as well as their structural and optical properties.
In Section 3 of the chapter, the effect of an external electric field on cathodoluminescence
from the nanocrystal/polymer structure will be studied, giving special emphasis to the
doped ZnO nanocrystals in Poly(4,4- diphenylene diphenylvinylene) (PDPV) matrix and
PBET/ITO structures. Electric field-induced color switching of cathodoluminescence from
nanocrystals/polymer is reviewed. The assumed mechanism of electric field-tunable
cathodoluminescence implies the presence of radiative recombination channels, which are
sensitive to the electric field through the band bending at the crystal surface. The
experiments on a reversible quenching of the UV near-band-edge emission under visible

Cathodoluminescence

122
illumination confirm an appearance of the recombination channel after recharging oxygen-
vacancy states in the surface depletion zone. It has been shown that the UV near-band-edge
emission can be modulated at frequencies of hundreds of hertz. Limiting performance
bounds for potential future devices fabricated from nanocrystals with different radiative
emission rates are considered.
2. Structural and optical properties of ZnO (MgO) nanocrystals and
nanocomposites
Zinc oxide is a II-VI compound semiconductor with a wide direct band gap of 3.3 eV at
room temperature (RT), which makes it attractive for optoelectronic devices in the near UV
region (Kang et al., 2006; Look, 2001; zgr et al. 2005). ZnO has a large exciton binding
energy of about 60 meV, which promises efficient RT exciton emission. The unique physical
and electronic properties of ZnO, such as large bond strength and large damage threshold
for laser irradiation, (Minne et al., 1995; Yamamoto et al., 1999) and its high exciton binding
energy allow to fabricate an excitonic laser operating at room temperature (Bagnall et al.,
1997; Ohtomo et al., 2000; Yu et al., 1997). A lower pumping threshold for laser operation is
expected when exciton-related recombination rather than electronhole plasma
recombination occurs. The use of a ZnO-based quantum structure with a higher exciton
binding energy could decrease significantly the current threshold, which is the key point for
current injection laser. The modified density of states in the ZnO-based low-dimensional
quantum structure can confine both excitons and photons, thus achieving high-efficiency
stimulated emission (Sun et al., 2000). The RT cathodoluminescence (CL) spectrum of ZnO
along with the exciton emission band at near 380 nm contains a deep-level defect related
emission band at 500650 nm. The deep-level emission band, as usual, is considered
consisting of the green (500520 nm), yellow-orange (560 600 nm), and red (650 nm) bands.
The green emission band is frequently observed and the early studies it was attributed to
copper impurities (zgr et al., 2005). However at present, oxygen vacancies have been
assumed to be the most likely candidate for recombination centers involved in the green
luminescence of ZnO (Cheng et al., 2006; Leiter et al., 2001; Studenikin et al., 1998b;
VanDijken et al., 2000a; Vanheusden et al., 1996a, 1996b). UV (exciton) and visible (defect)
emissions are in competition with each other and it is presumed that the deep level emission
centers are preferable channels of the electron-hole recombination. Such behavior of the
luminescence bands gives rise to interesting physical phenomena, which creates an ability to
control the intensity of the UV emission, and could be also of practical relevance. In practice
the dependence of the excitonic band on external and internal perturbations is widely used
to modulate its intensity. The modulations of the excitonic PL by means of acoustic
(Takagaki et al., 2003), terahertz waves (Klik et al., 2005) and visible light (Chen et al. 1996;
Shih H. Y. et al., 2011; Kurbanov et al., 2008a; Kurbanov et al., 2008b ) as well as the excitonic
PL and CL by electrical field (Hagn M., et al., 1995; Panin et al., 2005a; Schmeller et al., 1994;
Zhang S. K., et al., 2001; Zimmermann et. al., 1997; Zimmermann et al., 1998; Zimmermann
et al., 1999) were reported. The opportunity to control the characteristics, in particular,
reflectance, of the sample under study by the external periodic perturbations serves as a
basis for the modulation spectroscopy and allow to study electron states and features of the
semiconductor band structure (Aigouy et al., 1997; Motyka et al., 2006; Rowland et al., 1998).
In ZnO nanocrystals the electron paramagnetic-resonance signal assigned to singly ionized
oxygen vacancy (V
0
+
) was observed to be photosensitive for photon energies down to as low
Nanocrystal Composites Tunable in an Electric Field

123
as 2.3 eV (Vanheusden et al., 1996a, 1996b). Upon illumination, the V
0
+
density was observed
to grow. Moreover, it has been found that the intensity of the green emission in ZnO
powder correlates well with the paramagnetic single-ionized oxygen-vacancy (V
0
+
) density.
Nevertheless, the effect of the subband excitation on emission from ZnO is still unclear and
remains unexplored in spite of the observed correlation between the density of V
0
+
and the
green emission intensity and the sensitivity of V
0
+
to a visible light.
MgO (E
g
= 7.3-7.7 eV) (Roessler & Walker, 1967; Johnson, 1954) solved in ZnO can produce a
large band gap Mg
x
Zn
1-x
O alloy, which is well suitable for quantum structures operating in
the ultraviolet spectral range (Ohtomo et al., 1998; Sharma et al., 1999). The radiative
recombination of electronhole pairs in terms of the quantum confined Stark effect was
observed in ZnO/Mg
x
Zn
1-x
O structures (Makino et al. 2002). According to the phase
diagram of ZnO/MgO binary systems (Raghavan et al., 1991), the thermodynamic solid
solubility of MgO in a ZnO matrix is less than 4 mol%. The crystal structure of ZnO
(hexagonal, a = 3.24 and c = 5.20 ) is vastly different from that of MgO (cubic, a = 4.24
). However, since the ionic radius of Mg
2+
(0.57) is almost the same as that of Zn
2+
(0.60 )
(Shannon, 1976), Zn
2+
can be replaced by Mg
2+
in the ZnO matrix. The synthesis of cubic
Mg
x
Zn
1-x
O nanocrystals under pressure was reported (Baranov et al., 2005a; Baranov et al.,
2010). Stable cubic Mg
x
Zn
1x
O solid solutions (0.33 < x < 0.68) with continuous band gap up
to 3.74 eV have been synthesized under high pressures and high temperatures.
2.1 The growth techniques
The growth techniques of NCs are very important since they influence their structure, shape
and distribution of nanocrystal sizes, stoichiometry, structure of the surfaces or interfaces,
and their optical properties. Therefore, before studying in the next sections the electric field
induced optical properties of NCs, we want first to briefly look to the chemical growth
techniques, paying special attention to the techniques of self-assembling (Fu et al, 2003; Fu et
al., 2005; Panin et al., 2003b; Panin et al 2004d; Panin et al., 2005c; Panin et al., 2007d;
Kurbanov et al., 2008c) as well as structural and luminescent properties of as-grown and
annealed NCs. Nanostructuring by physical techniques based on lithography and etching
does not usually produce NCs of sizes small enough for the detection of size effects, even if
one uses electron beam techniques. The advantage of these techniques is their compatibility
with microelectronics techniques. However, the drawback top / bottom of lithographic
techniques in their low productivity and poor optical quality of structures produced by
etching. The chemical techniques are the ones most used for the preparation of NCs. We will
mention a few of them which have been used in preparation of ZnO-based composites.
2.1.1 Colloidal synthesis of ZnO nanoparticles
Colloidal synthesis, by the reduction of metal salts in solutions with organic ligands, are
frequently used to produce metal nanoclusters (e.g. gold). As for II-VI semiconductors,
nanoparticles of ZnO, CdSe, CdS, etc. are produced from reagents containing the
nanocrystals constituents. One reagent contains the metal ions (e.g. Zn
2+
) and the other
provides the oxygen (O
2-
) or the chalcogenide (e.g. Se
2
). The size of the nanocrystals is
controlled by the temperature of the solutions and the concentrations of the reagents and the
stabilizers. Synthesis of ZnO nanoparticles with sizes of 28 nm is fairly detailed, (Baranov

Cathodoluminescence

124
et al., 2008; Sakohara et al., 1998; Van Dijken et al., 2001; Wong et al., 2001; Meulenkamp,
1998; Pesika et al., 2003; Spanhel & Anderson, 1991; Tokumoto et al., 2003; Hosono et al.,
2004) and the conditions for nanoparticle manufacturing by precipitation of zinc acetate
with alkali from alcoholic solutions are also well studied. In particular, particle growth rate
curves were obtained and the effects of several factors were determined, such as the water
proportion in the solution, the nature of the solvent, the starting solution concentration, and
precipitation temperature.
A typical absorption spectrum from suspended ZnO nanoparticles is shown in Fig 2.1 The
simplest way to estimate the particle size in colloidal ZnO solutions is to determine the
transition position in optical absorption spectra. The absorption onset of ZnO nanocrystals
appears at 337nm, which is blueshifted relative to 382 nm absorption onset of bulk ZnO. The
spectrum shows that these uncapped ZnO nanoparticles were well dispersed in ethanol, and
that the band gap absorption was much more effective compared to the absorption by
surface defect states (Chen et al., 1997).

Fig. 2.1. Optical absorption spectra of colloidal ZnO particles prepared in ethanol. In the
inset, the same data in other coordinates. (Baranov et al., 2008)
The inset in Fig. 2.1 illustrates the experimental data in the (h)
2
h coordinates. The
existence of a linear segment on the (h)
2
versus h curve verifies the existence of a direct
electron transition which defined the particle adsorption edge. Extrapolation of this segment
to axis h gave a value of 3.6 eV for the band gap of suspended ZnO nanoparticles.
Compared to the band gap of micron-sized ZnO (E
g
= 3.3 eV), the band gap of ZnO
nanoparticles was increased by the quantum confinement effect. The method which we used
is to equate E
g
with the wavelength at which the absorption is 50% of that at the excitonic
peak (or shoulder), so called
1/2
. To convert measured values of
1/2
into particle size an
expression (1) can be used:
1240/
1/2
= a + b/D
2
- c/D (1)
where
1/2
in nm, diameter D in and the values a= 3.556, b =799.9, and c = 22.64, which
gives a good description of the experimentally found size dependence for 25 < D < 65
(Meulenkamp, 1998). The mean particle size calculated using expression 1 was ~3.0 nm.
Comparison of the data derived from the absorption spectra with TEM data and theoretical
calculations demonstrates good convergence.
ZnO nanoparticles synthesized from aqueous solution (Kurbanov et al., 2007a; Wang et al.,
2005) and deposited on a substrate show usually the larger size (100-1000 nm) and cone-
shape (Fig. 2.2 (a)).

125

Fig. 2.2. Typical SEM images of ZnO nanocrystals (a) as-prepared and (b) annealed in air at
500 C. The inset shows the specklike defects revealed after annealing at higher
magnification. (Kurbanov, et al., 2009a)
As-prepared ZnO nanocrystals display a strong UV luminescence accompanied with a weak
yelloworange emission at around 565 nm. With increasing the excitation power a violet PL
band at ~400 nm starts to appear and it dominates over the UV near-band-gap emission at
high excitation power. While the UVPL band intensity displays saturation behavior, the
violet band intensity grows superlinearly. The violet luminescence most likely arises from
defects associated with zinc vacancy or zinc vacancy related complexes.

Fig. 2.3. Normalized (a) CL and (b) PL spectra of ZnO nanoparticles: (black) as- prepared
and (red) after annealing in air at 500
o
C for 1h. (Kurbanov et al., 2009b)
Figs. 2.3 (a) and (b) show the normalized room temperature CL and PL spectra of the ZnO
nanoparticles before and after annealing in air at 500
o
C for 1h, respectively. The annealing
resulted in a strong increase in intensity of the green band centered at 505 and 520 nm in the
CL and PL spectra, respectively. This band became dominant emission band in the visible
region. The decrease in intensity of the violet band at ~400 nm after annealing is clearly
observed in the spectra obtained at the high laser power excitation (80 mW) (Fig. 2.3 (b)).
The UV emission from nanocrystals is ascribed to near-band- edge luminescence of ZnO.
The variation of the UV peak energy as a function of temperature described by the Varshni
equation also confirms this statement (Kurbanov et al., 2009b). Because the ZnO
nanocrystals were prepared without intentional doping, the violet, green and yelloworange
emissions are attributed to native defect centers and their complexes. Namely, the yellow
orange emission band around 560580 nm from the as-prepared ZnO nanocrystals has been
assigned to oxygen interstitial related defects. The green band peaked at 505520 nm
appearing after the high-temperature annealing and being dominant in the visible emission
is attributed to oxygen vacancy related centers. The observed differences in the green band
positions in CL and PL spectra could be attributed to specific features of photon and
electron beam excitation, particularly, to a different penetration depth of photons and
electrons. The first principles investigations show that under oxygen-rich conditions the zinc

Cathodoluminescence

126
vacancy and oxygen interstitials are the dominant defect types and they have low formation
enthalpies than other defects (Erhart et al., 2006; Kohan et al., 2000). Based on these
calculations, it is reasonable to suggest that the violet emission from ZnO nanocrystals
grown in thermal equilibrium conditions is originated from Zn vacancy related defects or
their complexes. Moreover, as was mentioned above, in most cases the violet emission band
is accompanied with the yelloworange band, attributed to oxygen interstitials (Kumar et
al., 2006; Hua et al., 2007; Jeong et al., 2003; Wu et al., 2001; Zhao et al., 2005;). The first
principles studies predict that for Zn vacancy, the transition level (0 /-1) lies at ~0.45 eV
(Kohan et al., 2000) or ~0.3 eV (Erhart et al., 2006) above the valence band and the transition
level (-1/-2) is located at ~0.8 eV (Kohan et al., 2000) or 0.7eV (Erhart et al., 2006) above the
valence band. Thus a transition between the conduction band and the single charged Zn
vacancy acceptor level would give rise to luminescence at 405 nm (3.06 eV), in reasonable
agreement with the predicted transition energy.
A postgrown annealing treatment significantly improved the UV emission efficiency and
resulted in the clear appearance of a low temperature emission band around 3.31 eV (so-called
A-line). Spatially and wavelength resolved CL measurements revealed a spotlike distribution
of the A-line emission on a nanocrystal surface with a strong correlation between the emission
around 3.31 eV and the specklike defects that appeared on the nanocrystal surface after
annealing. At low temperatures, bound exciton emissions from ZnO nanocrystals are the
dominant radiative recombination channels. The energy range between 3.375 and 3.367 eV is
normally assigned to ionized donor bound excitons, while the neutral donor bound excitons
are positioned between 3.366 and 3.36 eV. The free exciton emission is usually observed at
3.377 eV. According to the conventional classification for the bulk ZnO crystals, the lines
situated at lower energies, between 3.33 and 3.31 eV, are attributed to the two-electron satellite
(TES) recombination of the donor bound excitons (Kang et al., 2006; Meyer et al., 2004; Teke et
al., 2004). However, in recent years many research groups reported on an observation of a
strong emission line in the TES region the so called A-line. The position of the A-line varies
from 3.31 to 3.333 eV depending on the growing methods of ZnO. The A-lines intensity is
strongly enhanced in ZnO powders when compared to bulk ZnO (Fallert et al., 2007). The
origin of the A-line emission is currently under discussion. Possible assignments of the A-line
include: neutral excitons bound to nitrogen impurities (Look et al., 2002; Yang et al., 2006),
donor-acceptor pair recombination (Zhang et al., 2003), free-to neutral- acceptor transitions
(Cao et al., 2007), two-electron transition of exciton bound to neutral donor (Studenikin et al.,
2000), the first longitudinal optical (LO) phonon replica of free excitons (Hirai et al., 2008) or
excitons bound to surface or structural defects ( Fallert et al., 2007; Fonoberov et al., 2006; He et
al., 2007; Meyer et al., 2004; Teke et al., 2004). Recently, on a basis of cathodoluminescence
measurements, it was suggested that this line is composed of two overlapped bands related to
point defects and the first LO phonon replica of the free exciton (Mass et al., 2008).
Wavelength resolved CL images as well as, spatially resolved CL spectra from ZnO
nanocrystals show a direct correlation between the A-line emission at 3.311 eV and the
annealing induced specklike defects (Kurbanov et al., 2009a). The specklike defects appear
on all faces of the annealed nanocrystals and lead to a spotlike distribution of the emission.
Figure 2.3 (b) shows a SEM image of the annealed ZnO nanocrystals. The annealing of the
nanocrystals leads to the formation of specklike defects on their surface, which appear as
dark spots and are not observed on the as-prepared nanocrystal faces.

127

Fig. 2.4. (Left) Spatially resolved CL (a, b) and integral CL(c) and PL (d) spectra at 90 K.
Spatially resolved CL measurements were performed on ZnO nanocrystal face: specklike
defect region (a) and defect-free region (b); (Right) Wavelength resolved CL images of the
ZnO nanocrystal face obtained at 90 K for (a) 3.358 eV (DX-line) and (b) 3.313 eV (A-line)
energies as well as a SE image of the ZnO nanocrystal face with specklike defects (c).
The spatially resolved CL spectra, as well as, the integral PL and CL spectra obtained from a
large number of ZnO nanocrystals at 90 K are presented in Fig. 2.4. It can be seen that in
contrast to the PL spectrum the DX and A-lines in the integral CL spectrum are not well
resolved; whereas, the A-LO line at around 3.24 eV appears clearly [Fig. 2.4, curve (c)] The
spatially resolved CL spectrum from the nanocrystal surface region without specklike
defects peaks at 3.35 eV; it is broadened and displays no A-LO line [Fig 2.4, curve (b)]. A
spectral maximum of the CL spectrum from the specklike defect region is redshifted and
located at 3.34 eV [Fig. 2.4 (a).] Moreover, it shows the pronounced A-LO peak at around
3.24 eV. Obviously, the A and A-LO lines appear only in the CL spectra collected locally
from the regions with the specklike defects, or a large number of nanocrystals with the
specklike defects. Fig. 2.5 shows the wavelength resolved CL images taken at the DX (3.358
eV) and A-line (3.311 eV) energies at 90 K. The CL image monitored at the bound exciton
energy exhibits an uniform distribution and no extraordinary features are observed [Fig. 2.4
(a)]. In contrast, the emission detected at the A-line energy is spotlike and localized [Fig. 2.4
(b)]. The distribution of the spots coincide with position of the specklike defects in the
secondary electron image [Fig. 2.4 (c)].
2.1.2 ZnO nanorods
Numerous techniques have recently been developed to synthesize quasi-1D ZnO
nanostructures, (Baranov et al. 2004; Liu et al., 2004; Xia et al., 2003; Yang et al., 2002; Yao et
al., 2002; Yin et al., 2004; Wang et al., 2004; Zhang et al., 2003) including growth with or
without catalysts by a vapourliquid solid epitaxial (VLS) mechanism (Yang et al., 2002;
Yao et al., 2002; Wang et al., 2004), microwave plasma growth (Zhang et al., 2003) and
solution methods (Liu et al., 2004; Yin et al., 2004). An important issue in the fabrication of
nanostructures is the control of the shape and size of nanomaterials, both of which affect
their optical and electrical properties. ZnO nanorods grown from a NaCl (NaClLi
2
CO
3
) salt
mixture were described by (Baranov et al., 2005b; Baranov et al. 2004). The salt composition
and the size of the Zn-containing precursor particles play a key role in the synthesis and
define their size and properties.

Cathodoluminescence

128

Fig. 2.5. (Left) HRSEM images of ZnO nanorods grown from the NaClLi
2
CO
3
salt mixture
at 700 C for 1 h and (Right) TEM images of ZnO nanorods grown from the NaCl mixture at
700 C for 1 h. Insetsselected area of electron diffraction and the ZnO atom-scale
resolution image (HRTEM).
Figures 2.5 show HRSEM and TEM images of ZnO nanorods synthesized from the NaCl
Li
2
CO
3
salt mixture as well as their selected area diffraction analysis. The presence of
Li
2
CO
3
allows to decrease the diameters of ZnO nanorods. The product prepared at 600 C
for 3 h consists of quite uniform nanorods with diameters ranging from 8 to 40 nm and of
length from 200 to 500 nm. As the temperature increased to 800 C, the diameters of the
nanorods increased drastically and ranged from 200 to 500 nm. The results indicate that
individual nanorods are single crystalline and possess the wurtzite structure (see the inset
in figure 2.5, right).

Fig. 2.6. (a) CL spectra taken at room temperature (V = 15 kV) from ZnO nanorods grown
from (a) the NaCl mixture at 500 C (dashdot line) and 700 C (solid line), (b) the NaCl
Li
2
CO
3
salt mixture at 700 C (solid line) (the dashed lines are Gaussian multi-fit
approximations) (Baranov et al., 2005b).
The CL spectra for the ZnO nanorods are shown in figure 2.6. Two peaks at around 3.25 eV
and 2.5 eV and a high-energy shoulder at around 3.65 eV are observed. The high-energy
bands 3.65 eV and 3.25 eV correspond to the near band-gap emission (NBE) of small and
large rods, respectively. The high energy shoulder in the spectrum indicates some
dispersion of small size rods. Strong quantum confinement effect could be attributed to the
real size limitation by an potential barrier formed due to the charge surface states or the
exciton dead layer (Pekar, 1958; Combescot et al., 2001), which increases with nanocrystal

129
size (Fonoberov et al., 2004). The UV shift of PL from nanocrystals which are large in
comparison with the exciton Bohr radius in ZnO (~1 nm (Fonoberov et al., 2004; Gill et al.,
2002)) has been reported. The UV NBE shift of about 40 meV was observed for surprisingly
large (24 nm) ZnO nanocrystals (Hur et al., 2005).
2.1.3 CVD grown ZnO nanorods and tetrapods
Chemical vapor deposition (CVD) and vapor transport (VT) techniques are widely used for
the growth of nanorods and tetrapods. The effect of growth conditions on their morphology,
size and properties has been well studied (Lyapina et al., 2008) and high efficient light
emitting diodes based on ZnO nanorods were reported (Lee et al., 2011).

Fig. 2.7. SEM micrographs of ZnO nanorods and tetrapods grown on Si substrates.
Figure 2.7 shows typical ZnO nanostructures grown on a Si substrate. The supersaturation
of the zinc vapor plays a key role in determining the morphology of the structures. The
nanorods have a hexagonal shape and its size can be well controlled by the zinc vapor. The
tetrapods are not bonded to the substrate surface, which implies that they were formed in
the gas phase. The lower layer consists of small tetrapods with legs ~ 23 m in length and ~
350 nm in diameter. In the upper layer, the tetrapods have elongated legs on the order of
5m in length, with a diameter of 650 nm at the junction and 100 nm at the tip. The rods of
the tetrapods have long whiskers, up to 10 m in length, with a sharp transition from a leg
thickness of 500 to 80100 nm.

Fig. 2.8. CL spectra of tetrapods grown in (1) oxygen-deficient and (2) oxygen-enriched
conditions (Lyapina et al., 2008).

Cathodoluminescence

130
Figure 2.8 shows the RT CL spectra of tetrapods obtained in different conditions. The UV
peak at 383 nm is due to free excitons. The green band centered at 505 nm is markedly
stronger. The intensity ratio between the UV and green emissions is larger in the samples
obtained at the highest oxygen flow rate and reduced zinc vapor concentration.
2.1.4 ZnO/MgO nanocrystals and nanocomposites
The techniques for growing nanocrystals inside a matrix are very developed, partly as a
consequence of the industrial production of color filters and photochromic glasses based on
copper halides, such as CuCl, CuBr, CuI, etc. Nanocrystals of II-VI compounds (ZnO, CdS,
CdSe, ZnSe, etc.), starting from supersaturated viscous solutions, are also frequently
incorporated in glass matrices for applications in optical filters due to the ease in controlling
the dot size. ZnO/MgO composites can be prepared by using aqueous solution and
methanol solution techniques (Panin et al., 2004c; Panin et al., 2005b). Fig. 2.9 (Left) shows
SEM images of the ZnO/MgO particles prepared by these techniques. The sizes of the
individual ZnO and MgO particles prepared from the aqueous solution are approximately
100 and 500 nm, respectively, while the size of the particles prepared from the methanol
solution is about 10 nm.

Fig. 2.9. (Left) Scanning electron microscopy images of ZnO/MgO nanoparticles prepared
by using (a) aqueous solution and (b) methanol solution techniques. (Right) RT CL spectra
of the ZnO/MgO particles annealed in (a) N
2
and (b) O
2
atmospheres.
(right panel) shows CL spectra of the large-size ZnO/MgO particles annealed in an oxygen
or nitrogen environments. Annealing in oxygen environment inhibits the green
luminescence and enhances the near-band-gap emission.

131
2.1.5 ZnO/MgO core-shell nanocrystals
ZnO/MgO core-shell nanocrystals were synthesized by chemical solution deposition (CSD)
technique using a colloidal zinc oxide solution and alcoholic solution of magnesium acetate
as described in (Baranov et al., 2008).

Fig. 2.10. XRD patterns from powdered nanoparticles of (a) ZnO annealed at 400 C and (b),
(), and (d) ZnO/MgO composites, annealed at 400, 500 and 600 C, respectively. (Right) (a)
TEM micrograph and (b) electron diffraction pattern of the ZnO/MgO nanocomposite
annealed at 500 C.
X-ray powder diffraction and electron diffraction show that, although the synthesis
temperature was rather low, nanoparticles have a crystal structure. The XRD pattern of
ZnO sample (Fig. 2.10, curve (a)) consistent with the hexagonal phase of ZnO crystals; the
unit cell parameters insignificantly differ from literature data: a = 3.249 and c = 5.206 .
The coherence length for zinc oxide determined from diffraction peak broadening agrees
well with the particle size derived by digitizing TEM micrographs (Fig. 2.10, (image) (a))
and is within 5 12 nm. X-ray powder diffraction and TEM data for samples annealed at
400, 500, and 600 indicate that the optimal temperature range for ZnO/MgO composite
manufacturing is 400500C. The X-ray diffraction patterns contain both peaks associated
with the cubic MgO structure (with Miller indices 111 and 200) and those associated with
zinc oxide (wurtzite; indices 100, 002, 101, 102, and 110) (Fig. 2.10, curves b and c). These
peaks are considerably broadened, but there is no anisotropy in directions 100 and 002;
that is, particles have near-spherical shapes, which are also seen in TEM micrographs (Fig.
2.10, image (a)). The ZnO coherence length (crystallite size) in the composite with MgO
varies as a function of synthesis parameters more widely than for pure zinc oxide;
crystallite sizes range from 8 to 14 nm. The unit cell parameters calculated for MgO and
ZnO phases almost coincide with the literature values; this means that, there is no
noticeable interaction between magnesium and zinc oxides with solid solution formation
at the low annealing temperatures. It confirms also by the electron diffraction pattern for
the nanocomposite annealed at 500C which contains both indices attributed to the cubic
MgO structure ([220], [200]) and those associated with wurtzite zinc oxide ([100], [002],
[101], and [110]) (Fig. 2.10, image (b)). A different situation is observed for samples
annealed at 600C (Fig. 2.10, curve d) or higher temperatures. In this case, wurtzite phases
vanish and there are only peaks of newly formed solid solution Mg
1-x
Zn
x
O; the init cell
parameters of the cubic phase increase systematically, which corresponds to a partial
substitution of zinc ions (with a larger ionic radius) for magnesium ions.

Cathodoluminescence

132

Fig. 2.11. Cathodoluminescence spectra of the ZnO/MgO nanocomposites annealed at (a)
400, (b) 500, () 600, and (d) 700 C (Baranov et al., 2008).
Cathodoluminescence spectra of annealed samples are shown in figure 2.11. The spectra of
the samples annealed at 400 and 500C show two peaks at 376 and 512 nm which are
attributed to the free exciton luminescence and the oxygen-vacancy-related green emission
from ZnO nanoparticles. The strong blue shift of the excitonic luminescence could be due to
the quantum confinement effect in ZnO nanocrystals. The relationship between nanocrystal
size and band gap can be obtained using a number of models (Efros & Rosen, 2000;
Andersen et al., 2002; Hyberstsen, 1994). The effective mass model for spherical particles
with a Coulomb interaction term (Brus, 1984; Brus, 1986) where the band gap E*[eV] can be
approximated by Eq. 2 (Pesika et al., 2003):
E*=E
g
bulk
+[(
2
2
)/2er
2
][(1/m
e
m
0
)+(1/m
h
m
0
)]-1.8e/4
0
r, (2)
where E
g
bulk
=3.3 eV is the ZnO bulk band gap, r is the particle radius, m
e
=0.26 is the
effective mass of the electrons, m
h
=0.59 is the effective mass of the holes, m
0
is the free
electron mass, =8.5 is relative permittivity of ZnO,
0
is the permittivity of the free space,
is Planks constant divided by 2, and e is the charge of the electron. The size of the ZnO
particles estimated from Eq. 2 was 8 nm that well agreed with TEM and X-ray diffraction
data for the nanocomposites studied. The relative intensity of the green luminescence
increases with an increase of annealing temperature (Fig. 2.11, curve b). This behavior of
luminescence could arise from an increase in density of single-ionized oxygen vacancies
after annealing as a result of an increase of the electron concentration in the ZnO particles.
The spectra of the samples annealed at 600 and 700C are shown in figure 2.11 (c, d). For the
sample annealed at 600C, edge luminescence is not observed but several peaks in green,
yellow and red spectral regions are appeared. After annealing at 700C, a peak in the deep
UV region (at 307 nm) appears, confirming the formation of a solid Mg
1x
Zn
x
O solution, for
which the UV shift of edge luminescence is characteristic. The peaks at around 600 and 700
nm can be attributed to Mg- or oxygen-related defects in the solid solution which together
with oxygen vacancy defects reduces transparency of the nanocomposite in the visible
range. The hexagonal ZnO (core) and cubic MgO (shell) phases in ZnO/MgO
nanocomposites are thermally stable and high transparent up to 500C. The annealing at
temperatures 600 -700C led to Mg doping and formation of the Mg
1x
Zn
x
O solid solution
with the radiative defect centers, emitted in green, yellow and red spectral regions.

133
The luminescent properties of nanocomposite consisting of ZnO nanotetrapods (NT) and
MgO particles were also investigated. The nanocomposites were prepared from ZnO
tetrapods (300-5000 nm in length and 20-500 nm in diameter) grown by CVD method
(Lyapina et al., 2008) and capped by MgO nanoparticles (3-50 nm) which were obtained by
thermolysis of jellylike product containing mixture of magnesium acetate and sodium
hydroxide alcohol solution (Baranov et al., 2008). The nanocomposite was annealed at 500C
in air. The low temperature annealing was used to prevent the diffusion of Zn
2+
cations to
MgO lattice and vice versa Mg
2+
cations to ZnO lattice and keep the cubic MgO and
hexagonal ZnO structures (Baranov et al., 2008; Panin et al., 2005b).

Fig. 2.12. The normalized PL spectra of (a) ZnO nanotetrapods (solid line) and (b) ZnO
nanotetrapods covered by MgO (dot line).
Fig. 2.12 shows PL spectra of ZnO tetrapods and the tetrapods capped with MgO particles.
The annealed tetrapods demonstrate high intensity of the green luminescence attributed to
the oxygen-vacancy-related defects. The tetrapods annealed with capping MgO
nanoparticles show however the suppressed green emission and the relatively-enhanced
excitonic luminescence (Fig. 2.12, curve (b)). To study the electronic and optical properties of
the samples on nano-scale through the lens detector (TLD) and CL measurements were
carried out (Ryu et al., 2002; Panin et al., 2003a). Figure 2.13 (a) shows an HRSEM image of
the tetrapod and (b) profiles of CL intensity and the TLD signal obtained from the structure
by scanning an electron beam along the A-B line. The enhanced TLD signal indicates the
electron depletion of a ZnO surface with formation of the space charge region. The intensity
of CL from the depletion zone is suppressed significantly (about 25 %).

Fig. 2.13. (a) An HRSEM image of the nanotetrapod structure and (b) profiles of the CL
intensity and the TLD signal from the structure obtained by an electron beam scanning
(along the A-B line in (a)).
The green luminescence from ZnO nanocrystals are generally explained by the formation of
single ionized oxygen vacancies (Vanheusden et al., 1996b). The mechanism of the observed
suppression of the green luminescence from the nanocomposite structure could be

Cathodoluminescence

134
explained by the electron depletion of ZnO nanocrystals due to the band bending on the
interface between ZnO and MgO capping agent. In the part of this depletion region where
the Fermi level E
F
passes below the V
0
+
/V
0
++
energy level, all oxygen vacancies are in the
nonradiative V
0
++
state and the green luminescence from the nanocrystals is suppressed
(Panin et al., 2008). In case of uncapped tetrapods the electron density is relatively high and
the green emission from such samples is well appeared.
3. Tuneable luminescence from nanocrystals and composites
3.1 ZnO Nanocrystals capped by Rh6G
ZnO-based polymer and molecular composites have demonstrated the attractive optical and
mechanical properties. In particular, it is found that capping of ZnO nanocrystals with
polymer, organic dye, etc. can effectively passivate the surface defects and decrease the
surface-related visible emission (Borgohain et al., 1998; Du et al., 2006; Harada et al., 2003;
Tong et al., 2004; Yang et al., 2001). The remarkable increase of the near-band-edge emission
to the visible emission intensity ratio (NBE/VB) in ZnO nanocrystals coated with
Rhodamine 6G (Rh6G) organic dye was reported by (Kurbanov et al., 2007b). These results
show that capping with Rh6G leads not only to passivation of the ZnO nanocrystals surface
but also to formation of dye monomers mostly, due to advanced surface of ZnO
nanocrystals.

Fig. 3.1. RT normalized CL (Left) and PL (Right) spectra for ZnO nanocrystals impregnated
with the various Rh6G concentration solutions: (Left) (a) uncapped, (b) 10
-6
mol%, (c) 10
-5

mol%, (d) 10
-4
mol% and (e) 10
-3
mol% and (Right) (a) uncapped, (b) 10
-4
mol%, (c) 10
-3
mol%
and (d) 10
-2
mol%. The spectra were normalized to (a) UV and (b) visible emission bands.
Fig. 3.1 shows the RT CL and PL spectra of ZnO nanocrystals impregnated with Rh6G
solution of various concentrations. Both the CL and the PL spectra of the uncapped
nanocrystals show two emission peaks the sharp peak at 382 nm and a broad peak at
~510530 nm. The luminescence at 382 nm is attributed to the near-band-edge emission (free
exciton) and the broad emission peak at 510530 nm is contributed to deep level defects,
probably ionized charge states of oxygen vacancies (Studenikin et al., 1998b; VanDijken et
al., 2000a; Van Dijken et al., 2001; Vanheusden et al., 1996a, 1996b). After capping with dye
the emission spectra of ZnO nanocrystals display a new band at 560570 nm. Its intensity
increases with increasing dye concentration and it can be attributed to Rh6G luminescence.
Moreover, at high dye concentrations a second band at 625 nm, which is also related to dye
molecules, appears (Fig. 3.1 (right panel) d). However, the visible emission intensity around
510 nm decreases with increasing dye concentration and at higher dye concentrations, the

135
emission arisen from Rh6G becomes dominant over the visible luminescence of ZnO. The
capping gave rise to increasing the intensity ratio of the UV emission to the visible emission
of the ZnO nanocrystals. After capping, the (NBE/VB) intensity ratio for the CL spectrum
increases from 1.96 to 3, and that for the PL spectrum increases from 1.22 to 3.6.
Improvement of the (NBE/VB) intensity ratio indicates that dangling bonds and defect
states in the surface layer of ZnO nanocrystals are significantly passivated by dye. The
visible luminescence of ZnO nanocrystals located around 510520 nm (green emission) is
affected by the surface (Matsumoto et al., 2002; Norberg & Gamelin, 2005; Ramakrishna &
Ghosh, 2003; Shalish et al. 2004; VanDijken et al., 2000a; Van Dijken et al., 2001). A direct
relationship between the size and the excitonic emission intensity in ZnO crystals in the 40
2500 nm size range was reported (Matsumoto et al., 2002; Shalish et al. 2004; Van Dijken et
al., 2001). Moreover the green emission intensity was reported to display a direct
correlation with surface hydroxide concentrations (Liu et al., 1992; Norberg & Gamelin,
2005; Van Dijken et al., 2000b). On the other hand, it has been found that surface OH groups
can form H-bridges with dye molecules through their OH groups and/or nitrogen (Ying et
al., 1993; Wood & Rabinovich, 1989). The evaporation of the residual ethanol solvent is
accompanied with the increased interaction between dyes and hydroxides. Statistically
neighboring molecules are bonded to the hydroxide groups, resulting in stabilization of dye
molecules at the ZnO surface. The obtained results indicate that Rh6G molecules present as
monomers at ZnO nanocrystals at lower dye concentrations. Fig. 3.2 shows PL spectra for
the ethanol solutions containing Rh6G of 10
-5
, 10
-4
, and 10
-3
mol%. The Rh6G solutions
display only one PL band at ~560 nm resulting from monomers (Innocenzi et al., 1996). With
increasing the dye concentration, this band exhibits redshift, which is attributed to
formation of dimers and increasing their amount (Bojarski, 2002; Innocenzi et al., 1996). The
aggregate formation process is clearly displayed in the PL spectrum of the dye deposited
onto a flat quartz glass surface (Fig. 3.2 (curve d)). The PL spectrum contains peaks near 665,
625 and 530 nm. Using the results (Bojarski, 2002) we assume that the PL bands at 625 and
665 nm are formed due to dimer and aggregate emissions. The broad PL peak at 530 nm is
related to the emission of the quartz glass. PL of organic dye on a quartz glass surface
undergoes a strong quenching effect owing to strong aggregation and as a result, the dye
emission intensity on the quartz glass surface is many orders of magnitude less then that in
the ethanol solutions as well as on the ZnO nanocrystal surface.

Fig. 3.2. Normalized PL spectra of Rh6G in ethanol solution for the different concentrations:
(a) 10
-5
mol%, (b) 10
-4
mol%, (c) 10
-3
mol% and (d) Rh6G deposited onto a flat quartz glass
surface (Kurbanov et al., 2007b).

Cathodoluminescence

136
Based on these results, we can infer that the emission band at 625 nm appearing in the ZnO
nanocrystal PL spectra at higher dye concentrations is caused by dimers and the spectrum
redshift caused by increasing the dye concentration is evidence of the dimer formation
process. However, the dye emission from the capped ZnO nanocrystals displays basically
monomer-like behavior. The large surface to- volume ratio of nanocrystals contributes to
isolating the dye molecules from each other that reduces the formation of dimers or
complexes, especially at low dye concentrations. Similar behavior of materials with
advanced surface structure to prevent the dye dimerization process has been observed in
artificial opal (Kurbanov et al., 2007c), porous aluminum (Levitsky et al., 2002) and mesa
structured titanium dioxide (Vogel et al., 2004) impregnated with dye. The dye covering the
ZnO surface absorbs also the emission at around 500 nm. It is known that Rh6G has strong
absorption bands near 500530 nm and in the UV region (Bojarski, 2002; Ethiraja et al., 2005;
Innocenzi et al., 1996). The findings indicate that the excitation of Rh6G at the surface of
ZnO nanocrystals during our experiment may occur through two channels, which are an
energy transfer from the ZnO deep level emission centers and a direct laser excitation. A
significant increase in the suppression of the green luminescence appears at higher dye
concentrations when the dimer related PL band at 625 nm has appeared. The Rh6G dimers
have two absorption bands, higher intensity one at the short wavelength side of the
monomer band (around 500 nm) and lower intensity one overlapping with the monomeric
one at 560 570 nm, which reabsorbs the emission of monomers. An increase of the dimer
concentration might also lead to an increase in the absorption of the visible emission from
ZnO nanocrystals and a decrease in the luminescence intensity from monomers. The surface
passivation of ZnO nanocrystals due to the organic dye capping improved their UV-to-
visible luminescence intensity ratio. Furthermore, a large surface area of nanocrystals helps
to suppress the aggregate formation process of dye molecules, and Rh6G on ZnO
nanocrystal surface displays a monomer-like emission, position of which depends on the
dye concentration. The results indicate that ZnO nanocrystals capped with Rh6G organic
dye hold promise for potential electro-optic applications.
3.2 CL from PBET structure
Organic semiconductors have the potential to provide a compact, low-cost light-emitting
and laser diodes over a broad range of wavelength throughout the UV-visible spectrum. The
large exciton binding energy in the conjugated polymers (Van der Horst et al., 2002) and
efficient excitonic recombination can be the basis of high quantum yield devices operating at
room temperature. Injecting electrons and holes into the charge states of the conjugated
polymers leads to electroluminescence (Burroughes, 1990). These carriers should be
balanced at the junction of the emitting layer to yield the maximum exciton formation. Light
emission of poly [3-(2-benzotriazolo) ethylthiophene] (PBET) at a different electron beam
current density has been investigated by the CL spectroscopy (Panin et al., 2004a). An
electron beam was used to inject directly electrons and holes in the polymer and induced an
electric field.
PBET was synthesized by (Ahn et al., 2001), to increase quantum efficiency by introducing
benzotriazole, an electron-withdrawing moiety, to the thiophene. Singlet excitons are
usually generated in polymer either by photoexcitation or by electrical injection of negative
and positive charge carriers to the lowest unoccupied molecular orbital (LUMO) and the

137
highest occupied molecular orbital (HOMO), respectively. Due to large exciton binding
energy and the lattice relaxation (Bredas et al., 1996) energy of photons emitted by
polythiophene is usually significantly lower than the HOMOLUMO band gap, which is
around 3.58 eV (Van der Horst et al 2002). Both photoluminescence and electroluminescence
spectra of the polymer show yelloworange luminescence (Ahn et al., 2002).

Fig. 3.3. CL spectra of PBET/ITO structure obtained at 10 kV accelerating voltage of incident
electrons and at different electron beam current densities: (A) 0.4 A/cm
2
, (B) 30 A/cm
2
and
(C) 2 kA/cm
2
(Panin et al., 2004a).
Room temperature CL spectra of the PBET/ITO structure under various electron beam
excitations are shown in Fig. 3.3. Under the low-excitation conditions with an incident beam
current density of 0.4 A/cm
2
the spectrum (Fig. 3.4, A) is similar to the photoluminescence
spectrum previously reported (Ahn et al., 2002). The peak centered at 580 nm with a
shoulder at 620 nm were attributed to the * transition of the conjugated thiophene
segments. Under higher excitation with a beam current density of 30 A/cm
2
(Fig. 3.3, B)
three peaks centered at 340, 565 and 620 nm are revealed. Under the highest excitation used
in our experiment with a beam current density of 2 kA/cm
2
the broad ultraviolet-blue bands
are detected. In addition, a blue shift of green luminescence in the spectrum up to 529 nm
and a red shift of the UV peak to 370 nm are observed. The peak at 347 nm could be
attributed to the direct interband radiative transition. The absorption spectrum of the PBET
samples has a strong absorption band at 282 nm (Ahn et al., 2002).

Fig. 3.4. (Left) CL spectra of the PBET/ITO structure acquired at 300 and 80 K at a 400
A/cm
2
electron beam current density. (Right) CL spectra of the structure obtained after the
electron beam irradiation at the 400 A/cm
2
electron beam current density for different
fluences: (A) 1.5 10
21
, (B) 3.0 10
21
and (C) 5.5 10
21
cm
2
.
A relative intensity of the UVblue band increases with increasing electron beam current
density. Fig. 3.4 (Left) shows CL spectra of the PBET/ITO structure obtained under the high
electron beam excitation at different temperatures. The UV band centered at 350 nm along

Cathodoluminescence

138
with greenorange luminescence bands are clearly observed. The relative intensity of the
bands depends strongly on temperature. At 80 K the polymer structure shows the high
intensity of the UV luminescence, which is substantially higher than that of the green
orange luminescence. Some red shift of a 620 nm peak is also observed. Note that no shifts
for the UV band and no blue band appearances are detected under such excitation
conditions. To investigate the origin of the blue band we measured the luminescence of the
polymer structure after the electron beam irradiation with the high electron beam current
density for various fluences and temperatures. As the fluence of the electron beam
irradiation increases, the intensity of the broad violetblue luminescence at 370470 nm
increases, while an intensity of the greenorange luminescence decreases (Fig. 3.4 (Right)).
The blueviolet band is detected under room temperature excitation, but it has not appeared
under the highest electron beam current density excitation at 80 K. The results indicate that
the structural/electron change in the polythiophene derivative may occur under the room
temperature electron beam irradiation with a high fluence. The most important observation
is the appearance of UV and blueviolet emissions centered at 347 and 420 nm. The wide
spectra with violet and UV emissions have been previously demonstrated in insulating
polymers at high electric fields (Chayet et al. 1997; Massines et al., 1997 ). The high-energy
photon emission in these systems was attributed to the so-called impact ionization, which
occurs when the injected and thermally generated carriers are accelerated by a strong
electric field to cause collision excitations and subsequent radiative recombination.
Theoretical models (Artbauer, 1996) suggest that the number of these collisions increases
with electric field, which can be arisen from the electron beam induced charging. Our
experimental results indicate that an increase of the electron beam current density results in
charging of the polymer and an induced electric field which increases the UV CL intensity.
The charging effect with formation of the built-in electric field in the polymer structure was
clearly observed in a secondary electron mode and could be controlled by the electron beam
current adjustment. The UV luminescence suggests the recombination of hot
nonequilibrium carriers in the presence of the strong electric field. Impact excitation induced
by the electric field produces a large number of excited carriers in excess of equilibrium
(hot carriers). These excited states decay thus approaching their number at equilibrium
via several competing mechanisms. Formation of singlet excitons, which decay radiatively
emitting green luminescence, is one such possible mechanism. However, the strong electric
field inhibits this type of decay by dissociation of the excitons, thereby enhancing the direct
HOMOLUMO interband radiative transition with UV emission. Under such conditions
direct radiative transition is favorable over exciton recombination. The saturation of the
green luminescence at high current density is a further evidence for the partial inhibition of
exciton recombination. The appearance of the blue luminescence at high current densities
could be due to the electron beam induced phase transformation in PBET at room
temperature. When the polymer backbone conformations are coupled with that of the
sidegroups the first-order phase transition is occasionally realized. Such polymers should
show the bistable states, which are separated by a potential barrier in free energy. The phase
conversion between the two states in some temperature region is thermally forbidden but
may be possible by electron beam excitation. In the hysteretic temperature region of the
phase transition the switching of the electronic phase was previously demonstrated by
pulsed photoexcitation of electron-hole pairs in polythiophene (Hosaka et al., 1999). The
photo-induced change in the absorption and luminescence spectra arising from the

139
structural change, which is generally termed photochromism, has been observed for many
photochromic molecules with photoactive -electron units and thin films of the
semiconducting -conjugated polymer (Koshihara et al. 1992). The structural change
induced by the photoexcitation is considered to be minimal with least steric hindrance but
significantly affects the -conjugation length or the electronic structure. The photoexcited
electron-hole pairs or their lattice relaxed analogs (polarons) are responsible as precursory
excitations for the observed photoinduced phase transition. Origin of the blue luminescence
from the polymer might be attributed to the electron beam induced change in the ring
torsion angle by the interchain interaction in a substituted benzotriazole derivative segment.
Stimulated emission from the PBET/ITO structure in a broad range of wavelengths are
observed at high current density (Panin et al., 2004b).

Fig. 3.5. a) RT spectra of stimulated emission (A) and lasing (B) from PBET obtained under
electron beam excitation with different power densities: 1.5 MW/cm
2
(A) and 2 MW/cm
2
(B)
and b) the intensity of emission as a function of the beam excitation density. (Panin et al., 2004b)
Figure 3.5 (a) shows RT spectra of stimulated emission (curve A) and multimode lasing in
broad range of wavelengths (curve B) from the device structure obtained under electron beam
excitation with high power densities. When the electron beam current density increases, the
narrow emission lines appeared and the FWHM strongly decreases. A finely structured
spectrum with features as narrow as 0.5 nm is observed at 2 MW/cm
2
power density. The
narrow lines appear above a threshold power density and indicate stimulated emission. The
dependence of the emission intensity on the pumping is shown in Fig. 3.5 (b). The broad
spontaneous emission depends approximately linearly on excitation density. In contrast,
stimulated emission shows the changeover to superlinear behavior. From the superlinear
dependence we estimate an 1.3 MW/cm
2
threshold power density. The narrow lines
appearing in the spectrum under high excitation are attributed to laser modes. These modes
could be arising from self-assembled cavities formed in the polymer film due to recurrent light
scattering (Cao et al., 1999) by domain boundaries. Small (1020 nm) nanostructures in a
fractal-like organization and elongated self-assembled aggregates of these structures forming
extended domains (1002000 nm) have been previously demonstrated in these polymer films
using scanning tunneling microscopy and near-field optical measurements (Micheletto et al.,
2001). It was shown that these structures are the source of the electroluminescence emission
pattern which was observed due to domain boundaries scattering. Laser emission from these
resonators results in a small number of discrete narrow peaks in the emission spectrum. As the
pump power increases further, the gain increases and it exceeds the loss in the lossier cavities.
Laser oscillation in those random cavities adds more discrete peaks to the emission spectrum.

Cathodoluminescence

140
3.3 ZnO Nanocrystal- PDPV composite
The combination of conjugated polymer and inorganic semiconductor nanoparticles is an
attractive field in fabrication of light-emitting devices. Such nanocomposites allow combining
the efficient luminescence from inorganic semiconductor with good mechanical and optical
properties of polymer films (Friend et al., 1999; Vanheusden et al., 1996b). Poly(4,4-
diphenylene diphenylvinylene), (PDPV) (Eg=3.0 eV) (Feast et al., 1995) is a conjugated
polymer for electroluminescence (EL) in the green region (Cacialli et al., 1997). The oxide
semiconductor ZnO is a promising material for EL in the blue-UV region (Lee et al., 2011;
Mordkovich et al., 2003; Panin et al., 2004c;). The luminescence efficiency of ZnO nanoparticles
can be higher than that of ZnO epitaxial films and the particle-size dependent band gap allows
controlling the excitation spectrum. ZnO nanocrystals doped by Mg, W, Fe, Mn, Y, Eu, V etc.
show the high efficiency in multicolor low-energy CL and are prospective to use in advanced
flat panel displays, solid state lighting applications and smart solid state electronics (Baranov
et al., 2006; Mordkovich et al., 2003; Panin et al., 2007a; Panin et al., 2007b; Panin et al., 2007c).
PDPV (cis/trans~50:50) was prepared by condensation polymerization of 4,4-dibensoyl
biphenyl following the procedure described elsewhere (Feast et al., 1995). The particles were
prepared by dissolving the mixture of analytic-grade ZnCO
3
and (MgCO
3
)4Mg(OH)25H
2
O
(Aldrich) in a mole Zn/Mg ratio in nitric acid and then precipitating by adding a
2M(NH
4
)2CO
3
solution. The precipitate was decomposed in a vacuum furnace at 320 C and
finally oxidized at 550 C in air. The doping effect of ZnO by Mg has been observed after
annealing of ZnO/MgO samples at temperatures above 500 C (Panin et al., 2005b).This
procedure leads to the formation of 1260 nm sized Mg doped ZnO crystals (Panin et al.,
2004c). PDPV was dissolved in chloroform and ZnO:Mg powder was dispersed in
chloroform separately by treatment in an ultrasonic bath for 30 min. The PDPV mixed with
the particles was deposited on a Si/SiO
2
substrate with gold electrodes separated by 5 m
by spin coating at 1000 rpm. The same procedure was used to get separate PDPV and
ZnO:Mg particle films (Fig. 3.6).

Fig. 3.6. An HRSEM image of the ZnO:Mg particle film (Panin et al., 2008).
Figs. 3.7 show CL spectra of the pristine PDPV film and the pristine ZnO:Mg particle film.
PDPV has a well pronounced CL with the maximum emission in the green region, which is
consistent with the luminescence results obtained for this polymer (Cacialli et al., 1997; Feast et
al., 1995). The CL spectrum of ZnO:Mg nanoparticles reveals a more complicated structure with
several peaks at 3.54; 2.83; 2.02 and 1.72 eV, respectively. The intensity of CL at 1.72 eV is higher

141
with respect to other picks, which indicates that the red emission from ZnO:Mg nanoparticles
related to impurities or traps (Falcony et al., 1998; Studenikin et al., 1998a) is the dominant one.

Fig. 3.7. CL spectrum of the PDPV film (Left) (inset shows chemical structure of PDPV) and
the pristine ZnO:Mg particles (right).
3.3.1 Switching of cathodoluminescence from ZnO nanoparticle/polymer
nanostructure by an electric field
The effect of reversible switching between blue-green and red emission in PDPV mixed with
ZnO:Mg nanoparticles by an electric field was reported by (Panin et al., 2005a). Figure 3.8
(a) shows the CL spectrum of the PDPVZnO:Mg composite film without application of an
electric field. The resulting spectrum combines CL from all the materials involved, which
caused the broadening of PDPV spectrum with the pronounced peaks at 3.50, 2.86, and 1.79
eV. The main maximum of this asymmetric spectrum in the blue-green region (2.86 eV)
results from pronounced blue-green emission from both PDPV and ZnO:Mg nanoparticles.
As can be seen from Fig. 3.8 (b), application of a positive bias (+5 V) to the electrode
suppresses the blue-green emission and shifts the emission maximum to the red region. The
effect is found to be reversible with respect to the application of the electric field, namely
then field was off, the emission maximum returned to the blue-green region.

Fig. 3.8. CL of PDPVZnO:Mg composite film (a) without electrodes and (b) with electrodes
biased with (+5 V) or without (0 V) application of an electric field (Panin et al., 2005a).

Cathodoluminescence

142
3.3.2 The mechanism of the electric field-tuneable luminescence from the ZnO
nanoparticle/polymer structure
Before starting the analysis of the luminescence switching mechanism in PDPVZnO:Mg
composite films, we have to note that the similar effect has been found in PPV films doped
with ruthenium dinuclear complex (Welter et al., 2003). The band gap of ruthenium complex
is about 0.5 eV smaller than that of the polymer. Therefore, the mechanism for the formation
of the excited state in the polymer involves the ruthenium complex in a stepwise electron
transfer process. In the case of the PDPV ZnO:Mg composite the band gap of ZnO is higher
than that of the polymer and the latter mechanism of the electric-field switching between
blue-green and red emissions does not work (as well as the mechanism proposed for the
emission from PPVcarbon nanotube composite ( Woo et al., 2000). Fig. 3.9 shows the
positions of the molecular orbitals and band structures of the materials used in our
composite structure. As can be seen, there is a direct contribution to the green emission from
LUMO-HOMO radiative recombination in PDPV and contributions to the blue, green and
red emissions from radiative deep levels recombination in ZnO.

Fig. 3.9. Energy band diagram of a biased Au/PDPV/ZnO:Mg/Au structure (Panin et al.,
2005a).
There are several models to explain the blue-green and red emission in ZnO. The most
acceptable models assume that the defect centers responsible for green luminescence are the
singly ionized oxygen vacancy centers (Vanheusden et al., 1996a, 1996b) or donoracceptor
level transitions (Egelhaaf & Oelkrug, 1996; Reynolds et al., 2001; Studenikin et al., 2002).
The acceptor level (Zn vacancy) is located 2.5 eV below the conduction band edge (Bylander
et al., 1978; Egelhaaf & Oelkrug, 1996), while the donor level (oxygen vacancy) is known as a
shallow level at 0.050.19 eV, leading to an emission band centered around 508540 nm. The
blue-green emission in ZnO might also be associated with a transition within a self-activated
center formed by a double-ionized zinc vacancy (V
Zn
)
-2
and the single-ionized interstitial Zn
+

at the one or two nearest-neighbor interstitial sites (Studenikin et al., 1998c). At last the blue-
green emission can be ascribed to a substitution of Zn by extrinsic impurities such as Cu or
Mg in the crystal lattice (Dingle, 1969; Mordkovich et al., 2003). The blue-green emission in
the doped ZnO particles can be attributed to recombination of V
0

electrons with excited
holes in the valence band (Fig. 3.10), while the red emission in the particles at 670 nm (E
d
=1.84 eV) could originate from the complex defect-related centers such as deep donor and
deep acceptor centers associated with V
Zn
(Fig. 3.10) (Ohtomo et al., 1999; Studenikin et al.,
1998b; Mitra et al., 2001).

143

Fig. 3.10. (Left) Electron energy level diagram for red and blue-green radiation centers in
ZnO. (Right) The energy-band diagrams of small (SP) and large (LP) particles in cross-
sections.
The mechanism for the formation of the excited states in the PDPVZnO:Mg composite
structure implies the presence of channels of radiative recombination, which can be controlled
by electric field. Figure 3.10 (right panel) shows the energy-band diagrams for small (SP) and
large (LP) particles in cross-sections. Application of an electric field causes depletion effects,
which are known to affect the green luminescence of ZnO crystals (Vanheusden et al., 1996a;
1996b). A high density of electrically active surface states enhances the effect of band bending at
a semiconductor surface, which creates an electron depletion region of width W in the particle.
In the part of this region where the Fermi level E
F
passes below the V
0
+
/V
0
++
energy level, all
oxygen vacancies are in the nonradiative V
0
++
state. The CL maximum at ~2.86 eV indicates that
both radiative recombination mechanisms including oxygen-vacancy related centers in ZnO
and direct LUMOHOMO transfers in PDPV coexist at zero electric field. It should be noted
that band bending might be reduced by both polymer ZnO surface passivation and strong
electron beam illumination. This effect results in an attraction of minority carriers to the surface,
where they become trapped, and converts some of the V
0
++
centers to the V
0
+
state. As for the
red emission related to the recombination via impurity-oxygen involved centers in ZnO:Mg, the
radiative recombination mechanism also works and can be recognized as the shoulder of the
main spectra at ~650 nm. Complete band bending suppression at both interfaces in the case of
the AuPDPVZnO:MgAu structure can be obtained by applying a bias voltage to electrodes.
PDPV is a p-type conductor similar to many other conjugated polymers. It means that
application of positive bias to the AuZnO-PDPV increases the barrier and the green emission
from PDPV is reduced because of charge separation at the polymernanocrystal interface
(Greenham et al., 1996). Furthermore, the interface band bending creates the electron depletion
region (W) at the ZnO particle surfaces and the conversion of the radiative V
0
+
centers to the
nonradiative V
0
++
state. As a result, blue emission becomes dominant. This mechanism is
reversible and when the bias becomes zero, the green radiative recombination channel from the
V
0
+
related states as well as PDPV emission should become dominant again in agreement with
our experimental data. Note that the band bending in our experiment does not affect red
emission, which is attributed to the deeper levels. A depletion region width (W) on the surface
of ZnO particles depends on the surface state density and the charge carrier density in particles
as it can be seen from equation 3 :

Cathodoluminescence

144
W=(2
Zn0
V
bi
/eN
d
)
(3)
where
Zn0
- is dielectric constatnt of ZnO, V
bi
potential barrier, e-electron charge, N
d

donor density. The size of particles and their charge density can control the luminescence
spectrum. It is noteworthy that the CL spectrum of the PDPVZnO:Mg film contains peaks
at ultraviolet-blue, blue-green, and red regions. This multicolor CL and the ability to tune
the intensity of the bands by an electric field makes this polymer-nanoparticles composite
promising for the creation of flexible tunable light emitters.
3.4 Optical tuning the luminescence of nanocrystals
The possibility of optical control of luminescence from nanocrystals was investigated by
(Shih et al., 2011; Kurbanov et al., 2008a, 2008b). The visible light illumination effect on near-
band-edge emission intensity from ZnO nanocrystals excited by UV He-Cd laser radiation
(325 nm) was reported by (Kurbanov et al., 2008a). A cw Ar
+
-laser irradiation (488 nm) in
addition to the UV excitation can effectively (up to 75%) reduce the exciton emission from
ZnO. This reduction in integrated intensity (I/I)(quenching) of the near-band-edge
emission is reversible and has a pronounced dependence on temperature and ratio of Ar
+
and He-Cd laser power densities. The mechanism of the observed quenching effect was
proposed to be similar the mechanism for the electric field tuned emission from ZnO
nanocrystals in PDPV (Section 3.3). It implies an appearance of the additional recombination
channel under a visible light illumination due to the recharging of oxygen-vacancy states in
the surface depletion zone. The ZnO nanocrystals investigated in this work have the regular
cone form and the size of 100500 nm (Kurbanov et al., 2007a; Wang et al., 2005). PL studies
were carried out on a SPEX spectrometer equipped with a 0.75 m grating monochromator
using a 50 mW cw He-Cd laser operating at the wavelength of 325 nm as the excitation
source. PL spectra were measured in the backscattering configuration and the registration of
the PL signal was carried out by using a conventional lock-in technique with a mechanical
chopper. A 12 mW cw Ar
+
-laser operating at the wavelength of 488 nm (2.54 eV) was used
as an additional illumination source. Both the laser beams were focused on the sample by
employing quartz lenses. The beam spot size after the lens was estimated to be about 100 m
in diameter and the Ar
+
-laser beams spot covered the He-Cd-lasers spot. In order to vary
the excitation intensity (power density), the spot size was kept constant while the input
power was attenuated by the neutral filters. The corresponding maximum intensities of
laser beams at 325 and 488 nm were estimated to be about 500 and 120 W/cm
2
, respectively.
The samples were mounted on the cold finger of a closed-cycle helium cryostat and the
sample temperature was controlled in the range from 10 K up to room temperature. Figures
3.11 (a) and (b) display the RT PL spectra collected from ZnO nanocrystals under single (He-
Cd laser) (black) and double (He-Cd and Ar
+
-lasers) (red) illuminations at different
excitation power densities of He-Cd laser. The quenching effect depends strongly on the
power-density ratio of the employed lasers (I
488
/I
325
)-with increasing the (I
488
/I
325
)

ratio the
magnitude of the quenching increases. A decrease in power density of the He-Cd laser by 2
orders of magnitude, at the constant Ar
+
-laser intensity of 120 W/cm
2
, results in an increase
in the integrated intensity quenching (I/I) of the UV emission from 0.07 (7%) to 0.14 (14%).
The observed quenching effect is reversible; the PL restores the initial intensity and form
instantly after the turning off the Ar
+
-laser illumination regardless of excitation power
density.

145

Fig. 3.11. (Left) RT PL spectra collected from CSD ZnO nanocrystals under single (He-Cd
laser) (black) and double (He-Cd and Ar
+
-lasers) (red) laser illuminations at different
excitation power densities of He-Cd laser: (a) I
325
~500 W/cm
2
and I
488
~120 W/cm
2
and (b)
I
325
~50 W/cm
2
and I
488
~120 W/cm
2
. (To protect the photomultiplier tube from the scattered
Ar
+
-laser radiation around 488 nm the spectrometers slit was shut down). (Right) PL
spectra collected from CSD ZnO nanocrystals under single (He-Cd laser) (black) and double
(He-Cd and Ar
+
-lasers) (red) laser illuminations at different excitation power densities of
He-Cd laser at 10 K: I
325
~0.25 W/cm
2
and I
488
~120 W/cm
2
.
The UV emission quenching effect was found to depend strongly on temperature and the
laser power density. With the decrease in temperature to 10 K the magnitude of the
quenching increases and under the He-Cd laser power density of 50 W/cm
2
, it reaches to
0.38 (38%). The reducing of the He-Cd laser power density by 2 orders of magnitude at the
fixed Ar
+
-laser power density increases the quenching value up to 0.55 (55%) (Kurbanov et
al., 2008a). The highest effect achieved in these experiments was ~75%. It was obtained at 10
K when the power-density ratio of the Ar
+
and He-Cd lasers was 500 (Fig. 3.11, right panel).
The reversible behavior of the quenching effect remains also at low temperatures. However
the time required reestablishing the initial PL intensity increases with a decrease in
temperature. At 10 K the recovery time takes several minutes. Moreover the recovery time
was found to depend on an illumination time by Ar
+
-laser. The long term illumination
results in prolongation of the recovery time of the PL intensity. The restoration of the PL
initial intensity could be significantly accelerated by using a short-term increasing of
temperature.
3.4.1 Electro-optical processes in nanocrystals under double excitation
The additional Ar
+
-laser illumination of ZnO nanocrystals creates a new partway for excited
charge-carrier recombination, which results in quenching of the near-band-edge emission
intensity. Obviously, the energy of Ar
+
-laser radiation (2.54 eV) is not enough to create new
defects in ZnO lattice. However such low-energy photons can effect on charge states of the
existing defects. As was identified in previous studies, the major defect related PL band at
around 500 nm is originated from oxygen-vacancy center and the quenching effect also was
more pronounced in samples with higher relative intensity of this band. Oxygen vacancies
in ZnO can occur in three different charge states: the V
0
state which has captured two
electrons and is neutral relative to the lattice, the singly ionized V
0
+
state with one electron,
and the V
0
++
state which has no electrons and is doubly positively charged with respect to
the lattice. Only the V
0
+
state is radiative and paramagnetic and consequently observable by
EPR measurements. It is well known that the ZnO particle surface layer contains an electron
depletion region created due to the surface states. The existence of such region results in the

Cathodoluminescence

146
band bending at the surface. In the fraction of the depletion region where the Fermi level
passes below the V
0
+
/V
0
++
energy level, all oxygen vacancies will be in the diamagnetic V
0
++

state while oxygen vacancies in V
0
+
state will exist in the particle core. The first principle
studies have shown (Lany & Zunger, (2005); Kohan et al., 2000; Van de Walle, 2001; Zhang
et al., 2001) that the oxygen vacancy is a negative-U center, i.e., V
0
+
state is deeper (farther
from the conduction band) than either V
0
++
or V
0
for any Fermi-level position. As already
mentioned the intensity of the green luminescence in ZnO correlates very well with the
paramagnetic single-ionized oxygen-vacancy density (Vanheusden et al., 1996a, 1996b). V
0
+

is found to be photosensitive for photons with an energy of >2 eV and this effect was
explained by converting some of the V
0
++
centers to the paramagnetic V
0
+
state. Considering
these results and fact that the photon energy of Ar
+
-laser radiation is 2.54 eV, the observed
quenching effect of the near-band-edge emission from ZnO nanocrystals is well interpreted
in terms of recharging of the oxygen-vacancy centers. Since V
0
++
centers can be easily
formed in the depletion region, consequently their number depends on the depletion region
width (see Eq.3, Sec. 3.3.2). As can be seen, the width of the depletion region is inversely
proportional to the square root of the free-carrier concentration in the nanocrystal and
proportional to the potential at the surface. It is well known that in the case of high
excitation, the surface potential depends on excitation intensity (Studenikin et al., 1998a). At
high intensities the width of the depletion region is smaller than at low intensities. On the
other hand an increase in temperature leads to additional narrowing of the depletion region
due to the increase in the free carrier concentration. The simultaneous action of these factors
(i.e., the increase in excitation intensity and temperature) results in the strong decrease in
the depletion region width, consequently, the concentration of V
0
++
centers. These
speculations are in the good agreement with a strong increase in the quenching effect with a
decrease in excitation power density and temperature. The temperature lowering from 295
to 10 K and a decrease in excitation intensity in 3 orders of magnitude (from 500 to 0.5
W/cm
2
) resulted in a large enhancement of the quenching effect (from 7% up to 53%) at the
fixed Ar
+
-laser power density of 120 W/cm
2
. Even though the Ar
+
-laser illumination
quenches well the UV emission, its impact on the visible emission intensity is observed very
week. Although at room temperature under the double illumination the visible emission
intensity slightly increases, at low temperatures it is rather unremarkable. This circumstance
may be originated from different excitation efficiencies of the UV and visible emissions
depending on temperature and excitation power density. One can see a decrease in
temperature results in an increase in ratio of intensity of the UV emission to that of the
visible emission. It increases from 2.25 to 66.5 with the decrease in temperature from 295 to
10 K at low excitation power density. Such behavior indicates that at low temperatures the
visible emission is excited with relatively low efficiency than UV emission and the effect of
Ar
+
-laser irradiation on the visible PL, probably, also decreases. On the other hand, Ar
+
-laser
illumination could induce mostly the nonradiative states. It is known that there exists
another type of single-ionized oxygen-vacancy-related center: V
0
+
complex, (Vanheusden et
al., 1997) which can be formed in the presence of nearby interstitial oxygen (Frenkel pair)
(Hoffmann & Hahn, 1974). This center unlike the isolated V
0
+
is not a green luminescent site,
even though it shows a somewhat similar nature. Moreover the V
0
+
complex may act as a
quenching center for radiative recombination in ZnO. The CSD technique used to grow ZnO
nanocrystals provides the oxygen rich samples (Kurbanov et al., 2007a). This circumstance
allows us to infer that in the ZnO nanocrystals

147
prepared in the oxygen rich conditions, some of V
0
++
centers under Ar
+
-laser illumination
creates V
0
+
-complex which does not take part in green emission but acts as nonradiative,
quenching center for UV PL. The reversible behavior of the quenching effect and
dependence of the recovery time on temperature and exposure time by the visible light
confirm the proposed recharging model of the oxygen-vacancy-related centers in the ZnO
nanocrystal depletion region. As was reported (Vanheusden 1996a), the EPR signal from V
0
+

decays when illumination was interrupted and at 294 K this decay is instantaneous, but at
150 K it requires several minutes. In our experiments the recovery process of the UV
emission intensity could be also accelerated by a short-term rising of temperature. These
observations indicate that there is some energetic barrier, escaping of which the system can
return into the initial state. The existence of energetic barrier for decay of photogenerated
V
0
+
into the other charge states was predicted by Walle (Van de Walle, 2001) based on the
first principles investigation. For a further understanding of the mechanism of the observed
UV PL band quenching, we investigated the quenching effect dependence on the excitation
power density in details. The integrated PL intensity versus the excitation power density
under (a) the single (He-Cd laser) and (b) the double (He-Cd-and Ar
+
-lasers) illuminations
as well as (c) the quenching (I/I) of the UV emission at 10 K are plotted in Fig. 3.12. The
both PL intensity dependences display a tendency to saturation at high excitation level. The
luminescence intensity I versus excitation power density can be expressed as
I =I
0
(4)
In this relation I
0
is the power density of the excitation laser radiation, is the constant of
proportionality (some authors define it as an emission efficiency), and the exponent
represents the radiative recombination mechanism. For excitonic recombination, 1<<2, for
band-gap emission, i.e., electron-hole bimolecular recombination, ~ 2, and is less than 1
when an impurity is involved in the transitions, as well as for donor-acceptor transitions
(Bergman et al., 2004; Jin et al., 1997). Using Eq. (4) to fit the data of the single and double
laser illuminations we found ~0.61 and ~0.7, respectively. The obtained values of the
exponent are less 1 and they could be interpreted as evidence that the UV emission
originates from native donor-acceptor transitions or impurities. However all results
undoubtedly indicate that UV PL has exciton nature. As is well known, a saturation effect of
the excitonic emission under high-power excitation is not rarely observed phenomenon
(Bergman et al., 2004; Jin et al., 1997). The saturation of the excitonic emission from GaN
powder excited in ambient air at RT was ascribed to a thermally activated nonradiative
process due to laser heating of particles (Bergman et al., 2004). A slow increase and
saturation of the PL intensity of free excitons in a GaInAsSb/GaAlAsSb single-quantum-
well structure under higher excitation intensity observed at 10 K was attributed to the
exciton screening effect (Jin et al., 1997). It seems in our experiment that both suggested
saturation mechanisms are present.
The visible light illumination increases the exponent from 0.61 to 0.70 by of approximately
15%. This growth undoubtedly is not result of a decrease in the laser heating. It originates
rather from a decrease in exciton relative population due to opening an additional channel
of recombination. Although this channel is basically nonradiative, the decrease in exciton
relative population can reduce the exciton screening effect and provide a faster increase in
the PL intensity with excitation power than that in the case of the single He-Cd laser

Cathodoluminescence

148
illumination. It may be expected that use of the observed quenching effect allows effectively
modulate the UV emission from ZnO and an employment of high power lasers with
wavelength in the 450550 nm region could provide it.

Fig. 3.12. Integrated exciton PL intensity (a) without and (b) under Ar
+
-laser illumination as
well as (c) the quenching of the PL intensity as a function of He-Cd laser excitation power
density at 10 K. The dash lines are the fit to a power law (see the text) (Kurbanov et al.,
2008a).
3.4.2 The optical modulation of UV PL from ZnO nanocrystals
The optical modulation of ultraviolet (UV) photoluminescence from ZnO nanocrystals
excited by HeCd laser (325 nm) with visible Ar
+
laser radiation (488 nm) was reported by
(Kurbanov et al., 2008b). The effective reversible quenching of the UV luminescence
intensity was achieved. The quenching efficiency was found to depend on temperature, the
ratio of HeCd and Ar
+
laser intensities, and the frequency of modulation. The observed
quenching effect was used to modulate UV emission by chopped visible Ar
+
laser radiation.
A sufficiently large modulation depth was obtained.
Figure 3.13 (solid line) shows the PL spectrum of ZnO nanocrystals obtained at 10K under
radiations with the HeCd laser. The spectrum displays a dominant near-band-edge
emission peak at 368.6 nm and two weak emission bands at 373.6 and 382.8 nm as well as a
weak visible emission band. The peak around 368 nm is related to the PL emission of a
donor bound exciton (Kang et al., 2005; Look, 2001; zgr et al. 2005). The bands at 373.6
and 382.8 nm, as indicated by arrows in Fig. 3.13, might be attributed to donoracceptor pair
recombination (Tomzig & Helbig, 1976) or exciton phonon replicas (Reynolds & Collins,
1969) or two photon transitions (Studenikin et al., 2000). An additional radiation using the
Ar
+
laser leads to the strong quenching of the UV emission up to approximately 55% with
some changes in deep-level PL [Fig. 3.13 (dashed line)].
The Ar
+
-laser-radiation-induced quenching effect was used to modulate the intensity of the
UV emission from ZnO nanocrystals. For this objective, the ZnO nanocrystals at 10K were
simultaneously excited using an unmodulated cw HeCd laser beam and a chopped Ar
+

laser beam. The power densities of the UV and visible lasers were 0.5 and 120W/cm
2
,
respectively. The UV peak intensity variations at 368.6 nm were observed using a SPEX
monochromator and a photomultiplier tube on a Tektronix oscilloscope. The intensity
variations of the exciton emissions at different modulation frequencies of the Ar
+
laser are
presented in Fig. 3.13 (right panel). The curve (a) (Fig. 3.13. (Right)) represents the dc PL
signal from the ZnO nanocrystals obtained under the single cw HeCd laser excitation. The
curves (b), (c), and (d) are PL signals modulated by Ar
+
laser radiation in various frequency

149
regions: low (20 Hz), middle (200 Hz), and high (400 Hz), respectively. With increasing Ar
+

laser beam modulation frequency, both the amplitude of the PL signal and its mean level
decrease. The curve (e) represents the dc PL signal obtained when nanocrystals were
illuminated with an unmodulated cw Ar
+
laser beam. It was the lowest level that was
achieved under Ar
+
laser radiation. The decrease in the modulation depth of the PL intensity
with increasing frequency at low temperature is caused by the long recovery time of the
quenching effect. As can be seen from Fig. 3.13, the complete modulation {i.e., the change in
the PL intensity from the initial level to the lowest level induced by cw Ar laser radiation
[curve (e)]} was observed at low (10 20 Hz) modulation frequencies. It seems reasonable to
expect that the complete modulation of the UV emission at RT could be obtained at higher
frequencies. However, a high-power Ar
+
laser or a pulse laser operating in the 480 500 nm
range should be employed owing to the small magnitude of the quenching effect at 300 K.

Fig. 3.13. (Left) PL spectra of ZnO nanocrystals obtained at 10K under HeCd laser excitation
(soid line) and combined HeCd and Ar
+
laser radiations (dashed line). (The intensities of the
HeCd and Ar
+
lasers are approximately 0.50 and 120 W/cm
2
, respectively.) The inset shows
the PL spectra in the wavelength range between 365 and 375 nm. (Right) Schematics of PL
signals from ZnO nanocrystals at 10K observed using oscilloscope. The curve (a) indicates the
dc PL signal from ZnO nanocrystals under cw HeCd laser excitation and the curves (b), (c),
and (d) represent the PL signals modulated by Ar
+
laser radiation under cw HeCd laser
excitation at approximately 20, 200, and 450 Hz, respectively. The curve (e) represents the dc
PL signal under both cw HeCd and Ar
+
laser radiations.
4. Conclusion
In this chapter recent progress in multicolor luminescence from semiconductor
nanocrystal composites tunable in an electric field have been highlighted. The effect of an
external electric field on cathodoluminescence from semiconductor nanocrystals and
nanocomposites with different radiative emission rates was described, giving special
emphasis to ZnO nanocrystals in MgO and polymer matrix. The structural and optical
properties ZnO nanocrystals and nanocomposites, including the core/shell nanocrystals
were investigated. An ability to control the relative intensity of the near-band-gap
emission (UV, violet-blue) and the deep-level luminescence (green, red) from ZnO
nanocrystals and polymer composites by an electron-hole pumping and an electric field
was demonstrated. It allows to adjust the luminescence of the composite in a broad visible
wavelength range. The effect of an external electric field on cathodoluminescence

Cathodoluminescence

150
from the nanocrystal/polymer structure was studied, giving special emphasis to the
doped ZnO nanocrystals in Poly(4,4- diphenylene diphenylvinylene) (PDPV) matrix and
PBET/ITO structures. Cathodoluminescence of the polythiophene derivative structure
was investigated under various electron beam current excitations. UV and blue bands in
the spectrum of the polymer structure at the high electron beam current density are
observed. The intensities of these bands increase as an electron beam current density
increases while the greenorange luminescence is saturated. The induced electron beam
field in the PBET/ITO suggests inhibiting the green luminescence by dissociation of the
excitons, thereby enhancing the direct interband radiative transition with UV emission.
The room temperature electron beam irradiation with a large fluence at a high current
density results in the broad blue luminescence that may be attributed to the
electron/structural changes in the polythiophene derivate film under an induced electric
field. These effects could be used to obtain white light luminescence from the
polythiophene derivative composites. Electric field-induced color switching of
cathodoluminescence from ZnO:Mg nanocrystals/PDPV composite from blue-green to
red is considered. The assumed mechanism of electric field-tunable cathodoluminescence
implies the presence of radiative recombination channels, which are sensitive to the
electric field through the band bending at the crystal surface. Optical control
luminescence through the impact of visible Ar
+
-laser illumination (488 nm) on the UV
emission from ZnO nanocrystals excited by He-Cd laser (325 nm) at various excitation
intensities and temperatures has been investigated. It was found that a visible light
illumination simultaneously with UV excitation results in a decrease in the near-band-
edge emission intensity. The experiments on a reversible quenching of the UV near-band-
edge emission under Ar
+
-laser illumination confirm an appearance of the recombination
channel after oxygen-vacancy charging in the ZnO surface depletion zone. The quenching
effect of the UV emission observed in ZnO samples is suggested to be due to the
recharging of oxygen-vacancy states under a visible light irradiation. The strong
quenching of UV luminescence from ZnO nanocrystals under Ar
+
-laser illumination (488
nm) and an opportunity to modulate the UV PL intensity by visible light irradiation was
demonstrated. The quenching effect depends on the intensity ratio of the visible and UV
lasers, temperature, and the relative intensity of the green luminescence band. The highest
quenching effect (75%) was achieved at I
488
/I
325
~500 at 10 K. It was shown that the UV
near-band-edge emission is modulated at frequencies of hundreds of hertz. It is
reasonable to expect that the complete modulation of the UV emission at RT could be
obtained at higher frequencies when using a high-power Ar
+
laser or a pulse laser
operating in the 480 500 nm range. In this context the quenching effect may find
application in ZnO based optoelectronic devices and optical communication systems.
5. Acknowledgment
This work was supported by a National Research Foundation of Korea (NRF) grant funded
by the Ministry of Education, Science and Technology (MEST) No. 2011-0000016 as well as
by the Leading Foreign Research Institute Recruitment Program through NRF funded by
MEST No. 2010-00218, and a Russian Ministry of Science and Education grant No.
02.740.11.5215. Author is grateful to all colleagues who took part in joint work, especially
A.N. Baranov, O.O. Kapitanova, S.S. Kurbanov, A.N. Aleshin, I. A. Khotina, and T.W. Kang.

151
6. References
Ahn S.H., Czae M.Z., Kim E.R., Lee H., Han S.H., Noh J. & Hara M. (2001). Synthesis and
Characterization of Soluble Polythiophene Derivatives Containing Electron-
Transporting Moiety. Macromolecules, 34, pp. 2522-2527.
Ahn T., Lee H. & Han S.-H. Effect of annealing of polythiophene derivative for polymer
light-emitting diodes. Appl. Phys. Lett. 80, pp. 392-394
Aigouy L., Holden T., Pollak F. H., Ledentsov N. N., Ustinov W. M., Kopev P. S. & Bimberg
D. (1997). Contactless electroreflectance study of a vertically coupled quantum dot-
based InAs/GaAs laser structure. Appl. Phys. Lett. 70, pp. 3329-3331.
Andersen K.E.; Fong C.Y. & Pickett W.E. (2002). Quantum confinement in CdSe
nanocrystallites. J.Non. Cryst. Solids 99, pp.1105-1110.
Artbauer J. (1996). Electric strength of polymers. J. Phys. D 29, pp. 446-456.
Bagnall D.M., Chen Y.F., Zhu Z., Yao T., Koyama S., Shen M.Y & Goto T. (1997). Optically
pumped lasing of ZnO at room temperature. Appl. Phys. Lett. 70, pp. 2230-2232.
Baranov A N, Chang C H, Shlyakhtin O A, Panin G N, Kang T W & Oh Y J (2004). In situ
study of the ZnONaCl system during the growth of ZnO nanorods.
Nanotechnology 15, pp. 1613-1619.
Baranov A N, Solozhenko V L, Chateau C, Bocquillon G, Petitet J P, Panin G N, Kang T W,
Shpanchenko R V, Antipov E V & Oh Y J (2005a). Cubic MgxZn1xO wide band
gap solid solutions synthesized at high pressures.: J. Phys.:Condens. Matter 17, pp.
33773384
Baranov A N, Panin G N, Kang Tae Wong & Oh Young-Jei (2005b). Growth of ZnO
nanorods from a salt mixture. Nanotechnology 16, pp. 19181923
Baranov A. N.; Panin G. N.; Yoshimura Masahiro & Oh Young-Jei (2006). Growth and
magnetic properties of Mn and MnSn doped ZnO nanorods. J. Electroceram., 17,
pp. 847-852.
Baranov A. N., Kapitanova O. O., Panin G. N. & Kang T. V. (2008). ZnO/MgO
Nanocomposites Generated from Alcoholic Solutions. Russ. J. of Inorg. Chem., 53
(9), pp. 1366-1370.
Baranov A. N.; Kurakevych O. O.; Tafeenko V. A.; Sokolov P. S.; Panin G. N. &
Solozhenkopp V. L. (2010). High-pressure synthesis and luminescent properties of
cubic ZnO/MgO nanocomposites. J. Appl. Phys. 107, pp. 073519-1-073519-5.
Bergman L., Chen X. B., Morrison J. L., Huso J. & Purdy A. P. (2004). Photoluminescence
dynamics in ensembles of wide-band-gap nanocrystallites and powders. J. Appl.
Phys. 96, pp. 675-682.
Borgohain K. & Mahamuni S. (1998). Luminescence behaviour of chemically grown ZnO
quantum dots. Technol. 13, pp. 1154-1157.
Bojarski P. (2002). Concentration quenching and depolarization of rhodamine 6G in the
presence of fluorescent dimers in polyvinyl alcohol films. Chem. Phys. Lett. 278,
pp. 225-232.
Bredas J.-L., Cornil J. & Heeger A.J. (1996). The exciton binding energy in luminescent
conjugated polymers. Adv. Mater. 8, pp. 447-452.
Bredas J. L., Logdlund M. & Salaneck W. R. (1999). Electroluminescence in conjugated
polymers Nature, 397, pp. 121-128.
Bylander E. G. (1978). Surface effects on the lowenergy cathodoluminescence of zinc oxide.
J. Appl. Phys. 49, pp. 1188-1195.

Cathodoluminescence

152
Burroughes J.M., Burroughes J. H., Bradley D. D. C., Brown A. R., Marks R. N., Mackay K.,
Friend R. H, Burns P. L. & Holmes A. B. (1990). Nature (London) 347, pp. 539-541.
Brus L.E. (1984). Electronelectron and electronhole interactions in small semiconductor
crystallites: The size dependence of the lowest excited electronic state. J. Chem.
Phys. 80. pp. 4403-4409.
Brus L.E. (1986). Electronic wave functions in semiconductor clusters: experiment and
theory. J. Phys. Chem. 90, p. 2555-2560.
Cacialli F., Friend R. H., Haylett N., Daik R., Feast W. J., Dos Santos D. A. & Bredas J. L.,
(1997). Efficient green light emitting diodes from a phenylated derivative of poly(p-
Phenylene-Vinylene). Synth. Metals, 84, pp. 643-644.
Cao B. Q., Lorenz M., Rahm A., von Wenckstern H., Czekalla C., Lenzner J., Benndorf G. &
Grundmann M. (2007). Phosphorus acceptor doped ZnO nanowires prepared by
pulsed-laser deposition. Nanotechnology 18, pp. 455707.
Cao H., Zhao Y. G., Ho S. T., Seelig E. W., Wang Q. H. & Chang R. P. H. (1999). Random
Laser Action in Semiconductor Powder. Phys. Rev. Lett. 82, pp. 2278-2281.
Chayet H; Pogreb R. & Davidov D. (1997). Transient UV electroluminescence from poly(p-
phenylenevinylene) conjugated polymer induced by strong voltage pulses. Phys.
Rev. B 56 (20), pp. 12702-12705.
Chen I. J.; Chen T. T.; Chen Y. F. & T. Y. Lin (1996). Nonradiative traps in InGaN/GaN
multiple quantum wells revealed by two wavelength excitation. Appl. Phys. Lett.
89, 142113-142115.
Chen W., Wang Z. G., Lin Z. J. & Lin L. (1997). Absorption and luminescence of the surface
states in ZnS nanoparticles J. Appl. Phys. 82, p. 3111-3115.
Cheng W.; Wu P.; Zou X. & Xiao T. (2006). Study on synthesis and blue emission mechanism
of ZnO tetrapodlike nanostructures. J. Appl. Phys. 100, pp. 054311(4).
Combescot M, Combestcot R & Roulet B (2001). The exciton dead layer revisited. Eur. Phys.
J. B 23 139-151.
Dingle R. (1969). Luminescent Transitions Associated With Divalent Copper Impurities and
the Green Emission from Semiconducting Zinc Oxide. Phys. Rev. Lett. 23, pp. 579-
581.
Djurisic A. B., Leung Y. H., Tam K. H., Ding L., Ge W. K., Chen H. Y. & Gwo S. (2006).
Green, yellow, and orange defect emission from ZnO nanostructures: Influence of
excitation wavelength. Appl. Phys. Lett. 88, pp. 103107(3).
Du X.W., Fu Y.S., Sun J., Han X. & Liu J. (2006). Complete UV emission of ZnO nanoparticles
in a PMMA matrix. Semicond. Sci. Technol. 21, pp. 1202-1206.
Efros A.L. & Rosen M. (2000). The electronic structure of semiconductor nanocrystals. Annu.
Rev. Mater. Res, 30 pp. 475-521.
Egelhaaf H. J. & Oelkrug D. (1996). Luminescence and nonradiative deactivation of excited
states involving oxygen defect centers in polycrystalline ZnO. J. Cryst. Growth 161,
pp. 190-194.
Erhart P., Albe K. and Klein A. (2006).First-principles study of intrinsic point defects in ZnO:
Role of band structure, volume relaxation, and finite-size effects. Phys. Rev. B 73,
pp. 205203(9).
Ethiraja A.S., Hebalkara N., Kharrazia Sh., Urban J., Sainkar S.R., Kulkarn S.K. (2005).
Photoluminescent core-shell particles of organic dye in silica. J. Lumin. 114, pp. 15-
23.

153
Falcony C., Ortiz A., Garcia M. & Helman J. S. (1988).Photoluminescence characteristics of
undoped and terbium chloride doped zinc oxide films deposited by spray
pyrolysis.J. Appl. Phys. 63, pp. 2378-2388.
Fallert J., Hauschild R., Stelzl F., Urban A., Wissinger M., Zhou H., Klingshirn C. & Kalt H.,
(2007). Surface-state related luminescence in ZnO nanocrystals. J. Appl. Phys. 101,
pp. 073506(4).
Feast W. J., Millichamp I. S., Friend R. H., Horton M. E., Phillips D., Rughooputh S. D. D. V.
& Rumbles G. (1985). Optical absorption and luminescence in poly(4,4-
diphenylenediphenylvinylene). Synth. Metals 10, pp. 181-191.
Fonoberov V. A., Alim K. A., Balandin A. A., Xiu F. & Liu J. (2006). Photoluminescence
investigation of the carrier recombination processes in ZnO quantum dots and
nanocrystals. Phys. Rev. B 73, pp. 165317(9).
Fonoberov V A & Balandin A. A. (2004). Radiative lifetime of excitons in ZnO nanocrystals:
The dead-layer effect. Phys. Rev. B 70, pp. 195410(5).
Friend R. H., Gymer R. W., Holmes A. B., Burroughes J. H., Marks R. N., Taliani C., Bradley
D. D. C., Dos Santos D. A., Bredas J. L., Logdlund M. & Salaneck W. R. (1999).
Electroluminescence in conjugated polymers. Nature, 397, pp. 121-128.
Fu D. J., Panin G. N. & Kang T. W. (2003). GaN Pyramids Prepared by Photo-Assisted
Chemical Etching. Journal of the Korean Physical Society, Vol. 42, pp. S611-S613.
Fu Dejun; Park Young Shin; Panin Gennady N. & Kang Tae Won (2005). Formation of
Hexagonal GaN Pyramids by Photo Assisted Electroless Chemical Etching. Jap.
Journal of Applied Phys, Vol.44, No.11, pp. L342-L344.
Gaspar C., Costa F. & Monteiro T. (2001). Optical characterization of ZnO. J. Mater. Sci.:
Mater. Electron. 12, pp. 269-271.
Gill B. & Kavokin A. V. (2002). Giant exciton-light coupling in ZnO quantum dots. Appl.
Phys. Lett. 81, pp. 748-750.
Greenham N. C., Peng Xiaogang & Alivisatos A. P. (1996). Charge separation and transport
in conjugated-polymer/semiconductor-nanocrystal composites studied by
photoluminescence quenching and photoconductivity. Phys. Rev. B 54, pp. 17628-
17637.
Hagn M., Zrenner A., Bhm G. & Weimann G. (1995). Electric-field-induced exciton
transport in coupled quantum well structures, Appl. Phys. Lett. 67 (2), pp. 232-234.
Harada Y. & Hashimoto S. (2003). Enhancement of band-edge photoluminescence of bulk
ZnO single crystals coated with alkali halide. Phys. Rev. B 68, pp. 045421(4).
He H. P., Tang H. P., Z. Z. Ye, Zhu L. P., Zhao B. H., Wang L. & Li X. H. (2007).
Temperature-dependent photoluminescence of quasialigned Al-doped ZnO
nanorods. Appl. Phys. Lett. 90, pp. 023104(3).
Hirai T., Ohno N., Harada Y., Horii T., Sawada Y. & Itoh T. (2008). Spatially resolved
cathodoluminescence spectra of excitons in a ZnO microparticle. Appl. Phys. Lett.
93, pp. 041113(3).
Hoffmann K. & Hahn D. (1974). Electron Spin Resonance of Lattice Defects in Zinc Oxide.
Phys. Status Solidi A 24, pp. 637-648.
Hosaka N., Tachibana H., Shiga N., Matsumoto M., & Tokura Y. (1999). Photoinduced Phase
Transformation in Polythiophene. Phys. Rev. 82 , pp. 1672-1675.
Hosono Eiji, Fujihara Shinobu, Kimura Toshio & Imai Hiroaki (2004). Non-Basic Solution
Routes to Prepare ZnO Nanoparticles. J. Sol-Gel Sci. Technol. 29, pp. 71-79.
Hyberstsen M.S. (1994). Absorption and emission of light in nanoscale silicon structures.
Phys. Rev. Lett. 72, p. 1514-1517.

Cathodoluminescence

154
Hua G., Zhang Y., Ye Ch., Wang M. & Zhang L. (2007).Controllable growth of ZnO
nanoarrays in aqueous solution and their optical properties. Nanotechnology 18,
pp. 145605
Hur T B, Hwang Y H & Kim H. K. (2005). Quantum confinement in VolmerWeber-type
self-assembled ZnO nanocrystals. Appl. Phys. Lett. 86, pp. 193113(3).
Innocenzi P., Kozuka H. & Yoko T. (1996). Dimer-to-monomer transformation of rhodamine
6G in solgel silica films. J. Non-Crystal. Sol. 201, pp. 26-36.
Yamamoto A., Kido T., Goto T., Chen Y., Yao T. & Kasuya A. (1999). Dynamics of
photoexcited carriers in ZnO epitaxial thin films. Appl. Phys. Lett. 75, pp. 469-471.
Jeong S.H., Kim B.S. & Lee B.T. (2003). Photoluminescence dependence of ZnO films grown
on Si(100) by radio-frequency magnetron sputtering on the growth ambient. Appl.
Phys. Lett. 82, pp. 2625-2627.
Jin Sh., Zheng Y. & Li A. (1997). Characterization of photoluminescence intensity and
efficiency of free excitons in semiconductor quantum well structures. J. Appl. Phys.
82, pp. 3870-3873.
Johnson P.D. (1954). Some Optical Properties of MgO in the Vacuum Ultraviolet. Phys. Rev.
94, pp. 845-846.
Yu P., Tang Z.K., Wong G.K.L., Kawasaki M., Ohtomo A., Koinuma H. & Segawa Y. (1997).
Ultraviolet spontaneous and stimulated emissions from ZnO microcrystallite thin
films at room temperature. Solid State Commun. 103, pp. 459-463.
Kang T. W.; Yuldashev Sh. & Panin G. N. (2006). Electrical and optical properties of ZnO
thin films and nanostructures. Chapter 4, In Handbook of Semiconductor
Nanostructures and Nanodevices, edited by A. A. Balandin and K. L. Wang,
American Scientific, Los Angeles.
Kasai P. H. (1963). Electron Spin Resonance Studies of Donors and Acceptors in ZnO. Phys.
Rev. 130, pp. 989-995.
Klik M. A. J.; Gregorkiewicz T.; Yassievich I. N.; Ivanov V. Yu. & Godlewski M. (2005).
Terahertz modulation of the blue photoluminescence in ZnSe. Phys. Rev. B 72, pp.
125205(5).
Kohan A. F., Ceder G., Morgan D. & Van de Walle Chris G. (2000). First-principles study of
native point defects in ZnO. Phys. Rev. B 61, pp. 15019-15027.
Kroger E. A. & Vink H. J. (1954). The Origin of the Fluorescence in SelfActivated ZnS, CdS,
and ZnO. J. Chem. Phys. 22, pp. 250-252.
Koshihara S., Tokura Y., Takeda K. & Koda T. (1992). Reversible photoinduced phase
transitions in single crystals of polydiacetylenes. Phys. Rev. Lett. 68, pp. 1148-1151.
Kumar B., Gonga H., Vicknesh S., Chua S.J. & Tripathy S. (2006). Luminescence properties of
ZnO layers grown on Si-on-insulator substrates. Appl. Phys. Lett. 89, pp. 141901(3).
Kurbanov S., Panin G., Kim T. W. & T. W. Kang (2007a). Thermo- and Photo-annealing of
ZnO Nanocrystals. Jpn. J. Appl. Phys., Part 1 46, pp. 4172-4174
Kurbanov S.S., Panin G.N., Kim T.W. & Kang T.W. (2007b). Luminescence of ZnO
nanocrystals capped with an organic dye. Optics Communications 276, pp. 127
130.
Kurbanov S.S., Panin G. N., Park Y. S., Kang T. W. & Kim T. W. (2007c). Photo- and
Cathodoluminescence Studies of ZnO-Filled Opal Nanocomposites. Journal of the
Korean Physical Society, Vol. 50, No. 3, pp. 617-621.
Kurbanov S. S., Panin G. N., Kim T. W. & Kang T. W. (2008a). Impact of visible light
illumination on ultraviolet emission from ZnO nanocrystals. Phys. Rev. B 78, pp.
045311-1-045311-6

155
Kurbanov Saidislam, Panin Gennady, Kang Tae Won & Kim Tae Whan (2008b). Modulation
of Excitonic Emission from ZnO Nanocrystals by Visible Light Illumination. Jap. J.
Appl. Phys. Vol. 47, No. 5, pp. 37603762.
Kurbanov S.S.; Panin G.N.; Kim T.W. & Kang T.W. (2008c). ZnO filled opal arrays: Photo-
and cathodoluminescence studies. Solid State Communications, 145 (11-12), pp.
577-581.
Kurbanov S.; Panin G. & Kang T.M. (2009a). Spatially resolved investigations of the
emission around 3.31 eV (A-line) from ZnO nanocrystals. Appl. Phys. Lett. 95, pp.
211902-1- 211902-3.
Kurbanov S.; Panin G.N.; Kim T.W. & Kang T.W. (2009b). Strong violet luminescence from
ZnO nanocrystals grown by the low-temperature chemical solution deposition.
Journal of Luminescence 129, pp. 10991104.
Lany S. & Zunger A. (2005). Anion vacancies as a source of persistent photoconductivity in
II-VI and chalcopyrite semiconductors. Phys. Rev. B 72, pp. 035215(13).
Lee Sang Wuk; Cho Hak Dong; Panin Gennady & Kang Tae Won (2011). Vertical ZnO
nanorod/Si contact light-emitting diode. Appl. Phys. Lett. 98, pp. 093110-1-093110-
3
Leiter F. H.; Alves H. R.; Hofstaetter A.; Hofmann D. M. & Meyer B. K. (2001). The Oxygen
Vacancy as the Origin of a Green Emission in Undoped ZnO. Phys. Status Solidi B
226, pp. R4-R5.
Levitsky I.A., Liang J. & Xu J.M. (2002). Highly ordered arrays of organicinorganic
nanophotonic composites. Appl. Phys. Lett. 81, pp. 1696-1698.
Liu B & Zeng H C (2004) Room Temperature Solution Synthesis of Monodispersed Single-
Crystalline ZnO Nanorods and Derived Hierarchical Nanostructures. Langmuir 20,
pp. 4196-4204.
Liu M., Kitai A. H. & Mascher P. (1992). Point defects and luminescence centres in zinc oxide
and zinc oxide doped with manganese. J. Lumin. 54, pp. 35-42.
Look, C. (2001) Recent advances in ZnO materials and devices. Mater. Sci. Eng., B 80, pp.
383-387.
Look D. C., Reynolds D. C., Litton C. W., Jones R. L., Eason D. B. & Gantwell G. (2002).
Characterization of homoepitaxial p-type ZnO grown by molecular beam epitaxy.
Appl. Phys. Lett. 81, pp. 1830-1832.
Lyapina O. A., Baranov A. N., Panin G. N., Knotko A. V. & Kononenko O. V. (2008).
Synthesis of ZnO Nanotetrapods. Inorganic Materials, Vol. 44, No. 8, pp. 845851.
Makino T., Tamura K., Chai C.H., Segawa Y., Kawasaki M., Ohtomo A. & Koinuma H.
(2002). Radiative recombination of electronhole pairs spatially separated due to
quantum-confined Stark and FranzKeldish effects in ZnO/Mg
0.27
Zn
0.73
O quantum
wells. Appl. Phys. Lett. 81, pp. 2355-2357.
Mass J., Avella M., Jimnez J., Rodrguez A., Rodrguez T., Callahan M., Bliss D. & Wang B.
(2008). Cathodoluminescence study of ZnO wafers cut from hydrothermal crystals.
J. Cryst. Growth 310, pp. 1000-1005.
Massines F., Tiemblo P., Teyssedre G. & Laurent C. (1997),On the nature of the luminescence
emitted by a polypropylene film after interaction with a cold plasma at low
temperature. J. Appl. Phys. 81, pp. 937-943.
Matsumoto T., Kato H., Miyamoto K., Sano M. & Zhukov M. (2002). Correlation between
grain size and optical properties in zinc oxide thin films. Appl. Phys. Lett. 81, pp.
1231-1233.

Cathodoluminescence

156
Meulenkamp E. A. (1998). Synthesis and Growth of ZnO Nanoparticles. J. Phys. Chem. B
102, pp. 5566-5572.
Meulenkamp E. A. (1998). Size Dependence of the Dissolution of ZnO Nanoparticles. J.
Phys. Chem. B 102, pp. 7764-7769.
Meyer B. K., Alves H., Hofmann D. M., Kriegseis W., Forster D., Bertram F., Christen J.,
Hoffmann A., Straburg M., Dworzak M., Haboeck U. & Rodina A. V. (2004).
Bound exciton and donoracceptor pair recombinations in ZnO. Phys. Status Solidi
B 241, pp. 231-260.
Micheletto Ruggero Yoshimatsu Nobuki, Yokokawa Masatoshi, An Taekyung, Lee Haiwon
& Okazaki Satoshi (2001). Optical study of a polymeric LED with a nano-sized
electrode realized by a modified SNOM setup. Opt. Commun. 196, pp. 47-53.
Minne S.C., Manalis S.R. & Quate C.F. (1995). Parallel atomic force microscopy using
cantilevers with integrated piezoresistive sensors and integrated piezoelectric
actuators. Appl. Phys. Lett. 67, pp. 3918-3920.
Mitra A., Thareja R. K., Ganesan V., Gupa A., Sahoo P. K. & Kulkarni V. N., (2001). Synthesis
and characterization of ZnO thin films for UV laser. Appl. Surf. Sci. 174, pp. 232-
239.
Mordkovich V. Z., Hayashi H., Haemori M., Fukumura T. & Kawasaki M. (2003). Discovery
and Optimization of New ZnO-Based Phosphors Using a Combinatorial Method.
Adv. Funct. Mater. 13, pp. 519-524.
Motyka M., Sek G., Kudrawiec R., Misiewicz J., Li L. H. & Fiore A. (2006). On the
modulation mechanisms in photoreflectance of an ensemble of self-assembled
InAs/GaAs quantum dots. J. Appl. Phys. 100, pp. 073502(5).
Norberg N.S.& Gamelin D.R. (2005). Influence of Surface Modification on the Luminescence
of Colloidal ZnO Nanocrystals. J. Phys. Chem. B 109, pp. 20810-20816.
Ohtomo A., Kawasaki M., Koida T., Masubuchi K., Koinuma H., Sakurai Y., Yoshida Y.,
Yasuda T. & Segawa Y. (1998). Mg
x
Zn
1x
O as a IIVI widegap semiconductor alloy.
Appl. Phys. Lett. 72, pp. 2466-2468.
Ohtomo A., Shiroki R., Ohkubo I., Koinuma H. & Kawasaki M. (1999). Thermal stability of
supersaturated Mg
x
Zn
1x
O alloy films and Mg
x
Zn
1x
O/ZnO heterointerfaces. Appl.
Phys. Lett. 75, pp. 4088-4090.
Ohtomo A., Tamura K., Kawasaki M., Makino T., Segawa Y., Tang Z.K., Wong G.K.L.,
Matsumoto Y., & Koinuma H. (2000) Room-temperature stimulated emission of
excitons in ZnO/(Mg,Zn)O Superlattices. Appl. Phys. Lett. 77, pp. 2204-2206.
zgr .; Alivov Ya. I.; Liu C.; Teke A.; Reshchikov M. A.; Doan S.; Avrutin V.; Cho S.-J. &
Morko H. (2005). A comprehensive review of ZnO materials and devices. J. Appl.
Phys. 98, pp. 041301(103).
Panin G. N.; Kang T.W. & Jang M. S. (2003a). Spatially Resolved Study of Luminescent and
Electrical Properties of ZnO/GaAs Structures. Journal of the Korean Physical
Society, Vol. 42, pp. S357-S360.
Panin G.N.; Kang T.W.; Kim T.W.; Park S.H.; Si S.M.; Ryu Y.S. & Jeon H.C. (2003b).
Semiconductor quantum dots created by postgrowth treatment. Physica E 17, pp.
484 488
Panin G.N., Kang T.W. & Lee H. (2004a). Light emission from the polythiophene
derivative/ITO structure under electron beam excitation. Physica E 21, pp. 1074
1078.

157
Panin Gennady N., Kang Tae Won & Lee Haiwon (2004b). Electron beam induced light
emission from the polythiophene derivative/ITO structure. phys. stat. sol. (b) 241,
No. 12, pp. 28622865
Panin G. N., Baranov A. N., Oh Young-Jei & Kang Tae Won (2004c). Luminescence from
ZnO/MgO nanoparticle structures prepared by solution techniques. Curr. Appl.
Phys. 4, pp. 647-650
Panin G. N.; Park Y. S.; Kang T. W.; Kim T. W.; Wang K. L. & Bao M. (2004d). Self-
Assembled GaN Quantum Dots in GaN/AlxGa1-xN Structures Grown by PAMBE.
Journal of the Korean Physical Society, Vol. 45, pp. S840-S843.
Panin G. N., Kang T.W., Aleshin A. N., Baranov A. N., Oh Y.-J. & Khotina I. A. (2005a).
Electric field switching between blue-green and red cathodoluminescence in
poly.4,4- diphenylene diphenylvinylene. mixed with ZnO:Mg nanoparticles. Appl.
Phys. Lett. 86, pp. 113114(3).
Panin G.N., Baranov A.N., Oh Y.-J., Kang T.W. & Kim T.W. (2005b). Effect of thermal
annealing on the structural and the optical properties of ZnO/MgO nanostructures.
Journal of Crystal Growth 279, pp. 494500
Panin G. N.; Park Y. S.; Kang T. W.; Kim T. W.; Wang K. L. & Bao M. (2005c).
Microstructural and optical properties of self-organized GaN quantum-dot
assemblies. J. Appl. Phys., 97, pp. 043527.
Panin Gennady. N.; Baranov Andrey. N.; Kononenko Oleg. V.; Dubonos Sergey. V. & Tae
Won Kang (2007a). Resistance Switching Induced by an Electric Field in ZnO:Li, Fe
Nanowires. AIP Conf. Proc. 893, pp. 743 -744.
Panin, G.N., Baranov, A.N., Kang, T.W., Kononenko, O.V., Dubonos, S.V., Min S.K. & Kim
H.J. (2007b). Electrical and magnetic properties of doped ZnO nanowires. MRS
Symp. Proc, 957, pp. 49-54.
Panin G.N.; Baranov A.N.; Kang T.W.; Min S.K. & Kim H.J. (2007c). Spatially-resolved study
of magnetic properties of Mn-doped ZnO quantum wires. Journal of the Korean
Physical Society 50 (6), pp. 1711-1715.
Panin G. N.; Kim H. J.; Kim S. Y.; Jung J. H.; Kim T. W.; Jeon H. C.; Kang T. W. & Kim M. D.
(2007d). Formation and Optical Properties of ZnSe Self-assembled Quantum Dots
in Cl-doped ZnSe Thin Films Grown on GaAs (100) Substrates. Solid State
Phenomena, 124-126, pp. 567.
Panin Gennady N., Baranov Andrey N., Khotina Irina A. & Kang Tae W. (2008).
Luminescent Properties of ZnO/MgO Nanocrystal/Polymer Composite structure.
Journal of the Korean Physical Society, Vol. 53, No. 5, pp. 2943-2946.
Pekar S I 1958 Sov. Phys.JETP 6, pp. 785
Pesika S., Stebe K. J. & Searson P. C. (2003). Determination of the Particle Size Distribution of
Quantum Nanocrystals from Absorbance Spectra. Adv. Mater. 15, pp. 1289-1291.
Raghavan S., Hajra J.P., Iyengar G. N. K. & Abraham K.P. (1991). Terminal solid solubilities
at 9001000C in the magnesium oxide-zinc oxide system measured using a
magnesium fluoride solid-electrolyte galvanic cell. Thermochim. Acta 189, pp. 151-
158.
Ramakrishna G., G. & Ghosh H.N. (2003). Effect of Particle Size on the Reactivity of
Quantum Size ZnO Nanoparticles and Charge-Transfer Dynamics with Adsorbed
Catechols. Langmuir 19, pp. 3006-3012.
Reynolds D. C. & Collins T. C. (1969). Excited Terminal States of a Bound Exciton-Donor
Complex in ZnO. Phys. Rev. 185, pp. 1099-1103.

Cathodoluminescence

158
Reynolds D. C., Look D. C. & Jogai B. (2001). Fine structure on the green band in ZnO. J.
Appl. Phys. 89, pp. 6189-6191.
Riehl N. & Ortman O. Z. (1952) Elektrochem. 60, pp. 149 [in German].
Roessler D.M. & Walker W.C. (1967). Electronic Spectrum and Ultraviolet Optical Properties
of Crystalline MgO. Phys. Rev. 159, pp. 733-738.
Rowland G. L., Hosea T. J. C., Malik S., Childs D. & Murray R. (1998). A photomodulated
reflectance study of InAs/GaAs self-assembled quantum dots. Appl. Phys. Lett. 73,
pp. 3268-3270.
Ryu M. K.; Lee S.; Jang M. S.; Panin G. N. & Kang T. W. Kang (2002). Postgrowth annealing
effect on structural and optical properties of ZnO films grown on GaAs substrates
by the radio frequency magnetron sputtering technique. J. Appl. Phys., Vol. 92, No.
1, pp. 154-158.
Sakohara S., Ishida M. & Anderson M. A. (1998). Visible Luminescence and Surface
Properties of Nanosized ZnO Colloids Prepared by Hydrolyzing Zinc Acetate. J.
Phys. Chem. B 102, pp. 10169-10175.
Schmeller A., Hansen W., and Kotthaus J. P. Trinkle G. & Weimann G. (1994) Franz-Keldysh
effect in a two-dimensional system. Appl. Phys. Lett. 64 (3), pp. 330-332.
Shalish I., Temkin H., H. & Narayanamurti V. (2004).Size-dependent surface luminescence
in ZnO nanowires. Phys. Rev. B 69, pp. 245401(4).
Shannon R.D. (1976). Acta Crystallogr., Sect. A: Cryst. Phys., Diffr., Theor. Gen. Crystallogr.
32, pp. 751.
Sharma A.K., Narayan J., Muth J.F., Teng C.W., Jin C., Kvit A., Kolbas R.M. & Holland O.W.
(1999). Optical and structural properties of epitaxial Mg
x
Zn
1x
O alloys. Appl. Phys.
Lett. 75, pp. 3327-3329.
Spanhel L. & Anderson M. A. (1991). Semiconductor clusters in the sol-gel process:
quantized aggregation, gelation, and crystal growth in concentrated zinc oxide
colloids. J. Am. Chem. Soc. 113, pp. 2826-2833.
Shih H. Y.; Chen Y. T.; Huang N. H.; Wei C. M. & Chen Y. F. (2011)Tunable
photoluminescence and photoconductivity in ZnO one-dimensional nanostructures
with a second below-gap beam. J. Appl. Phys. 109, 103523-1-5.
Sun H.D., Makino T., Tuan N.T., Segawa Y., Tang Z.K., Wong G.K.L., Kawasaki M., Ohtomo
A., Tamura K. Koinuma H. (2000). Stimulated emission induced by excitonexciton
scattering in ZnO/ZnMgO multiquantum wells up to room temperature. Appl.
Phys. Lett. 77, pp. 4250-4252.
Studenikin S. A., Golego N. & Cocivera M. (1998a). Optical and electrical properties of
undoped ZnO films grown by spray pyrolysis of zinc nitrate solution. J. Appl.
Phys. 83, pp. 2104-2111.
Studenikin S. A., Golego N. & Cocivera M. (1998b). Fabrication of green and orange
photoluminescent, undoped ZnO films using spray pyrolysis. J. Appl. Phys. 84, pp.
2287-2295.
Studenikin S. A., Golego N. & Cocivera M. (1998c). Density of band-gap traps in
polycrystalline films from photoconductivity transients using an improved Laplace
transform method. J. Appl. Phys. 84, pp. 5001-5004.
Studenikin S. A., Cocivera M., Kellner W. & Pascher H. (2000). Band-edge
photoluminescence in polycrystalline ZnO films at 1.7 K. J. Lumin. 91, pp. 223-232.
Studenikin S. A. & M. Cocivera (2002). Time-resolved luminescence and photoconductivity
of polycrystalline ZnO films. J. Appl. Phys. 91, pp. 5060-5065.

159
Takagaki Y., Wiebicke E., Riedel A., Ramsteiner M., Kostial H., Hey R. & Ploog K. H. (2003).
Hybrid optical modulator based on surface acoustic waves fabricated using imprint
lithography and the epitaxial lift-off technique. Semicond. Sci. Technol. 18, pp. 807-
811.
Teke A., zgr ., Doan S., Gu X. & Morko H. (2004). Excitonic fine structure and
recombination dynamics in single-crystalline ZnO.Phys. Rev. B 70, pp. 195207(10).
Tokumoto M. S., Pulcinelli S. H., Santilli C. V. & Valrie Briois (2003). Catalysis and
Temperature Dependence on the Formation of ZnO Nanoparticles and of Zinc
Acetate Derivatives Prepared by the SolGel Route. J. Phys. Chem. B 107, pp. 568-
574.
Tomzig E. & Helbig R. (1976). Band-edge emission in ZnO. J. Lumin. 14, pp. 403-415.
Tong Y.H., Liu Y.C., Lu S.X., Dong L., Chen S.J. & Xiao Z.Y. (2004). The Optical Properties of
ZnO Nanoparticles Capped with Polyvinyl Butyral. J. Sol Gel Sci. Technol. 30, pp.
157-161.
Van de Walle C. G. (2001). Defect analysis and engineering in ZnO. Physica B (Amsterdam)
308-310, pp. 899-903.
Van der Horst J.W., Bobbert P.A. & Michels M.A.J. (2002). Electronic and optical excitations
in crystalline conjugated polymers. Phys. Rev. B 66, pp. 035206(7).
VanDijken A., Meulenkamp E. A., Vanmaekelbergh D. & Meijerink A. (2000a). The
luminescence of nanocrystalline ZnO particles: the mechanism of the ultraviolet
and visible emission. J. Lumin. 87-89, pp. 454456.
Van Dijken A., Meulenkamp E.A., Vanmaekelbergh D. & Meijerink A. (2000b). The Kinetics
of the Radiative and Nonradiative Processes in Nanocrystalline ZnO Particles upon
Photoexcitation. J. Phys. Chem. B 104, pp. 1715-1723.
Van Dijken A., Makkinje J. & Meijerink A. (2001). The influence of particle size on the
luminescence quantum efficiency of nanocrystalline ZnO particles. J. Lumin 92, pp.
323-328.
Vanheusden K.; Seager C. H.; Warren W. L.; Tallant D. R. & Voigt J. A. (1996a). Correlation
between photoluminescence and oxygen vacancies in ZnO phosphors. Appl. Phys.
Lett. 68, pp. 403-405.
Vanheusden K.; Warren W. L.; Seager C. H.; Tallant D. R. & Voigt J. A. (1996b). Mechanisms
behind green photoluminescence in ZnO phosphor powders. J. Appl. Phys. 79, pp.
7983-7990.
Vanheusden K., Seager C. H., Warren W. L., Tallant D. R., Caruso J., Hampden-Smith M. J.
& Kodas T. T. (1997). Green photoluminescence efficiency and free-carrier density
in ZnO phosphor powders prepared by spray pyrolysis. J. Lumin. 75, pp. 11-16.
Vogel R., Meredith P., Harvey M.D. & Rubinsztein-Dunlop H. (2004). Absorption and
fluorescence spectroscopy of rhodamine 6G in titanium dioxide nanocomposites.
Spectrochim. Acta Part A 60, pp. 245-249.
Wang X, Ding Y, Summers C J & Wang Z L (2004). Large-Scale Synthesis of Six-Nanometer-
Wide ZnO Nanobelts. J. Phys. Chem. B 108, pp. 8773-8777.
Wang H., Xie Ch. & Zeng D. (2005) Controlled growth of ZnO by adding H
2
O. J. Cryst.
Growth 277, pp. 372-377.
Welter S., Brunner K., Hofstraat J. W. & De Cola L. (2003). Electroluminescent device with
reversible switching between red and green emission. Nature, 421, pp. 54-57.
Wong. E. M., Hoertz P. G., Liang C. J., Shi Bai-Ming, Meyer Gerald J. & Searson Peter C.
(2001). Influence of Organic Capping Ligands on the Growth Kinetics of ZnO
Nanoparticles. Langmuir 17, pp. 8362-8367.

Cathodoluminescence

160
Woo H.S., Czerw R., Webster S. Carroll D. L., Ballato J., Strevens A. E., OBrien D. & Blau W.
J. (2000). Hole blocking in carbon nanotubepolymer composite organic light-
emitting diodes based on poly (m-phenylene vinylene-co-2, 5-dioctoxy-p-
phenylene vinylene). Appl. Phys. Lett. 77, pp. 1393-1395.
Wood D. L. & Rabinovich E.M. Study of Alkoxide Silica Gels by Infrared Spectroscopy.
Appl. Spectrosc. 43, pp. 263-267.
Wu X.L., Siu G.G., Fu C.L. & Ong H.C. (2001). Photoluminescence and cathodoluminescence
studies of stoichiometric and oxygen-deficient ZnO films. Appl. Phys. Lett. 78, pp.
2285-2287.
Xia Y, Yang P, Sun Y, Wu Y, Mayers B, Gates B, Yin Y, Kim F & Yan H (2003). One-
Dimensional Nanostructures: Synthesis, Characterization, and Applications. Adv.
Mater 15, pp. 353-389.
Yang P, Yan H, Mao S, Russo R, Johnson J, Saykally R, Morris N, Pham J, He R & Choi H. J.
(2002). Controlled Growth of ZnO Nanowires and Their Optical Properties. Adv.
Funct. Mater. 12, pp. 323-331.
Yang C.L., Wang J.N., Ge W.K., Guo L., Yang S.H. & Shen D.Z. (2001). Enhanced ultraviolet
emission and optical properties in polyvinyl pyrrolidone surface modified ZnO
quantum dots. J. Appl. Phys. 90, pp. 4489-4493.
Yang X. D., Xu Z. Y., Sun Z., Sun B. Q., Ding L., Wang F. Z. & Ye Z. Z. (2006). Recombination
property of nitrogen-acceptor-bound states in ZnO. J. Appl. Phys. 99, pp. 046101(3).
Yao B D, Chan Y F & Wang N (2002). Formation of ZnO nanostructures by a simple way of
thermal evaporation. Appl. Phys. Lett. 81, pp.757-759.
Yin M, Gu Y, Kuskovsky I L, Andelman T, Zhu Y, Neumark G F & OBrien S (2004). Zinc
Oxide Quantum Rods. J. Am. Chem. Soc. 126, pp. 6206-6207.
Ying J.Y., Benziger J.B. & Navrotsky A. (1993). Structural Evolution of Alkoxide Silica Gels
to Glass: Effect of Catalyst pH. J. Am. Ceram. Soc. 76, pp. 2571-2582.
Zhang B. P., Binh N. T., Segawa Y., Wakatsuki K. & Usami N. (2003). Optical properties of
ZnO rods formed by metalorganic chemical vapor deposition. Appl. Phys. Lett. 83,
pp. 1635(3).
Zhang S. B., Wei S.-H. & Zunger A. (2001). Intrinsic n-type versus p-type doping asymmetry
and the defect physics of ZnO. Phys. Rev. B 63, pp. 075205(7).
Zhang S. K., Santos P. V. & Hey R. (2001). Photoluminescence modulation by high-
frequency lateral electric fields in quantum wells. Appl. Phys. Lett. 78, pp. 1559-
1561.
Zhang X H, Xie S Y, Jiang Z Y, Xie Z X, Huang R B, Zheng L S, Kang J Y & Sekiguchi T. (2003).
Microwave plasma growth and high spatial resolution cathodoluminescent spectrum of
tetrapod ZnO nanostructures. J. Solid State Chem. Chem. 173, pp. 109-113.
Zhao Q.X., Klason P., Willander M., Zhong H.M., Lu W. & Yang J.H. (2005). Deep-level
emissions influenced by O and Zn implantations in ZnO. Appl. Phys. Lett. 87, pp.
211912(3).
Zimmermann S.; Govorov A. O.; Hansen W.; Kotthaus J. P.; Bichler M. & Wegscheider W.
(1997). Lateral superlattices as voltage-controlled traps for excitons, Phys. Rev. B 56,
pp. 1341413421
Zimmermann S.; Schedelbeck G.; Govorov A. O.; Wixforth A.; Kotthaus J. P.; Bichler M.;
Wegscheider W.; & Abstreiter G. (1998). Spatially resolved exciton trapping in a
voltage-controlled lateral Superlattice. Appl. Phys. Lett., Vol. 73, No. 2, pp. 154-156.
Zimmermann S.; Wixforth A.; Kotthaus J. P.; Wegscheider W.; & Bichler M. (1999). A
Semiconductor-Based Photonic Memory Cell. Science 283, pp. 1292-1295.
7
Cathodoluminescence from Amorphous and
Nanocrystalline Nitride Thin Films Doped with
Rare Earth and Transition Metals
Muhammad Maqbool
1
, Wojciech M. Jadwisienczak
2
and
Martin E. Kordesch
2

1
Ball State University
2
Ohio University
USA
1. Introduction
Rare earth (RE) ion luminescence has long been used in laser and optical fiber communications
technology. Bulk RE doped oxides were widely used in color phosphors for Cathode Ray
Tubes. The wide band gap (WBG) semiconductors and insulators have been used for visible
emission at 300 K from RE ions since the reports first by Zanata (Zanatta and Nunes 1998) for
Er in silicon nitride (photoluminescence) and then shortly thereafter by Steckl (Steckl and
Birkhahn 1998) for Er in GaN. The III-nitrides were emerging as Light Emitting Diodes and
semiconductors at the time, and it was reported that the luminescence intensity of the RE ions
was improved by a wide band gap host. Silicon (band gap 1.1 eV) is not suited for most
visible RE ion emission. Glasses and oxides were used for the infra red (IR), especially for the
emission from Er
3+
ion at ~ 1.5 m. In 1998, RE ion incorporation into a crystalline host was
often accomplished by ion implantation at low atomic concentrations, or by in situ doping
(again with low atomic percentages) of the RE ions. It was believed that the quality of the host
lattice was essential to the RE luminescence. However, Zanata and Nunes observed green
room temperature luminescence from an Er-doped silicon nitride film deposited by reactive
sputtering in nitrogen. Visible emission was observed with an estimated 10 at. % dopant
concentration in an amorphous material. Both the amorphous and crystalline hosts discovered
by the Zanata and Steckl groups set in motion the (enduring) pursuit of practical visible light
emission devices using RE ions in wide band gap materials.
In 1999, Gurumurugan (Gurumurugan, Chen et al. 1999) sputter deposited amorphous AlN
doped with Er (3.4 at.%) and observed the full range of Er emission lines in
cathodoluminescence (CL) in amorphous AlN. There is no visible defect band in the large gap
of AlN (~6 eV), in contrast to sputtered GaN which suffers from an intense, broad yellow
emission from defects. The AlN band gap is almost identical with the crystalline AlN bandgap,
and the bandgap was not reduced significantly by the RE dopant. The AlN film was
transparent, amorphous, and could be deposited at 300 K. Gurumarugan et al. also showed that
the amorphous AlN films could be heated to 1300 K without crystallization. The heating
improved the RE emission intensity. Also, it was noted that in these films the intensity of some

Cathodoluminescence

162
of the Er
3+
transitions decreased with rising temperature, and some increased with rising
temperature.
The crystalline III-nitrides are now extensively used for LEDs, lasers and solid state
lighting, high power electronic devices and other passive electronics uses related to the
thermal conductivity of AlN. The light emission from direct transitions in GaN and In-Al-
Ga-N alloys has displaced the III-nitride RE devices in lighting applications. The low
intensity and long lifetime of the RE
3+
transitions have also hindered the application of RE
luminescence for on-chip communications. Thin film materials doped with REs are used
for phosphors, and specialty applications such as in situ thermometry.
2. Growth, doping and measurement techniques
2.1 Sputter deposition
Radio frequency (RF) sputtering is a physical vapour deposition process that uses RF
excitation to cause gas molecules to collide with a target, removing the target material by
the mechanical impact of the sputtering gas. The RF sputtering process is well known and
used for many industrial applications. The details can be obtained from several textbooks.
RF magnetron sputtering is a variation most often used for the efficient deposition of thin
films.
In the sputtering process, a target of a metal or insulator might be made from solid material
or pressed powders. The sputter gas used is often Argon (Ar). The gas pressure is adjusted
so that a plasma of Ar ions is created by the RF field above the target. In magnetron
sputtering, a system of magnets is used to confine the RF field and the ions so that the target
is used efficiently. In this example, the sputter gas is inert, and physically collides with the
target surface to eject a target atom that is then condensed onto the substrate. In this simple
example, the kinetic energy of the Ar ion is used to eject the target atom. No reaction takes
place, and ideally the condensed film is composed only of the target atoms. For AlN or
GaN, sputterd with Ar, the condensed film is ideally a stoichiometric film of AlN or GaN.
Sputtered films are often disordered, because the process does not generate or require the
input of large amounts of heat such as thermal evaporation or chemical vapor deposition.
The substrate can be cold. As a consequence, the condensed atoms do not have sufficient
energy on the substrate to organize into a crystalline lattice.
In reactive sputtering, the target material is sputtered in a reactive gas. The RF plasma
creates ions of the reactive gas, so that impact with the target results in a compound of the
target and sputter gas formed by a chemical reaction which then condenses onto the
substrate. For nitrogen and aluminium, for example, the nitrogen decomposes into N
ions, and reacts with the Al metal target to form AlN. For simple compounds, where
there are few alternative compositions and structures, the stoichiometric product is often
the most likely compound to be deposited. The purity of the initial components
determines the composition of the deposited film. In the previous example, reactive
sputtering of Al in nitrogen, water vapor or oxygen in the sputter gas can produce
aluminium oxide along with AlN.
A very significant aspect of sputter deposition is the option to grow unusual phases or
compositions that could not be grown in equilibrium processes at high temperature.
Nanocrystalline Nitride Thin Films Doped with Rare Earth and Transition Metals

163
Because the plasma is hot, reactions are fast and complete; rapid thermal quenching by
condensation onto a (relatively) cold substrate preserves the composition achieved in the
plasma. Phase segregation can be avoided as long as the films are not heated to the point
where the different phases are free to form. Alternatively, sputtering onto a heated
substrate can produce polycrystalline films and even epitaxy.

Fig. 2.1. RF magnetron sputtering system. a) actual system; b) schematic diagram. From
(Ebdah 2011), used with permission.
Practical considerations can encourage the use of a particular target. Al sputtered in
nitrogen is an ideal system. Only AlN forms from the reaction. The Al itself is
inexpensive. Targets can even be made from industrial grade Al plate or bars (however,
these materials usually contain small amounts of Cr, and chromium oxide CL is
observed), making the use of multiple targets simple and convenient. In Kordeschs
laboratory at Ohio University, multiple targets were fabricated from 2 inch diameter (50
mm) Al bars or from targets purchased from commercial vendors. Small plugs of the RE
or transition metals were pressed into the aluminium targets. In this way, multiple
dopants or multiple dopant percentages could be obtained with very simple methods.
Zanatta used plates or chips of metal on the silicon target. For materials that do not melt,
or alloy with the sputter target, this method could also be used.
Gallium was sputtered in our laboratory at Ohio University from a pressed powder target
and also from a liquid metal target. In this case, the powder target was more difficult to
use with regard to the dopant material, because there is no good mechanical contact
between the powder and the metal insert. For the liquid target, some droplet ejection
occurs during sputtering. The existence of metal droplets on the thin film surface is
detrimental to device performance, especially if contact or insulating layers are deposited
over the luminescent layer.

Cathodoluminescence

164

Fig. 2.2. Left: Target with (clockwise from top) Sm, Er, and Tb pieces during sputtering.
Right: clockwise from top: Liquid metal Ga target in a copper cup, GaN powder target in a
copper cup, aluminium target with Mn , Tb and Cu plugs, aluminium target with RE plug
removed, and Tm dendrites pressed into the target. The circular depression in the targets at
about the radius is the racetrack, where the plasma is confined by the magnets to
improve the efficiency of the sputter process. All targets are 50 mm diameter.
In practice, the targets used were mostly 50 mm diameter, 6 mm thick disks of aluminium.
The RE or transition metal plugs were obtained commercially as rods of 3 12 mm diameter.
Plugs were cut from the rods and pressed into holes in the aluminium target. Manganese,
for example, could only be purchased as a powder, so the powder was pressed into the 6
mm diameter hole. Some RE metals were obtained as dendrites or split fragments. These
could be pressed into the softer aluminium target by force, even if voids were left around
the RE material. Typical power levels were from 100-200 Watts RF (13.56 MHz) power at 2-
10 mTorr pressures of nitrogen. This is the equivalent of 5-10 Watts/cm
2
. The substrates
could be heated from the back by using a quartz lamp with a parabolic reflector focussed
onto the back of the substrate, or cooled by clamping the substrates to a copper cold finger
cooled with dry ice. We have not used liquid nitrogen to cool the samples. First, the lower
temperature will condense water onto the sample, and second, there are stresses that come
from warming a film grown at low temperature to 300 K, just as there are stresses in films
cooled from a high temperature deposition.
Aluminum, gallium, beryllium, scandium, boron, silicon, silicon nitride, aluminium oxide,
zinc, hafnium, titanium and alloy targets were used. Dopant RE metals such as Er, Tb, Gd,
Tm, Sm, Pr and Ho were used, as well as Cr, Cu, Mn, Mg, Si, C, Ti, Sn, Ag. The sputter
target is always below the sample/substrate to avoid particulates formed in the sputter
process from falling onto the substrate (see Fig. 2.1).
An important consideration for RE doping in the nitrides is the stability of the host material.
Many RE nitrides decompose when in contact with water or humid air, which causes a
reaction resulting in ammonia and a RE oxide. AlN, BN, GaN, SiN and ScN are not soluble
in water. In our experiments, beryllium nitride was also stable in humid air. There is

165
considerable controversy over the role of oxygen in promoting the luminescence of RE ions
in the nitrides. It is not known if the RE nitrides in a nitride host decompose or remain
bonded to the N atom.
Finally, it should be pointed out that no electronic doping is observed in the amorphous
nitride films. Partially, this is due to the fact that most of the amorphous nitride films are
insulators, with the exception of n-type GaN. Electronic doping with Mg or Si was not
successful, because the dopants react to form nitrides in the plasma, making the dopants
commonly used in crystalline GaN for example, ineffective in amorphous nitrides.
2.2 Cathodoluminescence measurements
2.2.1 CL at room temperature and above
For CL measurements at 300 K and above, a simple system build in a six-way stainless steel
cross with copper sealed flanges was used. The system was pumped with a 450 L/sec
turbomolecular pump. A CRT electron gun with deflection plates was adapted to a multi-
pin electrical feedthrough. The electron gun was powered with a laboratory-built power
supply. The electron beam voltage could be varied up to 2.8 kV. Focus controls allowed the
beam to be focussed onto the sample and to observe the spot on the sample through a
vacuum viewport for optimization. The CL spot on the sample was then focussed onto the
entrance slits of a monochromater using a lens.
The sample was illuminated with the electron beam from one side, the viewport was at 90
degrees from the electron gun. The sample was placed at 45 degrees to both arms of the
cross. An aperture was placed in front of the sample between the sample and electron gun
to block the visible light spot from the filament. The deflectors on the electron gun make it
possible to steer the electron beam through the aperture while the direct light from the hot
filament is blocked by the aperture.

Fig. 2.3. Experimental setup for CL.
A commercial Luminoscope was also used for some CL measurements. The Luminoscope is
meant to be used for geological specimens, and is equipped with a vacuum chamber that
replaces the stage of a light-optical microscope. A window allows microscopic observation of

Cathodoluminescence

166
the sample surface during electron illumination. A gas discharge electron source is used to
generate an electron beam of up to 30 kV. The Luminoscope can be used to take color
micrographs of the sample surface. A drawback to spectroscopic measurements in the
Luminoscope is the necessity to remove the gas discharge spectrum (He) from the CL spectra.

Fig. 2.4. Left: Lumiscope sample holder. Right: Lumiscope control and electronics.
2.2.2 CL at low temperature
Low temperature CL spectroscopic analysis was measured in our laboratory at cryogenic
temperatures and excitation conditions using experimental setup shown in Fig.2.5. Typically
samples were mounted on the cold finger of a closed-cycled helium refrigerator operating
down to 6K. The CL was generated by the Staib Instruments, Inc. Electron Gun EK-20-R
equipped with a beam blanker (repetition rate: DC to 1 MHz) electron gun system being in
common vacuum (of 510
-7
Torr) with the cryostat. The electron beam was incident upon
the sample at a 45 angle from an electron gun. The CL depth of the excitation could be
easily varied by varying the electron acceleration voltage between 500eV up to 20 keV.

Fig. 2.5. Front view of a low temperature cathodoluminescence experimental setup
operating between 6K- 330K and electron acceleration voltage up to 20 kV. Insets show color
emissions from different RE-doped III-nitrides.

167
2.2.3 CL in the electron microscope
Cathodoluminescence attachments can be obtained for both scanning electron microscopes
(SEM) and transmission electron microscopes (TEM). In the simplest SEM versions, a
parabolic mirror is mechanically inserted into the specimen chamber above the sample, with
a hole in the mirror for the passage of the electron beam. The light from the sample CL is
reflected onto a photomultiplier tube and used to form a scanned image based on the CL
yield. In advanced systems, and monochromater is added to the optical system so that
spectroscopically resolved images can be generated. Similar systems are used in TEM, but
without a CL image. Only spectroscopy is possible in the TEM. CL based images can be
obtained in scanning TEM, STEM.

Fig. 2.6. Left: GaN crystallites in the SEM. A bit of a-GaN is shown in the stippled circle.
Right: The same area, showing CL from the a-GaN and c-GaN. This image was one of the
first experimental observations of a-GaN. (Hassan 1998).
3. Cathodoluminescence of rare earth doped amorphous nitrides
In the last two decades it has been shown that the incorporation of RE elements as dopant
atoms into III-nitride semiconductors such as GaN, AlN and their alloys, both crystalline and
amorphous, leads to a temperature-stable luminescence whose wavelength is nearly
independent of the specific semiconductor host (O'Donnell 2010). Luminescence from these
materials having amorphous morphology grown on variety of substrates, and doped by
implantation or during the growth with Ce (Aldabergenova, Osvet et al. 2002) , Pr (Maqbool,
Ahmad et al. 2007; Maqbool, Richardson et al. 2007; Maqbool and Ahmad 2009), Sm (Zanatta,
Ribeiro et al. 2001; Weingartner, Erlenbach et al. 2006; Maqbool and Ali 2009), Eu
(Aldabergenova, Osvet et al. 2002), (Weingartner, Erlenbach et al. 2006), (Caldwell, Van Patten
et al. 2001), Gd (Maqbool, Ahmad et al. 2007),(Maqbool, Kordesch et al. 2009), Tb
(Aldabergenova, Osvet et al. 2002), (Weingartner, Erlenbach et al. 2006),(Richardson, Van
Patten et al. 2002), (Jadwisienczak, Lozykowski et al. 2000), Dy (Weingartner, Erlenbach et al.
2006), Ho (Maqbool, Ali et al.; Aldabergenova, Frank et al. 2006; Maqbool, Kordesch et al.
2009), Er (Gurumurugan, Chen et al. 1999; Chen, Gurumurugan et al. 2000; Dimitrova, Van
Patten et al. 2000; Dimitrova, Van Patten et al. 2001; Zanatta, Ribeiro et al. 2005), Tm (Maqbool,
Kordesch et al. 2009) and Yb (Weingartner, Erlenbach et al. 2006) has been reported so far. In
particular, RE doped GaN-nitrides based electroluminescent devices (ELDs) have been shown
to have a versatile approach for the fabrication of variety of electrically driven optical light

Cathodoluminescence

168
sources with narrow line-width emissions from the ultraviolet to the infrared (Steckl 1999).
Thus optoelectronic devices utilizing 4f
n
transitions appearing in the nitrides forbidden band
gap window are practically viable with these semiconductors (O'Donnell 2010). Recently it
was demonstrated that the low voltage-operation of current-injected red emission from a
crystalline p-type/Eu-doped/n-type GaN epilayers light emission diode operating at room
temperature (Nishikawa, Kawasaki et al. 2009; Kasai, Nishikawa et al. 2010; Nishikawa,
Furukawa 2010; Dierolf 2011) together with demonstrations of stimulated emission from Eu
3+

doped GaN and AlGaN layers on a Si substrate proved in principal that optoelectronic devices
covering the UV, visible and IR regions might be fabricated (Park and Steckl 2004; Park and
Steckl 2005; Park and Steckl 2006).
These new results suggest a novel way to realize III-nitride semiconductors-based red
emitting, current driven light emitting devices, as well as other primary colors and their
mixture, monolithic devices. In the past the feasibility of using the RE-doped amorphous III-
nitrides (a-III-nitrides) for light emitting applications was also demonstrated (Dimitrova, Van
Patten et al. 2000; Dimitrova, Van Patten et al. 2001; Richardson, Van Patten et al. 2002;
Maqbool, Kordesch et al. 2009), (Kim, Shepherd et al. 2003; Kim and Holloway 2004).
However, to make these devices commercially viable, the internal quantum efficiency of the
active RE-doped layers has to be significantly improved. The future success of the RE-doped
III-nitrides optoelectronics, both crystalline and amorphous based will most probably depend
on engineering of multilayer structures. In these devices an enhancement of RE
3+
ion
luminescence intensity can be achieved through e.g. (a) modification of the RE
3+
center
environment and (b) localization of carriers in the vicinity of emitting RE
3+
ion center. The
former one can be achieved through engineering stress/strain parameters during the growth
process, whereas the last one will results from carriers confinement in the quantum structures.
In general, the majority available research papers focus on RE-doped crystalline III-nitride
semiconductors with less emphasis on a-III-nitrides used as hosts for optically, electrically
and/or magnetically active RE dopants. The amorphous semiconductors including a-III-
nitrides as hosts for RE doping have many of the desirable qualities of the crystalline materials;
however they also offer unique features rooted in the nature of the amorphous matrix (Adachi
1999; Singh 2003; Street 2010). The most important are that a-III-nitrides can be achieved at
higher growth rates and lower temperatures, substrate selection and resulting lattice mismatch
is not a significant obstacle here, they do not easily recrystallize when subjected to thermal
processing and that they can adapt the RE ions in concentrations far beyond those given by the
solubility limit found in their crystalline counterparts. Furthermore, it is known that a-III-
nitrides show a natural tendency to develop a state free gap [(Chen and Drabold 2002; Drabold
2010). Also a-III-nitrides, unlike other amorphous systems or glasses which look locally very
similar to the crystal, apparently have local environments very different from the main
crystalline morphology what make them potentially useful as hosts for optically and electrically
active RE dopants. It was demonstrated that the RE ions luminescence intensity, when intended
for optoelectronic applications, can be increased by proper structural tailoring of a-III-nitride
matrices. At the same time the RE
3+
ion excitation and de-excitation processes are more
dependent on the local RE ion environment than in crystalline semiconductors due to the lack
of long range ordering in amorphous host (Zanatta, Ribeiro et al. 2006). The f-f transitions are
forbidden, but they are partially allowed when the wave functions are mixed with those of
opposing parity. This always occurs on localized ions in non-central symmetrical lattices. The

169
forbidden nature of these transitions results in long radiative lifetimes when the selection rules
are relaxed. Placing trivalent rare earth ions in amorphous solids can still result in characteristic
emission from these intra-4f transitions despite the lack of a crystalline host structure (O'Donnell
2010). In both cases RE luminescence depends very little on the nature of the host and the
ambient temperature. The 4f orbitals of RE ions incorporated into semiconductors, including a-
III-nitrides, are so deeply buried within the electronic shell that the energy levels of the 4f
n

configuration are only slightly perturbed compared to free ion energy levels indicating large
independence from the chemical surrounding and thus the choice of the host material
(Wybourne and Smentek 2007). RE ions when doped to a-III-nitrides are known to exhibit sharp
and well-defined luminescence which arises mostly from internal transitions between 4f levels
within RE
3+
ion. Because of the shield provided by the outer 5s
2
and 5p
6
shells, the wavelength
involved in these intra-4f transitions is less temperature sensitive and rather weakly depends on
the a-III-nitride host. On the other hand the mechanism behind the activation of RE are known
to be sensitive to a local host details such as atomic structure and optical band gap which
defines which RE transitions can be excited.
3.1 Rare earth doping
Doping of a-III-nitrides is generally achieved in situ ; however it can be also done by RE ions
implantation (Aldabergenova, Frank et al. 2006). In the former case doping typically takes
place during sputtering process in which RE-doped a-III-nitride is formed from Ga, Al and RE
atoms removed from metal targets by momentum transfer from an RF plasma. The sputtering
growth of a-III-nitrides is explored much less than other technologically important amorphous
semiconductors e.g a-Si doped with RE impurities (Iacona, Franzo et al. 2009). Ion
implantation has been used extensively to ex situ dope III-nitride crystalline semiconductors
with variety of rare earth ions. Excellent review concerning this issue is available (O'Donnell
2010). There are a limited number of papers on RE ions implantation to a-III-nitrides and
resulting luminescence spectra (Jadwisienczak, Lozykowski et al. 2000). Ion implantation is a
well established technique, which provides good control of the concentration profile in the
sample. However, because of the unavoidable damage associated with the implantation
process, typically a high temperature annealing step is imperative. This is due to the fact that
when heavy atoms like RE metals are implanted to an amorphous semiconductor they can
introduce both a chemical change in the substrate, in that they can introduce a different
element than the host or induce a nuclear transmutation, and a structural change, in that the
morphology of the host can be damaged or even destroyed by the energetic collision cascades
induced by implanted heavy RE ions (Kucheyev, Williams et al. 2004).
It is known that the thermal anneal of the a-AlN films affects not only the luminescent
properties of the films, but also their optical band gap and chemical characteristics. As a
result of thermal treatments of RE-doped a-III-nitrides at increasing temperatures the
optical-absorption edge of amorphous films may change indicating the Al and N bonding
environment modification (Zanatta, Ribeiro et al. 2005). Based on the available experimental
results it is clear that RE-doped a-III-nitrides subjected to isochronal thermal annealing
treatments indeed most likely underwent partial removal of both structural and chemical
disorder resulting in modification of RE ions local environments as it is shown in Figs. 3.1
and 3.2 for Er-doped a-AlN films (Gurumurugan, Chen et al. 1999; Jadwisienczak,
Lozykowski et al. 2000; Weingartner, Erlenbach et al. 2006).

Cathodoluminescence

170

Fig. 3.1. Optical transmission spectra of undoped and Er-doped a-AlN thin films. Inset
shows the photon energy dependence of the square of the absorption coefficient for
undoped and Er-doped a-AlN thin films. After Ref. (Gurumurugan, Chen et al. 1999)
It is documented in the literature that thermally annealed amorphous semiconductors
experience some structural reordering (not necessarily crystallization) (Gurumurugan, Chen
et al. 1999). Figure 3.2 shows evolution of cathodoluminescence (CL) spectra of Tb-doped a-
AlN films subjected to thermal annealing in nitrogen ambient up to 1100 C.

Fig. 3.2. Evolution of CL spectra in Tb-doped a-AlN films as a function of annealing
temperature. After Ref. (Gurumurugan, Chen et al. 1999).

171
It is believed that when this takes place the dangling or broken bonds and/or the atomic
rearrangements of the amorphous material may have great influence on the electronic states
of amorphous host. The available reports show that the conducted x-ray diffraction analysis
of as-grown or thermally treated RE-doped a-III-nitride thin films has not revealed any
microscopic morphology changes in studied materials (see Fig. 3.3).

Fig. 3.3. XRD spectra of a-AlN thin films doped with Tb
3+
ions. (a) unannealed and (b)
annealed for 30 min at 1050 C in nitrogen. Ref. (Jadwisienczak, Lozykowski et al. 2000).
As reported by Zanatta et al. (Zanatta, Khan et al. 2007) a-SiN doped with Sm and Tb exhibit
significant enhancements of 4f-shell radiative recombinations upon thermal annealing due
to a decrease in the density of deep and/or tail defects (suppression of the nonradiative
processes). A similar observation was made for a-AlN:Ho system (Aldabergenova, Frank et
al. 2006). Furthermore, in that study, the growth of small crystallites in initially mostly a-
AlN:Ho host was observed after annealing above 900 C. See Fig 3.4.
In order to avoid the implantation induced amorphous material morphological changes and
defects due to ion implantation, simultaneous co-deposition of different RE metals during
RF sputtering growth was demonstrated (Maqbool, Kordesch et al. 2009). This approach is
an extension of a typical a-III-nitrides RF sputtering growth technique where RE ions
concentration in resulting film is controlled by sputtering metals to target surface ratio. It is
possible using this procedure to control the RE concentration in the sample over a wide
range. Furthermore, this technique gave an opportunity to derive mixed RE-doped a-III-
nitride systems where more complex interaction schemes defining energy transfer processes
can be studied. However, there exist only limited papers reporting on this issue up to date
(Maqbool, Kordesch et al. 2009). The single RE ion and multiple RE ions doping to a-III-
nitrides should have typical doping effects in these hosts in the sense of controlling the
electronic properties of the material and determining the Fermi level position. In contrast to
unhydrogenated a-Si and hydrogenated a-Si:H where it was proposed that several different
REs turned out to act as acceptors or donors (Tessler 1999) no similar postulate was
considered in the case of RE-doped a-III-nitrides to date. It is uncertain at present if other
than trivalent RE ions in a-III-nitrides can exist; however such a scenario is feasible

Cathodoluminescence

172
especially if one considers a plethora of possible RE ions configurations in amorphous
matrix as well as the fact that RE ions are strong oxygen and other impurities getters.
Systematic RE ion doping studies with controlled co-dopant concentrations are needed to
validate this consideration in the future.

Fig. 3.4. High resolution micrograph showing small AlN crystallites embedded in as grown
a-AlN film. The insert shows the pertaining diffraction pattern. After (Aldabergenova,
Frank et al. 2006).
3.1.1 Rare earth excitation
The excitation processes of RE ions in a-III-nitrides, similarly to crystalline hosts, can be
generally divided into two categories: direct and indirect excitation processes
(Jadwisienczak, Lozykowski et al. 2000; Lozykowski, Jadwisienczak et al. 2000). The direct
exci tati on process occurs i n sel ective exci tati on of 4f
n
el ectrons by photons
(photoluminescence, PL) selective excitation or in cathodoluminescence (CL) and
electroluminescence (EL) by collision with hot electrons. The indirect excitation process
occurs via transfer of energy to the 4f
n
electron system from electronhole pairs generated by
photons with higher energy than the band gap (PL excited above band gap), injected in
forward bias p-n junctions, or generated by hot carriers in CL and EL. An excitation
mechanism in CL and EL involves direct impact excitation of RE
3+
ions by hot electrons, as
well as an energy transfer from the generated electronhole pairs or by impact excitation (or
ionization) involving impurity states outside the 4f shell, with subsequent energy transfer to
this shell. The most important, from an applications point of view, is the excitation of the RE
ions by energy transfer processes from electron-hole (e-h) pairs or excitons. This process
most probably involves the rare earth ion isovalent traps (Lozykowski, Jadwisienczak et al.
2000). Since there is no charge involved, the isoelectronic center forms the bound states by
short range central cell potential. After an isoelectronic trap has captured an electron or a

173
hole, the isoelectronic trap is negatively or positively charged, and by Coulomb interaction it
will capture a carrier of the opposite charge creating a bound exciton. There are three
possible mechanisms of energy transfer (Lozykowski 1993). The first is the energy transfer
process from excitons bound to structured isoelectronic centers to the core electrons. This
takes place as a result of the electrostatic perturbation between the core electrons of the RE
structured impurity and the exciton, effective-mass-like particles. The second mechanism is
the transfer of energy to the core electrons involving the structured isoelectronic trap
occupied by electron (hole) and free holes (electrons) in the valence (conduction) band. The
third mechanism is the transfer through an inelastic scattering process in which the energy
of a free exciton near a RE structured trap is given to the localized core excited states. If the
initial and final states are not resonant, the energy mismatch must be distributed in some
way, e.g., by phonon emission or absorption. If the atomic core excitations are strongly
coupled to the host phonons, the energy transfer probability is likely to be higher
(Lozykowski 1993). Strong phonon coupling may also be desirable in ensuring that
relaxation down the ladder of the core excited state occurs quickly, thus preventing back
transfer. However, for efficient radiative recombination, the phonon coupling should not be
strong, in order to prevent core de-excitation by nonradiative multiphonon process. It is
natural to assume that when isoelectronic impurity atoms, in this case RE ions, are
incorporated in a-III-nitride host, they modify local vibrational properties in a definite way.
Thus, there should be observed distinct vibrational frequencies associated with localized
motion of the RE impurity in addition to the host lattice vibrations. In III-nitrides the
threshold energy for electron damage in amorphous films is considerably smaller contrary
to crystalline compounds (Zanatta, Ribeiro et al. 2005). Moreover, energetic electron
irradiation may create defects that are not possible with photons with energies in the visible
energy range. These additional defects, mainly broken bonds and atomic displacements, act
as nonradiative centers that considerably reduce the luminescence efficiency of a-III-nitride
semiconductor. On the other hand, RE ions are high efficiency recombination centers that
effectively compete with other non-radiative processes taking place in a semiconductor host.
As a result, when irradiating the RE-doped a-III-nitrides with up to a few keV energy
electrons, most of the electron-hole pairs recombine preferentially through the RE ions.
Luminescence after photon excitation, on the other hand, behaves in a different manner and
is very strongly excitation wavelengths dependent (Gurumurugan, Chen et al. 1999;
Jadwisienczak, Lozykowski et al. 2000; Zanatta, Ribeiro et al. 2005).
In photoluminescence each absorbed photon with energy higher than the band gap
produces a single electron-hole pair while in cathodoluminescence, high energy single
electrons generate a huge number of hot eh pairs, reducing the energy from tens of keV
to zero (Jadwisienczak, Lozykowski et al. 2000). In general, the optical excitation process
demands that the 4f electrons are excited absorbing energy. In a-III-nitride materials with
large band gap like a-GaN and a-AlN this can occur with a direct absorption to one of the
transitions of upper energy of the RE
3+
ion degenerated with the conduction band or via
transfer of energy from a defect related Auger resonant process between the RE
3+
ion states
and the dipole formed by the s-like conduction band states and p-like dangling bond states.
The PL of RE
3+
ion in a-III-nitrides is obtained by an indirect excitation process via transfer
of energy to the 4f
n
electron system from e-h pairs. In electron beam excitation (CL),
however, the RE
3+
ions are excited by direct impact with hot electrons, as well as by energy
transfer processes from the generated eh pairs or by impact excitation (or ionization)

Cathodoluminescence

174
involving other impurities (or complex defects) with subsequent energy transfer to the RE
4f-shell electrons. Generally the excitation by energetic electrons produces emission via all
possible luminescence mechanisms available in a semiconductor. Another factor which may
play a role in excitation and emission processes is the charged nature of excitation:
uncharged photons in PL versus negatively charged electrons in CL. The excitation depth in
PL and CL are also different due to the strong absorption of the excitation photons in a-III-
nitrides within a few tens of nanometers layer; whereas the electron penetration depth is
electron acceleration energy dependent and can reach up to a few hundreds nanometers
(Jadwisienczak, Lozykowski et al. 2000). Figure 3.5 and Fig. 3.6 show examples of CL spectra
of the rare earth (Tm, Tb, Dy, Sm, Eu, Er and Yb) doped a-AlN layers [(Gurumurugan, Chen
et al. 1999; Jadwisienczak, Lozykowski et al. 2000; Weingartner, Erlenbach et al. 2006). The
transitions corresponding to the strongest emission lines are indicated by their energy level
assignments. It was reported in these studies that all RE-doped a-AlN layers show (even
untreated) pronounced RE
3+
ion optical spectra at room temperature.

Fig. 3.5. Survey of CL spectra of RE-doped a-AlN thin films. After Ref. (Weingartner,
Erlenbach et al. 2006).
The RE ion excitation process responsible for observed EL from RE-doped a-III-nitrides is very
similar to described above CL. Thus CL and EL spectra, in general, shows similar spectra
features. It was reported that in cases of the Er and Tb-doped a-AlN alternating-current thin-
film electroluminescent devices the EL and CL spectra are very similar (Dimitrova, Van Patten
et al. 2000; Dimitrova, Van Patten et al. 2001; Richardson, Van Patten et al. 2002).
Typical transient analysis of luminescence observed from RE-doped III-nitride crystalline
semiconductors stimulated by different means provided detailed insight into energy migration
and excitation processes observed in these technologically important materials. There are a very
limited number of papers reporting on luminescence (CL, EL) kinetics of RE-doped a-III-nitride
up to date (Richardson, Van Patten et al. 2002). It is known that the nonradiative decays of

175
excited RE ions in semiconductors, including RE-doped a-III-nitrides, are phonon dependent
(Jadwisienczak, Lozykowski et al. 2000). Furthermore, at sufficiently high concentrations of RE
ions, nonradiative decay can occur via ion pair cross-relaxation or energy migration and finally
transfer to quenching centers (Jadwisienczak, Lozykowski et al. 2000). For low RE
concentration, where ionion pair relaxation or energy migration and transfer to quenching
centers has low probability, excited RE electronic states gave up energy radiatively by purely
electronic or phonon assisted transitions or nonradiatively by the emission of a single phonon
or multiphonon. Nonradiative relaxation between RE
3+
ion J-manifolds in solid state host,
including reported CL kinetics of Tb-doped a-AlN , typically requires the participation of
several phonons and occurs at slow rates, while relaxation between crystal field levels of a J-
manifold by single or two phonon processes is much faster. It is known that the rate of
multiphonon emission is strongly dependent upon the number of phonons required to conserve
energy, and hence on the size of the energy gap to the next lowest energy level. The detailed
investigations of nonradiative decay of RE ions in a-III-nitrides have not been done yet;
however it is expected that that multiphonon transitions involving the emission of a maximum
of a few phonons will effectively compete with radiative transitions. Furthermore, it is expected
that the temperature dependence of the decay rates will show that relaxation occurs mainly by
high energy phonons. It is worth noting that energy separation of J-crystal field levels shows
only small changes with the host (Jadwisienczak, Lozykowski et al. 2000; Jadwisienczak,
Lozykowski et al. 2000). Differences in the rate of nonradiative decay from a particular RE
energy level arise from the phonon energy spectrum of the host material and the strength of the
RE ionlattice coupling. Local vibrational modes related to RE ions, their complexes, or others
impurities present in a-III-nitride host will also play an important role in the above-discussed
process; however these assumptions will have to be critically confirmed in the future.

Fig. 3.6. CL spectra of a-AlN:Tb recorded at cryogenic and room temperature. The inserts show
the total integral intensity of the dominant 4f-shell transition lines as a function of temperature.
After Ref. [(Jadwisienczak, Lozykowski et al. 2000)].

Cathodoluminescence

176

Fig. 3.7. CL spectra of a-AlN:Er recorded at cryogenic and room temperature. The inserts
show the total integral intensity of the dominant 4f-shell transition lines as a function of
temperature. After Ref. [(Gurumurugan, Chen et al. 1999)].
The local environment of the RE
3+
ions, and consequent luminescence intensity, in a-III-
nitrides can be greatly influenced by the presence of nitrogen atoms, intrinsic impurities,
larger clusters and thermal treatments stimulating morphological changes at micro-scale.
This in turn affects the optical activity of RE
3+
ion by enhancing or quenching
luminescence originating from 4f-shalls transitions. The vulnerability of amorphous
semiconductors to defects generation in CL and EL is rather straight forward considering
the threshold energy for electron damage in amorphous networks being much smaller (1
keV) comparing to crystalline semiconductors (100 keV) (Zanatta, Ribeiro et al. 2005). This
fact creates an interesting research opportunity for RE-doped a-III-nitrides in the search
for more effective energy transfer process between RE ions and amorphous host. It is
known that the III-nitride quantum structures with dimensions close to the excitonic Bohr
radius exhibit electronic and optical properties affected by the confinement of electrons in
one, two, or three dimensions (Ihn 2010). The principal consequences of quantum
confinement are an increase in the band gap energy and increased probability of radiative
transitions. Confinement of carriers in real space causes their wavefunctions to spread out
in momentum space, increasing the probability of radiative processes due to greater
wavefunction overlap. Because of the modification of their band structure, doping such
confined systems with RE
3+
ions can also help to overcome some of the nonradiative de-
excitation problems associated with RE-doped III-nitride. Quantum confinement affects
also carrier lifetimes and their degree of localization in real space. This in turn modifies
the Auger back-transfer processes that limit luminescence efficiency and increases the
interaction probability between the confined carriers and the RE
3+
ions. In general, the
RE
3+
ion radiative quantum efficiency strongly depends on the carrier mediated energy
transfer processes, which have to compete with nonradiative recombination channels
abundant in a-III-nitrides. It is known that RE
3+
ions induce significant local site distortion

177
due to their large radii as well as gettering effect (O'Donnell 2010). It was theoretically
shown that the Coulomb excitation of 4f electrons near interface of heterostructures,
where certain degree of disorder is expected, is more effective than a similar excitation in
the bulk semiconductors (Zegrya and Masterov 1995; Zegrya and Masterov 1996; Zegrya
and Masterov 1998; Masterov and Gerchikov 1999). Also, as discussed above, the thermal
annealing process of RE-doped a-III-nitride changes grain boundary, where RE ions most
probably reside, resulting in change of RE ion local environments and their optical
activities due to relaxed 4f-shell electron-lattice coupling. In that sense one may assume
that RE
3+
ions located at the surface of the a-III-nitride nano-grains may act as grain
boundary activators. It has now been well established that semiconductor nano-clusters
can act as efficient luminescence sensitizers for RE ions in wide band gap solid states
(Iacona, Franzo et al. 2009). Moreover, it appears that it is not necessary for the
semiconductor inclusions to be crystalline amorphous nano-clusters are at least as
efficient as nano-crystals (Kenyon and Lucarz). The customary observation is that
absorption of photons by semiconductor nano-clusters results in excitation of RE ions via
an efficient transfer mechanism to unexcited RE ions. Such a process is significant for two
reasons: firstly, the effective absorption cross-section of the RE ion is increased by several
orders of magnitude (Kenyon, Lucarz; Kik 2003), and secondly, it becomes possible to
excite the RE ions via the broad-band absorption of the semiconductor nano-clusters
(Kenyon, Chryssou et al. 2002; Kik 2003; Iacona, Franzo et al. 2009). Moreover, the
prospect of engaging amorphous nano-clusters to activate RE ions in a-III-nitrides
removes the prerequisite to control tightly the crystallinity and size distribution of nano-
cluster sensitizers thus greatly simplifying material processing. More studies are
necessary to clear this point. It is our belief that the quenching mechanism should not
depend on the form of sample preparation rather than on the final RE ion local symmetry
inducing site degeneracy required for observing peculiar intra 4f-shell transitions.
3.2 Amorphous nitrides: Examples
Most of the metal nitrides are M
3+
N
3-
compounds, so that no special considerations are
necessary for RE emission. There is still some controversy about the site of the RE ion in
crystalline nitrides. In an amorphous solid, bonding to the N as a (3+) ion is the most
obvious choice. However, it is the case that sharp lines are observed from RE ions in the
amorphous hosts, and in some cases the analogue of crystal field splitting. There are
probably several types of sites for the RE ion, but not enough to broaden the emission
peaks excessively.
Because the emission lines in RE ions are well known, in most cases the experimental task is
to see which transitions are observed in each host. Post growth heat treatment of the as-
gown nitrides often improves the intensity of the RE emission in the nitrides. Because there
is no annealing in the sense of crystal structure or long range lattice improvement, the
heating process can cause the removal or transport of materials unrelated to the nitride
structure. There could be removal of hydrogen, for example, or diffusion of oxygen. The
reaction of water trapped in the film due to background gasses in the vacuum or in the heat
treatment gas during post growth thermal treatment would include the removal of N atoms
as ammonia, affecting the basic M-N bonds. Some rearrangement of the short range nitride
structure is possible.

Cathodoluminescence

178
Even though there can be much larger RE concentrations in sputter deposited nitride films
compared to ion implanted films, there is no dramatic increase in intensity with higher RE
concentrations. It must be assumed that a large number of the RE ions are not optically
active in the as-grown amorphous films. Heat treatment or oxygen additions increase the
number of active RE ions. Co-doping with other (3+) ions sometimes also improves the
emission intensity of RE ion luminescence. One of the possibilities is that the (3+) impurity
occupies the inactive sites and displaces the RE ions onto active sites. Optical interactions
are also possible.
3.2.1 Amorphous silicon nitride
The bulk of studies on amorphous silicon nitride doped with RE ions are due to Zanata and
co-workers. Silicon is multivalent, so that a-SiN may well result in a RE
3+
ion without any
further need for charge compensation.
3.2.1.1 Erbium, Holmium, Dysprosium Samarium and Praseodymium
Zanatta and co-workers studied Pr, Sm, Dy, Ho, and Er in amorphous SiN films.

Fig. 3.8. Cathodoluminescence spectra of RE
3+
-doped a-SiN films deposited by co-
sputtering. The spectra have been acquired at room temperature by using 15 keV electrons
and a current density of 10 nA. All spectra were normalized and vertically shifted for
comparison. From (Zanatta, Ribeiro et al. 2004).

179
The CL spectra of the present RE-doped a-SiN films are shown in Fig. 3.8. In addition to the
levels close to 490 nm these new transitions correspond to Er
3+
ions (
4
G
11/2
to
4
I
15/2
at 390 nm,
2
H
9/2
to
4
I
15/2
at 410 nm); Ho
3+
ions (
5
G
5
to
5
I
5
at 430 nm,
3
H
5
to
5
I
5
at 360 nm); Sm
3+
(
4
K
11/2
to
6
H
5/2
at 380 nm); and Pr
3+
ions (
1
S
0
to
1
D
2
at 370 nm,
1
S
0
to
1
I
6
or
3
P
1
at 395 nm). The
variations observed in the luminescence intensity probably occur because of differences in
the RE content; Er (0.5 at.%), Ho (0.3 at.%), Dy (0.3 at.%), Sm (0.5 at.%), and Pr (0.2 at.%).
(Zanatta, Ribeiro et al. 2004)
3.2.1.2 Samarium and terbium
Most of the luminescence features present in the 400850 nm wavelength range (Fig. 3.9)
correspond to optical transitions due to the Sm
3+
and Tb
3+
ions (Dieke 1968): Sm1 at 565
nm (
4
G
5/2

6
H
5/2
), Sm2 at 605 nm (
4
G
5/2

6
H
7/2
), Sm3 at 650 nm (
4
G
5/2

6
H
9/2
), Tb1 at
485 nm (
5
D
4

7
F
6
), Tb2 at 545 nm (
5
D
4

7
F
5
), Tb3 at 590 nm (
5
D
4

7
F
4
), and Tb4 at
625 nm (
5
D
4

7
F
3
). In addition to these, we also observe the Sm4 light emission at 725
nm (
4
G
5/2

6
H
11/2
), which is associated with the superposition of the
4
G
5/2

6
H
11/2
transition
due to Sm
3+
ions and the
5
D
0

7
F
0
,
5
D
0

7
F
1
, and
5
D
0

7
F
2
transitions of Sm
2+
ions. The
infrared contribution Sm5 at 810 nm corresponds to the
5
D
0

7
F
4
transition and is
exclusively due to Sm
2+
ions (Dieke 1968).

Fig. 3.9. Smarium and Terbium CL in amorphous silicon nitride. From (Zanatta, Khan et al.
2007).
3.2.2 Amorphous aluminum nitride
3.2.2.1 Erbium and terbium
The first RE spectrum in amorphous Aluminum Nitride (a-AlN) was that of Gurumurugan
et al. (Gurumurugan, Chen et al. 1999). The spectra are shown in Figures and 3.7
(Jadwisienczak, Lozykowski et al. 2000).
3.2.2.2 Europium
Europium doped a-AlN was studied by Caldwell et al. (Caldwell, Van Patten et al. 2001).
The most significant result of this study was that thermal activation of the RE ion in a-AlN,
which was found generally to increase the luminescence yield by a factor of up to 100, could

Cathodoluminescence

180
be matched or exceeded by the addition of oxygen during the growth of the film by
sputtering. The luminescence yield increased over 600 times with the addition of oxygen to
the sputter gas.

Fig. 3.10. CL spectra of a-AlN:Eu processed as follows: a) no oxygen, and no heat treatment,
b) no oxygen, heated to 923 K. c) grown with 1.6% oxygen in nitrogen, d) grown in 3.8%
oxygen in nitrogen, e) grown in 20% oxygen, with the balance nitrogen. From (Caldwell,
Van Patten et al. 2001).

Fig. 3.11. CL from some of the types of a-AlN:Eu films described in Fig. 3.10.
500 550 600 650 700
0
1
2
3
4
5
A
B
C
D
E
x 29
x 7
Wavelength (nm)
R
e
l
a
t
i
v
e

I
n
t
e
n
s
i
t
y

(
a
.
u
.
)
x 615
5
D0
7
F1
5
D0
7
F2
5
D0
7
F3
5
D0
7
F4

181
3.2.2.3 Holmium and gadolinium
Data for Ho and Gd in a-AlN are given by Maqbool (Maqbool 2005).

Fig. 3.12. CL from codoped Gd and Ho a-AlN films.

Material
Transition
Number
Transition
Wavelength
(nm)
Relative
Intensity
AlN:Ho 1
5
G
5

5
I
8
362 0.355
2
5
G
4

5
I
8
394 0.530
3
5
F
1

5
I
8
461 0.300
4
5
S
2

5
I
8
549 1.000
5
5
F
3

5
I
8
659 0.277
6
5
S
2

5
I
7
758
AlN:Gd 1
6
P
7/2

8
S
7/2
314
Table 3.1. Summary of Ho
3+
and Gd
+3
ions emission from a-AlN:Ho and a-AlN:Gd.

Ho relative
concentration
Gd relative
Concentration
Relative Intensity of
549 nm peak
% increase in
intensity
1 0 1 0 (reference)
1 1 1.275 27.5
1 2 1.564 56.4
1 3 1.747 74.7
1 4 2.074 107.4
1 5 2.047 104.7
1 6 2.054 105.4
Table 3.2. Effect of Gd concentration on the luminescence of Ho
5
S
2

5
I
8
transition at 549
nm.

Cathodoluminescence

182
3.2.2.4 Praseodymium
Data for Pr in a-AlN are given by Maqbool(Maqbool 2005).

Material Transition Number
Transition
Assignment
Wavelength
(nm)
Intensity(a.u)
AlN:Pr 1
1
S
0

1
D
2
335 162
2
1
S
0

1
I
6
385 438
3
3
P
2

3
H
4
439 362
4
3
P
0

3
H
4
488 1002
5
3
P
0

3
H
4
504 1024
6
3
P
1

3
H
5
526 4577
7
3
P
0

3
H
5
573 767
8
3
P
0

3
H
6
618 643
9
3
P
0

3
F
2
652 1227
10
3
P
0

3
F
3
710 410
11
3
P
0

3
F
4
738 520
Table 3.3. Summary of Pr
3+
ion emissions from Pr-doped a-AlN.

Pr Transition
Wavelength
(nm)
Gd Relative
Concentration
Relative
Intensity
% Increase in
Intensity

3
P
0

3
H
4
490 0 1 0 (reference)
1 1.58 58
4 4.37 337

3
P
0

3
H
6
618 0 1 0
1 1.73 73
4 4.14 314

3
P
0

3
F
2
649 0 1 0
1 1.66 66
4 3.4 240
Table 3.4. Effect of Gd concentration on the luminescence of Pr
3+
ion transitions.

183
3.2.2.5 Thulium
Data for Tm in a-AlN are given by Maqbool (Maqbool 2005).

Material
Transition
Number
Transition
Assignment
Wavelength
(nm)
Relative
Intensity(a.u)
AlN:Tm 1 D
2

3
H
6
371 0.126
2
1
D
2

3
F
4
467 1.000
3
1
D
2

3
H
6
480 0.648
4
1
D
2

3
H
5
528 0.092
5
1
G
4

3
F
4
650 0.126
6
1
D
2

3
H
4
685 0.061
7
3
H
4

3
H
6
802 0.563
8
3
H
4

3
H
6
808 0.539
Table 3.5. Summary of Tm
3+
ion emissions from Tm-doped AlN.
3.2.2.6 Samarium
Data for Sm in a-AlN are given by Maqbool (Maqbool 2005).

Material
Transition
Number
Transition
Assignment
Wavelength
(nm)
Relative
Intensity(a.u)
AlN:Tm 1
4
G
5/2

6
H
5/2
564 0.425
2
4
G
5/2

6
H
7/2
600 1.000
3
4
G
5/2

6
H
9/2
648 0.686
4
4
G
5/2

6
H
11/2
711 0.312
5 Gd impurity 314 0.276
6 Cr impurity 692 0.220
Table 3.6. Summary of Sm
3+
ion emissions from Sm-doped a-AlN.
3.2.3 Amorphous gallium nitride
Amorphous GaN usually has a large defect luminescence in the yellow region of the visible
spectrum. In Fig. 3.13, an a-GaN:Er thin film is shown in CL. This film was heated to 950
C in nitrogen for 30 minutes. The Er
3+
ion transitions are clearly visible on the yellow
defect emission band. The defect band problem is much less in crytsalline GaN, which was
extensively investigated by several groups, most notably Steckl and his coworkers.

Cathodoluminescence

184

Fig. 3.13. CL from a-GaN: Er.
Maqbool (Maqbool, Richardson et al. 2005) investigated Pr
3+
ion luminescence in a-GaN.
The data are given in Table 3.7 .

Material
Transition
Number
Transition
Assignment
Wavelength
(nm)
Relative
Intensity(a.u)
3
3
P
2

3
H
4
418 559
4
3
P
0

3
H
4
493 5687
6
3
P
1

3
H
5
532 1895
8
3
P
0

3
H
6
621 4015
9
3
P
0

3
F
2
650 4227
10
3
P
0

3
F
3
713 1508
11
3
P
0

3
F
4
736 1627
Table 3.7. GaN:Pr
3.2.4 Other amorphous nitrides
3.2.4.1 BN
Sputtered boron nitride was shown to be a useful host for RE ions by Maqbool et al.
(Maqbool, Richardson et al. 2005) and Allen et al. (Allen S. 2000).

185

Fig. 3.14. Left: CL from sputter deposited BN:Tb and pure BN. Right: CL spectra of BN and
BN:Tb, as deposited and heated to 900 K. At large Tb concentrations, the blue BN
luminescence is not visible.
Maqbool et al. (Maqbool, Richardson et al. 2005) investigated the luminescence of Pr in BN.
Data are given below in Table 3.8 (Maqbool 2005)

Material
Transition
Number
Transition
Assignment
Wavelength
(nm)
Relative
Intensity(a.u)
3 BN transition 415 372
4
3
P
0

3
H
4
492 973
6
3
P
1

3
H
5
544 752
8
3
P
0

3
H
6
628 842
9
3
P
0

3
F
2
651 748
10
3
P
0

3
F
3
713 257
11
3
P
0

3
F
4
736 289
Table 3.8. BN:Pr
3.2.4.2 BeN
Amorphous Beryllium Nitride, produced by reactive sputtering of Be in nitrogen, was
studied with samarium (Zanatta, Richardson et al. 2007).
PL and CL excitation was achieved with 532 nm photons and 10 keV electrons, respectively.
Light emission is due to the Sm
3+
ions and correspond to the following transitions (see inset
on the left in Fig. 3.15):
4
G
5/2

6
H
5/2
(A),
4
G
5/2

6
H
7/2
(B),
4
G
5/2

6
H
9/2
(C), and
4
G
5/2

6
H
11/2
(D). A photograph illustrating the cathodoluminescence image of the Sm-doped a-
BeN sample is presented in the upper-right corner of Fig. 3.15.

Cathodoluminescence

186

Fig. 3.15. Room-temperature luminescence spectra of a Sm-doped a-BeN film. From(Zanatta,
Richardson et al. 2007)
4. Transition metal cathodoluminescence
Elements in the periodic table that make a bridge between the first two groups and last six
groups of the main elements are defined as transition metals (TM). This bridging role has
given these elements the name transition metals. These metals consist of many elements,
however, we shall limit our discussion to a few of them, mainly
22
Ti,
24
Cr,
39
Y, and
74
W .
These metals are specially characterized by their common feature: partial occupancy of the
d-shells. This partially filled d shell configuration assigns special properties to these
elements, including their strong suitability for use as optical and luminescent materials. To
study their optical and luminescent properties it will be useful to understand some of their
basic characteristics. Housecroft (Housecroft et al 2007) and Mackay (Mackay 1996) reported
some of the basic characteristics of transition metals. The discussion of these characteristics
in section 4.1 follows Housecroft and Mackay.
4.1 Characteristics of transition metals
4.1.1 Density and metallic radii
The transition elements are much denser than the s-block elements and show a gradual
increase in density from scandium to copper. This trend in density can be explained by the
small and irregular decrease in metallic radii coupled with the relative increase in atomic mass.

187
4.1.2 Melting and boiling points
The melting points and the molar enthalpies of fusion of the transition metals are both high
in comparison to main group elements. This arises from strong metallic bonding in
transition metals which occurs due to delocalization of electrons facilitated by the
availability of both d and s electrons.
4.1.3 Ionization energies
In moving across the series of metals from scandium to zinc a small change in the values of
the first and second ionization energies is observed. This is due to the build-up of electrons
in the immediately underlying d-sub-shells that efficiently shields the 4s electrons from the
nucleus and minimizing the increase in effective nuclear charge from element to element.
The increases in third and fourth ionization energy values are more rapid. However, the
trends in these values show the usual discontinuity half way along the series. The reason is
that the five d electrons are all unpaired, in singly occupied orbitals. When the sixth and
subsequent electrons enter, the electrons have to share the already occupied orbitals
resulting in inter-electron repulsions, which would require less energy to remove an
electron. Hence, the third ionization energy curve for the last five elements is identical in
shape to the curve for the first five elements, but displaced upwards by 580 kJ mol
-1
.
4.1.4 Electronic configuration
The electronic configuration of the atoms of the first row transition elements are basically
the same. It can be seen in Table 4.1 that there is a gradual filling of the 3d orbitals across the
series starting from scandium. This filling is, however, not regular, since at chromium and
copper the population of 3d orbitals increase by the acquisition of an electron from the 4s
shell. This illustrates an important generalization about orbital energies of the first row
transition series. At chromium, both the 3d and 4s orbitals are occupied, but neither is
completely filled in preference to the other. This suggests that the energies of the 3d and 4s
orbitals are relatively close for atoms in this row.
In the case of copper, the 3d level is full, but only one electron occupies the 4s orbital. This
suggests that in copper the 3d orbital energy is lower than the 4s orbital. Thus the 3d orbital
energy has passed from higher to lower as we move across the period from potassium to
zinc. However, the whole question of preference of an atom to adopt a particular electronic
configuration is not determined by orbital energy alone. In chromium it can be shown that
the 4s orbital energy is still below the 3d which suggests a configuration [Ar]3d
4
4s
2
.
However due to the effect of electronic repulsion between the outer electrons the actual
configuration becomes [Ar]3d
5
4s
1
where all the electrons in the outer orbitals are unpaired.
Table 1 gives some of the physical properties and free atom electronic configuration of
transition metals.
4.1.5 Oxidation states
Oxidation states of transition metals are very important to study the spectroscopy and
luminescence from these metal ions. The partially filled d-shell electrons play important
role in the oxidation states of transition metals. To fully understand the phenomena of

Cathodoluminescence

188
oxidation states of transition metals, we have to understand how the unpaired d-orbital
electrons bond. There are five orbitals in a d subshell manifold. As the number of
unpaired valence electrons increases, the d-orbital increases, the highest oxidation state
increases. This is because unpaired valence electrons are unstable and eager to bond with
other chemical species. This means that the oxidation states would be the highest in the
very middle of the transition metal periods due to the presence of the highest number of
unpaired valence electrons. To determine the oxidation state, unpaired d-orbital electrons
are added to the 2s-orbital electrons since the 3d-orbital is located before the 4s-orbital in
the periodic table. For example: Scandium has one unpaired electron in the d-orbital. It is
added to the 2 electrons of the s-orbital and therefore the oxidation state is +3. So that
would mathematically look like: 1s electron + 1s electron + 1d electron = 3 total electrons
= oxidation state of +3.
The formula for determining oxidation states would be (with the exception of copper and
chromium):

=
Highest Oxidation State for a Transition metal
Number of Unpaired d electrons Two s orbital electrons

Element Group
Density
(g/cm
3
)
M.P.
(C)
B.P.
(C)
Radius
(pm)
Free atom
configuration
ionization
energy
(kJ mol
-1
)
Oxidation
state
Sc 3 2.99 1541 2831 164 [Ar] 3d
1
4s
2
631 +3
Ti 4 4.50 1660 3287 147 [Ar]3d
2
4s
2
658 +2, +3, +4
V 5 5.96 1890 3380 135 [Ar]3d
3
4s
2
650
+2, +3, +4,
+5
Cr 6 7.20 1857 2670 129 [Ar]3d
5
4s
1
653 +2, +3, +6
Mn 7 7.20 1244 1962 137 [Ar]3d
5
4s
2
717
+2, +3, +4,
+6, +7
Fe 8 7.86 1535 2750 126 [Ar]3d
6
4s
2
759 +2, +3
Co 9 8.90 1495 2870 125 [Ar]3d
7
4s
2
758 +2, +3
Ni 10 8.90 1455 2730 125 [Ar]3d
8
4s
2
737 +2
Cu 11 8.92 1083 2567 128 [Ar]3d
10
4s
1
746 +2
Zn 12 7.14 420 907 137 [Ar]3d
10
4s
2
906 +2
Table 4.1. Physical properties and free atom electronic configuration of transition metals.

189
Scandium is one of the two elements in the first transition metal period which has only one
oxidation state (zinc is the other, with an oxidation state of +2). All the other elements have
at least two different oxidation states. Manganese, which is in the middle of the period, has
the highest number of oxidation states, and indeed the highest oxidation state in the whole
period since it has five unpaired electrons.
It was mentioned previously that both copper and chromium do not follow the general
formula for transition metal oxidation states. This is because copper has 9 d-electrons, which
would produce 4 paired d-electrons and 1 unpaired d-electron. Since copper is just 1
electron short of having a completely full d-orbital, it steals an electron from the s-orbital,
allowing it to have 10 d-electrons. Likewise, chromium has 4 d-electrons, only 1 short of
having a half-filled d-orbital, so it steals an electron from the s-orbital, allowing chromium
to have 5 d-electrons.
4.2 Luminescence from pure TM doped in nitride semiconductors
Luminescence and spectroscopic properties of materials play an important role in optical
device fabrication, display technologies and the laser industry; transition metals are not
exceptions. The optical properties of transition metal ion solids have been studied for
many years. Hidalgo (Hidalgo, Mendez et al. 1998) and Muller (Muller, Zhou et al. 2009)
studied that partially occupied d-shells of these elements play important role in the
luminescence and spectroscopy of these elements. Thurbide (Thurbide and Aue 2002),
Gedam (Gedam, Dhoble et al. 2007), Grinberg (Grinberg, Barzowska et al. 2001) and
Lapraz (Lapraz, Iacconi et al. 1991) reported that in many host materials, these partially
occupied d-shells give rise to several important technological applications and in
particular, production of high resistivity materials and photonic devices through
transition-metal doping has been widely used. Maqbool (Maqbool, Wilson et al. 2010) and
Martin (Martin, Spalding et al. 2001) showed that thermal activation, the oxygen effect
and co-dopants are good tools to obtain high efficiency and improved luminescence from
these metal ions.
Recent

progress toward nitride-based light-emitting diodes and electroluminescent devices
(ELDs) has

been made using crystalline and amorphous nitride semiconductors doped

with
a variety of transition metals: Richardson (Richardson, Van Patten et al. 2002), Maqbool
(Maqbool, Main et al. 2010; Maqbool, Wilson et al. 2010) and Caldwell (Caldwell, Martin et
al. 2001). The amorphous III-nitride semiconductors are equally important as their
crystalline counterpart because the amorphous material can be grown at room temperature
with little stress due to lattice mismatch. They may also be more suitable for waveguides
and cylindrical and spherical laser cavities because of the elimination of grain boundaries at
low temperature growth. Cathodoluminescence, spectroscopy and the effects of various
factors on the luminescence of a few TMs are given below.
4.2.1 Chromium doped in amorphous aluminum nitride
Thin films of a-AlN:Cr were prepared and deposited on Si (100) substrate by the method of
plasma magnetron sputtering at low temperature as described earlier. The x-ray diffraction
(XRD) analysis confirmed that the deposited films were amorphous.

Cathodoluminescence

190
Figure 4.1 shows the XRD analysis of the a-AlN:Cr films deposited on flat Si(100) substrate.
Only one peak can be observed in the film at 69.1 which corresponds to Si(100). No other
peak is present in the figure, indicating that the films deposited on flat silicon substrates are
amorphous. Thermal activation of the films at 1200 K has not changed the structure of the
films.

Fig. 4.1. XRD analysis of the a-AlN:Cr films deposited on flat Si (100) substrate.
Cathodoluminescence (CL) of the Cr-doped a-AlN was obtained to study the suitability of
this TM for various applications in optical and display technology. Figure 4.2 shows the CL
spectrum of the as-deposited and thermally activated a-AlN:Cr in 600 nm to 750 nm range.

Fig. 4.2. CL spectra of room temperature and thermally activated a-AlN:Cr films.

191
A strong emission has occurred at 702 nm indicated by a sharp peak in Fig. 4.2. This peak
corresponds to
4
T
2

4
A
2
transition. It is clear from the figure that thermal activation
enhances the luminescence intensity six times. A small peak at 661 nm is also observed.
This peak corresponds to Ho
+3
ion, indicating the presence of holmium impurity in the
films.
Figure 4.2 reports that the intensity of the emission is not only strong but an enhancement
in the luminescence is possible by various means like thermal activation. The strong
intensity enabled us to see the red emission in CL apparatus directly with naked eye
when the excitation current is reasonably high. Six times increase in the emitted light
intensity makes it possible that a-AlN:Cr can serve as a potential candidate for a laser
production at 702 nm and other optical devices applications. Moreover the significant
increase in the intensities of luminescence from Cr
+3
ions by thermal activation can be
explained on the basis of luminescence from the triply ionized chromium ions.
Luminescence occurs from Cr
+3
ions and not from Cr
+2
or Cr
+1
. During the film deposition
it is most likely that some of Al
+3
of AlN may be replaced by Cr
+3
but there are also
chances for imperfections and defects giving rise to Cr
+2
and Cr
+1
during film growth.
These ions do not contribute to luminescence. The smaller the number of these ions, more
will be Cr
+3
ions and hence luminescence will be higher. When these films are activated
thermally at a higher temperature then most of Cr
+2
and Cr
+1
impurities ionize and
convert to Cr
+3
ions giving a path to enhanced luminescence. Moreover when the films
are transferred to the furnace and thermally activated after removed from the deposition
chamber, they are exposed to air. Thus oxidation of the surface of the film cannot be
ignored. Oxygen enhances the luminescence of TM ions giving rise to the enhanced
luminescence after thermal activation of the films. Chen (Chen, Chen et al. 2000), Little
and Kordesch (Little and Kordesch 2001) and Suyver (Suyver et al. 2005) have reported
such results in other materials as well.
The results show that amorphous AlN:Cr is a promising candidate for its use in optical and
photonic devices and communication tools. The strong red-IR emission makes this material
a potential candidate for making laser cavities, quantum dots and other wave-guided
applications. Due to the high penetration ability of near infrared light in human tissues, it
can also be used in biomedical applications.
4.2.2 Tungsten doped in amorphous aluminum nitride
Another important member of the TM family is Tungsten (W). Tugsten has the highest
melting point and lowest vapor pressure of all metals. It has a very high tensile strength.
Our investigations revealed that along with other physical properties tungsten can also be
used for visible light emission applications. The cathodoluminescence of tungsten shows
that it gives a very broad emission under cathode ray excitation. The emission spectrum is
so broad that it covers the entire visible range of the electromagnetic spectrum from 350 nm
up to 700 nm. However the dominant portion of the spectrum comes in blue region with a
peak at 491 nm. Another peak in blue is also observed at 429 nm. Because of the huge
portion of the spectrum and its peaks lying in blue the films also looked blue in appearance
when directly exposed to the electron beam in CL. This broad CL emission from tungsten is
shown in Fig. 4.3.

Cathodoluminescence

192

Fig. 4.3. CL spectrum of amorphous AlN:W
4.2.3 Interaction of co-doped Gd
+3
with AlN:W
Due to its broad emission W
+3
is also able to interact with other ions when co-doped with it.
For this purpose it was also co-doped with holmium and gadolinium separately to test for
luminescence enhancement in the green emission from holmium ions or luminescence
enhancement in the AlN:W by gadolinium. It was observed that no change has occurred in Ho
luminescence by the addition of W. However the luminescence from Gd has enhanced W
luminescence. Figure 4.4 shows how the addition of Gd
+3
affects the light emission from W
+3
.
Figure 4.4 shows luminescence from amorphous AlN films co-doped with 1 piece of W and
1-3 pieces of Gd. It is clear from the figure that the addition of Gd has a dominant effect on
the luminescence from W. Luminescence from AlN:1W2Gd is 3 times the luminescence from
AlN:1W1Gd and luminescence from AlN:1W3Gd is almost 6 times the luminescence from
AlN:1W1Gd. This fact shows a huge increase in intensities in the W luminescence by Gd.
Figure 4.4 also reveals some other information. It can be observed that the main peak in blue
shifts with the relative increase in Gd concentration. With one piece of Gd the main peak of
W appears at 506 nm. With two pieces of Gd added to AlN:W the peak gets bigger but shifts
to 534 nm. However with the addition of another piece of Gd the peak gets bigger but shifts
back to 523 nm. A possible explanation for this shift in the peaks may be the activation of
new radiative energy levels in W by the Gd. Further, the intensity of Gd peak is reduced
with higher concentration of Gd. Table 4.2 is giving the increase in the W intensity with the
concentration of Gd and also the intensity of Gd peak with the concentration of Gd.

193

Fig. 4.4. The effect of Gd concentration on the luminescence of a-AlN:W films.
4.2.4 Interaction of co-doped Ho
+3
with AlN:W
Tungsten is added to Holmium (Ho) in order to observe any enhancement in the green
luminescence from Ho. Figure 4.5 shows the effect of varying W concentration on Ho. A
comparison between amorphous AlN:1Ho1W, AlN:1Ho3W and thermally activated
AlN:1Ho1W is given in this figure. It is clear from the figure that there is no effect of
increasing W concentration on the Ho luminescence. The W concentration is increased 3
times which made the W peak broad but no enhancement in Ho. Further, if we compare this
figure with the pure W spectrum then a shift can be seen in the W emission wavelength just
like that happened due to Gd. And we see from the figure that in the presence of Ho,
tungsten emits a single blue light with a wavelength of 461 nm rather than emissions at 429
nm and 491 nm in pure tungsten. A possible explanation may be the holmium has affected
the energy level distribution in tungsten. Further it can also be given in explanation that this
shift of wavelength did not occur in rare-earth elements discussed in previous chapters but
the transition element W has suffered a lot from this shift. Since rare-earth elements are
favorable for their internal f-f transitions. These f-levels lying inside the other shells and
hence any external change barely affect these transitions. However it is not true for
transition metals and they can be significantly affected by other dopant impurities.
Moreover this figure also gives us the effect of thermal annealing on the luminescence of W
and Ho. It can be easily deduced from the figure that thermal annealing has activated Ho
luminescence more than W. There is also existence of a ruby impurity, which is considerably
enhanced in luminescence by thermal activation.

Cathodoluminescence

194

Fig. 4.5. The effect of thermal activation and W concentration on Ho luminescence.
Gd Concentration
Intensity of Gd
luminescence (a.u)
Intensity of W
luminescence (a.u)
Percent increase in
luminescence of W
1 2207 208 Reference
2 1318 705 339
3 538 1159 557
Table 4.2. The effect of Gd
+3
concentration on the W and Gd emission.

195
4.2.5 Cathodoluminescence and thermal activation of AlN doped with Yttrium (Y)
Cathodoluminescence of cold deposited Yttrium doped AlN films were characterized for CL
in the same way described earlier for other materials. Figure 4.6 represents the luminescence
from AlN:Y films when exposed to cathode rays in CL assembly. This figure confirms
emission from amorphous AlN:Y films in UV , blue and in bluish green regions. There is
also a peak in IR which may correspond to Yttrium or possibly a ruby impurity. The
wavelengths correspond to these emissions are 360 nm, 421 nm, 518 nm and 705 nm
respectively. The emission 705 nm could be from AlN:Y or from Cr
+3

The bluish green and
blue emissions are the dominant in intensity and that is why the films appear blue to naked
eye when exposed to electron beam in CL. However we already know from the previous
materials investigations that thermal annealing tremendously enhance the luminescence
from Cr
+3

and hence we can analyze this peak obtaining the CL after performing thermal
annealing.
Figure 4.7 gives a comparison of CL spectra from amorphous AlN:Y before and after
thermal activation. Films were thermally activated at 900 C for one hour and then
characterized for CL. From the figure it is clear that thermal activation has almost no effect
on the Yttrium luminescence at 360 nm, 421 nm and 518 nm. However the peak at 705 nm is
strongly enhanced by thermal activation. This enhancement in the peak at 705 nm is more
than three and half times the luminescence from the same peak in the film, which is not
thermally activated. Further, this peak was less than half the intensity of peak at 518 nm
before thermal activation but after activation it is about one and half time more intense than
the peak at 518 nm. One can guess at this stage that the peak at 705 is most probably due to
chromium oxide (Ruby) rather than emission from yttrium ions itself.

Fig. 4.6. CL spectrum of amorphous AlN:Y

Cathodoluminescence

196

Fig. 4.7. Comparison of the CL spectra of thermally activated and as deposited amorphous
AlN:Y films.
4.2.6 Titanium doped amorphous aluminum nitride
Titanium is one of the important members of the TM family. It is lightweight, strong,
corrosion resistant metal and the ninth most plentiful amongst all the elements in the earths
crust. This metal, when doped in nitride semiconductors, is proved to be a very good source
to make laser cavities and optical devices. The cathodoluminescence of Ti
+3
doped in
amorphous AlN shows a broad emission over a wide range from 650 nm up to 900 nm with
a peak around 760 nm. Figure 4.8 shows the CL spectrum of a-AlN:Ti. The broad emission is
clear from the spectrum.

197

Fig. 4.8. Cathodoluminescence spectrum of amorphous AlN:Ti.
5. Applications and special geometries
5.1 Amorphous AlN:Tm (Ti) on optical fiber in a cylindrical and ring geometry
The broad emission from a-AlN:Ti, as discussed in the previous section, makes it hard to use
it directly for any particular wavelength application. However it has a great potential for
narrowing the peak by a resonance emission to produce a laser out of Ti
+3
. The resonance
emission to produce a laser will definitely need a cavity. Rather than making a traditional
cavity, however, we made a cavity by depositing a-AlN:Ti, around optical fiber and obtained a
microlaser with emission wavelength of 780.5 nm. The fiber acts as a cavity when the light
emitted from the deposited film circulates around the fiber. Only those light waves will
enhance each other whose wavelength is an integral multiple of their path around the fiber.
Figure 5.1 represent thin film deposition and light propagation around optical fiber, where
5.1(a) shows longitudinal view of the fiber containing thin film around it. The dark red
region around the fiber is the a-AlN:Ti, film deposited uniformly around the fiber. Figure
5.1(b) is a cross sectional view of the fiber with the film, showing light propagation in
whispering gallery mode (WGM), The diameter s of this section is 12 m. The dark red
region around the fiber is the a-AlN:Ti film deposited on the fiber. The thickness d of this
film is 4 micron. This makes the total diameter of the fiber and the film around it to be 20
micron (D = 2d + s = 20 m). The white lines in the film represent the propagation of light in
the film. The pattern of light propagation is restricted to the a-AlN:Ti film only, without
touching the fiber itself. Such arrangement is known as whispering gallery. Figure 5.1(c)
shows pump laser coupled to fiber and microring emission. Optical fiber with a-AlN:Ti film
was held such that its longitudinal axis is perpendicular to the pump laser beam direction.
The pump laser hits the thin films around the fiber directly, exciting a-AlN:Ti for infrared
emission, that generates whispering gallery modes in the fiber. The laser produced in the
fiber cavity comes out as a microring emission from the fiber.

Cathodoluminescence

198

Fig. 5.1. Whispering gallery mode microlaser cavity formation around optical fiber.
Figure 5.2 shows the band narrowing and laser action in a-AlN:Ti films deposited around
the optical fiber excited by 532 nm Nd:YAG laser. The power of the excitation laser is varied
between 7.5 mW and 30 mW. The figure shows two emission spectra. The broad spectrum
showing just the fluorescence emission from the deposited films was obtained when 15 mW
of power from the Nd:YAG laser is used for excitation. This broad emission spectrum is
about 20 nm wide and no lasing action is observed in this spectrum under 15 mW excitation
power. The sharp and very intense emission with narrow bandwidth shows lasing action in
the film. The laser is produced in the whispering gallery mode at 780.5 nm. We also
observed the secondary modes in this spectrum, confirming the resonances in the lasing
action. The primary mode lasing and the secondary mode emission peak show gain of ~ 20
at an input power around 30 mW.

199
Figure 5.3 gives increase in the observed near infrared laser in a-AlN:Ti with the increasing
power of Nd:YAG pump laser. The curve shows that no laser action can be achieved in a-
AlN:Ti if the power P of excitation laser is less than a threshold value P
0
. For P > P
0
an
almost linear increase in the titanium microlaser emission was observed with increasing
excitation laser power. The threshold power P
0
was found to be about 23.5 mW. For P < P
0

we just observed a fluorescence emission from the fiber with an emission width of 35 nm.
750 760 770 780 790 800
0
500
1000
1500
2000
2500
3000
I
n
t
e
n
s
i
t
y

(
a
.
u
)
Wavelength (nm)
22.5 mW
30 mW

Fig. 5.2. Laser action at 780.5 nm in a-AlN:Ti thin films around optical fiber. The Equally
spaced peaks at 776.4 nm and 785.5 nm are the secondary modes of laser. Mode spacing is
4.6 nm.

Cathodoluminescence

200
5 10 15 20 25 30
0
500
1000
1500
2000
2500
3000
Threshold Power = 24 mW
I
n
t
e
n
s
i
t
y

(
a
.
u
)
Pump Laser Power (mW)

Fig. 5.3. Increase in the a-AlN:Ti laser intensity with the pump laser power. A threshold
power of 24 mW can be observed.
In micro-ring cavities, the thin a-AlN:Ti film deposited around the optical fiber supports
whispering gallery modes. Frolov (Frolov, Fujii et al. 1998) (Frolov, Vardeny et al. 1998) and
Maqbool and Kordesch (Maqbool, Main et al. 2010) figured out that the angular momentum
mode number (n) for a WGM is given by equation (1), where D is the diameter of the optical
fiber with thin film deposited on it,
n
is the wavelength of the WGM and m is the index of
refraction for the a-AlN film.
n = D m /
n
(1)
Khoshman and Kordesch (Khoshman and Kordesch 2005) worked on amorphous AlN and
found that the refractive index in the near infrared region (780.5 nm) is 1.95. This value of
the index of refraction of amorphous AlN is obtained using the films deposited in the same
deposition system and the same conditions that we used for our work. Thus, m = 1.95 is the
most appropriate value to use in our work. Our results satisfy the given equation for the
integers n = 157. Because the film around the optical fiber is thick enough to satisfy the
WGM condition (d/D 0.2), we did not observe waveguide modes. The a-AlN:Ti doped
optical fiber was placed in vertical direction so that the ring laser formation occur in a
horizontal axis. The resonance wavelength
n
should also support the mode separation
equation, given by;
=
n
-
n+1
=
2
/ D m (2)

201
Using n = 156, the outer diameter D = 20 micron, and m = 1.95, equation (2) gives = 4.9
nm. From figure 5.2 we find that our experimental results give = 4.6, which is in quite
good agreement with the theoretical calculations.
The a-AlN:Ti laser, we produced, is very important from biomedical applications point of
view. Along with its uses in optics and photonics this laser can also be very beneficial in
health sciences applications, particularly for diagnostic and therapeutic purposes.
Researchers have reported that near infrared light with a wavelength between 700 nm and
900 nm has minimum absorption and the greatest penetration in body tissues (Noriyuki,
Ohdan et al. 1997; Cerussi, Shah et al. 2006). Our laser produced at 780.5 nm is in this range
and hence, can be used for diagnosis of deep tissues abnormality and tumors, and laser
surgery of deep body tissues due to its high penetration ability in the human body.
5.2 Textiles
The deposition of amorphous AlN onto textiles was examined by Kordesch and Richardson
(Kordesch and Richardson 2003). The non-woven textiles (clean room suit material) is made
by thermally binding polymer threads. The threads are bound by many pads so that no
loose threads are lost. The material is sensitive to heat, so that it is a good test for the
successful deposition of a RE doped Al coated textile. In this case the dopant was Tb.

Right: Cathodoluminescence from the a-AlN:Ti coated textile. The voltage and current density used
was 2.8 kV and 0.15uA/cm
2
. Field of View 30mm.
Fig. 5.4. Left: a-AlN:Ti film on a thermally bonded textile pad. Field of view is 3mm.
5.3 Thermometery
A novel use of a-AlN:Ti thin films has been developed by Richardson and co-workers
(Carlson, Khan et al. 2011; Wang, Carlson et al. 2011).
While the example given in Figure 5.5 is necessarily photoluminescence because the intention
of the study is to determine the heating due to a gold particle in a liquid or biological
environment, Er based thermometry could be used in CL and in other environments. Several
other RE based thermometry systems are possible(Alden, Omrane et al.; Lai, Feng et al.).

Cathodoluminescence

202

Fig. 5.5. Schematic of the Er thermometer. The Er:AlGaN film is deposited onto a substrate.
The gold particle on the film surface is excited with laser light through the objective of a
microscope. The photoluminescence spectrum of the Er peaks at 540 and 565 nm are used to
determine the temperature of the film surrounding the substrate. From (Carlson, Khan et al.
2011), with permission.
6. Acknowledgments
We would like to thank all of the former members of our research teams who have worked
on Rare Earth luminescence in the amorphous nitrides. In particular, we thank Professors
Henryk J. Lozykowski, A. Ricardo Zanatta, and Hugh H. Richardson. This work was
funded by several agencies, including grants from the Ballistic Missile Defence Organization
and the Defence Advanced Research Projects Agency administered by the Office of Naval
Research: N00014-96-1782 entitled Growth, Doping and Contacts from Wide Band Gap
Semiconductors and grant N00014-99-1-0975 entitled Band-Gap Engineering of the
Amorphous In-Ga-Al Nitride Semiconductor Alloys for Luminescent Devices from the
Ultraviolet to the Infrared.
7. References
Adachi, S. (1999). Optical Constants of Crystalline and Semiconductors:Numerical Data and
Graphical Information, Springer.
Aldabergenova, S. B., G. Frank, et al. (2006). "Structure changes of AlN: Ho films with
annealing and enhancement of the Ho3+ emission." Journal Of Non-Crystalline Solids
352(9-20): 1290-1293.
Aldabergenova, S. B., A. Osvet, et al. (2002). "Blue, green and red emission from Ce3+, Tb3+
and Eu3+ ions in amorphous GaN and AlN thin films." Journal Of Non-Crystalline
Solids 299: 709-713.

203
Alden, M., A. Omrane, et al. "Thermographic phosphors for thermometry: A survey of
combustion applications." Progress In Energy And Combustion Science 37(4): 422-461.
Allen S., R., H.H. and Kordesch. M.E. (2000). Luminescence of Transition and Lanthanide
Metal Ions in Sputtered BN Thin Films, Ohio Univerity.
Caldwell, M. L., A. L. Martin, et al. (2001). "Emission properties of an amorphous AlN: Cr3+
thin-film phosphor." Applied Physics Letters 78(9): 1246-1248.
Caldwell, M. L., P. G. Van Patten, et al. (2001). "Visible luminescent activation of amorphous
AlN: Eu thin-film phosphors with oxygen." Mrs Internet Journal Of Nitride
Semiconductor Research 6(13): 1-8.
Carlson, M. T., A. Khan, et al. (2011). "Local Temperature Determination of Optically Excited
Nanoparticles and Nanodots." Nano Letters 11(3): 1061-1069.
Cerussi, A., N. Shah, et al. (2006). "In vivo absorption, scattering, and physiologic properties
of 58 malignant breast tumors determined by broadband diffuse optical
spectroscopy." Journal Of Biomedical Optics 11(4).
Chen, H., K. Y. Chen, et al. (2000). "Band gap engineering in amorphous AlxGa1-xN:
Experiment and ab initio calculations." Applied Physics Letters 77(8): 1117-1119.
Chen, H., K. Gurumurugan, et al. (2000). "Visible and infrared emission from GaN: Er thin
films grown by sputtering." Mrs Internet Journal Of Nitride Semiconductor Research 5:
art. no.-W3.16.
Chen, K. Y. and D. A. Drabold (2002). "First principles molecular dynamics study of
amorphous AlxGa1-xN alloys." Journal Of Applied Physics 91(12): 9743-9751.
Dieke, G. H. (1968). Spectra and Energy Levels of Rare Earth Ion in Crystals. New York, Wiley.
Dierolf, V., et al., eds. (2011). Rare-Earth Doping of Advanced Materials for Photonic Applications.
Dimitrova, V. I., P. G. Van Patten, et al. (2001). "Photo-, cathodo-, and electroluminescence
studies of sputter deposited AlN: Er thin films." Applied Surface Science 175: 480-483.
Dimitrova, V. I., P. G. Van Patten, et al. (2000). "Visible emission from electroluminescent
devices using an amorphous AlN: Er3+ thin-film phosphor." Applied Physics Letters
77(4): 478-479.
Drabold, D. A. a. E., S., eds. (2010). Theory of Defects in Semiconductors, Springer.
Ebdah, M. (2011). Engineering the Optical, Structural, Electrial and Magnetic Properties of
Oxides and Nitrides of In-Ga-Zn Thin Films Using Nanotechnology. Physics and
Astronomy. Athens, Ohio University. PhD: 155.
Frolov, S. V., A. Fujii, et al. (1998). "Cylindrical microlasers and light emitting devices from
conducting polymers." Applied Physics Letters 72(22): 2811-2813.
Frolov, S. V., Z. V. Vardeny, et al. (1998). "Plastic microring lasers on fibers and wires."
Applied Physics Letters 72(15): 1802-1804.
Gedam, S. C., S. J. Dhoble, et al. (2007). "Dy3+ and Mn2+ emission in KMgSO4Cl phosphor."
Journal Of Luminescence 124(1): 120-126.
Grinberg, M., J. Barzowska, et al. (2001). "Inhomogeneous broadening of the dominant Cr3+
sites in LiTaO3 system." Journal Of Luminescence 94: 85-90.
Gurumurugan, K., H. Chen, et al. (1999). "Visible cathodoluminescence of Er-doped
amorphous AlN thin films." Applied Physics Letters 74(20): 3008-3010.
Hassan, Z. (1998). Growth, Characterizationation and Fabrication of GaN-based Device
Structures. Physics and Astronomy. Athens, Ohio University. PhD: 1-158.
Hidalgo, P., B. Mendez, et al. (1998). "Luminescence properties of transition-metal-doped
GaSb." Physical Review B 57(11): 6479-6484.
Housecroft, C., and Sharpe, A.G. (2007) Inorganic Chemistry (3rd Edition), Upper Saddle
River, NJ, Prentic Hall.

Cathodoluminescence

204
Iacona, F., G. Franzo, et al. (2009). "Er-based materials for Si microphotonics." Optical
Materials 31(9): 1269-1274.
Ihn, T. (2010). Semiconductor Nanostructures: Quantum states and electronic transport, Oxford
University Press.
Jadwisienczak, W. M., H. J. Lozykowski, et al. (2000). Visible emission from AlN doped with Eu,
Tb, and Er ions. 2001 Ieee International Symposium On Compound Semiconductors: 489-494.
Jadwisienczak, W. M., H. J. Lozykowski, et al. (2000). "Luminescence of Tb ions implanted
into amorphous AlN thin films grown by sputtering." Applied Physics Letters 76(23):
3376-3378.
Kasai, H., A. Nishikawa, et al. (2010). "Improved Eu Luminescence Properties in Eu-Doped
GaN Grown on GaN Substrates by Organometallic Vapor Phase Epitaxy." Japanese
Journal Of Applied Physics 49(4).
Kenyon A.J. and Lucarz, F. "A Critique of the Existing Model for Excitation Exchange
Between Silicon Nanostructures and Erbium Ions in SIlica." from
http://discovery.ucl.ac.uk/1306794.
Kenyon, A. J., C. E. Chryssou, et al. (2002). "Luminescence from erbium-doped silicon
nanocrystals in silica: Excitation mechanisms." Journal Of Applied Physics 91(1): 367-374.
Khoshman, J. M. and M. E. Kordesch (2005). Spectroscopic ellipsometry characterization of
amorphous aluminum nitride and indium nitride thin films. Physica Status Solidi C -
Conferences and Critical Reviews, Vol 2, No 7. 2: 2821-2827.
Kik, P. G. a. P., A. (2003). Towards the First Silicon Laser, Kluwer Academic Publishers.
Kim, J. H. and P. H. Holloway (2004). "Room-temperature photoluminescence and
electroluminescence properties of sputter-grown gallium nitride doped with
europium." Journal Of Applied Physics 95(9): 4787-4790.
Kim, J. H., N. Shepherd, et al. (2003). "Visible and near-infrared alternating-current
electroluminescence from sputter-grown GaN thin films doped with Er." Applied
Physics Letters 83(21): 4279-4281.
Kordesch, M. E. and H. H. Richardson (2003). Electroluminescent textiles using sputter-
deposited amorphous nitride-Rare-Earth ion coatings. Electronics On Unconventional
Substrates-Electrotextiles And Giant-Area Flexible Circuits. 736: 61-66.
Kucheyev, S. O., J. S. Williams, et al. (2004). "Dynamic annealing in III-nitrides under ion
bombardment." Journal Of Applied Physics 95(6): 3048-3054.
Lai, B. Y., L. Feng, et al. "Optical transition and upconversion luminescence in Er(3+) doped and
Er(3+)-Yb(3+) co-doped fluorophosphate glasses." Optical Materials 32(9): 1154-1160.
Lapraz, D., P. Iacconi, et al. (1991). "Thermostimulated Luminescence And Fluorescence Of
Alpha-Al2o3-Cr-3+ Samples (Ruby) - Influence Of The Cr-3+ Concentration."
Physica Status Solidi A-Applied Research 126(2): 521-531.
Little, M. E. and M. E. Kordesch (2001). "Band-gap engineering in sputter-deposited ScxGa1-
xN." Applied Physics Letters 78(19): 2891-2892.
Lozykowski, H. J. (1993). "Kinetics Of Luminescence Of Isoelectronic Rare-Earth Ions In Iii-V
Semiconductors." Physical Review B 48(24): 17758-17769.
Lozykowski, H. J., W. M. Jadwisienczak, et al. (2000). "Photoluminescence and
cathodoluminescence of GaN doped with Pr." Journal Of Applied Physics 88(1): 210-222.
MacKay, K.M., et al., (1996) Introduction to Modern Inorganic Chemistry, Boca Raton, CRC
Press.
Maqbool, M. (2005). Growth Characterization and Luminescence and Optical Properties of
Rare Earth elements and Transition Metals doped in Wide Gap Nitride
Semiconductors. Physics and Astronomy. Athens, Ohio University. PhD: 1-180.

205
Maqbool, M. and I. Ahmad (2009). "Ultraviolet spectroscopy of Pr+3 and its use in making
ultraviolet filters." Current Applied Physics 9(1): 234-237.
Maqbool, M., I. Ahmad, et al. (2007). "Direct ultraviolet excitation of an amorphous AlN:
praseodymium phosphor by codoped Gd3+ cathodoluminescence." Applied Physics
Letters 91(19).
Maqbool, M., G. Ali, et al. "Nanocrystals formation and intense, green emission in thermally
annealed AlN:Ho films for microlaser cavities and photonic applications." Journal
Of Applied Physics 108(4).
Maqbool, M. and T. Ali (2009). "Intense Red Catho- and Photoluminescence from 200 nm Thick
Samarium Doped Amorphous AlN Thin Films." Nanoscale Research Letters 4(7): 748-752.
Maqbool, M., M. E. Kordesch, et al. (2009). "Electron penetration depth in amorphous AlN
exploiting the luminescence of AlN:Tm/AlN:Ho bilayers." Current Applied Physics
9(2): 417-421.
Maqbool, M., M. E. Kordesch, et al. (2009). "Enhanced cathodoluminescence from an amorphous
AlN:holmium phosphor by co-doped Gd(+3) for optical devices applications." Journal
Of The Optical Society Of America B-Optical Physics 26(5): 998-1001.
Maqbool, M., K. Main, et al. (2010). "Titanium-doped sputter-deposited AlN infrared
whispering gallery mode microlaser on optical fibers." Optics Letters 35(21): 3637-3639.
Maqbool, M., H. H. Richardson, et al. (2005). Cathodoluminescence of praseodymium doped
AlN, GaN and turbo static BN. GaN, AIN, InN and Their Alloys. 831: 417-421.
Maqbool, M., H. H. Richardson, et al. (2007). "Luminescence from praseodymium doped
AlN thin films deposited by RF magnetron sputtering and the effect of material
structure and thermal annealing on the luminescence." Journal Of Materials Science
42(14): 5657-5660.
Maqbool, M., E. Wilson, et al. (2010). "Luminescence from Cr(+3)-doped AlN films
deposited on optical fiber and silicon substrates for use as waveguides and laser
cavities." Applied Optics 49(4): 653-657.
Martin, A. L., C. M. Spalding, et al. (2001). "Visible emission from amorphous AlN thin-film
phosphors with Cu, Mn, or Cr." Journal Of Vacuum Science & Technology A-Vacuum
Surfaces And Films 19(4): 1894-1897.
Masterov, V. F. and L. G. Gerchikov (1999). "Mechanisms of excitation of the f-f emission in
silicon codoped with erbium and oxygen." Semiconductors 33(6): 616-621.
Muller, S., M. J. Zhou, et al. (2009). "Intra-shell luminescence of transition-metal-implanted
zinc oxide nanowires." Nanotechnology 20(13).
Nishikawa, A., N. Furukawa, et al. (2010). "Improved luminescence properties of Eu-doped
GaN light-emitting diodes grown by atmospheric-pressure organometallic vapor
phase epitaxy." Applied Physics Letters 97(5).
Nishikawa, A., T. Kawasaki, et al. (2009). "Room-Temperature Red Emission from a p-
Type/Europium-Doped/n-Type Gallium Nitride Light-Emitting Diode under
Current Injection." Applied Physics Express 2(7).
Noriyuki, T., H. Ohdan, et al. (1997). "Near-infrared spectroscopic method for assessing the
tissue oxygenation state of living lung." American Journal Of Respiratory And Critical
Care Medicine 156(5): 1656-1661.
O'Donnell, K. P., ed. (2010). Rare-Earth Doped III-Nitrides for Optoelectronic and Spintronic
Applications, Springer.
Park, J. H. and A. J. Steckl (2004). "Laser action in Eu-doped GaN thin-film cavity at room
temperature." Applied Physics Letters 85(20): 4588-4590.
Park, J. H. and A. J. Steckl (2005). "Demonstration of a visible laser on silicon using Eu-
doped GaN thin films." Journal Of Applied Physics 98(5).

Cathodoluminescence

206
Park, J. H. and A. J. Steckl (2006). "Visible lasing from GaN: Eu optical cavities on sapphire
substrates." Optical Materials 28(6-7): 859-863.
Richardson, H. H., P. G. Van Patten, et al. (2002). "Thin-film electroluminescent devices
grown on plastic substrates using an amorphous AlN: Tb3+ phosphor." Applied
Physics Letters 80(12): 2207-2209.
Singh, J. a. S., K., eds. (2003). Advances in Amorphous Semiconductors, Advances in Condensed
Matter Sciences, CRC Press.
Steckl, A. J. and R. Birkhahn (1998). "Visible emission from Er-doped GaN grown by solid
source molecular beam epitaxy." Applied Physics Letters 73(12): 1700-1702.
Steckl, A. J. a. Z., J.M (1999). "Optoelectronic Properties and Applications of Rare Earth."
MRS Bulletin 24: 33-38.
Street, R. A., ed. (2010). Technology and Applications of Amorphous Silicon, Springer.
Tessler, L. R. (1999). "Erbium in a-Si: H." Brazilian Journal Of Physics 29(4): 616-622.
Thurbide, K. B. and W. A. Aue (2002). "Chemiluminescent emission spectra of lead,
chromium, ruthenium, iron, manganese, rhenium, osmium and tungsten in the
reactive flow detector." Spectrochimica Acta Part B-Atomic Spectroscopy 57(5): 843-852.
Wang, D., M. T. Carlson, et al. (2011). "Absorption Cross Section and Interfacial Thermal
Conductance from an Individual Optically Excited Single-Walled Carbon
Nanotube." Acs Nano 5(9): 7391-7396.
Weingartner, R., O. Erlenbach, et al. (2006). "Thermal activation, cathodo- and
photoluminescence measurements of rare earth doped (Tm, Tb, Dy, Eu, Sm, Yb)
amorphous/nanocrystalline AlN thin films prepared by reactive rf-sputtering."
Optical Materials 28(6-7): 790-793.
Wybourne B.G. and Smentek, L. (2007). Optical Spectroscopy of Lanthanides, CRC Press.
Zanatta, A. R., A. Khan, et al. (2007). "Red-green-blue light emission and energy transfer
processes in amorphous SiN films doped with Sm and Tb." Journal Of Physics-
Condensed Matter 19(43).
Zanatta, A. R. and L. A. O. Nunes (1998). "Green photoluminescence from Er-containing
amorphous SiN thin films." Applied Physics Letters 72(24): 3127-3129.
Zanatta, A. R., C. T. M. Ribeiro, et al. (2001). "Visible luminescence from a-SiN films doped
with Er and Sm." Applied Physics Letters 79(4): 488-490.
Zanatta, A. R., C. T. M. Ribeiro, et al. (2004). "Photon and electron excitation of rare-earth-
doped amorphous SiN films." Journal Of Non-Crystalline Solids 338: 473-476.
Zanatta, A. R., C. T. M. Ribeiro, et al. (2005). "Optoelectronic and structural characteristics of
Er-doped amorphous AlN films." Journal Of Applied Physics 98(9).
Zanatta, A. R., C. T. M. Ribeiro, et al. (2006). "Thermally synthesized ruby microstructures
and luminescence centers." Journal Of Applied Physics 100(11).
Zanatta, A. R., H. H. Richardson, et al. (2007). "Amorphous BeN as a new solid host for rare-
earth-related luminescent materials." Physica Status Solidi-Rapid Research Letters 1(4):
153-155.
Zegrya, G. G. and V. F. Masterov (1995). "Mechanism Of The Intensification Of F-F
Luminescence In Semiconductors." Semiconductors 29(10): 989-995.
Zegrya, G. G. and V. F. Masterov (1996). Two novel mechanisms of f-f-luminescence
resonance excitation in semiconductors. Tenth Feofilov Symposium On Spectroscopy
Of Crystals Activated By Rare-Earth And Transitional-Metal Ions. 2706: 235-240.
Zegrya, G. G. and V. F. Masterov (1998). "Mechanism of generation of f-f radiation in
semiconductor heterostructures." Applied Physics Letters 73(23): 3444-3446.
Part 3
Application to Oxides and Minerals

8
Peculiarity of the Cathodoluminescence of
lpha- Alumina Prepared by Calcination of
Gibbsite Powder or Generated by Oxidation
of a Metallic FeCrAl Alloy
Djelloul Abdelkader and Boumaza Abdecharif
Laboratoire des Structures, Proprits et Interactions Inter Atomiques (LASPI
2
A)
Centre Universitaire de Khenchela 40000
Algeria
1. Introduction
Most of metallic materials functioning at high temperature need to have oxidation
resistance. This resistance can be achieved when the material develops, through oxidation,
an oxide film which acts as a diffusion barrier while keeping a good adherence. In this
respect, alpha alumina clearly acts as such. The oxides of aluminum have been the subject of
many investigations because of their commercial importance and scientific interest. The
thermal stability and optical properties of pure nanometer-sized alumina powder have
received much attention because of their intrinsic interest and commercial value.
Nanometer-sized alumina powders are widely applied today. One of its applications is in
fluorescent lamps due to the absorption of ultraviolet light. In fact, it can also emit the light
under excitation with a suitable wavelength. It is important to note that there are many
works about alpha alumina using X-ray diffraction, but there is a need for a more detailed
structural analysis. To achieve this more exhaustive structural characterization we have
used the Rietveld refinement method and cathodoluminescence (CL) measurements.
CL spectroscopy is widely used as a contactless and relatively nondestructive method to
provide microcharacterization of the optical and electronic properties of luminescent
materials. Nevertheless, it is used comparatively rarely for the investigation of oxide
semiconductor structures. The major advantage of CL spectroscopy in the case of such
structures is that most of the anticipated products of oxidation are luminescent, and it is
easy to get excitation across the bandgap of any dielectric with readily available electron
beam voltages. The emission occurs for all the luminescent mechanisms present in the
material.
Pure -Al
2
O
3
crystal is colorless and shows little absorption in the ultravioletvisible (UV
Vis) range. But various impurities (Ti, Mn, Cr, and Fe) even a trace level causes apparent
absorptions which are attributed to various emission centers (Jheeta et al., 2006).
The colors arise from very minor amounts of impurity (<1% of the Al
3+
replaced by other
cations) because the Al
2
O
3
structure apparently does not tolerate substitutions. However,

Cathodoluminescence

210
these trace substitutions can cause intense colors. Ruby is red because of its Cr
3+
content.
Yellow sapphire owes its color to Fe
3+
. Blue sapphire derives its color from Fe
2+
-Ti
4+
and
Fe
2+
-Fe
3+
intervalence charge transfer. Green sapphires contain a mixture of the blue and
green colors.
Strong well-known
2
E
4
A
2
lines of Cr
3+
(693 nm) with a long decay time characterize their
luminescence spectra. Besides that, much weaker narrow lines are present, which are
connected with Cr-pairs and more complicated complexes. The Mn
4+
ion is isoelectronic
with Cr
3+
, i.e., both of them have the same electronic structure of the open shell (3d
3

configuration). Thus, the spectroscopic properties of -Al
2
O
3
:Mn
4+
are similar to those of
ruby (-Al
2
O
3
:Cr
3+
). Octahedral Mn
4+
(3d
3
) would be expected to show the R-line
fluorescence characteristic of isoelectronic Cr
3+
and in approximately the same region. The
dominate defects for the visible emission might be different for -alumina powders formed
by heating any of the hydrates of aluminium to a sufficiently high temperature.
In the presence of lattice defects, extra luminescence emissions can be observed in the ultra-
violet (UV) region upon highly energetic excitation. The main intrinsic defects in the -
alumina crystals are oxygen vacancies in different charge states: a neutral vacancy, a
vacancy capturing one electron (a F
+
-center), and a vacancy capturing two electrons (a F-
center) (Kislov et al., 2004; Michizono et al., 2007; Yu et al., 2002). The observed UV
spectrum in - alumina can be deconvoluted into two distinct sub-band components: an F
+
-
center band, located at around 3.8 eV, and a less intense F-center band, located at around 3.0
eV (Brewer et al., 1980; Boumaza & Djelloul, 2010; Boumaza et al., 2010). Depending on the
defect introduction method one can create also F
2
-centers,
2
F
-centers and
2
2
F

-centers
(double oxygen vacancy with four, three and two trapped electrons respectively). -Al
2
O
3

crystals with defects in the oxygen sublattice are actively studied as promising storage
materials (Kortov & Milman, 1996). In this connection, it is interesting to study luminescence
properties of the nanostructured aluminium oxide and compare them with analogous
properties of crystalline samples.
In this chapter, we present X-ray diffraction (Rietveld analysis) and CL measurements of -
alumina powders formed by calcination of gibbsite or generated by oxidation of a metallic
FeCrAl alloy. The peculiarity of the cathodoluminescence under comparable conditions of
- alumina is discussed.
2. Materials and experimental methods
Gibbsite powder, Al
2
(OH)
6
, from Prolabo (no 20 984.298) was used. The powder is made of
platelet aggregates and was composed of 64.567% Al
2
O
3
and max.: 0.01% Fe
2
O
3
, 0.02% SO
4
,
0.002% heavy metals (as Pb), and 1.0% non precipitable by NH
4
OH (as SO
4
). The sample
experienced an ignition loss of 3334.5% at 1000 C and had a purity grade of 99.7%. Its
average particle size (20 m) was due to the agglomeration of crystallites. The specific
surface area of the original sample was 0.5 m/g.
The gibbsite platelets was calcined in ambient atmosphere (pO
2
= 0.21 atm) at 1573 K. The
cycle was as follows: heating up to an isothermal temperature at 5 K/min, maintaining for 24h
at the calcination temperature and fast cooling down to room temperature (air quench). The
calcination temperature was maintained for 24 hours to obtain a well-crystallised product.
Peculiarity of the Cathodoluminescence of lpha- Alumina
Prepared by Calcination of Gibbsite Powder or Generated by Oxidation of a Metallic FeCrAl Alloy

211
The present work was performed on ferritic ODS commercial FeCrAlY alloy PM2000 (20
wt% Cr, 5.8 wt% Al, 0.5 wt% Ti, 0.5 wt% Y
2
O
3
). This alloy is an alumina-forming alloy.
The specimens, with dimensions 25255 mm
3
, are cut from a rolled plate. In the case of
the PM2000, sample was oxidised in air at 1223 K for 72h. The thermogravimetric method
(TG) and differential thermal analysis (DTA) data were recorded under a dry air flow
with a heating rate of 10 K/min in a SETARAM TGDTA9216.18 thermal analyser. TG
measurements were corrected for temperaturedependent buoyancy by subtracting the
data of a measurement carried out on an inert sample. The crystalline structure of the
sample was investigated by XRD using a PANalytical XPert Pro MRD diffractometer
configured as follows: Cu tube operating at 40 kV and 30 mA ((K
1
) = 0.15406 nm, (K
2
)
= 0.15444 nm). The scan rate (2) was 1/min at a step size of 0.025. The data were
processed to realize the conditions of the software program Fullprof Suite for the
structure refinement.
The FTIR technique was used in the absorbance mode in the 200-4000 cm
-1
range. For oxides
all bands have characteristic frequencies between 200 and 1000 cm
-1
. For the FTIR
measurements, samples were prepared by grinding the oxide films scraped from the
substrate (PM2000). After calcination, 10 to 100 g of the powder was drawn, then grinded
with 23 2 mg of CsBr in order to obtain, a pellet of 200-250 m in thickness. After grinding,
the powder was placed in a mould (5 mm diameter) and a cold isostatic pressure (CIP) of
150 MPa was applied for 5 min.
The FTIR spectra are obtained using a Perkin-Elmer spectrometer at resolution of 8 cm
-1
. For
each sample, 120 scans were used. The apparatus is equipped with a system allowing the
reduction of the optical course in air in order to minimize the perturbations associated with
ambient air (water vapour and CO
2
). The uncertainty on the position of the various peaks is
equal to 2 cm
-1
.
The emitted light under electron beam excitation in a UHV system was analyzed through a
quartz window with a Jobin Yvon CP 2000 spectrograph and a CCD detector. The
wavelength range 200-1000 nm was investigated.
3. Results and discussions
3.1 Differential thermal analysis and thermal gravimetry analysis (DTA-TG) of gibbsite
Fig. 1 shows the typical TGDTA curves of the gibbsite. TG and DTA curves are indicated as
dotted and solid lines, respectively. Concerning the dehydration-dehydroxylation process of
gibbsite, the dehydration appears to occur in two steps (around 598 and 803 K respectively)
at higher temperature. The expected theoretical loss due to dehydration is 34.6%, the
experimental loss is 34.3% a little lower. This difference 0.3% is a bit larger than the
experimental uncertainty 0.1%, the starting gibbsite may be slightly dehydrated. The last
step (803 to 1273 K) gives no thermal event but appears as a continuous mass loss (about
2%) which corresponds to the elimination of residual hydroxyls. The formation of
alumina occurs between 1473 and 1533 K (MacKenzie et al., 1999). Finally, the structural
transformations to well-crystallised alumina are described by nucleation (T<1470 K) and
growth mechanisms (T>1470 K).

Cathodoluminescence

212

Fig. 1. TG and DTA curves of gibbsite.
3.2 Gibbsite
Gibbsite (-Al(OH)
3
) has monoclinic symmetry (a=0.8684, b=0.5078, c=0.9736 nm, =94.54)
with the space group P21/n, and the unit cell contains eight Al(OH)
3
units (Saalfeld &
Wedde, 1974). Gibbsite is characterized by the stacking of two-layer units (AA or BB) of
hydroxyl sheets with the sequence ABBAABBA... where hydroxyl sheets of the adjacent
Gibbsite layers face the c direction (Kogure, 1999). In Fig. 2a, the XRD pattern obtained on
the as received gibbsite powder shows a good agreement with the reference XRD pattern
(33-0018 JCPDS file).
3.2.1 Phase transitions induced by heat treatment of gibbsite
When heating up fine-grained gibbsite, most OH groups are eliminated, and various forms
of alumina are formed with the sequence: gibbsite ---Al
2
O
3
when temperature
increases. In order to study these phases, we performed XRD measurements on four
samples prepared from gibbsite calcined for 24h at 773, 1073 (62h), 1173 and 1573K
respectively.
According to Fig. 2b, at 773K ,the phase is expected. In spite of many investigations since
the1950s (Brindley & Choe, 1961; Kogure, 1999; Saalfeld, 1960; Stumpf et al., 1950; Yu et al.,
2002), the crystal structure of -alumina is still uncertain. Stumpth et al. (Stumpf et al., 1950)
assumed a cubic (not spinel) unit cell of lattice parameter a=0.795 nm (04-0880 JCPDS file).
On the other hand, two hexagonal structures have been suggested, either with the
parameters a=0.556 nm and c=1.344 nm (Saalfeld, 1960) or with a=0.557 nm and c=0.864 nm
(Brindley & Choe, 1961) (13-0373 JCPDS file). The two previous hexagonal unit cells may be
described respectively as a stacking of 6 and 4 close-packed oxygen layers, of approximately
the same thickness (0.224 and 0.216 nm) as the AlOH layers in gibbsite (0.212 nm). More

213
recently, Kogure (Kogure, 1999) proposed a hexagonal lattice with a=0.49 nm and an
undefined c length indicating that -alumina structure can be regarded as random close
packing of gibbsite-like layers.
For samples prepared at 1173 K, the phase is expected (see Fig. 2d). Contrary to the
phase, the crystal structure of -alumina is well known (see for example Ref. (Ollivier et
al., 1997) and references therein). -alumina is orthorhombic with the space group pna2
1

and results in ten independent atoms positions (four Al and six O). The experimental XRD
pattern at 1173K is specific of a pure -alumina (Fig. 2d). Nevertheless, the presence of
remnant -phase cannot be excluded as all the peaks also appear in structure. Note
that the experimental XRD patterns show well crystallized phases, in contrast with the
phase.
Fig. 3 give SEM images and XRD patterns of gibbsite powder (a) after calcination at (b)- 773
K; (c)- 1073 K; (d)- 1573 K, for 24 h.
When heating up at temperature above 1573K for 24h, gibbsite transforms into -alumina,
the stable structure. The XRD pattern of the 1573K sample (Fig. 6) shows that only -Al
2
O
3
is
present when compared with 421468 JCPDS file. In this structure, tetrahedral Al
3+
ions are
no longer present and only AlO
6
octahedron remain.

Fig. 2. XRD patterns: (a) Gibbsite (33-0018 JCPDS file), (b) alumina formed from gibbsite
calcined for 24h at 773K (expected phase) (JCPDS 13-0373 file), (c) alumina formed from
gibbsite calcined for 62h at 1073K, (d) alumina formed from gibbsite calcined for 24h at
1173K (expected phase) (JCPDS 88-0107 and 01-1305 file).

Cathodoluminescence

214

Fig. 3. SEM images: (a)- Gibbsite; (b)- 773 K; (c)- 1073 K; (d)- 1573 K.
3.2.2 General properties of alumina (Al
2
O
3
)
Aluminum oxide, commonly referred to as alumina, possesses strong ionic interatomic
bonding giving rise to its desirable material characteristics. It can exist in several crystalline
phases which all revert to the most stable hexagonal alpha phase at elevated temperatures.
This is the phase of particular interest for structural applications and the material available
from Accuratus.
The exceptional properties of alumina (Al
2
O
3
), such as great hardness, high thermal and
chemical stability, and high melting temperature, make it a very attractive material. The
crystalline -Al
2
O
3
phase (corundum or sapphire) is the single stable modification of
alumina. The crystalline -Al
2
O
3
has the band gap Eg8.5 eV and is widely used in optical
devices. Sapphire doped with chrome (ruby) or titanium is applied as an active medium in
laser systems. In microelectronics sapphire is used as a substrate for growing silicon and
gallium nitride (GaN). Alumina is a highly radiation-resistant material and is used as a
sensitive element in detectors when measuring the ionizing radiation parameters.

215
High purity alumina is usable in both oxidizing and reducing atmospheres to 1925C.
Weight loss in vacuum ranges from 10
7
to 10
6
g/cm
2
.sec over a temperature range of 1700
to 2000C. It resists attack by all gases except wet fluorine and is resistant to all common
reagents except hydrofluoric acid and phosphoric acid. Elevated temperature attack occurs
in the presence of alkali metal vapors particularly at lower purity levels.
The composition of the ceramic body can be changed to enhance particular desirable material
characteristics. An example would be additions of chrome oxide or manganese oxide to
improve hardness and change color. Other additions can be made to improve the ease and
consistency of metal films fired to the ceramic for subsequent brazed and soldered assembly.
Mechanical, thermal and electrical properties of Al
2
O
3
are summarized in table 1.

Units of Measure 94% Al
2
O
3
96% Al
2
O
3
99.5% Al
2
O
3

Mechanical
Density gm/cc 3.69 3.72 3.89
Porosity % 0 0 0
Color white white ivory
Flexural Strength MPa 330 345 379
Elastic Modulus GPa 300 300 375
Shear Modulus GPa 124 124 152
Bulk Modulus GPa 165 172 228
Poissons Ratio 0.21 0.21 0.22
Compressive
Strength
MPa 2100 2100 2600
Hardness Kg/mm
2
1175 1100 1440
Fracture
Toughness K
IC

MPam
1/2
3.5 3.5 4
Maximum Use
Temperature
C 1700 1700 1750
Thermal
Thermal
Conductivity
W/mK 18 25 35
Coefficient of
Thermal
Expansion
10
6
/C 8.1 8.2 8.4
Specific Heat J/KgK 880 880 880
Electrical
Dielectric
Strength
ac-kV/mm 16.7 14.6 16.9
Dielectric
Constant
At 25C, 1 MHz 9.1 9.0 9.8
Dissipation
Factor
At 25C, 1 MHz 0.0007 0.0011 0.0002
Volume
Resistivity
Ohmcm >10
14
>10
14
>10
14

Table 1. Mechanical, thermal and electrical properties of Al
2
O
3

(http://accuratus.com/alumox.html).

Cathodoluminescence

216
3.2.3 Oxidation of PM2000
The oxidation resistance of high-temperature alloys and metallic coatings is dependent on the
formation of a protective surface oxide. In an ideal case, the oxide layer should be highly stable,
continuous, slow growing, free from cracks or pores, adherent and coherent. -Al
2
O
3
is an oxide
which comes close to satisfying these requirements; the slow growth rate is related to its highly
stoichiometric structure and its large band gap which makes electronic conduction difficult.
One of the most crucial factors in the oxidation of alumina-formers is the temperature,
which must be high enough to promote the formation of -Al
2
O
3
in preference to the less
protective transition alumina. Another critical factor is the aluminum content which must be
sufficiently high to develop and maintain an alumina layer and prevent subsequent
breakaway oxidation.
The addition of chromium to Fe-Al alloys promotes the formation and maintenance of a
complete layer of -Al
2
O
3
by acting as a getter and preventing internal oxidation of the
aluminum (Wood, 1970).
Iron and chromium are the major impurities present in Al
2
O
3
scales formed on PM2000.
Primarily their oxides formed during the transient stage and were incorporated into the -Al
2
O
3

scale. Fe segregated to some -Al
2
O
3
grain boundaries, but not Cr. The Al
2
O
3
scale became
progressively purer with oxidation time. It is possible that the Fe in the Al
2
O
3
scale increases the
scaling rate and, in particular, enhances lateral growth that causes scale convolution.
The growth of -Al
2
O
3
scales that form on FeCrAl alloys during high temperature oxidation
is generally considered to be controlled by oxygen inward diffusion through oxide grain
boundaries (Mennicke et al., 1998; Quaddakkers et al., 1991). Aluminum also diffuses out,
which can cause growth within the scale (Golightly et al., 1979). The degree of Al outward
transport can be significantly reduced by the presence of reactive elements, such as Y, Hf or
Zr (Mennicke et al., 1998; Quaddakkers et al., 1991), which segregate to Al
2
O
3
grain
boundaries (Przybylski et al., 1987). However, the extent of outward growth seems to differ
appreciably among several reactive-element doped Fe based alloys.
The EDS results of the average Fe and Cr concentrations in the scale as a function of scale
thickness from different transmission electron microscopy (TEM) specimens are
summarized in Table 2.

Oxidation
condition
Scale
thickness
(m)
Oxide grain
size (nm)
Average [Cr]
(at%)
Average [Fe]
(at%)
Lattice
parameter (nm)
1000 C, 0.5 h 0.39 10736 4.120.95 4.491.67
a = 0.4950.004
c = 1.353 0.008
1000 C, 1 h 0.9 19144 0.340.28 1.910.49 a = 0.4750.004
c = 1.347 0.008 1000 C, 26 h 1.77 18653 0.240.28 0.600.32
1200 C, 2 h 2.94 29146 0.30.64 0.700.57 a = 0.4710.004
c = 1.323 0.008 1200 C, 120 h 4-5.5 1546423 0.270.20 0.060.06
Standard parameters for -Al2O3 are: a = 0.4758 nm, c = 1.2991 nm.
Table 2. Fe and Cr concentrations in -Al
2
O
3
scales and effect on lattice parameters (Hou et
al., 2004)

217
Fig. 4 give SEM images and XRD patterns of PM2000 after oxidation at 1023 K for 76 h and
1223 K for 72 h.
EDX analysis (Fig. 5) of the sample oxidized at 1123 K indicates that the film mainly consists
in aluminium and oxygen elements and very small amount of the substrate constituting
elements are observed. As the film thickness decreases, like sample oxidized at 1023 K, the
signature of the substrate increases due to interactions of the electrons with the underlying
substrate mater as the film is thinner than that formed at 1123 K.

Fig. 4. SEM images of the outer oxidized surface and XRD patterns of -Al
2
O
3
obtained by
oxidation of PM2000 at 1023 K for 76 h and 1223 K for 72 h.

Cathodoluminescence

218

Fig. 5. SEM images of the outer surface and EDX analyses of -Al
2
O
3
films obtained by
oxidation of PM2000 in air at 1023 K for 76 h (top) and 1123 K for 95 h (bottom).

219
3.2.4 Rietveld refinement of the structures
The Rietveld refinement of the structures was performed using the WinPlotr/FullProf suite
package (Rodrguez-Carvajal, 1993). The peak shape was described by a pseudo-Voigt
function, and the background level was modeled using a polynomial function. The profiles
were refined using the space group and structure models of Al
2
O
3
( 3 R C , JCPDS 46-1212)
(in this structure cations occupy the 12c sites and oxygen ions the 18e sites).
The XRD data for the 2 regions between 20 and 80 was used for the refinement. The
observed, Rietveld refined and difference patterns are shown in Figs. 6 and 7.
It was obvious that the agreement between the experimental data and the simulations was
excellent since the R
wp
(weighted residual error) factor was small ( 14.1%). The Rietveld
results (cell parameters, atom position, reliability factors and crystallite size (D)) are given in
Table 3 and 4. The size of -alumina crystallites (D=36 nm) obtained using PM2000 alloy
was smaller than that (D=43 nm) obtained using gibbsite precursor. Furthermore, the -
Al
2
O
3
from PM2000 has a greater lattice parameter (a=0.4763nm, c=1.3047nm) than that of
JCPDS file 46-1212 (a=0.4758 nm, c=1.2991 nm, c/a=2.730) and a greater c/a ratio (c/a
=2.739) than that of JCPDS file 46-1212. However, the -Al
2
O
3
from gibbsite has a smaller
lattice parameter (a=0.4752nm, c=1.2980nm) and a similar c/a ratio (c/a =2.731) than that of
JCPDS file 46-1212. Furthermore, the -Al
2
O
3
from gibbsite has a greater calculated density
(3.996 g/cm
3
) than that of -Al
2
O
3
from PM2000 (3.961 g/cm
3
). Phase transformations are
frequently accompanied by microstructural changes. This fact could explain the
crystallographic parameters differences between a two -alumina.

Lattice parameters
(nm)
atom Wyck. Site x y z Biso occupancy
a= 0.47523
c= 1.29805
Al 12c 3. 0 0 0.3521 0.2200 0.3333
O 18e .2 0.3065 0 0.2500 0.2400 0.5000
Table 3. The Rietveld refinement results: -Al
2
O
3
from gibbsite powder, R
p
= 16.7%,
R
wp
= 6.93%, Calc. density =3.996 g/cm
3
, D=43 nm.

Lattice parameters
(nm)
atom Wyck. Site x y z Biso occupancy
a= 0.47637
c= 1.30472
Al 12c 3. 0 0 0.3417 4.7208 0.9602
O 18e .2 0.3392 0 0.2500 2.7191 1.1186
Table 4. The Rietveld refinement results: -Al
2
O
3
from PM2000, R
wp
= 14.1%,
Calc. density =3.961 g/cm
3
, D=36nm.

Cathodoluminescence

220

Fig. 6. The observed and calculated diffraction patterns of -alumina powder prepared by
calcination at 1573K for 24h of gibbsite powder. Vertical bars indicate the calculated position
of the Bragg peaks. The blue curves in the bottom correspond to the differences between
experimental and calculated profiles.

Fig. 7. The observed and calculated diffraction patterns of -alumina obtained by oxidation
of PM2000 at 1223K for 72h.

221
3.2.5 Fourier transform infrared (FTIR) analysis
Fig. 8 compares the FTIR absorbance spectra of Al
2
O
3
obtained after calcination of
gibbsite and oxidation of PM2000. For the samples calcined at 1573 K, significant
spectroscopic bands at ~640, ~594, ~447 cm
-1
and ~386 cm
-1
appear which are identified to
be the characteristic absorption bands of -Al
2
O
3
(Barker, 1963). This is in good agreement
with XRD observations. Common bands exist in all cases, such as the broad OH band
centered around 3420 cm
1
, and the 1640 cm
1
H
2
O vibration band (Ma et al., 2008). The very
high surface area of these materials results in rapid adsorption of water from the
atmosphere because the FTIR samples were kept and grinded in air. Three peaks of very
weak intensities at 2850 cm
-1
, 2920 cm
-1
and 2960 cm
-1
are observed which are due to C-H
stretching vibrations of alkane groups. The absorption in ~2356 cm
1
is due to CO
2

molecular presence in air.

Fig. 8. FTIR absorption spectra of -alumina powder prepared by calcination of gibbsite
(black line) or generated by oxidation of PM2000 (red line).
3.3 Cathodoluminescence
CL is the phenomenon of light emission from specimens as a result of interaction with an
electron beam. In insulating crystals, the origin of the luminescence arises from impurity
atoms (e.g. transition metals or rare earths) in the crystal lattice. Using an electron
microscope to produce the electron beam, the spatial distribution of luminescent sites can be
observed with submicron spatial resolution, and correlated with features of the specimen
morphology or microstructure.
The mechanisms for CL are similar to those for photoluminescence, but the energy input
or excitation source is that of an electron beam rather than a visible or ultraviolet light
beam. When an energetic (keV range) electron beam propagates within a semiconductor
or insulator, the primary electrons lose energy by the creation of electron-hole pairs. These

Cathodoluminescence

222
electron-hole pairs then recombine via radiative and non-radiative processes. Only the
radiative recombination process which leads to the creation of a photon is viewed with
CL. Radiative recombination may be intrinsic (arising from electronic states of the perfect
crystal) or extrinsic (arising from electronic states that are localized at defects or
impurities in the crystal). Extrinsic luminescence thus provides information about defects
and impurities in the crystal lattice. For conciseness, the defects and impurities that give
rise to extrinsic luminescence are often denoted the luminescence centers. Each type of
luminescence center in a particular crystal has a characteristic emission spectrum. The
spectrum may contain both narrow lines and broad bands, depending on the energy level
structure of the luminescence center and the coupling of the center to the host lattice. CL
is advantageous compared to photoluminescence. CL can potentially give additional
information about local positions in the sample because the electron beam can be focused
on several nanometers. In addition, the CL system operates under UHV conditions of less
than 10
9
Torr. Hence, CL measurements can be performed in a contamination-free
environment, which is very effective in detecting weak luminescence. Moreover, the
depth dependent emission profiles can be examined in CL by controlling the accelerating
voltage.
3.3.1 Oxygen vacancy in Al
2
O
3

CL signal is a good signature of the material qualities and is used in this study to
characterize the point defects associated to oxygen vacancies in -Al
2
O
3
. The CL spectra of
-Al
2
O
3
formed from gibbsite and PM2000 are given in figure 9 and 12.
In figure 9, the CL spectra shows that wide band over the interval of (200600 nm) consists
of a series of overlapping bands. The main emission bands located at about 250 nm (4.96
eV), 281 nm (4.41 eV), 325 nm (3.81eV), 373 nm (3.32 eV) and 487 nm (2.54 eV) occur in alpha
alumina powder and also in alpha alumina films generated by oxidation of a metallic
FeCrAl alloy. We believe that the observed CL peak at 4.96 eV is related to the interband
transitions or to defect that is different in origin to the F or F
+
centers in -alumina. The
luminescence band at 4.41 eV is detected only if the excitation density is high and was
previously observed in -alumina by Kortov et al. (Kortov et al., 2008).
In -Al
2
O
3
(corundum structure) each O atom is surrounded by four Al atoms forming two
kinds of Al-O bonds of length 0.186 and 0.197 nm. This is why in corundum F-type centers
have low C
2
symmetry. Besides, an O vacancy has two nearest neighbor O atoms, forming
the basic O triangle with O-O band length of 0.249 nm in perfect corundum. Thus, the F-
type centers are surrounded by six nearest atoms which determine mainly their optical
properties.
Defects induced in Al
2
O
3
may be of various kinds: F centers (oxygen vacancy with two
electron), F
+
centers (oxygen vacancy with one electron), F
2
centers (two oxygen vacancies
with four trapped electrons),
2
F
centers (two oxygen vacancies with three electrons) and

2
2
F

centers (two oxygen vacancies with two electrons) (Ghamdi & Townsend, 1990).
In the case of -alumina and sapphire, there have been reported a number of F-type centers
including the F
+
and F centers (Evans, 1995). As for the luminescence of irradiation defects,
it is known that the luminescence of the F
+
center is observed at the UV region around 3.8

223
eV (325 nm) while that of the F centers is observed at a lower photon energy region around
3.0 eV (410 nm). Thus the presently observed luminescence from -alumina and sapphire
which is centered at 330 and 420 nm can be attributed to the F
+
and F centers. As for the
others, it is noted that the luminescence at 250 and 290 nm is also observed in the case of -
alumina. The luminescence intensity at these bands was found to be sensitive to thermal
annealing at higher temperatures, and then might be attributed to the effect of some
impurities such as OH.
In Ref. (Oster & Weise, 1994), the absorption bands of 220 and 260 nm in pure -Al
2
O
3

crystals have been attributed to F+ absorption bands. Due to the presence of C
2
-symmetry
in the F
+
-center (oxygen vacancy occupied by a single electron) in pure -Al
2
O
3
crystals,
the excited state is split into three levels,
1
B,
2
A, and
2
B, according to the theory of La et al.
(La et al., 1973), giving three polarized optical absorption bands located at 255, 229 and
200 nm.
The optical properties of these luminescent centres are well known. They possess absorption
and emission bands which are produced in the gap as summarised in Fig. 10. F+ centre is
characterised by three absorption bands at 6.3, 5.4 and 4.8 eV and emits at 3.8 eV (330 nm). F
centre absorbs at 6 eV and emits at 3 eV (415 nm).

Fig. 9. CL spectra of -Al
2
O
3
, formed by calcination of gibbsite powder at 1573 K for 24 h
(black line) and by oxidation of PM2000 at 1223K for 72h (red line), obtained in the region of
200 to 600 nm at room temperature.

Cathodoluminescence

224

Fig. 10. Schematic energy level diagram for absorption and emission of F- and F
+
-centres in
-Al
2
O
3
crystal.
In Ref. (Evans, 1994), synthetic sapphire single crystals grown by four different techniques
all showed an anisotropic 5.4 eV (230 nm) absorption broad band. There was not any other
absorption band presented in his cases. As for the nanometer-sized Al
2
O
3
powder apart
from normal lattice vacancy-type defects such as the F-, F
+
- and F
2
-type centers it also has
the surface defect because the nanometer powder has a larger specific surface area and there
exists a lot of dangling bonds in the surface.
Experimental measurements of absorption and luminescence energies for single-vacancy
and for dimer centers in Al
2
O
3
are collected in Table 5.

Absorption Luminescence
F center 6.0 3.00
F
+
center 6.3, 5.4, 4.8 3.80
F
2
4.1 2.40
2
F

3.5 3.26
2
2
F

2.7 2.22
Table 5. Experimental measurements of optical properties for single-vacancy and for dimer
centers in Al
2
O
3
, energies are in eV (Crawford, 1984; Evans et al., 1994).

225
The band of F-centers is absent in both samples due to the existence of impurity Cr (Aoki,
1996). For instance, the CL intensity at 3.81 eV (F
+
-center) of the -Al
2
O
3
formed from
gibbsite is approximately 5 times higher than that of the -Al
2
O
3
formed from PM2000
measured under the same excitation conditions. The main specific feature of the CL spectra
of the -Al
2
O
3
is the presence of a new emission band with the maximum at 3.32 eV. It is a
possible that the new emission band is related to surface F
s
-centers concentrating on
nanoparticle boundaries (Evans, 1995).
Luminescence spectra obtained for anion-deficient aluminum corundum exposed to
different types of excitation and stimulation exhibit luminescence with nanosecond (F
+
-
centers), microsecond (Ti
3+
and Al
+

i
) and millisecond (F and Cr
3+
) decay times (Surdo et
al., 2005). Significantly, the aforementioned centers, which actively participate in
relaxation processes, have considerably different decay times t and emission band
maxima h (Table 6).

Parameter F
+
F
i
Al
Cr
3+
Ti
3+
h (eV)
3.8 3.0 2.4 1.79 1.75
2 ns 34 ms 56 s 4 ms 3.5 s
Table 6. Basic parameters of the emission of most active centers (Surdo et al., 2001; Springis
& Valbis, 1984)
The band 487 nm (2.54 eV) can be related to the aggregate F
2
-centers produced by double-
oxygen vacancies and the centers formed by interstitial aluminum ions. It is known that
these centers are responsible for the green luminescence in highly disordered crystals of
aluminum oxide (Tale et al., 1996; Springis & Valbis, 1984).
3.3.2 Cr
3+
in Al
2
O
3
A classic example of the isolated luminescent centre is Cr
3+
in Al
2
O
3
(ruby) when the excited
electronic energy levels of the host are at much higher energy than those of the dopant ion.
The dopant ion colours the colourless host lattice red. If the concentration of the dopant ion
is low, the interaction between the dopant ions can be neglected. This is what we consider
here as an isolated luminescent centre.
Different impurities in corundum (-Al
2
O
3
) produce different color varieties. All colors of
corundum are referred to as sapphire, except for the red color, which is known as ruby.
Corundum has a trigonal lattice D
6
3d
structure. The crystals have an approximately
hexagonal closed packing structure of oxygen and metal atoms. The six oxygen ions are
octahedrally coordinated cations; and only two-thirds of the octahedral sites are filled. If
corundum has more than 1000 ppm Cr
3+
ions as impurities, it is referred to as a ruby. Rubies
can be used in solid-state lasers (Soukieh et al., 2004), and they fetch high prices in gem
markets. Chromium can be substituted for the aluminum in corundum and is present as
chromium oxide. The Cr
3+
ion is slightly larger than Al
3+
; therefore, it naturally enters easily
into the corundum structure. As a result, Cr
3+
ions form 3d
3
, with only three unpaired
electrons in the 3d orbitals. If the Cr
3+
ion is located in the Al
3+
site in corundum, it
coordinates the six oxygens into a distorted octahedral configuration (Nassau, 1983).

Cathodoluminescence

226
According to the ligand field theory (Figgis et al., 2000), splitting of the 3d
3
(Cr
3+
) orbital
should result in the spectroscopic terms
4
A
2
(A: no degeneracy) ,
4
T
2
,
4
T
1
(T: three fold
degeneracy), and
2
E (E: two fold degeneracy).
For Cr
3+
in Al
2
O
3
crystal, Cr
3+
substitutes for some of Al
3+
, and adopts octahedral ligand
coordination. The 3d levels are extremely host sensitive. The strong crystal field in Al
2
O
3

leads to the splitting of 3d electron orbits of Cr
3+
and produces the ground level:
4
A
2
, and the
excited states:
2
E,
4
T
2
, and
4
T
1
, etc. the transitions from
4
A
2
to
4
T
2
, and
4
T
1
are spin-allowed,
so these energy levels act as broad pumping levels. The
2
E is the narrow lowest excited
band, acting as emitting level. The unusual magnitude of this crystal field splitting extends
the lowest
2
E state 14400 cm
-1
(694 nm) above the ground state. Thus the
2
E-
4
A
2
transition of
Cr
3+
: Al
2
O
3
crystal lies in visible spectral region. Exciting any of the pumping bands of
4
T
2
,
and
4
T
1
results in fast relaxation to lowest
2
E excited state. At room temperature, the
fluorescence emitting from
2
E state appears as a sharp band with a peak at 694 nm
corresponding to the transition to the
2
E terminal state. The Cr
3+
ion has two strong
absorption bands in the visible part of the spectrum, which explain the red color, i.e., 2.2 eV
light can be absorbed to raise the chromium from the
4
A
2
ground level to the
4
T
2
excited
level as absorption in the yellow-green, and 3.0 eV light raises it to the
4
T
1
level as violet
absorption. In addition, the absorption decreases to zero in the red region below 2.0 eV.
Therefore, rubies have a red color with a slight purple overtone.
Chromium impurity in -Al
2
O
3
lattice is characterised by two bands of absorption (3.1
and 2.2 eV) and one fine emission structure peaked at 1.8 eV (693 nm) as summarised in
Fig. 11.

Fig. 11. Schematic energy level diagram for absorption and emission of Cr
3+
-center in -
Al
2
O
3
crystal.

227
In oxide insulators, a number of transition metal and rare earth impurities act as
luminescence centers. The trivalent chromium ion (Cr
3+
), with electronic configuration
3d
3
, is an efficient luminescence center in many light-metal oxides, including Al
2
O
3
and
MgO. The trivalent chromium ion enters substitutionally and is surrounded by an
octahedron of oxygen ions. In aluminum oxide, the surroundings of the chromium ion
are not quite cubic, as the oxygen octahedron is stretched along its trigonal symmetry
axis C
3
.
Chromium in -Al
2
O
3
lattice gives a luminescence in the visible domain. In CL, the narrow
band at 693 nm is attributed to chromium impurity (Ghamnia et al., 2003). In Al
2
O
3
:Cr
3+

(ruby) the apparent lifetime of the R-line emission may increase from the intrinsic value of
3.8 ms up to 12 ms (Auzel & Baldacchini, 2007).
The typical CL spectra of -Al
2
O
3
, formed by calcination of gibbsite powder at 1573 K for 24
h (black line) and by oxidation of PM2000 at 1223K for 72h (red line), obtained in the region
of 600 to 800 nm at room temperature are shown in Fig. 12. The sharp band at 693 nm (1.79
eV), with a radiative lifetime
R
~4 ms (de Wijn, 2007), as well as features at 706 nm (1.76 eV)
and 713 nm (1.74 eV) undoubtedly belongs to Cr
3+
emission in -alumina, and the subband
at 677 nm (1.83 eV) is attributed to the
2
E
4
A
2
transition of Mn
4+
ions in Al
2
O
3
(Jovanic, 1997;
Geschwind et al., 1962; Crozier, 1965).

Fig. 12. CL spectra of -Al
2
O
3
, formed by calcination of gibbsite powder at 1573 K for 24 h
(black line) and by oxidation of PM2000 at 1223K for 72h (red line), obtained in the region of
600 to 800 nm at room temperature.

Cathodoluminescence

228
Mn
4+
is known to emit doublet lines at 672 and 676 nm in -Al
2
O
3
(Kulinkin et al., 2000).
A similar emission was recently reported in -Al
2
O
3
microcones (Li et al., 2010). Thus, it can
be concluded that the incorporation of Mn
4+
ions in -Al
2
O
3
observed in the experiment is
irreversible and occurs during its formation from the Mn
4+
ions dissolved quite uniformly in
the bulk of the low-temperature polymorphic modifications of alumina (gibbsite --
-Al
2
O
3
). The Mn
4+
impurity emission at 677 nm is absent in -alumina films obtained by
oxidation of a metallic FeCrAl alloy. Taking into account high sensitivity of the method, this
indicates very low concentration of such ions.
4. Conclusion
Al
2
O
3
was prepared either by calcination of gibbsite and also generated by oxidation of a
metallic FeCrAl alloy. The Mn
4+
impurity emission at 1.83 eV is absent in -alumina thin
films obtained by oxidation of a metallic FeCrAl alloy. The band of F-centers is absent in
both samples due to the existence of impurity Cr. The difference in oxygen vacancies (F
+
-
centers) amount between Al
2
O
3
from gibbsite and from PM2000 was confirmed by CL
spectra.
5. References
Aoki, Y.; My, N. T.; Yamamoto, S. & Naramoto, H. (1996). Luminescence of sapphire and
ruby induced by He and Ar ion irradiation. Nuclear Instruments and Methods in
Physics Research Section B, Vol.114, No.3-4, (July 1996), pp. 276-280, ISSN 0168-
583X
Auzel, F. & Baldacchini, G. (2007). Photon trapping in ruby and lanthanide-doped materials:
Recollections and revival. Journal of Luminescence, Vol.125, No.1-2, (July-August
2007), pp. 25-30, ISSN 0022-2313
Barker, Jr, A.S. (1963). Infrared lattice vibrations and dielectric dispersion in corundum.
Physical Review, Vol.132, No.4, (November 1963), pp. 14741481, ISSN 0031-
899X
Boumaza, A. & Djelloul, A. (2010). Estimation of the intrinsic stresses in a-alumina in
relation with its elaboration mode. Journal of Solid State Chemistry, Vol.183, No.5,
(May 2010), pp. 10631070, ISSN 0022-4596
Boumaza, A.; Djelloul, A. & Guerrab, F. (2010). Specific signatures of -alumina powders
prepared by calcination of boehmite or gibbsite. Powder Technology, Vol.201, No.2,
(July 2010), pp. 177180, ISSN 0032-5910
Brewer, J.D.; Jeffries, B.T. & Summers, G.P. (1980). Low-temperature fluorescence in
sapphire. Physical Review B condensed matter and Materials Physics Vol.22, No.10,
(November 1980), pp. 49004906, ISSN 1098-0121
Brindley, G.W. & Choe, J.O. (1961). Reaction series gibbsitechi aluminakappa
aluminacorundum. American Mineralogist, Vol.46, No.7-8, (July-August 1961), pp.
771-785, ISSN 0003-004X
Crawford Jr., J.H. (1984). Defects and defect processes in ionic oxides: Where do we stand
today?. Nuclear Instruments and Methods in Physics Research Section B, Vol.1, No.2-3,
(February 1984), pp. 159-165, ISSN 0168-583X

229
Crozier, M.H. (1962). Optical Zeeman effect in the R
1
and R
2
lines of Mn
4+
in Al
2
O
3
. Physics
Letters, Vol.18, No.3, (September 1965), pp. 219-220, ISSN 0031-9163
de Wijn, H.W. (2007). Phonon physics in ruby studied by optical pumping and
luminescence. Journal of Luminescence, Vol.125, No.1-2, (July-August 2007), pp. 55-
59, ISSN 0022-2313
Evans, B.D.; Pogatshnik, G.J. & Chen, Y. (1994). Optical properties of lattice defects in -
Al
2
O
3
. Nuclear Instruments and Methods in Physics Research Section B, Vol.91, No.1-4,
(June 1994), pp. 258-262, ISSN 0168-583X
Evans, B.D. (1994). Ubiquitous blue luminescence from undoped synthetic sapphires. Journal
of Luminescence. Vol.60-61, (April 1994), pp. 620-626, ISSN 0022-2313
Evans, B.D. (1995). A review of the optical properties of anion lattice vacancies, and
electrical conduction in -Al
2
O
3
: their relation to radiation-induced electrical
degradation. Journal of Nuclear Materials, Vol.219, (March 1995), pp. 202-223, ISSN
0022-3115
Figgis, B.N. & Hitchman, M.A. (2000). Ligand Field Theory and its Applications, John Wiley &
Sons Inc, ISBN 978-0-471-31776-0, New York
Geschwind, S.; Kisliuk, P.; Klein, M.P.; Remeika, J.P. & Wood, D.L. (1962). Sharp-Line
Fluorescence, Electron Paramagnetic Resonance, and Thermoluminescence of Mn
4+

in -Al
2
O
3
. Physical Review, Vol.126, No.5, (June 1962), pp. 1684-1686, ISSN 0031-
899X
Ghamdi, A. Al. & Townsend, P.D. (1990). Ion beam excited luminescence of sapphire.
Nuclear Instruments and Methods in Physics Research B, Vol.46, No.1-4, (February
1990), pp. 133-136, ISSN 0168-583X
Ghamnia, M.; Jardin, C. & Bouslama, M. (2003). Luminescent centres F and F
+
in -
alumina detected by cathodoluminescence technique. Journal of Electron
Spectroscopy and Related Phenomena, Vol.133, No.1-3, (November 2003), pp. 55-
63, ISSN 0368-2048
Golightly, F.A.; Stott, F.H. & Wood, G.C. (1979). The Relationship Between Oxide Grain
Morphology and Growth Mechanisms for Fe-Cr-Al and Fe-Cr-Al-Y Alloys. Journal
of The Electrochemical Society, Vol.126, No.6, (June 1979), pp. 1035-1042, ISSN 0013-
4651
Hou, P.Y.; Zhang, X.F. & Cannon, R.M. (2004). Impurity distribution in Al
2
O
3
formed on an
FeCrAl alloy. Scripta Materialia, Vol.50, No.1, (January 2004) pp. 4549, ISSN 1359-
6462
http://accuratus.com/alumox.html/ Aluminum Oxide, Al
2
O
3
Material Characteristics
Jheeta, K.S.; Jain, D.C.; Fouran Singh; Ravi Kumar & Garg, K.B. (2006). Photoluminescence
and UVvis studies of pre- and post-irradiated sapphire with 200 MeV Ag
8+
ions.
Nuclear Instruments and Methods in Physics Research B, Vol.244, No.1, (March 2006),
pp. 187189, ISSN 0168-583X(02)
Jovanic, B.R. (1997). Shift under pressure of the luminescence transitions of corundum
doped with Mn
4+
. Journal of Luminescence, Vol.75, No.2, (September 1997), pp. 171-
174, ISSN 0022-2313

Cathodoluminescence

230
Kislov, A.N.; Mazurenko, V.G.; Korzov, K.N. & Kortov, V.S. (2004). Interionic potentials and
localized vibrations in Al
2
O
3
crystals with vacancies. Physica B: Condensed Matter,
Vol.352, No.1-4, (October 2004), pp. 172178, ISSN 0921-4526
Kogure, T. (1999). Dehydration sequence of gibbsite by electron beam irradiation in a TEM.
Journal of the American Ceramic Society, Vol. 82, No.3, (March 1999), pp. 716-720,
ISSN 0002-7820
Kortov, V.S.; Ermakov, A.E.; Zatsepin, A.F. & Nikiforov, S.V. (2008). Luminescence
properties of nanostructured alumina ceramic. Radiation Measurements, Vol.43,
No.2-6, (February-June 2008), pp. 341-344, ISSN 1350-4487
Kortov, V. & Milman, I. (1996). Some New Data on Thermoluminescence Properties of
Dosimetric Alpha-Al
2
0
3
Crystals. Radiation Protection Dosimetry, Vol.65, No.1-4,
(June 1996), pp. 179-184, ISSN 0144-8420
Kulinkin, A. B.; Feofilov, S. P. & Zakharchenya, R. I. (2000). Luminescence of impurity 3d
and 4f metal ions in different crystalline forms of Al
2
O
3
. Physics of the Solid State,
Vol.42, No.5, (May 2000), pp. 857-860, ISSN 1063-7834
La, S.; Bartram, R.H. & Cox, R.T. (1973). The F
+
center in reactor-irradiated aluminum oxide .
Journal of Physics and Chemistry of Solids, Vol.34, No.6, (June 1973), pp. 1079-1086,
ISSN 0022-3697
Li, P.G.; Lei, M. & Tang, W.H. (2010). Raman and photoluminescence properties of -Al
2
O
3

microcones with hierarchical and repetitive superstructure. Materials Letters, Vol.64,
No.2, (January 2010), pp. 161-163, ISSN 0167-577X
Ma, C.; Chang, Y.; Ye, W.; Shang, W. & Wang, C. (2008). Supercritical preparation of
hexagonal -alumina nanosheets and its electrocatalytic properties. Journal of
Colloid and Interface Science, Vol.317, No.1, (January 2008), pp. 148154, ISSN 0021-
9797
MacKenzie, K.J.D.; Temuujin, J. & Okada, K. (1999). Thermal decomposition of mechanically
activated gibbsite. Thermochimica Acta, Vol.327, No.1-2, (March 1999), pp. 103108,
ISSN 0040-6031
Mennicke, C.; Schumann, E.; Ruhle, M.; Hussey, R.J.; Sproule, G.I. & Graham, M.J. (1998).
The Effect of Yttrium on the Growth Process and Microstructure of -Al
2
O
3
on
FeCrAl. Oxidation of metals, Vol.49, No.5-6, (June 1998), pp. 455-466, ISSN 0030-
770X
Michizono, S.; Saito, Y.; Suharyanto; Yamano, Y. & Kobayashi, S. (2007). Surface
characteristics and electrical breakdown of alumina materials. Vacuum, Vol.81,
No.6, (February 2007), pp. 762765, ISSN 0042-207X
Monteiro, T.; Boemare, C.; Soares, M.J.; Alves, E.; Marques, C.; McHargue, C.; Ononye, L.C.
& Allard, L.F. (2002). Luminescence and structural studies of iron implanted -
Al
2
O
3
. Nuclear Instruments and Methods in Physics Research Section B: Beam
Interactions with Materials and Atoms, Vol.191, No.1-4, (May 2002), pp. 638-643, ISSN
0168-583(2)
Nassau, K. (1983). The Physics and Chemistry of Color: The Fifteen Causes of Color, John Wiley &
Sons Inc, ISBN 0471867764, NewYork

231
Ollivier, B.; Retoux, R.; Lacorre, P.; Massiot, D. & Frey, G. (1997). Crystal Structure of
Kappa Alumina an X-Ray powder diffraction, TEM and NMR study. Journal of
Materials Chemistry, Vol.7, No.6, (1997), pp. 1049 1056, ISSN 0959-9428
Oster, L.; Weise, D. & Kristiapoller, N. (1994). A study of photostimulated
thermoluminescence in C-doped alpha -Al
2
O
3
crystals. Journal of Physics D: Applied
Physics, Vol.27, No.8, (August 1994), pp. 1732-1736, ISSN 0022-3727
Przybylski, K.; Garrett-Reed, A.J.; Pint, B.A.; Katz, E.P. & Yurek, G.J. (1987). Segregation of Y
to Grain Boundaries in the Al
2
O
3
Scale Formed on an ODS Alloy. Journal of The
Electrochemical Society, Vol.134, No.12, (December 1987), pp. 3207-3208, ISSN 0013-
4651
Quadakkers, W.J.; Elschner, A.; Speier, W. & Nickel, H. (1991). Composition and growth
mechanisms of alumina scales on FeCrAl-based alloys determined by SNMS.
Applied Surface Science, Vol.52, No.4, (December 1991), pp. 271-287, ISSN 0169-
4332
Rodrguez-Carvajal, J. (1993). Recent Advances in Magnetic Structure Determination by
Neutron Powder Diffraction. Physica B Condensed Matter, Vol. 192, No. 1-2, (October
1993), pp. 5569, ISSN 0921-4526
Saalfeld, H. & Wedde, M. (1974). Refinement of the structure of gibbsite, Al(OH)3. Zeitschrift
fr Kristallographie, Vol.139, No.1-2, (April 1974), pp. 129135, ISSN 0044-2968
Saalfeld, H.N. (1960). Strukturen des Hydrargillitis und der Zwis- chenstufen biem
Entwassern. Neues Jahrbuch fr Mineralogie, Vol.95, No.7-8, (1960), pp. 187, ISSN
0077-7757
Soukieh, M.; Ghani, B.A. & Hammadi, M. (2004). Numerical calculations of intracavity dye
Q-switched ruby laser. Optics and Lasers in Engineering, Vol.41, No.1, (January 2004),
pp. 177-187, ISSN 0143-8166
Springis, M.J. & Valbis, J.A. (1984). Visible luminescence of colour centres in sapphire.
Physica status solidi. B, Vol.123, No.1, (May 1984), pp. 335-343, ISSN 0370-1972
Stumpf, H.C.; Russel, A.S.; Newsome, J.W. & Tucker, C.M. (1950). Thermal transformations
aluminas and alumina hydrates - Reaction with 44% Technical Acid, Industrial &
Engineering Chemistry Research, Vol.42, No.7, (July 1950), pp. 1398-1403, ISSN 0888-
5885
Surdo, A.I.; Kortov, V.S. & Pustovarov, V.A (2001). Luminescence of F and F
+
centers in
corundum upon excitation in the interval from 4 to 40 eV. Radiation Measurements,
Vol.33, No.5, (October 2001), pp. 587-591, ISSN 1350-4487
Surdo, A.I.; Pustovarov, V.A.; Kortov, V.S.; Kishka, A.S. & Zinin, E.I. (2005). Luminescence
in anion-defective -Al
2
O
3
crystals over the nano-, micro- and millisecond
intervals. Nuclear Instruments and Methods in Physics Research A, Vol.543, No.1, (May
2005), pp. 234238, ISSN 0168-9002
Tale, I.; Piters, T.M.; Barboza-Flores, M.; Perez-Salas, R.; Aceves, R. & Springis, M. (1996).
Optical Properties of Complex Anion Vacancy Centres and Photo-Excited
Electronic Processes in Anion Defective Alpha-Al
2
O
3
. Radiation Protection Dosimetry,
Vol.65, No.1-4, (June 1996), pp. 235-238, ISSN 0144-8420

Cathodoluminescence

232
Whittington, B. & Ilievski, D. (2004). Determination of the gibbsite dehydration reaction
pathway at conditions relevant to Bayer refineries. Chemical Engineering Journal,
Vol.98, No.1-2, (March 2004), pp. 89-97, ISSN 1385-8947
Wood, G. C. (1970). High-temperature oxidation of alloys. Oxidation of Metals, Vol.2, No.1,
(March 1970), pp. 11-57, ISSN 0030-770X
Yu, Z.Q.; Li, C. & Zhang, N. (2002). Size dependence of the luminescence spectra of
nanocrystal alumina. Journal of Luminescence. Vol.99, No.1, (August 2002), pp. 29-
34, ISSN 0022-2313
9
Cathodoluminescence Properties of SiO
2
:
Ce
3+
,Tb
3+
, SiO
2
:Ce
3+
, Pr
3+
and SiO
2
: PbS
Odireleng M. Ntwaeaborwa
1
, Gugu H. Mhlongo
1,2
, Shreyas S. Pitale
1
,
Mokhotswa S. Dhlamini
1,3
, Robin E. Kroon
1
and Hendrik C. Swart
1
1
Department of Physics, University of the Free State, Bloemfontein
2
National Centre for Nanostructured Materials, CSIR, Pretoria
3
Department of Physics, University of South Africa, Pretoria
South Africa
1. Introduction
Silicon dioxide (SiO
2
), also known as silica, is an oxide of silicon (Si) that is found in nature
in two different forms, namely amorphous and crystalline. Traditionally, amorphous SiO
2
is
used in many applications such as semiconductor circuits, microelectronics and optical
fibers for telecommunication. In modern age research, amorphous glassy SiO
2
has emerged
as a potential host lattice for a variety of rare-earth ions to prepare light emitting materials
or phosphors that can be used in different types of light emitting devices. SiO
2
based
phosphors can also be prepared by encapsulating semiconducting nanocrystals of zinc oxide
(ZnO) and lead sulphide (PbS). In addition, recent studies have demonstrated that when
semiconducting nanocrystals are incorporated in glassy SiO
2
activated with trivalent rare
earth ions (Ce
3+
,Tb
3+
,Eu
3+
,Pr
3+
) light emission from the rare-earth luminescent centres could
be increased considerably as a result of energy transfer from the nanocrystals to the rare-
earths, i.e. semiconducting nanocrystals act as sensitizers for radiative relaxation processes
in these centres.
Like any other oxide, amorphous SiO
2
is expected to be a good host matrix for rare-earth
ions for preparation of phosphors because oxides are more chemically stable than
traditional sulphide matrices. In addition, SiO
2
has been reported to have advantages such
as high transparency, dopant solubility and ease of production (Nogami et al., 2007;
Ntwaeaborwa et al., 2007,2008). In recent studies, the preparation of phosphors by
incorporation of semiconducting nanocrystals and rare-earths ions in amorphous SiO
2
has
commonly been achieved by using a sol-gel method. This method has been reported to have
more advantages over other wet chemistry and conventional glass processing methods
because of its potential to produce materials with high purity and homogeneity at low
temperatures (Hench & West, 1990; Ding, 1991). It extends the traditional glass melting
processes by allowing incorporation of semiconductor nanocrystals and rare-earth activators
at low temperatures and predetermined concentrations in such a way that the size and
shape of the particles can be controlled (Reisfeld, 2000) during nucleation and growth
processes. The sol-gel method was used in this study to prepare respectively red, green and
blue cathodoluminescent phosphors by incorporating trivalent rare-earth ions such as

Cathodoluminescence

234
prasedomium (Pr
3+
), terbium (Tb
3+
) and cerium (Ce
3+
) in amorphous SiO
2
. Orange-red
cathodoluminescence was also observed when PbS nanocrystals were incorporated in the
sol-gel derived SiO
2
. Cathodoluminescence intensity of these phosphors was shown to
increase either by co-doping with a different rare-earth ion or encapsulating ZnO
nanocrystals. The CL properties and the intensity degradation of these SiO
2
based
phosphors were evaluated for their possible application in low voltage cathodoluminescent
flat panel displays such as field emission displays (FEDs) and plasma display panels (PDPs).
This chapter is a review of cathodoluminescent properties of phosphor materials prepared
by incorporating Ce
3+
, Tb
3+
, Pr
3+
and PbS nanocrystal in amorphous SiO
2
host. The effects of
rare-eaths co-doping and encapsulating of ZnO nanocrystals on cathodoluminescent
intensity of these materials is also presented. In addition, a review of cathodoluminescence
intensity degradation when these phosphors were irradiated with a beam of electron is
discussed. It was demonstrated that when these phosphors were irradiated with the beam
of electrons for a long period of time they lose their CL intensity and this occured
simultaneously with desorption of oxygen (O) from the phosphor surfaces. In the process,
an oxygen deficient non-luminescent layer was formed on the surface whose formation
could be explained by an electron stimulated surface chemical reaction (ESSCR) model
proposed by Holloway and co-workers (Holloway et al., 1996, 2000). The desorption of
atomic species was explained by Knotek-Fiebelman electron stimulated desorption (ESD)
proposed by Knotek and Feibelman (Knotek and Feibelman, 1978). The objective of the
study was to investigate the effects of a prolonged electron beam irradiation, accelerating
voltage, and oxygen gas on the CL properties and the chemical stability of these phosphors.
2. Cathodoluminescence intensity degradation
Cathodoluminescence (CL) degradation can be defined as a process by which
cathodoluminescent phosphors, used mainly in cathode ray tubes (CRTs) or field emission
display (FEDs), lose their brightness (luminescence intensity) as a result of prolonged
irradiation by a beam of electrons during normal operation. The CL intensity degradation
of CRT/FED phosphors, with special reference to traditional sulphide based phosphors, is a
well documented phenomenon. When a beam of electrons is incident on a phosphor, a
number of physical processes can occur. These include emission of secondary electrons,
Auger electrons and back-scattered electrons. Hundreds of free electrons and free holes are
produced along the path of the incident electron (primary excitation electrons). As
illustrated in figure 1, the free electrons and holes may couple and produce electron-hole (e-
h) pairs that can diffuse through the phosphor and transfer their energy to luminescent
centres resulting in emission of visible photons (Stoffers, 1996; Raue, 1989). This process is
referred to as radiative recombination. Unwanted processes in which the e-h pairs
recombine non-radiatively by transferring their energy to killer centers (incidental
impurities and/or inherent lattice defects) are also possible. The e-h pairs can also diffuse to
the surface of the phosphor and recombine non-radiatively (Stoffers, 1996). In this case, a
non-luminescent oxide layer, which is known to reduce the CL intensity of the CRT/FED
phosphors, may form on the surface. For examples, it was demonstrated that when zinc
sulfide (ZnS) based phosphors were exposed to a prolonged irradiation by energetic beam
of electrons, the ZnS host dissociated into reactive ionic Zn
2+
and S
2-
species, which in turn
combined with ambient vacuum gases such as O
2
and H
2
O to form non-luminescent ZnO or

2
: Ce
3+
,Tb
3+
, SiO
2
:Ce
3+
, Pr
3+
and SiO
2
: PbS

235
ZnSO
4
layers or H
2
S gas (Swart el al., 1998; Itoh et al., 1989) as explained by the ESSCR
model (Holloway et al., 1996, 2000). In the case of oxide based systems, the electron beam
induced dissociation of atomic species is followed by desorption of oxygen from the surface.
For example, Knotek and Fiebelman observed desorption of surface oxygen when SiO
2
or
TiO
2
was irradiated with a beam of electrons and they developed a model called Knotek-
Fiebelman electron stimulated desorption (ESD) mechanism to explain the desorption
process. According to this model, O
+
species produced in the valence band of SiO
2
(TiO
2
)
desorb from the surface following the creation of an electron-hole pair in the L
2,3
level of Si,
Auger relaxation and emission (Knotek and Feibelman, 1978).
+
+
+
+
+
+
+
+ free hole
+
_
_
_
_
_
_
_
free electron
_
+
radiative
activator center
_
+
non-radiative
killer center
_
+
e-h pair

dead layer

dead layer
+
+
_
+
PHOSPHOR
_
+
_
_
back-scattered electrons
Auger electrons
secondary electrons
light emission
primary electrons

Fig. 1. Cathodoluminescence process in a phosphor grain.
3. Experiments
A sol-gel process was used to prepare rare-earths (Ce
3+
,Tb
3+
,Pr
3+
) doped and
semiconducting nanocrystals (ZnO and PbS) encapsulated SiO
2
based phosphors. The sol-
gel is a wet chemical technique commonly used to prepare glassy and ceramic materials at
low temperatures. It involves the preparation of a viscous solution (sol) from one or more
precursor materials and the conversion of the sol into a gel. There are two major chemical
reactions integral to the sol-gel process, namely hydrolysis (reaction of metal alkoxides with
water to form oxides) and condensation/polymerization (the formation of a three
dimensional network). In the presence of a catalyst, the conversion of the sol to gel can be
completed in a short time (within 1-2 hrs). For more information on the sol-gel process, the
reader is referred to the literature cited (Hench & West, 1990; Klein, 1985; Huang et al.,
1985).

Cathodoluminescence

236
3.1 Preparation of PbS and ZnO nanocrystals
Lead sulphide (PbS) nanocrystals were prepared by dissolving anhydrous lead acetate
(Pb(CH
3
COO)
2
) in boiling ethanol (C
2
H
5
OH) and the resulting solution was cooled and
then combined with the ethanol solution of sodium sulphide (Na
2
S) in ice water. The PbS
nanocrystals were formed during hydrolysis and condensation of dissolved species
according to the following reaction equation:
Pb(CH
3
COO)
2
+ Na
2
S PbS + 2CH
3
COO
-
+ 2Na
+

Similarly, ZnO nanocrystals were prepared by dissolving anhydrous zinc acetate
(Zn(CH3COO)
2
) in boiling ethanol and the resulting solution was cooled and then combined
with the ethanol solution of sodium hydroxide (NaOH) in ice water. The ZnO nanocrystals
were formed during hydrolysis and condensation of dissolved species according to the
following reaction equation:
Zn(CH
3
COO)
2
+ 2NaOH ZnO + 2CH
3
COO
-
+ 2Na
+
+ H
2
O()
The unwanted CH
3
COO
-
and Na
+
impurity ions were removed by centrifuging repeatedly
in a mixture of ethanol and heptane. The resulting PbS and ZnO precipitates were either
dispersed/suspended in ethanol for mixing with SiO
2
or were dried at 90
o
C in an oven for
characterization.
3.2 Preparation of rare-earths, PbS and ZnO nanocrystals incorporated SiO
2

A silica (SiO
2
) sol was prepared by hydrolyzing tetraethylorthosilicate (Si(OC
2
H
5
)
4
) or TEOS
with a solution of water, ethanol and dilute nitric acid (HNO
3
). The mixture was stirred
vigorously at room temperature. The SiO
2
was formed during the condensation reaction
according to the following equation:
Si(OC
2
H
5
)
4
+ 2H
2
O SiO
2
+ 4C
2
H
5
OH
The SiO
2
sol was divided into two parts and one part was mixed with the ethanol
suspension of PbS nanocrystals and the other part was mixed with RE(NO
3
) 6H
2
O (RE =
Ce/Tb/Pr) dissolved in ethanol before adding the ethanol suspension of ZnO nanocrystals
and the mixtures were stirred vigorously until thick viscous gels were formed. The gels
were dried at room temperature for 3 8 days, were ground using a pestle and mortar, and
were finally annealed at 600
o
C in air for 2 hours. Samples that were prepared include SiO
2
;
SiO
2
:Pr
3+
/Ce
3+
/Tb
3+
; SiO
2
:Ce
3+
,Tb
3+
; ZnO-SiO
2
:Pr
3+
; and SiO
2
:PbS powders. The molar
concentrations of rare-earth ions and ZnO in SiO
2
were varied while that of the PbS
nanocrystals was fixed at 0.34 mol%.
4. Results and discussions
4.1 X-ray diffraction
Figure 2 shows the X-ray diffraction patterns of ZnO and PbS nanocrystals. The patterns are
consistent with the hexagonal and cubic phases of ZnO and PbS referenced in JCPDS cards
number 36-14551 and 05-0592 respectively. The broadening of the diffraction peaks is
attributed to smaller particles. The average crystallite sizes estimated from the broadened

2
: Ce
3+
,Tb
3+
, SiO
2
:Ce
3+
, Pr
3+
and SiO
2
: PbS

237
XRD peaks were 5 and 6 nm in diameters for ZnO and PbS respectively. Figure 3 shows
XRD patterns of SiO
2
and ZnO-SiO
2
:Pr
3+
, all annealed at 600
o
C for two hours. The
concentrations of ZnO and Pr
3+
in SiO
2
host were 5 and 1 mol% respectively. All the spectra
are characterized by the well known broad diffraction of the amorphous SiO
2
peak at 2 =
20 25
o
. The absence of x-ray diffraction peaks from encapsulated 5 mol% of ZnO
nanocrystals and 1 mol% of Pr
3+
is probably due to their relatively low concentration and/or
high scattering background from amorphous SiO
2
.

Fig. 2. XRD patterns of (a) ZnO and (b) PbS nanoparticles.

Fig. 3. XRD patterns of SiO
2
and ZnO-SiO
2
:Pr
3+

Cathodoluminescence

238
4.2 Cathodoluminescence: Properties and intensity degradation
Cathodoluminescence data were recorded using S2000 Ocean Optics CL spectrometer
attached to a vacuum chamber of the PHI 549 Auger electron spectrometer (AES) either at
base pressure or after backfilling with oxygen gas. Figure 4 presents a simplified schematic
diagram of the AES system whose major components are vacuum chamber, housing for the
cylindrical mirror analyzer, and electron gun. Attached to the chamber are the fiber optics
CL spectrometer and the residual gas analyzer (RGA). The AES and CL spectrometers are
connected to two separate computers equipped with programs for recording the Auger and
CL data during electron beam irradiation. The Auger and CL data were recorded when the
samples were irradiated with a beam of electrons using different accelerating voltages and
beam currents.

Fig. 4. Schematic diagram of the PHI 549 Auger system.
4.2.1 SiO
2
:Ce
3+
,Tb
3+

SiO
2
:Ce
3+
,Tb
3+
(Ce
3+
=Tb
3+
= 0.5 mol%) powders were irradiated with a 54 mA/cm
2
beam of
electrons accelerated at 2 kV in a vacuum chamber of the Auger spectrometer maintained at
either 110
-7
or 110
-8
Torr O
2
for 10 hours. Figure 5 compares the normalized CL emission
spectra of SiO
2
and SiO
2
:Ce
3+
(0.5 mol%). The inset shows the CL spectrum of SiO
2
:Tb
3+
(0.5
mol%). The SiO
2
spectrum is characterized by a broadband blue emission with a maximum
at 445 nm and a satellite peak at 490 nm. The visible emission from SiO
2
is attributed to
carrier trapping by structural defects (Lin and Baerner., 2000; Han et al., 2002; Gu et al.,
1999) or charge transfer between Si and O atoms (Garca et al., 1995). Upon incorporation of
Ce
3+
and Tb
3+
the defects emission at 445 nm from SiO
2
was suppressed and blue and green
emissions from Ce
3+
and Tb
3+
were observed. The blue emission with a maximum at 489 nm
can be attributed to 4f5d transitions of Ce
3+
while the line emissions with the main

2
: Ce
3+
,Tb
3+
, SiO
2
:Ce
3+
, Pr
3+
and SiO
2
: PbS

239
emission at ~550 nm in the inset can be attributed to the 5D 7F transitions of Tb
3+
. Figure
6 shows the CL emission spectra of the co-activated SiO
2
:Ce
3+
,Tb
3+
before and after electron
irradiation. The insets are the photographs of the irradiated area in the beginning and at the
end of irradiation. The emission peaks are associated with radiative transitions of Tb
3+
ions.
The main emission peak was observed at ~550 nm and it is associated with
5
D
4

7
F
5

transitions of the Tb
3+
ions. The emission intensity of this peak was reduced by ~50% after
10 hours of irradiation as shown by the fading of green luminescence in the inset. The
synergies between the CL intensity degradation and the changes in the surface chemistry
were determined using the Auger electron and the X-ray photoelectron spectroscopy.

Fig. 5. Normalized CL intensity of SiO
2
and SiO
2
:Ce
3+
. The inset is the CL intensity versus
wavelength spectrum of SiO
2
:Tb
3+
.

Fig. 6. CL intensity spectra of SiO
2
:Ce
3+
,Tb
3+
before and after electron irradiation.

Cathodoluminescence

240
Figure 7 shows the Auger spectra of the SiO
2
powder before and after electron irradiation.
The main features in both spectra are the Si (76.9 eV) and O (505 eV) peaks. Note the
decrease in the O Auger peak intensity after 10 hours of continuous irradiation. The Auger
peak at 76.9 eV is associated with Si in SiO
2
. As a result of the prolonged electron irradiation
this peak shifted to 82.7 eV and its intensity was reduced slightly. Thomas (Thomas; 1974)
attributed the shift to change in the density of state in the valence band rather than the shift
in the binding energies of Si. The XPS data showed that with continuous irradiation an
Auger peak associated with elemental Si developed at 98.2 eV and there was also a
subsequent change in colour of the chathodoluminescence at the irradiated area.
Figure 8 shows the decrease/degradation of the CL intensity and the Auger-peak-to-peak-
heights (APPHs) of oxygen (O), silicon (Si) and adventitious carbon (C) as a function of
electron dose for the data recorded when the chamber was maintained at 110
-7
Torr O
2
.
During electron irradiation, the decrease in the CL intensity was simultaneous with rapid
desorption of oxygen from the surface as shown in the figure. Ce
3+
and Tb
3+
ions were not
detected probably due to their relatively low concentration. While the C peak was almost
unchanged, the Si peak was shown to decrease marginally but steadily. Similar trend was
observed when the chamber pressure was 110
-8
Torr O
2
.

Fig. 7. AES spectra of SiO
2
:Ce
3+
,Tb
3+
before and after electron irradiation.

2
: Ce
3+
,Tb
3+
, SiO
2
:Ce
3+
, Pr
3+
and SiO
2
: PbS

241

Fig. 8. Auger peak-to-peak-heights of O, Si, C; and the CL intensity as a function of electron
dose (Ntwaeaborwa et al., 2006).
Figure 9 and 10 compare respectively the normalized O APPHs and the CL intensity
degradation as functions of electron dose at the chamber pressures of 110
-7
and 110
-8
Torr
O
2
. As shown in figure 9, the rate of O desorption was faster at the low flux (10
-8
Torr O
2
) of
O
2
while that of the CL intensity degradation in figure 10 was faster at higher oxygen
pressure (10
-7
Torr O
2
). The data in figure 9 and 10 suggest that there is a correlation
between O desorption and the CL intensity degradation.

Fig. 9. Normalized O APPHs as a function of electron dose at 110
-7
and 110
-8
Torr O
2

(Ntwaeaborwa et al., 2007).

Cathodoluminescence

242

Fig. 10. Normalized CL intensity as a function of electron dose at 110
-7
and 110
-8
Torr O
2
(Ntwaeaborwa et al., 2006).
X-ray photoelectron spectroscopy (XPS) was used to investigate the correlation between the
CL intensity degradation and the changes on the surface chemistry. The XPS data were
recorded from the SiO
2
:Tb
3+
,Ce
3+
powders before and after electron irradiated. Figure 11
compares the high resolution Si 2p XPS peaks before and after irradiation. Note the shift of
0.2 eV to the right in the peak position, development of the new peak at 98.2 eV and the
narrowing of the spectrum after electron irradiation. While the peak shift and narrowing can
be attributed to surface charging, among other things, the development of the new peak can
be attributed to the chemical changes that occurred on the surface during irradiation.

Fig. 11. High resolution Si XPS peaks before and after electron irradiation.
Figures 12 and 13 show respectively the fitted high resolution Si 2p XPS peaks before and
after irradiation. The peaks at 102.5, 103.7 and 104.8 eV in figure 12 can be assigned to SiO
x

2
: Ce
3+
,Tb
3+
, SiO
2
:Ce
3+
, Pr
3+
and SiO
2
: PbS

243
(x < 2), SiO
2
and chemisorbed species (from atmospheric oxygen/moisture) respectively
(Nagpure et al.; 2011). Note the absence of the peak from chemisorbed species and the
appearance of two additional peaks at 100.6 eV and 98.2 eV in figure 12 that can be assigned
to SiC (Nagpure et al.; 2011) and elemental Si (Moulder et al.; 1992) respectively. The well
known effect of the prolonged electron beam irradiation of SiO
2
is the breaking
(dissociation) of Si-O bond and the subsequent desorption of oxygen. As shown in the XPS
spectrum after electron beam irradiation in figure 11, the depletion of oxygen was
accompanied by a growth of free (elemental) silicon (Si) at 98.2 eV. As a result of oxygen
desorption the surface was rich in Si and the remaining structure was probably an oxygen
deficient SiO
x
(0<x<2). The mechanism of desorption of oxygen from SiO
2
is beyond the
scope of this chapter and the readers are referred to the literature cited for further reading
(Carriere and Lang, 1977; Knotek and Feibelman, 1978; Thomas, 1974; Fiori and Devine,
1984). It is most likely that the oxygen deficient SiO
x
was non-luminescent and it therefore
contributed to the CL intensity degradation (Ntwaeaborwa et al., 2006, 2007).

Fig. 12. Fitted high resolution Si 2p XPS peak before electron irradiation.

Fig. 13. Fitted high resolution Si 2p XPS peak after electron irradiation.

Cathodoluminescence

244
4.2.2 ZnO-SiO
2
:Pr
3+
and SiO
2
:Ce
3+
,Pr
3+
In this section the effects of ZnO nanoparticle, Ce
3+
and Pr
3+
concentrations on
cathodoluminescence intensities of ZnO-SiO
2
:Pr
3+
and SiO
2
:Ce
3+
,Pr
3+
are discussed. Figure
14 shows the CL spectra of SiO
2
:Pr
3+
, with different concentrations (0.05 0.25 mol%) of Pr
3+
,
measured in the AES vacuum chamber at a base pressure of 10
-8
Torr. Note that relatively
low concentrations of Pr
3+
were used to avoid quenching at higher concentrations. The CL
intensity increased with concentration from 0.05 to 0.2 mol% and it decreased when the
concentration was increased to 0.25 mol% due to concentration quenching effect. It is well
known that at higher concentrations of light emitting ions, luminescence can be quenched as
a result of clustering of or cross relaxation between the ions (Blasse and Grabmaier, 1994;
Sol et al., 2005). The CL emission spectra are characterized by multiple line emissions in
the visible region of the electromagnetic spectrum. The main emission associated with
3
P
0
3
H
6
transitions of Pr
3+
was observed at 616 nm and minor emissions due to transitions
from
3
P
0
and
1
D
2
to
3
H
(J = 6, 5, 4)
and
3
F
(J = 2, 3, 4)
were also observed. These transitions were not
measured but were assigned according to the literature cited (Mhlongo et al., 2011a,
Soklska et al., 2000). The inset of figure 13 shows the maximum CL intensity of the 616 nm
peak as a function of Pr
3+
concentration illustrating the increase in intensity with
concentration from 0.05 to 0.2 mol% and a sudden decrease when the concentration was
increased to 0.25 mol%.

Fig. 14. CL spectra of SiO
2
:Pr
3+
with different concentrations of Pr
3+
.
Figure 15 compares the CL emission spectra of SiO
2
, SiO
2
:1mol%Pr
3+
and ZnO-
SiO
2
:1mol%Pr
3+
(ZnO = 5mol%) powders recorded when the powders were irradiated with
a beam of 2 keV electrons and a beam current of 20 A in the chamber kept at a base
pressure of 1.610
-8
Torr. With incorporation of Pr
3+
, the blue emission of SiO
2
at 445 nm
was suppressed and all the observed emissions were due to transitions in Pr
3+
. The main
emission at 616 nm was enhanced considerably by the incorporation of ZnO nanopareticles.
This emission is 2 more intense than the emission from SiO
2
:Pr
3+
suggesting that ZnO
contributed to this enhancement by transferring excitation energy to Pr
3+
ions. Note that

2
: Ce
3+
,Tb
3+
, SiO
2
:Ce
3+
, Pr
3+
and SiO
2
: PbS

245
because of severe charging upon irradiation with electrons, cathodoluminescence of ZnO
nanoparticles could not be measured. However it is a well known phenomenon that ZnO
nanoparticles display dual emission in the UV and visible region irrespective of the type of
excitation (i.e. UV photons or electrons) (Mhlongo et al., 2011b). Since both the UV and
defects emissions were not detected when ZnO nanoparticles were incorporated in SiO
2
:Pr
3+

and there was a subsequent increase in the red emission from Pr
3+
, it is reasonable to
attribute the increase to energy transfer from ZnO to Pr
3+
. The reader is referred to the
literature cited to read about the mechanism of energy transfer from ZnO to rare-earth ions
in glassy SiO
2
host (Bang et al., 2006; Ntwaeaborwa and Holloway, 2005; Mhlongo et al.,
2010)
Figure 16 shows the CL emission spectra of Ce
3+
-Pr
3+
co-activated SiO
2
phosphors. The
concentration of Pr
3+
was fixed at 0.2 mol% while that of Ce
3+
was varied from 0.2 2 mol%.
The spectra were recorded when the powders were irradiated with 2 keV electrons and a
beam current of 8.5 A in the AES vacuum chamber maintained at a base pressure of 1.2
10
-8
Torr. The CL spectrum of Pr
3+
singly doped SiO
2
consists of a major emission peak at
616 nm and minor peaks at 510 and 489. The CL spectrum of Ce
3+
consists of broad
emission peak with a maximum at 489 nm. Notice the suppression of the Pr
3+
emission in
co-activated SiO
2
:Ce
3+
,Pr
3+
and the change in PL intensity of the 489 nm peak with Ce
3+

concentration. The maximum intensity was observed when 1 mol% of Ce
3+
was co-doped
with 0.2 mol% of Pr
3+
as shown in the inset of figure 16. The intensity was quenched,
probably as a result of concentration quenching effects, when the Ce
3+
was increased to 1.5
and 0.2 mol%.

2
, SiO
2
:Pr
3+
and ZnO-SiO
2
:Pr
3+
(Mhlongo et al., 2010)

Cathodoluminescence

246

2
:Ce
3+
, SiO
2
:Ce
3+
,Pr
3+
and SiO
2
:Pr
3+
. The inset is the maximum CL
intensity of the 489 peak as a function of Ce
3+
concentration.
The fact that the increase in the blue emission at 489 nm was simultaneous with the
quenching of Pr
3+
emission from SiO
2
:Ce
3+
,Pr
3+
(Ce
3+
= 1 mol% and Pr
3+
= 0.2 mol%)
suggests that energy was transferred from Pr
3+
to Ce
3+
. Note that the energy associated with
the
1
P
0
3
H
6
transition of Pr
3+
is 2.0 eV (616 nm), which is too low to transfer to Ce
3+
at 2.5
eV (489 nm). On the other hand, the energy associated with the
3
P
0
3
H
4
transition of Pr
3+

occurs at the same energy position of 2.0 eV (489 nm) as the blue emission from Ce
3+
. It is
therefore reasonable to speculate that energy was transferred resonantly or by phonon
mediated processes from the
3
P
0
state of Pr
3+
to 5d
1
states of Ce
3+
. Before discussing the
proposed mechanism of energy transfer, it is important to discuss briefly the transitions
responsible for blue emission in Ce
3+
. Shown in figure 17 is the deconvoluted CL emission
spectrum of SiO
2
:Ce
3+
,Pr
3+
(Ce
3+
= 1 mol%) recorded during irradiation with 2 keV beam of
electrons and a beam current of 8.5 A at a base pressure of 1.2 10
-8
Torr. The 489 nm peak
is deconvoluted into two Gaussian peaks, one at 480 nm and the other one at 525 nm. The 4f
ground state of the Ce
3+
is split into two components (
2
F
5/2
and
2
F
7/2
) due to spin-orbit
interaction while the excited 5d
1
state can be split into 2-5 components by the crystal field
(Ntwaeaborwa et al., 2008; Mhlongo et al., 2011a). Here we consider a case where the 5d
1
is
split into two components, namely the upper
2
D
5/2
and the lower
2
D
3/2
. We speculate that
the absorbed excitation energy causes transition from the ground state to the upper
2
D
5/2

state followed by radiationless transition to the lower
2
D
3/2
and a subsequent radiative
transition to the
2
F
7/2
(480 nm) and
2
F
5/2
(525 nm) energy states.

2
: Ce
3+
,Tb
3+
, SiO
2
:Ce
3+
, Pr
3+
and SiO
2
: PbS

247

Fig. 17. Deconvoluted CL emission spectrum of SiO
2
:Ce
3+
.
The proposed mechanism of energy transfer from Pr
3+
to Ce
3+
is presented in figure 18.
According to this mechanism, the absorbed excitation energy causes transition from the
ground state (
3
H
4
) to higher energy states (4f5d) of Pr
3+
. This is followed by a radiationless
transition to the
3
P
2,1,0
states of Pr
3+
. In a Pr
3+
singly doped SiO
2
the
3
P
1,0
states will be
followed by radiative transitions to the
3
H
6,5,4
states. However, in the Ce
3+
-Pr
3+
co-activated
SiO
2
the
3
P
0
state of Pr
3+
will transfer energy resonantly or by phonon-assisted processes to
the 5d
1
states of Ce
3+
resulting in an enhanced blue emission during radiative transition to
the ground state. This is only possible if the transfer rate is faster than the radiative
transition to the
3
H
6,5,4
states. Similar mechanism was reported by Ntwaeaborwa et al.
(Ntwaeaborwa et al., 2008) for energy transfer between Eu
3+
and Ce
3+
.

Fig. 18. Simplified energy level diagrams of Ce
3+
and Pr
3+
illustrating energy transfer from
Pr
3+
to Ce
3+
.

Cathodoluminescence

248
The effect of accelerating voltage on the brightness of SiO
2
:Ce
3+
,Pr
3+
(Ce
3+
= 1 mol% and
Pr
3+
=0.2 mol%) was evaluated by varying the voltage from 1 5 kV when the beam current
was fixed at 8.5 A. Figure 19 shows the CL intensity as a function of accelerating voltage.
As shown in the inset, the CL intensity of the 489 nm peak is increasing linearly with voltage
from 1 to 5 kV. It is well known that the depth of penetration by incident electrons is
proportional to the accelerating voltage (Kumar et al., 2010), i.e. high the accelerating
voltage deeper will be the penetration. The proportional increase in the CL intensity
(brightness) with accelerating voltage suggests that the rate of generation of free electron
and holes was high at greater penetration depths and a subsequent increase of radiative
recombination. A lack of luminescence saturation at voltages up to 5 kV is a sign of good
prospect for this phosphor to be used in low voltage (1 10kV) field emission display
technology.

Fig. 19. The effect of accelerating voltage on the CL intensity of SiO
2
:Ce
3+
,Pr
3+
.
4.2.3 SiO
2
:PbS

In this section, cathodoluminescent properties and intensity degradation of SiO
2
:PbS
phosphor are discussed. The preparation of the samples was discussed in section 1.2. The
concentration of PbS nanoparticles in SiO
2
was 0.34 mol%. The CL data were recorded when
the SiO
2
:PbS powders were irradiated with a 2 keV beam of electrons in the Auger
spectrometer chamber at different oxygen pressures. The Auger-peak-to-peak-heights
(APPHS) of O, Si and C and CL intensity as a function of electron dose resembled that of
SiO
2
:Ce
3+
,Tb
3+
in figure 8 suggesting that the decrease in the CL intensity was simultaneous
with desorption of O. Figure 20 shows broadband CL emission spectra, before and after
electron irradiation, with a maximum at ~700 nm. While the CL emission from pure SiO
2

was observed at 445 nm (figure 5), the CL emission from pure PbS nanoparticles could not
be recorded due to excessive surface charging. It is however well known that light emission
from PbS nanoparticles associated with excitonic recombination is usually in the near
infrared region at 1100 1700 nm (Ntwaeaborwa et al., 2009a, which is blue-shifted from
PbS bulk emission at ~3000 nm due to quantum confinement effects. Therefore the orange-
red emission in figure 20 is neither coming from SiO
2
nor PbS nanoparticles. This emission

2
: Ce
3+
,Tb
3+
, SiO
2
:Ce
3+
, Pr
3+
and SiO
2
: PbS

249
is similar to emission associated with transitions in Pb
2+
ions in ZnS:Pb
2+
(Bol and Meijerink,
2001) and SiO
2
:Pb
2+
(Ntwaeaborwa et al., 2009b). Although the exact mechamism of this
emission is not known yet, it is reasonable to attribute the emission to transisions in Pb
2+
.
This can only be possible if Pb
2+
ions are selectively excited. Note that blue, green and red
emissions from Pb
2+
have been reported and the red emission, similar to the one in figure 20,
was attributed to
3
P
0,1
1
S
0
transition of Pb
2+
(Bol and Meijerink, 2001). As shown in the inset
of figure 20, the red-orange cathodoluminescence degraded by more than 50% following the
prolonged irradiation by 2 keV electrons. Similar mechanism that involves desorption of O
and a subsequent formation of less luminescent SiO
x
(x < 2) discussed in section 2.2.1 can be
used to explain the CL intensity degradation of the SiO
2
:PbS phosphor.

Fig. 20. CL emission spectra of SiO
2
:PbS before and after electron irradiation.
5. Conclusion
Green, blue and orange-red cathodoluminescence was observed, respectively, from
SiO
2
:Ce
3+
,Tb
3+
; SiO
2
:Ce
3+
,Pr
3+
and SiO
2
:PbS phosphors prepared by the sol-gel method.
Green emission from Tb
3+
was enhanced by Ce
3+
co-doping. The CL intensity of the green
emission degraded by 50% when irradiated with a beam of electrons for 10 hours and the
CL intensity degradation was simultaneous with desorption of O from the surface. The
degradation was attributed to the formation of oxygen deficient SiO
2
(x<2) layer. Enhanced
blue emission from Ce
3+
was observed from Ce
3+
-Pr
3+
co-doping while the red emission of
Pr
3+
suppressed. Possible mechanism of energy transfer from Pr
3+
to Ce
3+
was discussed.
Orange-red cathodoluminescence from SiO
2
:PbS was attributed to
3
P
0,1
1
S
0
transition of
Pb
2+
. SiO
2
is a potential host matrix for rare-earths and semiconducting nanocrystals for
preparation of phosphors that can be used in light emitting devices including cathode ray
tubes and field emission displays.

Cathodoluminescence

250
6. Acknowledgment
The authors would like to acknowledge the financial support from the South African
National Research Foundation and the Nanoscience cluster fund of the University of the
Free State.
7. References
[1] Nogami, M., Enomoto, T., Hayakawa, T. (2002). Enhanced fluorescence of Eu
2+
induced
by energy transfer from nanosized SnO
2
crystals in glass. J. Lumin. Vol. 97, No. 3-4,
pp. (147-152)
[2] Ntwaeaborwa, O.M., Swart, H.C., Kroon, R.E., Botha, J.R., Holloway, P.H. (2007).
Cathodoluminescence degradation of SiO
2
:Ce,Tb powder phosphors prepared by a
sol-gel process. J. Vac. Sci. Technol. A. Vol. 25, No. 4, pp. (1152-1155).
[3] Ntwaeaborwa, O.M., Swart, H.C., Kroon, R.E., Botha, J.R., Ngaruiya, J.M., Holloway,
P.H. (2008). Enhanced photoluminescence of rare-earth activators in sol-gel derived
SiO
2
by energy transfer from ZnO nanoparticles and co-activators, In:
Photoluminescence Research Progress, Wright H.K. and Edwards, G.V., pp. (287-306),
Nova publishers, 978-1-60456-538-6, New York.
[4] Hench, L.L., Wesk, J.K. (1990). The Sol-Gel Proces, Chem. Rev., Vol. 90, No. 1, pp. (33-72).
[5] Ding, J.Y., Day, D.E. (1991). Preparation of silica glass microsphere by sol-gel, J. Mat. Res,
Vol. 6, No. 1, pp. (168-174).
[6] Ntwaeaborwa, O.M.; Swart H.C.; Kroon, R.E.; Holloway P.H.; Botha, J.R. (2006),
Enhanced luminescence and degradation of SiO
2
:Ce
3+
,Tb
3+
powder phosphor
prepared by a sol-gel process, J. Phys. Chem. Sol.,Vol. 67, pp (1749-1753)
[7] Reisfeld, R., Gaft M., Saridarov, T., Panczer, G., Zelner, M. (2000). Nanoparticles of
cadmium sulfidewith europium and terbium in zirconia films having intensified
luminescence, Mat. Lett, Vol. 45, No. 3, pp. (154-156).
[8] Holloway, P.H., Sebastian, J., Trottier, T., Jones, S., Swart, H., Petersen, R.O. Degradation
mechanisms and vacuum requirements for FED phosphors. (1996). Mat. Res. Soc.
Symp. Proc, Vol. 424, pp. (425-431).
[9] Holloway, P.H., Trottier, T.A., Sebastian, S., Jones, S., Zhang, X.-M., Bang, J.-S., Abrams,
B., Thomes, W.J., Kim, T.J. (2000). Degradation of field emission display phosphors,
J. Appl. Phys., 88(1), pp. (483-488)
[10] Knotek, M.L. and Feibelman, P.J. (1978). Ion desorption by core-hole Auger decay, Phys.
Rev. Lett., Vol. 40, No. 14, pp. (964-967)
[11] Stoffers, C.; Yang, S.; Jacobsen, S. M. and Summers, C.J. (1996). Saturation of phosphor
under low voltage excitation, J. Soc. Inf. Display. Vol. 4, No. 4, pp. (337-341)
[12] Raue, R.; Vink, A.T. and Welker T. (1989). Phosphors screens in cathode ray tubes for
projection television Phillips Tech. Rev, Vol. 44, No. (11-12), pp. (335 347).
[13] Swart, H.C.; Oosthuizen, L.; Holloway, P.H. & Berning G.L.P. (1998). Degradation
behavior of ZnS phosphors under different experiment conditions, Surf. Interface
Anal., Vol. 26, No. 5, pp (337-342)
[14] Itoh, S.; Kimizuka, T. and Tonegwa T. (1989). Degradation mechanism for low voltage
cathodoluminescence sulphide phosphors, J. Electrochem. Soc., Vol. 136, No. 1819-
1823
[15] Klein, L.C. (1985). Sol-gel processing of silicates, Ann. Rev. Mater. Sci., 15,pp. (227-248)

2
: Ce
3+
,Tb
3+
, SiO
2
:Ce
3+
, Pr
3+
and SiO
2
: PbS

251
[16] Huang, H-H.; Orler, B. and Wilkes, G.L. (1985). Ceramers: Hybrid materials
incorporating polymeric/oligomeric species with inorganic glasses by a sol-gel
process 2. Effect of acid content on the final product, Polymer Bulletin., Vol. 14,
No. , pp. (557-564)
[17] Nagpure, I.M.; Pitale, S.S.; Tshabalala, K.G.; Kumar, V.; Ntwaeaborwa, O.M., Terblans,
J.J. and Swart, H.C. (2011). Luminescence response and CL degradation of
combustion synthesized spherical SiO
2
:Ce nanophosphor, Mater. Res. Bull.,
doi:10.1016/j.materresbulls.2011.08.051
[18] Thomas, S. (1973). Electron-irradiation effect in the Auger analysis of SiO
2
, J. Appl. Phys.,
Vol. 45, No. 1, pp (161-166).
[19] Lin, J. and Baerner K. (2000). Tunable photoluminescence in sol-gel derived silica
xerogels, Mat. Lett., Vol. 46, No. (2-3), pp (86-92)
[20] Han, Y.; Lin J. and Zhang H. (2002). Photoluminescence of organic-inorganic hybrid
SiO
2
xerogels, Mat. Lett., Vol. 54, No. (5-6), pp (389-396)
[21] Gu G.; Ong P.P. and Chu C. (1999). Thermal stability of mesoporous silica molecular
sieve, J. Phys. Chem. Sol., Vol. 60, No. 7, pp (943-947)
[22] Garca M.; Mondragn M.A.; Tllez S.; Campero A. and Castano V.M. (1995). Blue
emission and tetraethoxysilane and silica gels, Mater. Chem. Phys., Vol. 41, No. 1, pp
(15-17).
[23] Moulder, J.F.; Stickle, W.F.; Sobol, P.E. and Bomben, K. (1992). Handbook of X-ray
photoelectron spectroscopy, Perkin Elmer Corp., Minnesota (USA).
[24] Carriere, B. and Lang, B. (1977). A study of the charging effect and dissociation of SiO
2

surfaces by AES, Surf. Sci., Vol. 64, pp (209-223)
[25] Fiori, C. and Devine, R.A.B. (1984). Photon-induced oxygen loss in thin SiO
2
films, Phys.
Rev. Lett., Vol. 52, No. 23, pp (2081-2083)
[26] Blasse, G.; Grabmaier, B.C. (1994). Luminescent Materials, Springer Verlag, ISBN 3-540-
58019-0 New York
[27] Sol, J.G.; Baus, L.E.; Jaque D. (200). An Introduction to the Optical Spectroscopy of
Inorganic Solids, John Wiley and Sons, Ltd, ISBN 0-470-86886-4,Chichester.
[28] Mhlongo, G.H.; Ntwaeaborwa, O.M., Dhlamini, M.S.; Swart, H.C; Hillie, K.T. (2011a).
Effects of Ce
3+
concentration, beam voltage and current on the
cathodoluminescence of intensity of SiO
2
:Ce
3+
-Pr
3+
nanophosphor, J. Alloy.
Compd., Vol. 509, pp (2986-2992)
[29] Soklska, I.; Golab, S.; Baluka, J.; Ryba-Romanowski, W. (2000), Quenching of Pr
3+
in
single crystals of K
5
Pr
x
La
1-x
Li
2
F
10
., J. Lumin., Vol. 91, pp (79-86).
[30] Mhlongo, G.H.; Ntwaeaborwa, O.M., Swart, H.C.; Kroon, R.E.; Solarz, P.Ryba-
Romanowski, W., Hillie, K.T. (2011b). Luminescence dependence of Pr
3+
activated
SiO
2
nanophosphor on Pr
3+
concentration, temperature and ZnO incorporation, J.
Phys. Chem. C., Vol. 115, pp (17625 17632).
[31] Mhlongo, G.H.; Ntwaeaborwa, O.M.; Dhlamini, M.S.; Swart, H.C.; Hillie, K.T. (2010)
Cathodoluminescence properties of SiO
2
:Pr
3+
and ZnOSiO
2
:Pr
3+
, J. Mater. Sci., Vol.
45, pp (5228 5236).
[32] Kumar, V.; Mishra, V.; Pitale, S.S.; Nagpure, I.M.; Coetsee, E.; Ntwaeaborwa, O.M.;
Terblans, J.J.; Swart, H.C. (2010). Surface chemical reactions during electron beam
irradiation of nanocrystalline CaS:Ce
3+
phosphor, J. Appl. Phys., Vol. 107, pp
(123533-1 - 123533-6)

Cathodoluminescence

252
[33] Bol, A.A.; Meijerink, A. (2001). Luminescence of nanocrystalline ZnS:Pb
2+
, Phys. Chem.
Chem. Phys., Vol. 3 (11), pp (830-832).
[34] Ntwaeaborwa, O.M.; Kroon, R.E.; Kumar, V.; Dubroca T.; Ahn, J.-P.; Park, J.-K.; Swart,
H.C. (2009a), Ex-situ synthesis and optical properties of ZnO-PbS nanocomposites,
J. Phys. Chem. Sol., Vol. 70, pp (1438 1442)
[35] Ntwaeaborwa O.M.; Swart H.C.; Kroon, R.E.; Terblans, J.J.; Holloway, P.H. (2009b).
Synthesis, characterization and luminescent properties of ZnO-SiO
2
:PbS, J. Vac. Sci.
Technol. A., Vol. 27(4), pp (767 769)
10
Cathodo- and Photo- Luminescence of Silicon
Rich Oxide Films Obtained by LPCVD
Rosa Lpez-Estopier
1, 2
, Mariano Aceves-Mijares
3
and Ciro Falcony
4

1
Department of Electronics, ITSPR, Poza Rica, Veracruz
2
Department of Applied Physics, ICMUV, University of Valencia, Burjassot, Valencia
3
Department of Electronics, INAOE, Tonantzintla, Puebla
4
Department of Physics, CINVESTAV-IPN, Distrito Federal
1,3,4
Mxico
2
Spain
1. Introduction
Silicon technology dominates the electronics industry today, so it is highly desirable the
development of silicon-based components compatible with silicon technology, allowing
integration of electrical and optical components on a single chip. One promising approach to
the development of a silicon based light emitter is Silicon Rich Oxide (SRO), also called off-
stoichiometric silicon oxide. The interest on the optical properties of this material has grown
since it was demonstrated that SRO films subjected to high-temperature annealing exhibit
efficient photoluminescence (PL) (Iacona et al., 2000; Shimizu-Iwayama et al., 1996).
At present, different techniques have been employed to produce SRO films, these include
plasma enhanced chemical vapor deposition (PECVD) (Pai et al., 1986), low pressure chemical
vapor deposition (LPCVD) (Dong et al., 1978), silicon implantation into thermal oxide (SITO)
(Pavesi et al., 2000), reactive sputtering (Hanaizumi et al., 2003) and others. An indicator of the
Si content in this material is the parameter R
0
, which is the ratio of the partial pressure of the
precursor gases (in this case, N
2
O/SiH
4
) when it is prepared by gas phase deposition methods
like CVD (Chemical Vapor Deposition). Silicon excess ranging from 17% to 0% can be obtained
varying the R
0
from 3 to 100. Among CVD methods, the low pressure chemical vapor
deposition (LPCVD) is a very convenient approach for the deposition of SRO films since it
allows an exact variation of the Si content and produces higher PL than other methods.
Different applications of SRO have been proposed, such as: visible light emission devices
(Shimizu-Iwayama et al., 1996), non-volatile or electrically alterable memory devices (Calleja
et al., 1998), surge suppressors (Aceves et al., 1999), microsensors (Aceves et al., 2001), and
single electron devices (Yu et al., 2003).
SRO can be considered as a multi-phase material composed of a mixture of stoichiometric
silicon oxide (SiO
2
), off-stoichiometric oxide (SiO
x
, x<2) and elemental silicon. After thermal
treatment at temperatures above 1000C, off-stoichiometric oxide and silicon excess are
separated into Si nanoClusters (crystalline or amorphous depending on their size), defects
(oxidation states) and SiO
2
(Yu et al., 2006b). Transmission electron microscopy (TEM) studies

Cathodoluminescence

254
on the PECVD-SRO films with Si excess of 4-9% have confirmed the existence of Si
nanocrystals with sizes between 1-2 nm (Iacona et al., 2000). The observation of Si nanocrystals
in LPCVD-SRO films with Si excess of ~13% has also been reported (Irene et al., 1980).
Although no Si nanocrystals were observed in LPCVD-SRO with less Si excess (<6%) (DiMaria
et al., 1984), however, it was still proposed that amorphous Si nanodots exist in the SiO
2

matrix, which could not be distinguished by TEM (Nesbit, 1985). It was found that the growth
of the Si nanodots is a diffusion-controlled process with a diffusion coefficient as larger as
110
-16
cm
2
/s at 1100C (Nesbit, 1985); it is estimated that a diffusion length of 10-20 nm can be
obtained during the thermal annealing of 3 hours. This diffusion length is long enough for the
excess Si to precipitate into Si nanodots during the annealing process.
Depending of the silicon excess, SRO shows various properties, among others, variable
conductivity (Aceves et al., 1999), charge-trapping effect (Aceves, et al., 1996) and
photoluminescence (PL) (Shimizu-Iwayama et al., 1996). These properties depend on the
concentration and the exact nature of the excess Si. For example, the trapped charge density
and photoluminescence intensity of SRO increase with decreasing its Si excess. The light
emission is one promising property of SRO that makes it a good candidate for futures light-
emitting devices. It has been observed that strong visible light can be emitted from SRO
films deposited by LPCVD, its emission wavelength and intensity is tuned by the Si excess
and thermal annealing parameters (temperature and time). It has also been found that SRO
can trap both negative and positive charges, and the trapped charge density depends on the
same parameters.
Although this material has been extensively studied there is a considerable uncertainty
about the nature of the luminescence mechanisms. A clear understanding of the
luminescence mechanisms in SRO is the key for the production of silicon-based light-
emitting devices. Some authors attribute the luminescence to quantum confinement effects
(QCE) in Si-nCs (Si nanocrystals) (Chen et al., 2006; Iacona et al., 2000), to the emission of
several types of defects in the matrix or in the interface of SiO
2
/Si-nCs (Kenyon et al.1996;
Lockwood et al., 1996), including the decay of donor acceptor pairs, DAD (Lpez-Estopier et
al., 2011; Yu et al., 2006a).
Even though SRO has been strongly study by PL, there are only a few works that study the
cathodoluminescence (CL) of SRO. CL expands the possibilities to understand the emission
in SRO. For example point defects in silicon oxide and silicon could be observed trough CL,
allowing studying the distribution of such defects on the surface and inside the structure.
An advantage of this method is the possibility to observe optical transitions of electrons
from high energy levels (or band to band), which need to be exited with high energies, this
is very important for wide-gap materials.
CL and PL are similar techniques with some possible differences associated with the
excitation of the electron-hole pairs due to photons or more energetic electrons (Yacobi &
Holt, 1990). PL emission depends strongly on the excitation energy, and not all
luminescence mechanism could be excited. Cathodoluminescence, in general, leads to
emission by all the luminescence mechanism present in the solid due to the high energy
excitation.
In this chapter, different techniques, such as Fourier transform infrared spectroscopy (FTIR),
X-ray photoelectron spectroscopy (XPS), photoluminescence (PL) and cathodoluminescence

Cathodo- and Photo- Luminescence of Silicon Rich Oxide Films Obtained by LPCVD

255
(CL) were used to study SRO films obtained by LPCVD. In section 2, the experimental
details of the fabrication of SRO films and their characterization techniques used are
described. Section 3 gives the results of the structural and optical measurements. Also, in
this section, the discussion and analysis of the results are included. Finally, in Section 4,
conclusions are drawn. The structural and optical properties of SRO-LPCVD films were
related to better understand the process of light emission in these materials.
2. Experimental
SRO samples were deposited by Low Pressure Chemical Vapor Deposition (LPCVD) at 720
C on n-type Si substrates. The silicon excess in the SRO films was varied by adjusting the
ratio of partial pressure of the reactive gases, Silane (SiH
4
) and Nitrous Oxide (N
2
O).

2
4
[ ]
0
[ ]
10, 20 and 30
N O
SiH
P
R
P
(1)
With this R
0
variation a silicon excess of 11-5.5% is approximately obtained. The SRO layers
thickness is about 300 nm. After the deposition, one part of samples was keep as deposited
and other part was annealed at 1100 C in N
2
atmosphere during 180 minutes.
The PL emission spectra were obtained with a Fluoromax-3 Spectrofluorometer; all the films
were excited with 250 nm. Cathodoluminescence (CL) measurements were performed using
a Luminoscope equipment model ELM2-144, 0.3 mA current and energies of 2.5, 5, 10 and
15 keV were used. The luminescence spectra (PL and CL) were measured at the room
temperature. The IR spectra were measured with a Brucker FTIR spectrometer model V22,
which works in a range of 4000-350 cm
-1
with 5 cm
-1
resolution. The silicon excess in SRO
films was measured with a PHI ESCA5500 Xray photoelectron spectrometer (XPS) using a
monochromatic Al radiation source with energy of 1486 eV.
3. Results and discussion
3.1 Compositional and structural properties
The IR absorbance spectra of SRO samples are depicted in Fig. 1. Their characteristic
absorption bands are enumerated and identified in the Table 1. SRO films show the
absorption peaks associated with rocking at ~458 cm
-1
, bending at ~812 cm
-1
and stretching
at ~1080 cm
-1
vibration modes of the Si-O-Si bonds in SiO
2
(Ay & Aydinli, 2004; Pai et al.,
1986). In addition, the shoulder at ~1150 cm
-1
corresponds to the out-of-phase Si-O
stretching vibration (Pai et al., 1986). The insets of Fig. 1 show Si-N stretching vibration at
~886-990 cm
-1
(Ay & Aydinli, 2004; Daldosso et al., 2007); in addition Si-H bending and
stretching at ~886 and 2258 cm
-1
respectively (Iacona et al., 2000).
The as deposited samples exhibit a characteristic IR absorption at a lower frequency than that
of a thermal SiO
2
film (1080 cm
-1
), and the frequency of stretching vibration peak decreases as
R
0
decreases due to Si atoms replacing O atoms (Pai et al., 1986); this result is related to off-
stoichiometric of SRO, frequency of the Si-O stretching decreases with x of SiO
x
(silicon excess
increases). For all samples, the stretching frequency increases after annealing for all silicon
excess, this result suggest phase separation during the thermal annealing.

Cathodoluminescence

256

500 1000 1500 2000 2500 3000
0.00
0.02
0.04
0.06
0.08
0.10
0.12
0.14
0.16
0.18
0.20
900 1000 1100 1200 1300
0.00
0.03
0.06
0.09
0.12
0.15
0.18
2200 2250 2300
0.000
0.002
0.004
0.006
(6)
(5)
(3)
(2)
(a) As deposited
A
b
s
o
r
b
a
n
c
e

(
a
.

u
.
)
Wavenumber (cm
-1
)
SRO
30
SRO
20
SRO
10
(1)
500 1000 1500 2000 2500 3000 3500
SRO
30
SRO
20
SRO
10
(b) Annealed
Wavenumber (cm
-1
)
(1)
(2) (6)
(5)
900 1000 1100 1200 1300
0.00
0.03
0.06
0.09
0.12
0.15
0.18
0.21
A
b
s
o
r
b
a
n
c
e

(
a
.

u
.
)
(b)'
Wavelength number (cm
-1
)
(4) Si-N
(4) Si-N
(a)'
A
b
s
o
r
b
a
n
c
e

(
a
.

u
.
)
Wavenumber (cm
-1
)
A
b
s
o
r
b
a
n
c
e

(
a
.

u
.
)
(a)''
(7) Si-H
Wavenumber (cm
-1
)
2200 2250 2300
0.000
0.002
0.004
0.006
Wavenumber (cm
-1
)
A
b
s
o
r
b
a
n
c
e

(
a
.

u
.
)
(b)''

Fig. 1. FTIR absorption spectra of SRO films with R
0
= 10 (a) before and (b) after thermal
annealing. The inset shows Si-N and Si-H bonding in the 850 to 1300 cm
-1
and 2200 to 2300
cm
-1
range.

Vibration mode
Wavenumber [cm
-1
]
As deposited Annealed
SRO
30
SRO
20
SRO
10
SRO
30
SRO
20
SRO
10

(1) Si-O rocking 447 445 455 457 459 460
(2) Si-O bending 810 808 802 808 808 810
(3) Si-H bending 883 883 883 - - -
(4) Si-N stretching 980 984 940 984 991 954
(5) Si-O symmetric stretching 1060 1058 1055 1078 1078 1082
(6) Si-O asymmetric stretching 1176 1170 1159 1192 1190 1190
(7) Si-H stretching 2258 2258 2258 - - -
Table 1. Infrared vibrations modes observed in the as deposited samples.
It has been reported that SRO films deposited by the mixture of N
2
O and SiH
4
display
absorption bands associated with Si-H (660, 880, 2250 cm
-1
), Si-N (870-1030 cm
-1
), Si-OH
(940, 3660 cm
-1
) and N-H (1150, 1180, 3380 cm
-1
) vibrations in addition to the three
characteristic bands related to the Si-O-Si bonding arrangement (Fazio et al., 2005; Pai et al.,
1986). The presence of some nitrogen and hydrogen characteristic peaks was also observed
in the IR spectra of the films. Si-H bending and stretching were observed only in as
deposited samples; these peaks disappear after annealing. Si-N stretching was also found in
as deposited and annealing samples.


257

0 2 4 6 8 10 12 14 16 18 20 22 24
0
10
20
30
40
50
60
70
80
90
100
Si substrate
(a) SRO
30
Si in SiO
2
A
t
o
m
i
c

c
o
n
c
e
n
t
r
a
t
i
o
n

(
a
t
.

%
)
Sputter time (min)
As deposited
Annealed
O1s
Si2p
N1s
0 2 4 6 8 10 12 14 16 18 20 22 24
0
10
20
30
40
50
60
70
80
90
100
Si substrate
Si in SiO
2
(b) SRO
20
A
t
o
m
i
c

c
o
n
c
e
n
t
r
a
t
i
o
n

(
a
t
.

%
)
Sputter time (min)
As deposited
Annealed
O1s
Si2p
N1s

0 2 4 6 8 10 12 14 16 18 20 22 24
0
10
20
30
40
50
60
70
80
90
100
Si substrate
(c) SRO
10
As deposited
Annealed
Si in SiO
2
A
t
o
m
i
c

c
o
n
c
e
n
t
r
a
t
i
o
n

(
a
t
.

%
)
Sputter time (min)
O1s
Si2p
N1s

Fig. 2. Silicon, oxygen and nitrogen concentration in SRO films with different flow ratio (R
0
)
values as measured by XPS.
Fig. 2 indicates the composition measured by XPS of as deposited and annealed SRO films
with different flow ratio, R
0
= 30, 20 and 10. As it was expected, the silicon excess varies
depending on the flow ratio (R
0
). For SRO films with R
0
= 30, the silicon profile stays about
37.5 at.% and it increases until 47 at.% as R
0
value reduces to 10. A thin silicon dioxide layer
is formed at the SRO surface; which can be explained by the silicon oxidation at the end of
the deposition process. A nitrogen profile was also observed, however is negligible,
specially compared with that reported for SRO-PECVD films where the N incorporation is
about 10 at.% (Morales et al., 2007; Ribeiro et al., 2003). Table 2 summarizes the silicon excess
in the SRO films; silicon (Si), oxygen (O) and nitrogen (N) concentration in SRO is also
shown.

Cathodoluminescence

258
R
0

As deposited Annealed
Si excess
(at. %)
Concentration (%) SiOx Si excess
(at. %)
Concentration (%) SiOx
O Si N x=O/Si O Si N x =O/Si
30 4.54 60.90 37.87 1.23 1.61 4.21 61.15 37.54 1.31 1.63
20 5.38 59.67 38.71 1.62 1.54 5.09 59.29 38.42 2.29 1.54
10 13.50 50.89 46.83 2.28 1.09 14.50 48.09 47.83 4.07 1.01
Table 2. Silicon excess and atomic concentration of SRO films obtained by XPS.

98 100 102 104 106 108
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
As deposited
Annealed
I
n
t
e
n
s
i
t
y

(
a
.

u
.
)
Energy (eV)
(a) SRO
30
98 100 102 104 106 108
As deposited
Annealed
(b) SRO
20
Energy (eV)
98 100 102 104 106 108
As deposited
Annealed
(c) SRO
10
Energy (eV)

Fig. 3. Si 2p XPS peaks for as deposited and annealed samples with different flow ratio (R
0

30, 20 and 10).
The microstructure of SRO films was studied based on the analysis of the Si 2p spectra,
shown in Fig. 3. As can be seen, the effect of silicon excess and thermal treatment is evident.
After annealing and as silicon increases the shoulder in low energies becomes apparent,
indicating the contribution of different oxidation states according to the random bonding
model (RBM) (Chen et al., 2004, 2005).
The Si 2p XPS peaks were deconvoluted considering the five possible oxidation states for the
silicon: Si, Si
2
O, SiO, Si
2
O
3
, and SiO
2
, noted as Si
+0
, Si
+1
, Si
+2
, Si
+3
and Si
+4
respectively, shown
in Fig. 4. Each oxidation state has been fitted by using peaks constituted by Gaussians. The
energy positions of the different peaks of the Gaussians were centered with those previously
reported in the literature at ~99.8, 100.5, 101.5, 102.5 y 103.5 eV (Alfonsetti et al., 1993;
Bresch et al., 1993; Philipp, 1972; Wang et al., 2003; Yang et al., 2005). The full widths at half
maximum (FWHM) have been allowed to vary within a small range of values while
increasing in the order Si
0+
<Si
1+
<Si
2+
<Si
3+
<Si
4+
(Dehan et al., 1995; Liu et al., 2003). Si
0+

compounds were found in all samples in both before and after annealing due to silicon
excess.


259

100 102 104 106
0
2000
4000
6000
8000
10000
12000
14000
(a) SRO
30
, As deposited
Experimental
Fit
Si
n+
I
n
t
e
n
s
i
t
y

(
a
.

u
.
)
Energy (eV)
Ro=30, Annealed
100 102 104 106
Ro=20, Annealed
(b) SRO
20
, As deposited
Experimental
Fit
Si
n+
Energy (eV)
100 102 104 106
Energy (eV)
(c) SRO
10
, As deposited
Experimental
Fit
Si
n+
Ro=10, Annealed

Fig. 4. Illustration of Si 2p peak deconvolution for the SRO films with R
0
: (a) 30, (b) 20 and
(c) 10.
It would be useful to examine quantitatively the changes of the concentrations of the five
oxidation states with annealing. In this regard, as a first-order approximation, the relative
concentration (in percentage) of each oxidation state is obtained by calculating the ratio
n
Si Total
I I
(n = 0, 1, 2, 3, and 4), where

n
Si
I

is the peak area of the oxidation state Si
n+
and I
total

is the total area of the Si 2p peaks (
4
0
i
total
Si i
I I

) (Khriachtchev et al., 2002; Liu et al.,

2003).
Fig. 5 shows the effect of annealing in the concentrations of the five oxidation states for R
0

30, 20 and 10. After annealing concentration of Si
0+
, Si
1+
, Si
2+
increases while Si
3+
and Si
4+
compounds decrease in all silicon excess indicating the phase separation; this separation is
more evident in R
0
10. Then, when SRO films are thermally annealed and the silicon excess
is high, the silicon atoms diffuse around of a nucleation site and well defined silicon
nanocrystals are observed. However, for lower excess silicon films, the particles are far
enough and do not produced agglomeration, therefore the thermal diffusion causes the
silicon to be redistributed and form compounds with oxygen. These two mechanisms are
not mutually exclusive and both exist simultaneously, but depending on the silicon excess
will dominate a mechanism or the other. Because of this effect, concentration of Si
0+
is much
higher in SRO
10
than SRO
20
and SRO
30
; in addition the concentration of Si
1+
y Si
2+
is larger
for films with low silicon excess (SRO
20
and SRO
30
).
On the other hand, it is well known that in SRO with a large excess silicon (higher than
10%), Si-nCs of ~ 9 nm have been found by TEM (Transmission Electron Microscopy), as the
silicon excess decreases, the density and size of the nanocrystals decreases until there are no
Si-nCs (~ 5.5% silicon excess) (Yu et al., 2006b). Since the presence of Si
0+
compounds in

Cathodoluminescence

260
SRO
20,30
, agglomeration of amorphous compounds is more likely than elemental silicon
agglomeration (nCs), and its possible size is smaller than 2 nm, then they cannot be
observed by TEM.

Si
0+
Si
1+
Si
2+
Si
3+
Si
4+
0
10
20
30
40
50
60
70
80
90
100
As deposited
Annealed
C
o
n
c
e
n
t
r
a
t
i
o
n

(
%
)
Oxidation state
(a) SRO
30
Si
0+
Si
1+
Si
2+
Si
3+
Si
4+
As deposited
Annealed
(b) SRO
20
Oxidation state
Si
0+
Si
1+
Si
2+
Si
3+
Si
4+
As deposited
Annealed
(c) SRO
10
Oxidation state

Fig. 5. Changes in the concentration of the five Si oxidation states Si
n+
(n = 0, 1, 2, 3, and 4)
for different R
0
values.
3.2 Photoluminescence
PL spectra of as deposited and annealed films are depicted in Fig. 6. The as deposited
samples present a weak emission only in SRO
30
from ~460 to 650 nm. After annealing all
SRO films (R
0
= 10, 20 and 30) present a main emission from ~650 to 850 nm. This emission
range is characteristic of SRO annealed films deposited by LPCVD (Lpez-Estopier et al.,
2011; Yu et al., 2006a).The emission intensity increases when the silicon excess decrease.
The multi-Gaussian deconvolution of PL spectra was performed only for annealed samples,
and the set of band positions have been determined. As can be seen in Fig. 7, each spectrum
can be well fitted to a superposition of three Gaussian distributions: a main band and two
shoulders. Fit distributions are centered at 715, 780, 825 nm with FWHM of 101, 80, 47 nm
respectively. Distribution position and FWHM are the same for all R
0
values while intensity
varies according to the silicon excess. As silicon excess decreases the intensity of main
distribution increase; however, distributions centered at 780 and 825 nm are more intense to
SRO
20
than SRO
30
; because of this PL emission of SRO
20
shows a shoulder in this region.
Distribution centered at 715 nm is the highest emission (R
0
= 20 and 30), hence this
distribution is the main contribution of the PL emission. Since there are different
distributions that change with silicon excess can be assumed that PL emission is related with
at least three different types of emission centers (or emission mechanism).


261

400 500 600 700 800
0
100
200
300
400
500
600
700
800
400 500 600 700 800 900
Wavelength (nm)
SRO
30
SRO
20
SRO
10
(a) As deposited
P
L

i
n
t
e
n
s
i
t
y

(
a
.

u
.
)
SRO
30
SRO
20
SRO
10

(b) Annealed
Wavelength (nm)

Fig. 6. Photoluminescence of (a) as deposited and (b) annealed SRO films with different ratio
flows.

600 700 800
0
100
200
300
400
500
600
700
800
900
(a) SRO
30
P
L

i
n
t
e
n
s
i
t
y

(
a
.

u
.
)
Wavelength (nm)
600 700 800
Experimental
Gaussian Fit
Distributions
(b) SRO
20

Wavelength (nm)
600 700 800
(c) SRO
10

Wavelength (nm)
x5

Fig. 7. PL spectra and fits from SRO films with different silicon excess and annealed at 1100
C for 180 min. Symbols are experimental data, lines are the Gaussian fits and dash lines are
distributions.
3.3 Cathodoluminescence
CL spectra from SRO films (as deposited and annealed) with different silicon excess are
depicted in Fig. 8. In as deposited samples there is CL emission only in SRO
30
. The CL
spectra of SRO with thermal treatment consist of a broad emission in the visible and near
infrared region (NIR) from ~400 to 850 nm. After annealing, intensity of the blue band at
~460 nm increases with increasing the R
0
. On the other hand, the intensity of the red-NIR CL
band seems to have a maximum for R
0
=20.

Cathodoluminescence

262

300 400 500 600 700 800
0
500
1000
1500
2000
2500
3000
3500
4000
400 500 600 700 800 900
SRO
30
SRO
20
SRO
10
(a) As deposited
C
L

I
n
t
e
n
s
i
t
y

(
a
.

u
.
)
Wavelength (nm)
SRO
30
SRO
20
SRO
10
(b) Annealed
Wavelength (nm)

Fig. 8. Cathodoluminescence spectra from SRO films with different silicon excess: (a) as
deposited and (b) annealed. Energy excitation of 5 keV and 0.3 mA current was used.
Since CL emission has asymmetrical shape for all SRO samples, it can be assumed that CL
emission is due to different causes. Hence, multi-Gaussian deconvolution of CL spectra
was also obtained, shown in Fig. 9. The best fit of CL spectra requires 4 and 6 distributions
for SRO
30
and SRO
20
respectively. Distributions were obtained at about 460, 522, 643, and
714 nm for SRO
30
and 447, 541, 645, 714, 780 and 823 nm for SRO
20
. Distributions obtained
at 714, 780 and 823 nm in CL are centered in the same position than distributions obtained
from PL spectrum in SRO
20
. Furthermore, distribution centered at 714 nm was obtained
for Gaussian fit, in PL and CL in SRO
30
. Then, the red emission of the CL emission can be
ascribed to the same PL emissive centers. PL distributions in higher wavelength are not
observed in CL due to either destruction of the emissive centers or emissive center with
lower energies do not efficiently emit (Trukhin et al., 1999). The latter one could rise
because cathode excited electrons acquire so high energy that they arrive to the higher
emissive center where they emit in the blue region (higher energy), however almost none
of the excited electrons reach lower energy centers; then, the red emission is not likely to
occur. Therefore, there could be several different kinds of emission centers located at
different energy levels in SRO.
Depending on the emission wavelength, multiple luminescence centers have been
reported to act as radiative recombination centers in SiO
2
films. Luminescent emission at
460 nm (2.7 eV), 520 nm (2.4 eV) and 650 nm (1.9 eV) are mainly related to defects such as
Oxygen deficiency-related centers (ODC) or oxygen vacancies (Cervera et al., 2006;
Fitting, 2009; Gritsenko et al., 1999), E defect or peroxide radical (Goldberg et al., 1997)
and non-bridging oxygen hole centers (NBOHC) (Fitting, 2009; Fitting et al., 2001;
Gritsenko et al., 1999), respectively. Since CL measurements have shown luminescent
peaks (or distributions) close to those wavelengths, such defects could be inside the SRO
films.


263

400 500 600 700 800
0
500
1000
1500
2000
2500
3000
3500
4000
C
L

i
n
t
e
n
s
i
t
y

(
a
.

u
.
)
Wavelength (nm)
(a) SRO
30
400 500 600 700 800
(b) SRO
20
Experimental
Gaussian fit
Distributions

Wavelength (nm)

Fig. 9. Gaussian fit of CL experimental spectrum, the best fit requires 4 distributions for (a)
SRO
30
and 6 distributions for (b) SRO
20
. Symbols are experimental data, lines are the
Gaussian fits and dash lines are distributions.

400 500 600 700 800 900
0
500
1000
1500
2000
2500
3000
3500
4000
SRO
30
SRO
20
SRO
10
Thermal SiO
2
C
L

i
n
t
e
n
s
i
t
y

(
a
.

u
.
)

Wavelength (nm)
(a)

300 400 500 600 700 800 900
0
100
200
300
400
500
600
Experimental
Gaussian Fit
Distributions
C
L

i
n
t
e
n
s
i
t
y

(
a
.

u
.
)
Wavelength (nm)
(b) Thermal SiO
2

Fig. 10. (a) CL spectra of thermal SiO
2
and SRO annealed samples and (b) Gaussian fit of
thermal SiO
2
sample.
In order to compare CL spectra between SiO
2
and SRO, CL spectrum from thermal SiO
2
was
obtained, as shown in Fig. 10 (a). CL of SiO
2
results are helpful because a large fraction of
the SRO films consists of silicon oxide. As can be seen, spectrum from thermal SiO
2
is
similar to SRO
20
but it does not include the red to infrared emission. To compare emission
bands, Gaussian deconvolution was also obtained from thermal oxide; the best fit was
centered at 432, 520 and 653 nm, as shown in Fig. 10 (b). Clearly only the 650 nm peak in
both SRO
20
and SiO
2
are comparable. The other emissions are enhanced in both Ro 20 and
30. That could mean that emission in SRO with low silicon excess and without silicon nCs
are mainly due to silicon oxidation states.

Cathodoluminescence

264
In summary, luminescence peaks at 460, 520, and 620 nm are mainly due to defects such as
Oxygen deficiency-related centers (ODC), E defect, and non-bridging oxygen hole centers
(NBOHC), respectively (Cervera et al., 2006; Fitting et al., 2001; Inokuma et al., 1998).
Electroluminescence (EL) studies on SRO films with the same silicon excess as in this work
exhibit an emission band similar to our CL results with peaks at 450, 500, 550, and 640 nm
(Morales-Snchez et al., 2010).Then the mentioned defects should exist into the SRO films,
which are only excited with electron of high energy generated by CL or by EL.
The emission in red and NIR region could be associated to Si-clusters of less of 2 nm and
defects interaction. In (Morales-Snchez et al., 2008), the authors proposed that the band
gap is large when the size of the Si cluster is small, and then the energy difference between
the defects (localized state) and the Si clusters is big enough to produce the emission. This
result is similar to the decay of Acceptor-Donator pair in a crystalline semiconductor.
Therefore, we can assume that the absorption and emission processes in SRO films are
connected with electrons decay between donor acceptor pairs (localized states) inside the
silicon oxide band gap.
3.4 Depth profiling by different electron beam energy
It is well known that depth analysis can be obtained varying the electron energy (excitation
range of CL) (Goldberg et al., 1998; Wittry & Kyser, 1967). In this case the CL intensity
versus energy would remain constant when assuming a homogeneous depth distribution of
luminescence centers. Deviation of such a behavior is easy to detect and should be
interpreted.
The depth-resolved CL spectra for SRO annealed samples measured with electron beam
energies of 2.5, 5, 10 and 15 keV was obtained, shown in Fig. 11. As can be seen, there is
no band shift or any new emission bands by varying the beam energy, indicating specific
luminescence correlated with a homogeneous depth impurity distribution in the
samples.
400 500 600 700 800
0
500
1000
1500
2000
2500
3000
3500
4000
4500
(a) SRO
30
C
L

i
n
t
e
n
s
i
t
y

(
a
.

u
.
)
Wavelength (nm)
2.5 keV
5 keV
10 keV
15 keV
400 500 600 700 800
(b) SRO
20
Wavelength (nm)
400 500 600 700 800
(c) SRO
10

Wavelength (nm)

Fig. 11. Cathodoluminescence spectra of SRO annealed films at various electron energies.
The spectra were measured at the room temperature.


265
3.5 Effect of the electron excitation on the emissive centers
In order to study the effect of the cathode excitation on the emissive centers, PL spectra were
obtained before and after CL measurements with different energy electron doses;
additionally, the effect of time excitation on CL spectra was obtained.
Different SRO
20
and SRO
30
annealed samples at 1100C in N
2
by 180 minutes were used. One
sample was not cathode-excitated, and it was used as reference. CL measurements were
performed applying energy of 2.5, 5, 10 and 15 keV in each sample, respectively. In all cases
the current used in the experiment was fixed at 0.3 mA.

Fig. 12. CL spectra of SRO
20
after different times of electron bombardment. Different
samples were used: (a) sample excited with 2.5 keV, (b) sample excited with 5 keV, (c)
sample excited with 10 keV, (d) sample excited with 15 keV.
CL spectra of SRO
20
samples after different times of electron bombardment and different
energy are shown in Fig. 12. CL spectrum consists of a broad emission in the visible and
near infrared region (NIR) from ~400 to 850 nm. The different excitation energies produce
both characteristic bands of SRO
20
(red and blue). It can be observe in all the samples and
independent of excitation energy that the blue emission does not have a considerable change
while red band rapidly decrease or quenched with cathodo-excitation time. Therefore the
excitation with high energy electrons destroys or modifies the red emissive states, inclusive
with low energy (2.5 keV). Using electron energies greater than or equal to 10 keV, red band
practically disappears after 5 minutes of bombardment.

Cathodoluminescence

266
PL spectra obtained after constant electron beam excitation during 30 minutes continually
are shown in Fig. 13. After apply excitation energy of 2.5, 5 and 10 keV, the PL emission
decreases considerately (50%), and with 15 keV PL emission is reduced 80%. This indicates
that electron beam excitation modifies the structure of SRO due to the high energies used,
and emission mechanism or emission centers were destroyed. Also it can be observed that
after high energy excitation there is no new emission band.
400 500 600 700 800
0
100000
200000
300000
400000
500000
P
L

i
n
t
e
n
s
i
t
y

(
a
.

u
.
)
Wavelengh (nm)
Reference
2.5 keV
5 keV
10 keV
15 keV
SRO
20
samples annealed at 1100C by 180 min.

Fig. 13. PL spectra of SRO
20
samples after continuous cathode-excitation during 30 minutes.
Solid line is the reference sample emission (without electron beam irradiation). The symbols
represent the PL emission after cathodo excitation.
The evolution of CL spectra for SRO
30
annealed samples after 10minutes of constant electron
bombardment is shown in Fig. 14 (a), (b) and (c) with 5, 10 and 15 keV respectively. There is
no CL emission when these samples were excited with 2.5 keV. Samples excited with higher
energies (more than 5 keV) exhibit strong blue emission with a shoulder in higher
wavelength, characteristic emission of SRO
30
. Blue emission centered at ~460 nm increases
after 10 minutes of constant cathode excitation. The PL spectra of these samples were
obtained after constant electron beam excitation during 10 minutes, and shown in Fig. 15.
Similar to SRO
20
, after cathode-excitation the PL emission decreases considerately (~50%),
and as energy electron increase the PL emission reduces.
As mentioned before, blue emission centered at ~460 nm has been assigned to oxygen
deficiency. Also, it has been proposed that high energy excitation and long exposition time
are responsible of this band in SiO
2
(Chen et al., 1999; Goldberg et al., 1997; Inokuma et al.,
1998). Since blue emission increases slightly after constant electron excitation in SRO, it is
possible that oxygen deficiencies are created by excitation. However, we can assume that the
emission in the blue region is due to the non-stoichiometry of the SRO because blue
emission is not observed in PL neither before nor after of cathode excitation. Besides, blue
emission has also been found in thermoluminescence and electroluminescence in SRO
deposited by LPCVD (Morales-Snchez et al., 2010; Piters et al., 2010). Therefore, the
emission is due to band to band recombination through localized states in the SRO.


267
Degradation of the intensity in red region could be due to the passivation of luminescent
centers. It is possible that the electron beam introduces changes around the emission center
that affect the emission rate of emission of radiative and non radiative transitions.

400 500 600 700 800
0
100000
200000
300000
400000
500000
600000
C
L

i
n
t
e
n
s
i
t
y

(
a
.

u
.
)
Wavelength (nm)
Initial
10 minutes
(a) Excitation energy 5 keV
400 500 600 700 800
0
100000
200000
300000
400000
500000
600000
(b) Excitation energy 10 keV
C
L

i
n
t
e
n
s
i
t
y

(
a
.

u
.
)
Wavelength (nm)
Initial
10 minutes

400 500 600 700 800
0
100000
200000
300000
400000
500000
600000
(c) Excitation energy 15 kV
C
L

i
n
t
e
n
s
i
t
y

(
a
.

u
.
)
Wavelength (nm)
Initial
10 minutes

Fig. 14. CL spectra of SRO
30
annealed samples after different times of electron
bombardment. Different samples were used: (a) sample excited with 5 keV, (b) sample
excited with 10 keV, (c) sample excited with 15 keV.

Cathodoluminescence

268
500 600 700 800 900
0
10000
20000
30000
40000
50000
60000
SRO
30
annealed samples
P
L

i
n
t
e
n
s
i
t
y

(
a
.

u
.
)
Wavelength (nm)
Reference
2.5 kV
5 kV
10 kV
15 kV

Fig. 15. PL spectra of SRO
30
after cathode excitation at different beam energies. A sample
without electron beam irradiation was used as reference (solid line).
4. Conclusion
SRO films with different silicon excess and with and without thermal treatment were studied
using CL and PL. In order to understand their emission FTIR and XPS were used to study the
structural composition and the effect of thermal treatment. FTIR and XPS results show the
formation of si nanoaglomerates in all samples. Analysis of the XPS Si 2p peaks shows the
existence of some chemical structures corresponding to the Si oxidation states in the SRO films;
these oxidation states depend on the silicon excess. The concentration of each oxidation state
was estimated. According to variation of silicon compounds found in analysis of the Si 2p
spectra, we can assume that as silicon excess decrease Si-nCs density decreases and defects (or
silicon oxygen compounds) tend to predominate over Si nanoclusters or nanocrystals. In fact,
when the Si nanoclusters are very small, the interface will play important role due to the large
stress in the interface of the nanoclusters and the localized states will be form.
Photo and Cathodoluminescence (PL and CL) properties in Silicon Rich Oxide (SRO) films
with different silicon excess were studied. SRO deposited by LPCVD has shown high visible
PL and CL emission at room temperature mainly after thermal treatment at high
temperatures. Samples with lower silicon excess show maximum luminescence while
samples with higher silicon excess show low luminescence. Strong emission from 650 to 800
nm was found in PL while CL spectra show luminescence emission from 400 to 850 nm in
SRO annealed samples. CL spectra are wider than the PL one, because high energy
excitation leads the emission of all the emissive centers present in the SRO. Also was found
that red and NIR region emission is affected by electron beam.
Luminescent emission in blue-green region is mainly due to defect created by silicon excess,
such as oxygen deficiency-related centers (ODC), E defect, and non-bridging oxygen hole
centers (NBOHC). Emission in red and near infrared region is associated to some defects
that are acting as localized states and also to Si nanoaglomerates and defects interaction.


269
5. Acknowledgment
The authors express their thanks to the microelectronic laboratory technicians of INAOE for
their support and to the CONACYT for providing financial support for this work.
6. References
Aceves, M., Falcony, C., Reynoso-Hernandez, A., C. W., & Torres, A. (1996). The conduction
properties of the silicon/off-stoichiometry-SiO
2
diode. Solid-State Electronics, Vol. 39
No. 5, (May 1996 ), pp. 637-644, ISSN 00381101.
Aceves, M., Malik, A., & Murphy, R. (2001). The FTO/SRO/SI structure as a radiation
sensor, In: Sensors & Chemometrics, M. T. Ramrez-Silva, M. A. Romero Romo, & M.
E. Palomar Pardav (Ed.), 1-25, Research Signpost, ISBN 81-7736-067-1, India.
Aceves, M., Pedraza, J., Reynoso-Hernandez, J. A., Falcony, C., & Calleja, W. (1999). Study
on the Al/silicon rich oxide/Si structure as a surge suppressor, DC, frequency
response and modeling. Microelectronics Journal,Vol. 30, No. 9, (September 1999),
pp. 855-862, ISSN 00262692.
Alfonsetti, R., Lozzi, L., Passacantando, M., Picozzi, P., & Santucci, S. (1993). XPS studies on
SiOx thin films. Applied Surface Science, Vol. 70-71, No. part 1,(June 1993), pp. 222-
225, ISSN 01694332.
Ay, F., & Aydinli, A. (2004). Comparative investigation of hydrogen bonding in silicon
based PECVD grown dielectrics for optical waveguides. Optical Materials, Vol. 26
No. 1, (June 2004), pp. 33-46, ISSN 09253467.
Bresch, P., Stockmeier, T., Stucki, F., & Buffat, P. A. (1993). Physical properties of semi-
insulating polycrystalline silicon. I. structure, electronic properties, and electrical
conductivity. Journal of Applied Physics, Vol. 73, No. 11, (n. d.), pp. 7677-7689, ISSN
00218979.
Calleja, W., Aceves, M., & Falcony, C. (1998). EEPROM transistor fabricated with stacked
SiOx LPCVD films. Electronics Letters, Vol. 34, No. 13, (June 1998), pp. 1294-1296,
ISSN 00135194.
Cervera, M., Hernndez, M. J., Rodrguez, P., Piqueras, J., Avella, M., Gonzlez, M. A., &
Jimenez, J. (2006). Blue-cathodoluminescent layers synthesis by high-dose N
+
, C
+

and B
+
SiO
2
implantation. Journal of Luminescence, Vol. 117, No.1, (March 2006), pp.
95-100, ISSN 00222313.
Chen, C. M., Pan, H. C., Zhu, D. Z., Hu, J., & Li, M. Q. (1999). Cathodoluminescence and ion
beam analysis of ion-implanted combinatorial materials libraries on thermally
grown SiO
2
. Nuclear Instruments and Methods in Physics Research. Vol. 159 No.1-2,
pp. 81-88, ISSN 0168583X.
Chen, X. Y., Lu, Y. F., Tang, L. J., Wu, Y. H., Cho, B. J., Xu, X. J., Dong, J. R & Song, W. D.
(2005). Annealing and oxidation of silicon oxide films prepared by plasma-
enhanced chemical vapor deposition. Journal of Applied Physics, Vol. 97 No. 1,
(January 2005), pp. 014913-1-014913-1, ISSN 00218979.
Chen, X. Y., Lu, Y. F., Wu, Y. H., Cho, B. J., Song, W. D., & Dai, D. Y. (2004). Optical
properties of SiOx nanostructured films by pulsed-laser deposition at different
substrate temperatures. Journal of Applied Physics, Vol. 96, No. 6, (September 2004),
pp. 3180-3186, ISSN 00218979.

Cathodoluminescence

270
Chen, X. Y., Lu, Y. F., Wu, Y. H., Cho, B. J., Tang, L. J., Lu, D., & Dong, J. R. (2006).
Correlation between optical properties and si nanocrystal formation of si-rich si
oxide films prepared by plasma-enhanced chemical vapor deposition. Applied
Surface Science, Vol. 253, No. 5, (December 2006), pp. 2718-2726, ISSN 01694332.
Daldosso, N., Das, G., Larcheri, S., Mariotto, G., Dalba, G., Pavesi, L., Irrera, A., Priolo, F.,
Iacona, F. & Rocca, F. (2007). Silicon nanocrystal formation in annealed silicon-rich
silicon oxide films prepared by plasma enhanced chemical vapor deposition.
Journal of Applied Physics, Vol. 101, No. 11, (n. d.), ISSN 00218979.
Dehan, E., Temple-Boyer, P., Henda, R., Pedroviejo, J. J., & Scheid, E. (1995). Optical and
structural properties of SiOx and SiNx materials. Thin Solid Films, Vol. 266, No. 1,
(September 1995), pp. 14-19, ISSN 00406090.
DiMaria, D. J., Kirtley, J. R., Pakulis, E. J., Dong, D. W., Kuan, T. S., Pesavento, F. L., Theis,
T., & Cutro, J. A. (1984). Electroluminescence studies in silicon dioxide films
containing tiny silicon islands. Journal of Applied Physics, Vol. 56, No. 2, pp. 401-416,
ISSN 00218979.
Dong, D., Irene, E. A., & Young, D. R. (1978). Preparation and some properties of chemically
vapor-deposited Si-rich SiO
2
and Si
3
N
4
films. Journal of the Electrochemical Society,
Vol. 125, No. 5, (May 1978), pp. 819-823, ISSN 00134651.
Fazio, E., Barletta, E., Barreca, F., Neri, F., & Trusso, S. (2005). Investigation of a
nanocrystalline silicon phase embedded in SiOx thin films grown by pulsed laser
deposition. Journal of Vacuum Science and Technology B, Vol. 23, No. 2, (n. d.), pp.
519-524, ISSN 10711023.
Fitting, H. J. (2009). Can we make silica luminescent?. Optical Materials, Vol. 31, No. 12,
(October 2009), pp. 1891-1893, ISSN 09253467.
Fitting, H. J., Barfels, T., & Trukhin, A. N. (2001). Cathodoluminescence of crystalline and
amorphous SiO
2
and GeO
2
. Journal of Non-Crystalline Solids, Vol. 279, No. 1,
(January 2001), pp. 51-59, ISSN 00223093.
Goldberg, M., Barfels, T., & Fitting, H. J. (1998). Cathodoluminescence depth analysis in
SiO
2
-Si-systems. Fresenius' Journal of Analytical Chemistry, Vol. 361 No. 6-7, (n. d.),
pp. 560-561, ISSN 09370633.
Goldberg, M., Fitting, H. J., & Trukhin, A. (1997). Cathodoluminescence and
cathodoelectroluminescence of amorphous SiO
2
films. Journal of Non-Crystalline
Solids, Vol. 220, No. 1, (October 1997), pp. 69-77, ISSN 00223093.
Gritsenko, V. A., Shavalgin, Y. G., Pundur, P. A., Wong, H., & Lau, W. M. (1999).
Cathodoluminescence and photoluminescence of amorphous silicon oxynitride.
Microelectronics Reliability, Vol. 39, No. 5, (May 1999), pp. 715-718, ISSN 00262714.
Hanaizumi, O., Ono, K., & Ogawa, Y. (2003). Blue-light emission from sputtered Si:SiO
2

films without annealing. Applied Physics Letter, Vol. 82, No. 4, (January 2003), pp.
538-540, ISSN 00036951.
Iacona, F., Franzo, G., & Spinella, C. (2000). Correlation between luminescence and
structural properties of Si nanocrystals. Journal of Applied Physics, Vol. 87, No. 3,
(February 2000), pp. 12951303, ISSN 00218979.
Inokuma, T., Kurata, Y., & Hasegawa, S. (1998). Cathodoluminescence properties of silicon
nanocrystallites embedded in silicon oxide thin films. Journal of Luminescence, Vol.
80, No. 1-4, (December 1998), pp. 247-251, ISSN 00222313.


271
Irene, E. A., Chou, N. J., Dong, D. W., & Tierney, E. (1980). On the Nature of CVD Si-Rich
SiO
2
and Si
3
N
4
Films. J. Electrochem. Soc., Vol. 127, No. 11, pp. 2518-2521.
Kenyon, A. J., Trwoga, P. F., Pitt, C. W., & Rehm, G. (1996). The origin of photoluminescence
from thin films of silicon-rich silica. Journal of Applied Physics, Vol. 79, No. 12, (June
1996), pp. 9291-9300, ISSN 00218979.
Khriachtchev, L., Novikov, S., & Lahtinen, J. (2002). Thermal annealing of Si/SiO
2
materials:
Modification of structural and photoluminescence emission properties. Journal of
Applied Physics, Vol. 92, No. 10, (November 2002), pp. 5856-5862, ISSN 00218979.
Liu, Y., Chen, T. P., Fu, Y. Q., Tse, M. S., Hsieh, J. H., Ho, P. F., & Liu Y. C. (2003). A study on
si nanocrystal formation in Si-implanted SiO
2
films by x-ray photoelectron
spectroscopy. Journal of Physics D, Vol. 36, No. 19, (October 2003), pp. L97-L100,
ISSN 00223727.
Lockwood, D. J., Lu, Z. H., & Baribeau, J. (1996). Quantum confined luminescence in Si/SiO
2

superlattices. Physical Review Letters, Vol. 76, No. 3, (n. d.), pp. 539-541, ISSN
00319007.
Lpez-Estopier, R., Aceves-Mijares, M., Yu, Z., & Falcony, C. (2011). Determination of the
energy states of the donor acceptor decay emission in silicon rich oxide prepared
by low-pressure chemical vapor deposition. Journal of Vacuum Science and
Technology B: Microelectronics and Nanometer Structures, Vol. 29, No. 2, (March 2011),
ISSN 10711023.
Morales, A., Barreto, J., Domnguez, C., Riera, M., Aceves, M., & Carrillo, J. (2007).
Comparative study between silicon-rich oxide films obtained by LPCVD and
PECVD. Physica E: Low-Dimensional Systems and Nanostructures, Vol. 38, No. 1-2,
(April 2007), pp. 54-58, ISSN 13869477.
Morales-Snchez, A., Aceves-Mijares, M., Gonzlez-Fernndez, A. A., Monfil-Leyva, K.,
Juvert, J., & Domnguez-Horna, C. (2010). Blue and red electroluminescence of
silicon-rich oxide light emitting capacitors. Proceedings of SPIE - the International
Society for Optical Engineering, Vol. 7719, ISBN 978-081948192-4, Brussels, Belgium,
April 12, 2010.
Morales-Snchez, A., Barreto, J., Domnguez-Horna, C., Aceves-Mijares, M., & Luna-Lpez,
J. A. (2008). Optical characterization of silicon rich oxide films. Sensors and
Actuators, A: Physical, Vol. 142, No. 1, (March 2008), pp. 12-18, ISSN 09244247.
Nesbit, L. A. (1985). Annealing characteristics of si-rich SiO
2
films. Applied Physics Letters,
Vol. 46, No. 1, (n. d.), pp. 38-40, ISSN 00036951.
Pai, P. G., Chao, S. S., Takagi, Y., & Lucovsky, G. (1986). Infrared spectroscopic study of
SiOx films produced by plasma enhanced chemical vapor deposition. Journal of
Vacuum Science & Technology A: Vacuum, Surfaces, and Films, Vol. 4, No. 3, pp. 689-
694.
Pavesi, L., Dal Negro, L., Mazzoleni, C., & Franzo, G. (2000). Optical gain in silicon
nanocrystals. Nature, Vol. 408, No. 6811, (November 2000), pp. 440-444, ISSN
00280836.
Philipp, H. R. (1972). Optical and bonding model for non-crystalline SiOx and SiOxNy
materials. Journal of Non-Crystalline Solids, Vol. 8-10, No.C, (June 1972), pp. 627-632,
ISSN 00223093.

Cathodoluminescence

272
Piters, T., Aceves-Mijares, M., D., B. M., Berriel-Valdos, L., & Luna-Lpez, J. A. (2010). Dose
dependent shift of the TL glow peak in a silicon rich oxide (SRO) film. Revista
mexicana de fisica, Vol. S 57, No. 2, pp. 26-29.
Ribeiro, M., Pereyra, I., & Alayo, M. (2003). Silicon rich silicon oxynitride films for
photoluminescence applications. Thin Solid Films, Vol. 426, No. 1-2, (February 2003),
pp. 200-204, ISSN 00406090.
Shimizu-Iwayama, T., Yoichi, T., Atsushi, K., Motonori, K., Setsuo, N., & Kasuo, N. (1996).
Visible photoluminiscence from nanocrystal formed in silicon dioxide by ion
implantation and thermal processing. Thin Solid Films, Vol. 276, No. 1-2, (April
1996), pp. 104-107, ISSN 00406090.
Trukhin, A. N., Fitting, H. J., Barfels, T., & Czarnowski, V. (1999). Cathodoluminescence and
IR absorption of oxygen deficient silica - influence of hydrogen treatment. Journal of
Non-Crystalline Solids, Vol. 260, No. 1-2, (December 1999), pp. 132-140, ISSN
00223093.
Wang, Y. Q., Chen, W. D., Liao, X. B., & Cao, Z. X. (2003). Amorphous silicon nanoparticles
in compound films grown on cold substrates for high-efficiency
photoluminescence. Nanotechnology, Vol. 14, No. 11, (November 2003), pp. 1235-
1238, ISSN 09574484.
Wittry, D. B., & Kyser, D. F. (1967). Measurement of diffusion lengths in direct-gap
semiconductors by electron-beam excitation. Journal of Applied Physics, Vol. 38, No.
1, (n. d.), pp. 375-382, ISSN 00218979.
Yacobi, B. G., & Holt, D. B. (1990). Cathodoluminescence Microscopy of inorganic solids,
Springer, ISBN 0-306-43314-1, New York:, USA.
Yang, D. Q., Gillet, J. N., Meunier, M., & Sacher, E. (2005). Room temperature oxidation
kinetics of si nanoparticles in air, determined by x-ray photoelectron spectroscopy.
Journal of Applied Physics, Vol. 97, No. 2, (January 2005), pp. 024303-1-024303-6, ISSN
00218979.
Yu, Z., Aceves, M., Carrillo, J., & Flores, F. (2003). Single electron charging in si nanocrystals
embedded in silicon-rich oxide. Nanotechnology, Vol. 14, No. 9, (September 2003),
pp. 959-964, ISSN 09574484.
Yu, Z., Aceves-Mijares, M., Luna-Lpez, A., Du, J., & Bian, D. (2006b). Formation of silicon
nanoislands on crystalline silicon substrates by thermal annealing of silicon rich
oxide deposited by low pressure chemical vapour deposition. Nanotechnology, Vol.
17, No. 1, (October 2006), pp. 4962-4965, ISSN 09574484.
Yu, Z., Mariano, A., Luna-Lpez, A., Quiroga, E., & Lpez-Estopier, R. (2006a).
Photoluminescence and single electron effect of nanosized silicon materials, In
Focus on Nanomaterials Research, B. M. Caruta (Ed.), pp. 233-273, Nova Science
Publishers, ISBN 1-59454-897-8, New York, USA.
11
Textural Characterization of Sedimentary Zircon
and Its Implication
Yixiong Qian
1
, Wen Su
2
, Zhong Li
2
and Shoutao Peng
1
1
Research Institute of Exploration & Production, Sinopec, Beijing
2
Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing,
China
1. Introduction
Zircon with highly refractory, as accessory, detrital minerals occurs in virtually all
sediments and sedimentary rocks, has played a prominent and complex role in interpreting
the history of sediments and source history of a deposit, paleogeography, and tectonic
reconstructions (Qian Y.X.,et al.,2007a). The main use of CL imaging of Detrital zircon has
been as an adjunct to U-Pb dating of zircons, which allows identification of different types
of zircon domains ,then may be dated in situ within a spatial resolution of about 15~30m.
By using a SHRIMP(Sensitive High-Resolution Ion Microprobe) , this data analysis would
be a powerful tool in understanding origin of the zircon , that is , the various geological
processes(magmatism, deposition , metamorphism, hydrothermal alteration, metasomatic
leaching of Th, U, and Pb) , their surrounding source terrane of known age may be traced
and a geological evolution history of sedimentary basins may be established. In fact, the
analyzed sample wouldnt completely represent geological history by including evidence of
all the possible provenances and their relationships to each other due to natural complexity
of sample. It was mentioned that the analysis techniques involve the sampling protocol and
interpretation of data and then show the application of detrital zircon studies to: (1) analysis
of origin characteristics such as U-Pb isotopic analysis and composition, (2) determine on
the using age of stratigraphic successions and to help recognize time gaps in the geologic
records, (3) test regional paleogeographic reconstructions by origin analysis, and (4) reveal
geological history relative with the mineral chemistry of detrital zircon, and reveal
complexity in order to gain insight into natural processes.
2. Analysis of origin characteristics such as U-Pb isotopic data and
composition
The distribution of heavy minerals in sediments is affected by the following factors:
provenance, uplift and erosion, paleo-topographic feature, palaeo-climate and palaeo-
environment. The analysis of heavy minerals can be applied in explanation of the sedimentary
response to tectonic cycles based on the provenance and sedimentary environment.
The heavy minerals assemblage can be act as indicators of provenance. The assemblage of
zircon, tourmalie, apatite and little biotite with the well-development crystal shapes is

Cathodoluminescence

274
generally believed to come from granitic rocks; the assemblage of a great quantity of Garnet,
zircon, epidote, chlorite is derived from metamorphic rocks; while the assemblage of
magnetite, ilmentite, anatase, augite and hornblende is commonly be readily traced to balt-
igneous rocks. In addition, the characteristic of heavy minerals are not only indictor of the
composite of provenance, but also of physical-selection, mechanical abrasion of granular
and chemical dissolution of sedimentary rocks during the process of sedimentary and
transportation. Based on the primary analysis of sandstones with regard to sedimentary
environment and provenance, five heavy minerals assemblages can be reasonable classified
as follows: stable minerals assemblage: includes the most of Ti oxide , zircon and
Tourmalie; the relative stable minerals assemblage: predominantly composes of garnet
and apatite; un-stable minerals assemblage: be dominated by epidoteaugite and
Hornblende; the assemblage of the indication of initial depositional environment :be
associated with hematite, pyrite, glauconite, barite, and carbonate minerals; ore minerals
present in hydrothermal mineralization. In addition, the index of ZTR, which shows the
percent of zircon, toumalie, rutile, and oblique carbonate minerals in heavy minerals
assemblage, the high value ZTR, the great maturity of heavy mineral.
3. Reveal geological history relative with the mineral chemistry of detrital
zircon, and reveal complexity in order to gain insight into natural processes
The mineral zircon is extremely variable both in terms of external morphology and
internal textures. These features reflect the geologic history of the mineral, especially the
relevant episode(s) of magmatic or metamorphic crystallization (and recrystallization),
strain imposed both by external forces and by internal volume expansion caused by
metamictization, and chemical alteration. One of the major advantages of zircon is its
ability to survive magmatic, metamorphic and erosional processes that destroy most other
common minerals. Zircon-forming events tend to be preserved as distinct structural
entities on a pre-existing zircon grain. Because of this ability, quite commonly zircons
consist of distinct segments, each preserving a particular period of zircon-formation(or
consumption).
In common rocks, zircon ranges in size from about 20 to 200m .The elongation (length-to-
width) ratios is ranging from 1 to 5. which is commonly believed to reflect crystallization
velocity. Indeed, needle-shaped acicular zircon crystals are common in rapidly crystallized,
porphyritic, sub-volcanic intrusions, high-level granites, and gabbros. In addition, newly-
grown zircon crystals can themselves exhibit evidence for multiple stages of growth and
corrosion.
Sedimentary rocks may also contain a significant fraction of zircon. Although authigenic
zircon has been reported, sedimentary zircon is predominantly derived from weathered
igneous and metamorphic rocks. Detrital zircon in sedimentary rocks and sediments is
highly durable and records age information of crustal units that contributed to the sediment
load.
The External morphologies of zircons in heavy minerals are included the followings: colors,
opaque, lustre, roundness, coarse, dissolution, abrasion and elongation, based on the its
roundness of grains, which is to unravel the transportation distance of sediments, which has
been attributed to its provinces, two subgroups are classified as:

Textural Characterization of Sedimentary Zircon and Its Implication

275
1. Zircons with a high degree roundness: which are mostly consist of dark pink-rose with
few of yellow-pink colors ,sub-opaque , coarse glass luster, and sub-rounded and
rounded grains with obvious relics of abrasionwith elongation (length-to-width)
ratios ranging from 1 to 2, and preferential growth of grains 0.05mm~0.2mmin
response to transportation of sediments
2. Zircons with a medium degree roundness:which are dominantly composed by sub-
opaque or opaque, weak-diamond to coarse glass luster, pink-shallow pink colors and
sub-rounded and rounded grains, euhedral with obvious relics of abrasion of mostly
grains with few smooth surface of grainswith elongation (length-to-width) ratios
ranging from 1.5 to 3.5 and with a size from 0.05mm~0.35mm in diameter of grains
the grains with a rather high round would be taken a great parts among all of grains
with a well selection, indicating the obvious transportation of sediments
In simple way, three different types of detrital zircons would be termed :
1. the igneous zircon with six- prismatic concentric zone (Fig.1);
2. Metamorphic zircon with a internal structure of core-mantle-rimmed; either the core-
mantle-rimmed or the core- rimmed commonly appears to developed in metamorphic
zircon with a different component of core, mantle and rim(Fig.1). The origin and inherit
zircons may be observed in the core part of grain of zircon, while the metamorphic
zircons grows in the mantle or rimmed(Hermann et al.2001)equant and weak
zoning zircons are not likely to occur in granulite rocks; it is not uncommon to find
zircons with a fan-shape and oscillatory zoning in the process of serpentinization of
basic and ultra basic rocksthe zircons can retain a plane zoning in response to the
metamorphic re-crystallization either in re-melting migmatite or in a neocryst of ovoid
shape(Fig.1);
3. re-cycle sedimentary zircons with a feature of cathodoluminescence (CL) or no growth
zoning and metamict zones(Fig.1).
The zircons in the Lower Silurian sandstones, well Shun1 also shows the difference among
the samples, for example, 15 percent, the most lowest content of six- prismatic concentric
zone, 71 percent, the highest content of non-zoning and 33 percent, the rather high
cathodoluminescence (CL) are presented in Shun1-21indicating the stable re-cycle
sedimentary environment; while 20 percent of six-prismatic concentric zone33 percent of
structure of core-mantle-rimmed ,48 percent of non-zoning, and 25 percent of
cathodoluminescence (CL) in Shun1-22, illustrated that igneous and metamorphic zircon has
been majorly attributed to the detrial zircons(Fig.2).
It would be reasonable predicted that unaltered igneous zircon generally contains the
highest contents of UTh and Pb, which is a comparable to that of Metamorphic zircon and
Re-cycle sedimentary zircons, and the re-cycle sedimentary zircons would be various in
contents between the igneous zircon and Metamorphic. U as a un-compatible element,
would be accumulated in anatectic melting crystallization and partial differentitation; while
the contents for U and Th decreased in the metamorphic process due to being dispelled
from the crystal lattices during the recrystallization in solid. In addition, the contents for Pb
is supposed to relate the contents of U and Th and age of rocks, the higher of contents of U
and Th and the older of age of rocks; the greater of the accumulated content of radiogenic
lead in zircons.

Cathodoluminescence

276

1. Alteration developed along concentric fractures parallel to the boundary between the low-U
interior part of the zircon and the high-U outer shell; an outermost low-U rim has radial cracks,
which have allowed the access of the fluids(Z1-2(D); Partially preserved growth zoned zircon
penetrated by trangressive zones of recrystallization and with local development of
recrystallization or convolute zoning (ZS2-25);
2. Recrystallization and new growth of zircon in high-grade metamorphic rocks, Bands or other large
segment of homogeneously textured zircon (ZS2-26); strong variations in the relative development
of zoned domains, large uniform zone external and much finer oscillatory-zoned bands in internal
structure (Z1-3)
3. Late to post-magmatic recrystallization of zircon, Variable appearance of xenocrystic cores in
magmatic (Z1-2D,ZS-2)and high-grade meta-morphic rocks(Z1-3,ZS2-26),Z1-2(D),Z1-3); Variations
in growth zoning in magmatic zirconAppearance and texture of zircon in meteorite impact
structures(ZS1-21);

Fig. 1. Typical Zircon CL structures for different origin and SHRIMP U-Pb Concordia plots
with spot locations, from Donghe sandstones of DongheTang formation, the Upper
Devonian in western Tazhong area , or the Lower Silurian sandstones ,Tarim basin: igneous
zircons(upper),Metamorphic zircon(Middle),Re-cycle sedimentary zircons(down),different
scales in zircon megacryst,ranging from 0.5 mm down to a few 50um.


277

Fig. 2. Diagram showing the statistics of typical Zircon grains based on their sources from
Donghe(lift)and the Lower Silurian sandstones(right) in wells of Zhong1,11,12,13 and
Shun1,2. Tarim basin (Qian Y.X.,et al.,2007b)

Cathodoluminescence

278
4. Determine maximum age of stratigraphic successions and to help
recognize time gaps in the geologic records
The most reliable means of directly determining depositional sedimentary ages is through the
dating of interstratified volcanic rocks, or dating time-of-deposition authigenic xenotime
overgrowths on detrital zircon grains (Mc Naughton et al. 1999). Under certain circumstances,
the age of the youngest detrital zircon in a population can approach the age of deposition
(Nelson, 2001). Disconformities reveal information about coeval tectonics/uplift and sea-level
fluctuations. In the absence of fossils to identify gaps in the geologic record.
The results are typically the U-pb isotopic data of zircon from sensitive high resolution ion
microprobe(SHRIMP)analyses can provide the accurate age constrains for record of events
associated with the major geodynamic evolution of stage from Proterozoic to the Early
Paleozoic in Tarim basin and its surrounding orogenic belts.
The U-pb isotopic data of zircons can be divided into three groups , i.e. 1.8Ga,2.2Ga and
2.6Ga from Meso-Proterozoic to Neo- Proterozoic; 0.84Ga in Cambrian; and 477~439.8Ma
and 431.6~421.1Ma from Ordovician to The Lower Silurian(Fig.3), which have a
concordance with the records of rocks derived from that of wells or outcrops of Tarim basin
and its surrounding orogenic belts.

Fig. 3. Frequency histograms for Distribution of
206
Pb/
238
U ages of Zircon from Proterozoic(a)
to Neo-Meso Proterozoic(b) and Paleozoic(c) in western Tazhong area ,Tarim basin


279
5. Test regional paleogeographic reconstructions by origin analysis
Previously published research from the Kuqa Sub-basin along northern margin of the Tarim
Basin shows five tectonic-depositional phases from Triassic to Neogene time(Li et al.,2010).
In order to reveal more detailed information on the nature of provenance terrains and
tectonic attributes since late Mesozoic time, five typical sandstone samples from Jurassic
Neogene strata were collected for U-Pb dating of detrital zircons.
Geochronological constitution of detrital zircons of the Middle Jurassic sample is essentially
unimodal and indicates major contributions from the SouthTian Shan, wherein most
370~450Ma zircons probably resulted from tectonic accretion events between the Central Tian
Shan block and SouthTian Shan Ocean during Silurian and Devonian time, with sandstone
provenance tectonic attributes of passive continental margin. The Lower Cretaceous sample
shows a complicated provenance detritalzircon signature, with new peak ages of 290~330Ma
as well as 370(or 350)~450Ma showing evidence of arc orogenic provenance tectonic attribute,
probably refecting a new provenance supply that resulted from denudation process within the
SouthTian Shan and SouthTian Shan suture. There are no obvious changes within age
probability spectra of detrital zircons between the Cretaceous and early Paleogene samples,
which suggests that similar provenance types and basin-range framework continued from
Cretaceous to Early Paleogene time. However, unlike the Cretaceou sand early Paleogene
samples, an age spectra of the Miocene sample is relatively unimodal and similar to that of the
Pliocene sample, with peak ages ranging between 1392 and 1458Ma older than the comparable
provenance ages (peak ages about 370~450Ma) of the Middle Jurassic and Lower Cretaceous
samples. Therefore, we can conclude that the SouthTian Shan was rapidly exhumated and the
southern South Tian Shan had become the main source of clastics for the Kuqa Sub-basin since
the Miocene epoch. and the corresponding age-probability plots in Fig.4.
We contrast the detrital zircon age spectra from the Kuqa Subbasin with those of potential
provenance areas, discussing the implications for prove nance and paleogeographical
changes. Detrital zircons of the Middle Jurassic sample mostly range from 370 to 450Ma,
with a small number of Proterozoic-Archean ages.This geochronological constitution is
comparatively unimodal and indicates that the dominant source of detritus was likely from
the northern South Tian Shan and southern Central Tian Shan (Fig.5), probably resulting
from tectonic accretion events between the Central Tian Shan Block and the South Tian Shan
Ocean that occurred during Silurian and early Devonian time, with dominant sandstone
provenance tectonic attributes of passive continental margin discriminated by major
element composition of whole-rock samples.The Lower Cretaceous sample shows a
provenance with complicated detrital zircon age spectra, with new peak ages of 290-330Ma
as well as 370 (or 350)-450Ma. Refecting a new provenance supply produced by denudation
of the South Tian Shan and South Tian Shan suture. These grains likely refect
Carboniferous-Permian volcanism and the Silurian-Devonian tectonic events between the
South Tian Shan-Tarim and Central Tian Shan blocks, with dominant sandstone provenance
tectonic attributes of active continental margin or island arc (arc orogenic belt). In addition,
several clusters of Proterozoic-Archean ages from this sample probably reect that some
provenance regions may have been deeply exhumated.The lack of obvious changes in
detrital zircon age probability spectra between the Cretaceous and early Paleogene samples
suggests that similar provenance types and basin-range framework likely continued from
Cretaceous to early Paleogene time(Fig. 5).

Cathodoluminescence

280

Fig. 4. Probability plots and number histograms of U-Pb ages of detrital zircons from Middle
Jurassic-Neogene sandstone samples in Kuqa Subbasin. (a)-(e) 0-3700Ma grains; (a0)-(e0)
150-500Ma grains(after Li et al.,2010).


281

Fig. 5. Schematic map showing major provenance area changes from middle Jurassic (a),
through Cretaceous-Paleogene (b), to Neogene (c). Arrowheads, with width, indicate
directions, power and ranges of major inferred provenance supply. Structural deformation
and intra-continental shortening that occurred in the study area from Jurassic to Neogene is
ignored on the above maps(after Li et al.,2010)

Cathodoluminescence

282
6. References
Li, Z., Peng, S.T. (2010) Detrital zircon geochronology and its provenance implications:
responses to Jurassic through Neogene basin-rangeinteractions
alongnorthernmargin of the Tarim Basin,Northwest China. Basin Research,22
126~138
Mc Naughton NJ, Rasmussen B, Fletcher IR (1999) SHRIMP uranium-lead dating of
diagenetic xenotime in siliciclastic sedimentary rocks. Science, 285:78-80
Nelson DR (2001) An assessment of the determination of depositional ages for Precambrian
clastic sedimentary rocks by U-Pb dating of detrital zircon. Sed geol, 141-142
Qian Y.X., He Zh.L., Cai X.Y., et al. (2007a) Characteristics of heavy minerals from Upper
Devonian Donghe sandstone and Silurian sandstone in western Tazhong area ,
Tarim Basin , and their geological implications. Acta Petrologica et Mineralogica,
26(2)147~154 (in Chinese)
Qian Y.X., He Zh.L., Cai X.Y., et al. (2007b) Characteristics of and U-Pb ages of zircons and
its implication from zircon of Donghe sandstones of The upper Devonian and
Silurian sandstones in Tazhong area,Tarim basin. Acta Petrologica Sinica,
23(11)3003~3014 (in Chinese)
Part 4
Related Luminescence

12
Novel Problems in the Solid State
Cathodoluminescence of Organic Materials
Zheng Xu
1
and Suling Zhao
2

1
Institute of Optoelectronic Technology, Beijing Jiaotong University, Beijing,
2
Key Laboratory of Luminescence and Optical Information, Ministry of Education, Beijing,
P.R. China
1. Introduction
Luminescence is the unique efficient supplementary technology to make up a constant
deficit of solar light. Among all kinds of luminescence the electric field induced
luminescence is especially attractive due to its feasibility of manipulation and direct
transformation of electrical energy into light. For this trend cathodoluminescence is used for
display already approximately 100 years. But the trend of contemporary display is to make
flat panel display. For this reason many efforts were made to use cathodolumunescence,
e.g., VFD and FED although the latter achieves still no practical success. At the same time EL
finds applications in a certain field. In order to obtain blue color inorganic EL we proposed
layered optimization scheme, and very soon we convinced the energy of hot electrons from
the acceleration layer is high enough to excite luminescence directly. We realized this idea
by using insulator phosphors. But the probability is very low because the environment of
luminescent center is not regular near surface. In the layered optimization scheme we used
the secondary property of electrons (hot electrons) of inorganic materials in order to raise
the electron energy before they are accelerated in EL layer. In this acceleration layer, both
the number of electrons and the energy of electrons are increased. The correctness and
reasonableness of this idea was fully realized [1]. We tried to use organic materials to
substitute the inorganic materials in the layered optimization scheme [2-3], thus discovered
the solid state cathodoluminescence of organic luminescent materials. We call it as solid
state cathodoluminescence (SSCL) because the hot electrons are accelerated in solids instead
of in vacuum. SSCL is an independent branch of electric field induced luminescence. Soon
we found also the possibility of integrating SSCL with OEL[4], i.e., using a single electrical
source and a single luminescent material we can get two kinds of excitations for SSCL and
OEL, which is the unique integrated excitation among all kinds of luminescence.
2. Classification of electric field induced luminescence according to mode of
excitation
SSCL is a new kind of electric field induced luminescence. Up to now, electric field induced
luminescence play an import role in the field of flat panel display, for example field
emission display (FED), vacuum Fluorescent Display (VFD), electroluminescence (EL), light-
emitting diode(LED) and organic electroluminescence(OEL), even though some

Cathodoluminescence 286
technologies such as FED are still not practical at present. Their excitation mechanism is
basically classified as two types, impact excitation and carrier injection.
2.1 Impact excitation
In the impact excitation mechanism, the luminescence center is excited by hot electrons
collision which provides an efficient means of producing electronic transition to excited
states of the luminescence centers. Based on impact excitation, some technologies of field
emission display, vacuum fluorescence display, solid state cathodoluminscence, inorganic
electroluminescence are produced.
2.1.1 Field emission display (FED)
A field emission display (FED) is a low power, flat cathodoluminescence display that uses a
matrix-addressed cold cathode. Field emission display is a high-voltage display with a
triode structure consisting of anode, cathode, and gate electrodes to achieve high
illumination by applying a high voltage and a low current. In the field emission display, a
strong electric field is formed between a field emitter and gate electrodes disposed on a
cathode at a constant interval, so that electrons are emitted from the field emitter so as to
impact on phosphors of an anode, thereby emitting light. Field emission display technology
makes possible the thin panel as today's liquid crystal displays (LCD), offers a wider field-
of-view, provides the high image quality of today's cathode ray tube (CRT) displays, and
requires less power than today's CRT displays. However, FED display requires a high level
vacuum which is difficult maintain.
2.1.2 Vacuum fluorescent display (VFD)
The VFD is composed of three basic electrodes: the cathode (filaments), anodes (phosphor)
and grids under a high vacuum condition in a glass envelope. Electrons emitted from the
cathode are accelerated with positive potential applied to both grid and anode, which upon
collision with the anode excites the phosphor to emit light. The desired radiative patterns
can be achieved by controlling the positive or negative potentials on each grid and anode.
This voltage can be as low as 10V DC. The principle of operation is identical to that of a
vacuum tube triode. Electrons can only reach (and "illuminate") a given plate element if both
the grid and the plate are at a positive potential with respect to the cathode. This allows the
displays to be organized as multiplexed displays where the multiple grids and plates form a
matrix, minimizing the number of signal pins required. Compared to LCDs, they have
relatively high power consumption. Other problems include some segments of a VFD
display gradually becoming brighter or dimmer than others (caused by the phosphors
glowing less brightly as they get older) and flickering.
2.1.3 Solid state cathodoluminescence (SSCL)
Organic materials for example, Alq
3
[5-6 ] , MEH-PPV [2], Ir(ppy)
3
[7], etc, were used to
substitute the inorganic materials in the layered optimization scheme and discovered two
peaks of luminescence under AC bias. One is the typical organic electroluminescence peak
with longer wavelength. Another one is a new luminescence peak with short wavelength.
Both peaks constitutes the solid state cathodoluminescence (SSCL). The excitation

Novel Problems in the Solid State Cathodoluminescence of Organic Materials 287
mechanism of two emission peaks is impact excitation by hot electrons. The emission is not
due to the typical recombination electroluminescence of injected carriers because no holes
are injected to the organic materials in this kind of device. After primary electrons tunneled
into the insulator layer, they are accelerated in the insulator layer and then have the high
energy. These hot electrons collide organic materials to excite electrons from HOMO to
LUMO like the excitation by photons. These excited electrons relax fast to form excitons
with holes in HOMO. Then formed excitons recombine and emit light. Under high electric
field, parts of formed excitons should be dissociated and then the light should be emitted
directly recombined from electrons in LUMO and holes in HOMO. The excitation
mechanism of SSCL is same to that of typical vacuum cathodoluminscence except that the
hot electrons are accelerated not in vacuum but in solids.
2.1.4 Electroluminescence (EL)
Electroluminescence is the result of radiative recombination of electrons and holes in a
material, usually a semiconductor. The excited electrons release their energy as photons -
light. Electroluminescent devices are fabricated using either organic (OEL) or inorganic
electroluminescent (IEL) materials. It is well known that the sandwich type thin-film IEL.
The phosphor in these devices is not a powder but a thin continuous film prepared by
sputtering or vacuum evaporation. The luminescence activators are manganese or rare-earth
ions, atomic species doped in ZnS or other host materials with internal electronic transitions
that lead to characteristic luminescence. A thin film luminescence (TFEL) device generally
has a double insulator structure of electrode/insulator layer/active layer/insulator
layer/electrode. For excitation process of luminescence centers in TFEL, the hot electron has
been known to be the dominant carrier. Under AC bias, the accumulated electrons in the
interface region of the electrode and the insulator layer are accelerated to become hot
electrons. These hot electrons impact luminescence centers and EL results. A TFEL device
acts like a pure capacitor at low applied voltage; no light is emitted until the voltage reaches
a threshold value determined by the dielectric properties of the insulator and phosphor
films. Above this threshold a dissipative current flows, and light emission occurs. The
brightness increases very steeply with the applied voltage but is finally saturated. The light
output, or average brightness, is roughly proportional to the frequency up to at least 5 kHz,
and also depends on the waveform of the applied voltage.
Traditional IEL displays are bright, very fast in video response time and highly tolerant of
environmental extremes. However, the lack of full-color has limited their application for the
mainstream consumer television market. In order to get efficient inorganic
electroluminescence, we have proposed the layered optimization scheme. The correctness
and reasonableness of this idea was fully realized [1].
2.2 Inject electroluminescence
Injection electroluminescence results when an inorganic semiconductor pn junction (LED) or
a point contact is biased in the forward direction. This type of emission is the result of
radiative recombination of injected minority carriers, with majority carriers in a material.
Such emission has been observed in a large number of semiconductors. The wavelength of
the emission corresponds to an energy equal, at most, to the forbidden band gap of the

material. If a pn junction is biased in the reverse direction, so as to produce high internal
electric fields, other types of emission can occur, but with very low efficiency. LEDs present
many advantages over incandescent light sources including lower energy consumption,
longer lifetime, improved robustness, smaller size, faster switching, and greater reliability.
For room lighting LED are powerful enough and relatively expensive and require more
precise current and heat management than compact fluorescent lamp sources of comparable
output.
Electroluminescence in organic materials may be observed in a single organic layer
sandwiched between two electrodes, but the efficiency is not high. In order to enhance the
efficiency of OLED, multilayer structure devices are introduced. To explain the operation of
a typical multilayer OLED, the physical processes have to be identied [8]. Firstly Charge
carriers are injected from electrodes on the top and bottom of the OLED and transport in
organic materials. A minimum layer thickness has to be retained in order to prevent
minority carriers to migrate to the opposite contact. Opposite space charges extend from the
two contacts into the device. The zone in which they overlap and where recombination takes
place is only a few nanometers wide. The recombinative process is strongly inuenced by
traps. It results in a charge-transfer complex or an excited molecule, which is called a
(Frenkel-)exciton. Due to the spin multiplicity, excited singlet and triplet states are formed
with a ratio of one to three. The excitation energy can be transferred to neighboring or more
distant molecules of the same type or to molecules with a lower excitation energy. While
triplet excitons decay nonradiatively, luminescence is caused by the radiative decay of
singlet excitons. Nonradiative decay may take place at contacts and impurities.
2.2.1 Integrated excitation of SSCL and OEL
SSCL of organic materials is resulted from the recombination emission and exciton emission.
When the acceleration layer is an suitable material in which electrons can be accelerated and
holes can inject to the organic layer, it is possible to detect the luminescence of organic
materials due to the electron impact excitation(SSCL) and carrier injection (OEL). The
integration of the excitation of SSCL and OEL was achieved [5].
2.2.2 Mixed luminescence of EL and OEL
Electric field induced luminescence provides the way of direct transformation of electric
energy into light. According to the materials, there are two kinds of electroluminescence,
inorganic electroluminescence and organic electroluminescence. The inorganic
electroluminescence (IEL) suffers from the lack of bright blue luminescence, and the organic
electroluminescence (OEL) suffers from its stability. But they have their own advantageous
to be used in flat plate display. For example, OLED is widely considered as the potential
display technology. Therefore we hope to combine their superiority and minimize their
inferiority in order to improve their performance. To combine IEL and OEL, We tried to use
organic materials to substitute the inorganic materials in the layered optimization scheme
[2-3]. Firstly we used the primary property (simple electron conduction) of ZnO or ZnS [9]
as the electron transport material for enhancing the electroluminescence of organic layer and
achieved positive results. In both of these two kinds of electroluminescence the required
electric field is almost of the same order of magnitude. We tried to combine ZnS:Mn and

PPV layers together and really obtained the simultaneous emission from both layers[9]. On
the other hand, we successfully applied the layered optimization scheme to improve the
overall performance of IEL.
3. Characterization of SSCL
The spectral characteristics of SSCL of organic luminescent materials are the appearance of
two light emission peaks when the applied electric field increases. More detailed
examination of these two peaks shows that the longer wavelength peak is originated from
excitons and the shorter wavelength peak is attributed to transitions from LUMO to HOMO
levels directly. When the applied electric field increases, the longer wavelength peak
increases at first, then passes a maximum and monotonically drops down to zero. The
shorter wavelength peak gradually increases monotonically after the diminishing of the
longer wavelength peak. The reason of the variation lies in the ionization of excitons in an
electric field.
3.1 When the applied voltage increasesthe two bands varies according to ionization
of excitons
We used a series of organic phosphors, such as Alq
3
, PPV, MEH-PPV, C
9
-PPVto substitute
the inorganic layer in the layered optimization scheme. The device structure is Al/insulator
layer/organic layer/insulator layer/ITO. The methods to prepare different organic layers
were different. For example the sample of MEH-PPV with the structure Al/SiO
2
/MEH-
PPV/SiO
2
/ITO was prepared by means of spin casting. The thickness of MEH-PPV is about
50nm. The thin film SiO
2
was obtained by electron beam evaporation (EVA450, alliance
concept Co. Ltd., France) at the growth rate of 1/s under the high vacuum of 210
-6
Torr
with a thickness of about 200nm. The other sample of Alq
3
within the structure Al/SiO
2
/
Alq
3
/SiO
2
/ITO was prepared by means of thermal evaporation of Alq
3
. Excited by AC
voltage, these samples showed a similar spectral behavior of having dual spectral peaks and
shifting to short wavelength when the applied voltage was increased.
The spectrum of ITO/SiO
2
/Alq
3
/SiO
2
/Al is shown in Fig.1. It is a wide band emission
which can be fitted by the Gaussian fitting to two peaks [6]. One peak is 510nm, another one
is 451nm. In Fig.1, the electroluminescence of ITO/Alq
3
/Al is observed also under DC bias.
The peak locates at 510nm which is the emission from the singlet exciton recombination of
Alq
3
. In a typical OLED, electrons and holes inject to the fluorescent material from the
cathode and anode respectively, then they meet each other and singlet and triplet excitons
are formed. The recombination of singlet excitons gives the electroluminescence. In our
study, SiO
2
is used in this new kind of devices. As we known SiO
2
is an insulator. So it is
difficult to transfer electrons, especially holes. In such case, the mechanism of
electroluminescence of ITO/SiO
2
/Alq
3
/SiO
2
/Al is different to that of the typical OLED.
This means that the electroluminescence of ITO/SiO
2
/Alq
3
/SiO
2
/Al is not originated from
the carrier injection recombination.
According to above simulation, electrons which enter to Alq
3
layer have been accelerated in
SiO
2
of the device of ITO/SiO
2
/Alq
3
/SiO
2
/Al under AC bias. These electrons with high
energies should bombard Alq
3
, then Alq
3
is excited by this electron bombardment as excited
by light directly. In this case, more singlet excitons are formed. Along with the increasing of

the electric field on Alq
3
, singlet excitons should be dissociated into free electrons and holes
to LUMO and HOMO respectively. If these free electrons and holes recombine, they emit
450nm light. Therefore, we observed a new emission peak in the electroluminescence
spectrum of ITO/SiO
2
/Alq
3
/SiO
2
/Al besides the singlet exciton emission of 510nm.
Theoretically, in this kind of devices, the quantum efficiency is higher than that of typical
OLED devices because the impact excitation likes the excitation by light. There are almost no
triplet exciton formations under electron impact excitation. Therefore, the efficiency of this
kind device is larger than that of the typical OLED.
350 400 450 500 550 600 650
0.0
0.2
0.4
0.6
0.8
1.0
N
o
r
m
a
l
i
z
e
d

I
n
t
e
n
s
i
t
y
Wavelength (nm)
ITO/SiO2/Alq3/SiO/Al
ITO/Alq3/Al

Fig. 1. The electroluminescence spectrum of ITO/SiO
2
/Alq
3
/SiO
2
/Al (square) and ITO/
Alq
3
/ Al (triangle) under AC bias
The spectrum of Al/SiO
2
/MEH-PPV/SiO
2
/ITO consists of two peaks at 405 nm and 580 nm
(Fig.2)[10]. In ordinary organic electroluminescence we have only the long wavelength peak.
In SSCL, the spectrum shifts from the long wave length peak to blue one when the applied
voltage increases. The apperance of short wave length peak is a characteristic phenomenon
for SSCL due to the appearance of short wave length peak. The voltage range is divided into
three different parts according to the spectrum: short wavelength peak at 405 nm only, a
mixture of the two spectral components 405 nm and 580 nm, the long wavelength peak at
580 nm only. The 580 nm peak is attributed to the exciton emission, the 405nm peak
corresponds to the radiation related to the recombination of carriers in conduction and
valence band. When the applied voltage is increased, the spectrum shifts to blue.
We have also observed the solid state cathodoluminescence in organic phosphors, for
example, Ir(ppy)
3
[7]. Ir(ppy)
3
is a well known organic phosphor. Its phosphorescence
emission is a green emission with a peak at 520nm. We doped Ir(ppy)
3
in PVK, then
prepared PVK:Ir(ppy)
3
thin film on SiO
2
by the method of spin coating. SiO
2
layers and Al
cathode were prepared as before. Then the device of ITO/SiO
2
/ PVK:Ir(ppy)
3
/SiO
2
/Al
was realized. Under AC bias, the electroluminescence of this device which locates at

520nm was observed under different bias, as shown in Fig.3. This emission corresponds to
the triplet excition emission of Ir(ppy)
3
. As discussed above, the electroluminescence
essentially is the solid state cathodoluminescence, that is, the emission from the
bombardment of high energy electrons accelerated in SiO
2
. It is very surprise that there is
no the singlet exciton emission from PVK. It is possible that PVK is excited also by
electrons and singlet excitons are formed. These singlet excitons do not give the radiative
emission but transfer their energy to Ir(ppy)
3
directly. According to our previous study,
there is the spectrum overlap between the emission spectrum of PVK and the excitation
spectrum of Ir(ppy)
3
. So the Forster energy transfer is possible between PVK and Ir(ppy)
3
.
Therefore, there are two ways for the excitation of Ir(ppy)
3
. One is the directly excitation
of the high energy electrons, another one is the energy transfer from PVK. Therefore, the
efficiency should be very high theoretically.

400 500 600
0.0
0.2
0.4
0.6
0.8
1.0
I
n
t
e
n
s
i
t
y

(
a
.
u
.
)
Wavelength(nm)
40V
50V
80V
100V

2
/MEH-PPV/SiO
2
/Al under AC bias

300 400 500 600 700
0
5000
10000
15000
20000
25000
30000
35000
40000
I
n
t
e
n
s
i
t
y

(
a
.
u
.
)

Wavelength (nm)
145v
130v
110v
80v

2
/PVK:Ir(ppy)
3
/SiO
2
/Al under AC
bias
3.2 What is the nature of this emission? Cross proof of this emission being really a
new kind of luminescence
We prepared a sample of the structure Al/SiO
2
/MEH-PPV/ITO and discussed its emission
mechanism under AC bias. If the emission is originated from gaseous discharge, the
luminescence excited by it should be observed at both half periods of the applied voltage
because the light may propagate ether in front direction, or in tail direction. This is simply
because the amplitude of applied voltage remains the same in each of both half periods, the
gaseous discharge might occur equally during each half period although their amplitude
may be different. But experiment showed that emission occurs only when Al electrode is
negative. Really during this half period the accelerated electrons may bombard MEH-PPV
and in the opposite half period the electrons can not. Only the direct bombardment
produces luminescence. This spoke of the impossibility of gaseous discharge. The same
conclusion is true for symmetric structure.
This emission is not the injection luminescence as in p-n junction of LED and organic EL
also. Because the energy barrier for holes on the interface of Al/SiO
2
is higher than 5.2eV
and on the interface of ITO/ SiO
2
is 4.7eV, no holes can be injected into organic phosphor in
a symmetric structure and consequently no recombination of electrons with holes, i.e., no
LED or OEL is possible.
This emission is not also the self developed EL as in inorganic materials, in which the
acceleration of hot electrons, collision excitation and luminescence proceed in the same layer,
because the carrier mobility in organic phosphors is less than 10
-2
cm
2
/v.s or in special case
less than 1cm
2
/v.s. The kinetic energy of hot electrons accumulated in this material is lower
than the value required for excitation. More than this, the ordinary IEL may be observed only

in very limited class of semi-conductors, but the above mentioned new emission occurs not
only in semiconductors but also in insulators if it is an ordinary luminescent phosphors.
All these exclusions of trivial phenomena from these aspects stick out the SSCL as a new
type of luminescence.
3.3 In brightness waveforms the traditional method for determining phase relation
fails to be effective[11]
In order to make the SSCL mechanism clear, we measured the temporary changes in the
emission intensity along with the AC bias, which is the emission waveform, as shown in Fig 4.
The line with the triangle symbol is the excitation AC bias with the amplitude of 60 V and
frequency of 500 Hz. The lines with square and circle symbols are the waveforms of the
emissions of 405 nm and 580 nm, respectively. It is found that the waveforms of these two
emissions are different. The maximum intensity of 405 nm occurs under forward voltage, but
the maximum intensity of 580 nm occurs under reverse voltage. According to the traditional
phase detection method, the position of the maximum intensity is taken as the phase of the
emission. Apparently 405 nm is emitted before 580 nm according to their waveforms if we take
the position of the maximum brightness peak in the forward direction. When the bias is
increased to another higher value, the order of these two SSCL peaks appears to be reversed.
Certain important fact should be hidden behind this uncertainty of phase.
The detected intensities of 580 nm and 405 nm with increases in voltage are shown in Fig. 5.
The threshold bias of this device is 20 V. When the AC voltage increases, 580 nm first
enhances and then decreases. In the meantime, 405 nm is detected at the point which
corresponds to the voltage of the maximum intensity of 580 nm. Then, 405 nm increases
along with the increase in the voltage. We have concluded before that 405 nm emission is
due to the direct recombination of electrons and holes which results from the decomposition
of excitons. Their changes with changes in voltage can be described by the rate equation.
The intensity of 405 nm is:

1 0
[ ( ) ]
s
B G G E n o u o
-
= + (1)
where ( ) E o is the exciton decomposition velocity under electric field E,
1
o is the cross
section of the recombination emission, u is the average electron velocity in the conduction
band, G is the excitation rate, and G
-
is the number of holes. The simplest condition is G
-
=
G. They are functions of the electric field E.
The exciton formation rate is

0
2 0 0 0
1
[ ( ) ] ( )
dn
G G E n n E n
dt
o u o o
t
-
= + (2)
Here,
2
o is the cross section of the exciton formation. Because of the weak interaction
between molecules,
1
o and
2
o can be considered as constants. But but according to the
experiment, the recombination emission is difficult to detect in typical organic
electroluminescence devices, so
1
o is much less than
2
o .
0
n is the exciton concentration,
and t is the lifetime characteristic parameter of exciton emission.

0.000 0.002 0.004 0.006 0.008
-1.0
-0.5
0.0
0.5
1.0
-60
-30
0
30
60

D
r
i
v
i
n
g

v
o
l
t
a
g
e

(
V
)

I
n
t
e
n
s
i
t
y

(
a
.
u
.
)
Times (S)
Frontier
ITO
405nm 500Hz

0.000 0.002 0.004 0.006 0.008
-1.0
-0.5
0.0
0.5
1.0
-60
-30
0
30
60
D
r
i
v
i
n
g

v
o
l
t
a
g
e

(
V
)
I
n
t
e
n
s
i
t
y

(
a
.
u
.
)
Time (s)
ITO
580 nm 500 Hz
Frontier

Fig. 4. The emission waveform of ITO/SiO
2
/MEH-PPV/SiO
2
/Al
The intensity of exciton emission is

0 2 0 0
1
[ ( ) ] ( )
l
B n G G E n E n o u o o
t
-
= = +
(3)
Under forward bias, the electron injection from the cathode is very effective and electrons are
accelerated in the SiO
2
layer. So the voltage on MEH-PPV is relatively high. But for reverse
bias, the electron injection from ITO is not effective, and most of the voltage falls on the SiO
2

layer. The electric field on MEH-PPV is low. The simulation of the electric field on MEH-PPV
is shown in Fig. 6. At low electric fields, ( ) E o is nearly equal to zero. E increases, and G and
G* increase too. So the exciton emission increases. When E reaches a certain value, ( ) E o
begins to increase along with E. As E increases, the exciton emission increases slowly, then
decreases because of the decomposition of excitons. Meanwhile, the recombination emission is
detected thanks to the decomposition of excitons. The bias on the device is sine wave, and
should be changed from low to high. In such cases, 580 nm is detected firstly and then 405 nm
is detected along with the increase in the bias. The AC bias put on the device is a changing
bias. The voltage is not a constant but changes with time. We take V+ as the average voltage of
the organic layer in forward bias, and V as the average voltage in reverse bias. The actual bias
AV+ and AV on the organic layer is less than V+ and V because those other layers take part
of the bias, as shown in Fig. 6. The bias changes, and the injection current changes. Therefore,
the bias AV+ and AV changes and results in a changed electric field on the organic layer. This
is critical for the exciton decomposition which relates to the exciton emission and band-band
recombination.
Therefore, if we denote the phase of the emissions according to the fact shown in Fig.5, the
phase corresponding to the maximum intensity in Fig. 4 is not the actual phase of the
emissions. It is contrary to that found by the traditional phase detection method, which
means that the traditional phase detection method is not suitable for the SSCL. It is found
that the frontiers of their waveforms are consistent with their emission order. Thus if we use
the phase of the frontier to denote the phase or the valley phase (minimum intensity) of the
SSCL, we will obtain the right results. Therefore, the frontier phase instead of the maximum
intensity phase is the true phase of the SSCL.
40 50 60 70 80
0
5000
10000
15000
V
- V
+
V3
I
n
t
e
n
s
i
t
y

(
a
.
u
.
)
Driving voltage (V)
580 nm
405nm
V2
B
S
+
-B
S
- B
l
+
-B
l
-

Fig. 5. The dependence of SSCL on the driving voltage

Fig. 6. The simulative bias on the organic layer
4. The special feature of SSCL lies in the accumulation of hot electrons
energy
It is known that the SSCL are related to the electric field. The detected intensities of 580 nm
and 405 nm of MEH-PPV with increases in voltage are shown in Fig. 5. The threshold bias of
this device is 20 V. When the AC voltage increases, 580 nm first enhances and then
decreases. In the meantime, 405 nm is detected at the point which corresponds to the voltage
of the maximum intensity of 580 nm. Then, 405 nm increases along with the increase in the
voltage. We have concluded that 405 nm emission is due to the direct recombination of
electrons and holes which results from the decomposition of excitons. All these emissions
are due to the electron impact excitation. In this way, the SSCL are depending on the hot
electron energy.
4.1 Increment of hot electron energy in a stepwise way along the increase of electric
field intensity
Since the excitation of electroluminescence originates from impact by hot electrons, the
behavior of electroluminescence (EL) is chiefly determined by the energy of hot electrons.
Thus the diverse phenomena in EL inspire the study of electron energy. In the scheme of
layered optimization in which we used an acceleration layer located at the front of the
luminescent layer in order to form a thrust for the acceleration of hot electrons in the
electroluminescent layer, and hence enhance the energy of hot electrons. We observed really
the obvious superiority of energy enhancement of this layered optimization scheme over the
conventional sandwich structure. The intensity, efficiency, and percentage of short wave
length emission are increased
[
12-14]. Moreover, a second jump of light intensity appears at
higher electric field [15]. Here, three problems arise: 1) Does the additional acceleration layer
plays the role of providing thrust to the acceleration of electrons in luminescent layer really?
2) How does the density of energy states changes with electric field strength? 3) how is the
distribution of electron energy in dependence on electric field strength. Is it in accordance

with the experimental result of second jump of luminescence? We limited our research to
inorganic electroluminescence and investigated the behavior of electron energy (EE) to
clarify the reason for these obvious improvements of performance. In order to find the
energy distribution, the Boltzman equation should be solved. However, as it has no rigorous
analytical solution we need to ask help of calculation techniques such as numerical
calculation or Monte Carlo simulation. We chose the latter route. The initial electron energy
and electric field intensity in electroluminescent layer were assumed to vary in a large
scope. First, an analytical model

[16] of the conduction band is used. It consists of three
valleys (I, L, X) of the first conduction band, a Valley (X valley) of the second conduction
band, and also a valley of still higher energy. For the latter band we proposed a higher
conduction band for free electrons instead. The parameters for the free electron band are
taken from reference [17].
Second, once the analytical band structure is constituted, we calculated the electron energy
density state, as shown in Fig.7, which is defined as
0
( ) lim
E
Z
N E
E
A
A
=
A
, where Z A is the number
of energy states between E and E E + A [17]. In this process we took into account the
anisotropy of dispersion and averaged the weighted result obtained along different directions.

Fig. 7. Comparison of the overall density of states for our new analytical model of
conduction band and for full band model
Third, we found out the different scattering rates:
1. The acoustic and optic phonon scattering rates for lower and higher electron energies.
2. The equivalent and non-equivalent inter-valley scattering rates were cited from
reference[18].

3. The approximate scattering rate owing to the ionization of valence electron into the
conduction band was cited from [19, 20].
( )
ion
R E E
o
=

Where
10 1
5.14 10 ( ) R s ev
o
= , 3.5 E ev =
, and 5.183 o =
This process dominates the approach to energy stabilization and the beginning of
electron multiplication. Based on these kinds of scattering rates, we carried out the
Monte Carlo simulation. Electrons have a free flight time which may be determined
stochastically according to the total scattering rate. From the total scattering rates we
may work out a random number r (0<r<1) from the equation
0
( . ') '
t
S k t dt r =
}

. S( . ') k t

represents the probability of the existence of k
state within t. Because the variation of

the total scattering rates with time is very complicated, the idea of self-scattering
proposed by Rees is used. Then the free flight time is simplified to
1
ln
f
t r =
I
, where I
is the total scattering rate (including self-scattering). During this free flight time the
electron accumulates energy until this electron energy is balanced by scattering. At this
time this flight is ended. The scattering which is active in this flight is chosen according
to its relative magnitude. In the second flight we again chose a free flight time and new
dominant scattering process.
The temporary variation of electron energy shows that after nearly 300fs the energy of hot
electrons reaches a stationary value, which is responsible to the luminescence intensity.
Thus, we calculated the stationary value of hot electron energy and get the following
important results:
1. Initial electron energy might improve the stationary brightness only when the initial
electron energy in ZnS attains 8eV in an electric field either lower than 0.6Mv/cm or
greater than 4Mv/cm, as shown in Fig.8.
2. The distribution of hot electron energy under different electric fields is shown in Fig.9.
The energy of hot electrons centralizes at higher energy valleys and the position of
centralized energy consequently moves to the higher energy end when the electric field
is increased owing to the increase in electron energy after the acceleration in the electric
field and the balance with high energy scattering, including inter-valley scattering,. This
tendency is very important and greatly increases the effectiveness of hot electrons in
excitation.
3. The growth of hot electron energy proceeds in a stepwise way as shown in Fig. 10.
From Fig. 10, we can see that if the initial electron energy
0
c attains 8eV, the first
plateau in EE appears at an average energy of hot electrons of more than 2.0eV, then
after a sloping variation a second plateau in EE appears at an average energy of
electrons above 4.8eV. The first and second plateaus constitute a stepwise growth in the
electron energy. This behavior of EE is fully reflected in EL[15].

The complicated behavior of EE demonstrates the close dependence of electron transport on
material properties, including conduction band structure, different scattering characteristics,
the concrete structure of the device, and electric field strength. We obtained the satisfactory
answers to the previous three problems.

Fig. 8. Time evolution of electron energy at 0.6MV/cm (a) and 4MV/cm (b) field strength for
different initial energies of electrons
0
c in luminescent layer


Fig. 9. Distribution of electron energy in different electric fields

Fig. 10. Dependence of stationary electron energy on electric field strength when initial
electron energy attains 8eV
The reason of all these behaviors or physical processes may be understood as follows:
1. When the electric field strength is less than 0.6MV/cm, the strength of scattering is not
strong enough, and hence a part of the initial electron energy may be reserved and the
stationary energy of hot electrons is raised. When the electric field strength is greater
than 4MV/cm, although the strength of scattering is much increased, the free flight of
electrons is also strengthened correspondingly. As only a part of the initial electron
energy is required for energy balance, the stationary electron energy is thus increased.
When the electric field is in between 0.6MV/cm and 4 MV/cm, no influence of 8eV

initial electron energy is detected. That means the sum of free flight energy and initial
energy were balanced by total scattering.
2. In the process of electron transport the inter-valley scattering transfers the electrons
from lower energy valleys to higher energy ones. This process leads to the
concentration of hot electrons at a higher energy valley. During the further increase in
the electric field, these hot electrons will be transferred to another even higher valley.
Thus, the hot electron energy will be concentrated and shifted to another higher
energy valley simultaneously. This tendency agrees with the distribution of density
of states shown in Fig.7. When the electric field is high, most of electrons are
evacuated from lower states and moved to higher energy states and the result
appears in Fig.9.
3. When the electric field strength increases, the electron energy increases
correspondingly. We observed a sloping energy increase, but when the field strength is
high enough to start inter-valley scattering, the energy is concentrated and we observed
a plateau. After the valley is fulfilled, the increase in the slope starts again. Thus we
obtain a stepwise growth in hot electron energy. This phenomenon was observed in
[15].
4.2 Choice of better acceleration material
The energy of hot electrons depends on the electric field and used materials. We have
compared different materials as the acceleration materials and then compared the SSCL of
a same emission layer. The acceleration materials used are Si
3
N
4
, SiO
2
, ZnS or complex
accelerated layer SiO
2
/ZnS respectively [21-23]. The structure of all devices is same as
ITO/accelerated layer/MEH-PPV/ accelerated layer/Al. In these devices, two
luminescence peaks corresponding to the exciton emission and the extended band
emission respectively are observed. The electron acceleration ability of SiO
2
is better than
that of ZnS and Si
3
N
4
, but ZnS is better than SiO
2
when the ability of injection charges (as
shown in Fig. 11 and 12) is compared. In Fig.11, under same bias, the intensity of short
wavelength of MEH-PPV in the device of ITO/ SiO
2
/MEH-PPV/SiO
2
/Al is stronger than
that of ITO/ZnS/MEH-PPV/ ZnS/Al. So it is concluded that the acceleration ability of
SiO
2
is better than that of ZnS. We investigated the complex accelerated layer SiO
2
/ZnS.
The results showed that under low voltage, the performance of the complex accelerated
layer is superior than that of SiO
2
and inferior than that of ZnS. Under high voltage, the
complex accelerated layer is the best among different devices. Thus, SiO
2
is the main
accelerated layer, and ZnS is very useful to improve the performance of electrons injection
of SSCL device.

350 400 450 500 550 600 650 700
0.0
0.2
0.4
0.6
0.8
1.0
H: ITO/ ZnS / SiO
2
/MEH-PPV/SiO
2
/ZnS/Al
I ITO/ SiO
2
/MEH-PPV/SiO
2
/Al
J ITO/ ZnS /MEH-PPV/ZnS/Al
I
n
t
e
n
s
i
t
y

(
a
.
u
.
)
Wavelength(nm)
H
I
J

Fig. 11. The contrast of three devices on AC 65V
-0.05 -0.04 -0.03 -0.02 -0.01 0.00 0.01 0.02
-60
-40
-20
0
20
40
60
80
100
V
o
l
t
a
g
e

/

V
Time / s

Voltage
H ITO/ ZnS / SiO
2
/MEH-PPV/SiO
2
/ ZnS/Al
I ITO/ SiO
2
/MEH-PPV/ SiO
2
/Al
J ITO/ ZnS /MEH-PPV/ ZnS/ Al
Voltage
H
I
J

Fig. 12. The waveform of derivative luminescence (405nm) with applied voltage
5. The prospect of SSCL
Concerning above discussion, it is concluded that SSCL have more potential in the display
field. The excitation mechanism of SSCL is superior to the organic electroluminescence. In
organic electroluminescence, the maximum of the internal quantum efficiency is not more
than 1/4 because of the limitation of the photon transition selection rule. After electrons and
holes injected to the organic layer, only 1/4 number of electrons and holes can relax to form
singlet excitons and give the light emission. But SSCL behaves nearly the photon excitation.
In SSCL excitation, more singlet excitons are formed. There are almost no triplet excitons.
Therefore, the efficiency of SSCL is better than that of pure OLED. The effort of raising
luminescence efficiency is proposed to carry out in 4 aspects:
1. Examination of factors influencing the electron acceleration ability, e.g., crystal
orientation of materials.
Using the Monte Carlo simulation method we have found out that the distribution of hot
electrons concentrates and moves towards to higher energy end along the increase of energy.
This situation is obviously benefit for raising the effectiveness of excitation. Another very
important result is that the energy of hot electrons increases with electric field not
harmonically but in stepwise way. This result coincides with experimental result of second
jump in light intensity at higher electric field. The improvement of acceleration ability of the
inorganic layer in the device should enhance SSCL. For example, if electrons transfer along the
crystal orientation of the inorganic layer, under electric field, electrons should be accelerated
easily.
2. Improvement of light intensity by enlarging the percentage of singlet exciton emission
of organic materials.
From the consideration of excitation mechanism, it is beneficial to integrate of SSCL and
OEL in the same device and enlarge singlet exciton formations.
3. Extension of complex display panel used in SSCL to FED
We published a paper on Optics Letters proposing the combination of electron source of
FED and the complex screen in SSCL to form an improved variant of FED, in order to avoid
the gas from the target under electron bombardment, to absorb the incoming high energy
electrons b y injecting holes and consequently to hold the vacuum of the device.
4. Determination of trap depth of local electron source.
The release of electron from traps is the main source of the primary electron in SSCL
devices. How to enhance the number of primary electrons? We investigated the trap depth
of electrons. In luminescence the method of determining trap depth by analyzing glow
curve with the assumption of kinetics order being 1 or 2 has been used more than 60 years,
we pointed the large error of their results and proposed a quasi-equilibrium method which
traces the real physical processes and provide an accurate result.

6. References
[1] Xu Xurong et al. Proc.of 7th international workshop on EL, Beijing, 1994: 42
[2] Zheng Xu, Chong Qu, Feng Teng, et al, Appl. Phys. Lett.,2005, 86(6): 061911
[3] Zheng Xu, Feng Teng, Chong Qu, et al , Physica B: Condensed Matter, 2004, 348(14):
231
[4] Suling Zhao, Zheng Xu, Fujun Zhang, et al., Opt. Lett. , 2007, 32:2094
[5] Xu XL, Xu Z, Hou YB, et al, J. Appl. Phys., 2001, 89 (2): 1082
[6] Xu XL, Chen XH, Hou YB, et al, Chem. Phys. Lett., 2000, 25 (4): 420
[7] Dong Qian, Zhao Suling, Xu Zheng, et al, Acta Physica Sinica, 2008, 57(12):7896
[8] K. Book, V.R. Nikitenko, H. Bssler , et al, Synthetic Metals, 2001, 122(1):135
[9] Xiaohui Yang1 and Xurong Xu, Appl. Phys. Lett. ,2000, 77: 797
[10] Fujun Zhang, Suling Zhao, Zheng Xu, et al, Chin. Phys., 2007,16(5): 1464
[11] Suling Zhao, Zheng Xu, Fujun Zhang, et al, J. Appl. Phys., 2009,106023513
[12] X. Xu and G. Lei, Acta Polytech. Scand. , 1990, 170: 133
[13] X. Xu, G. Lei, and M. Shen, J. Cryst. Growth, 1990, 101: 1004
[14] X. Xu, G. Lei, and Z. Xu, J. Cryst. Growth, 1991, 117: 925
[15] W. J. Wang, Y. B. Hou, C. H. Wei, et al, Proceedings of the 1994 International Workshop
on Electroluminescence Science, Beijing, 1994 : 206.
[16] Q. F. He, Z. Xu, F. Teng, et al, Chin. Phys. Lett. , 2006, 23: 701
[17] Q. F. He, Z. Xu, D. A. Liu, et al, Solid-State Electron., 2006, 50: 456
[18] G. Weireick, T. M. Sanders, and W. G. White, Phys. Rev. , 1959,114: 53
[19] M. Reigrotzki, K. F. Brennan, R. Wang, et al, J. Appl. Phys., 1995, 83: 1456
[20] M. Reigrotzki, R. Redmer, I. Lec, et al, SAP, 1996, 80: 5054
[21] Li Yuan, Zhao Suling, Xu Zheng, et al., Acta Physica Sinica, 2007, 56(09):5526
[22] Li Yuan, Zhao Suling, Xu Zheng, et al., Spectroscopy and spectral analysis, 2008, 28(9):
1974
[23] Kong Chao, Xu Zheng, Zhao Suling, et al, Acta Physica Sinica, 2008, 57(12): 7891
13
A Lighting Mechanism for Flat Electron
Emission Lamp
Jung-Yu Li et al.
*

Green Energy and Environment Research Laboratories,
Industrial Technology Research Institute, Hsinchu,
Taiwan
1. Introduction
A novel lighting device, flat electron emission lamp (FEEL), has been developed for generating
planar light. The mechanism of FEEL is different from the mainstream techniques for lighting
and display. The basic idea for FEEL is, for the first time, to integrate the discharge mechanisms
of fluorescent lamp at cathode with the cathodoluminescent effect at anode. In that the electrons
are induced by gas discharge, accelerated toward anode under electrical field, and finally
impacting the phosphor-coated anode to emit visible light. The light spectrum radiated from
FEEL is thus, dominated by the detailed composition of the phosphor coated on the anode.
Unlike most of the fluorescent lamps, ultraviolet light is unnecessary in this case, and therefore
there is no need to have mercury inside the device. By taking into account the human factors,
such as safety and health, the uniform lighting obtained in FEEL further offers a prominent
characteristic which prevents the eyes from the uncomfortable glare and drastically improves
the persistence of vision. In this article we also demonstrate various prototype potential
applications featured by FEEL, namely the double-side lighting, transparency, colored gray-
scale image and heat insulation, etc. It is indicative that FEEL not only could have broad
potential markets for normal lighting, but also might play a prominent role in power-saving
building material and gray-scale ambiance lighting. More importantly, the unique mechanism
prevailing in FEEL shows the high flexibility of expanding its applicability albeit it is still in the
early stage of laboratory studies. We believed that it may become one of the competing
candidates for the next generation green lighting in the near future.
2. Motivation of the technique development
In the nineteenth century, the new electrical lighting, incandescent lamp, was invented and had
totally changed the human behavior and social environment in the early stage of electrical
power generation. Hereafter many different kinds of lighting and display technologies were

*
Ming-Chung Liu
1
, Yi-Ping Lin
1
,

Shih-Pu Chen
1
,

Tai-Chiung Hsieh
1,2
, Po-Hung Wang
1
,
Chang-Lin Chiang
1,2
, Ming-Shan Jeng
1
, Li-Ling Lee
1
, Hui-Kai Zeng
3
and Jenh-Yih Juang
1,2
1
Green Energy and Environment Research Laboratories, Industrial Technology Research Institute, Chutung, Taiwan
2
Department of Electrophysics, National Chiao Tung University, Hsinchu, Taiwan
3
Department of Electronic Engineering, Chung Yuan Christian University, Chung Li, Taiwan

Cathodoluminescence

306
developed vigorously for illuminating every corner in the world. At present, the major branches
of lighting techniques can be catalogued into four kinds of mechanisms: (1) Fluorescent Lamps
(FL) (2) Cathodoluminescence (CL) (3) Solid-state lightings (SSL) (4) Gas-discharge lamps
(GDL). All of these technologies have been developing for several decades and spread into
broad range of daily applications. The maturity and wide-spread industrialization of these
techniques have, consequently, led the manufacturing of most lighting and/or display
equipments to be confined within the concepts derived from these mechanisms.
In this chapter, a novel lighting device, flat electron emission lamp (FEEL), for generating
uniform planar light with competitive power efficiency in an environmentally recuperative
manner is described. The lighting mechanism prevailing in FEEL is different from that
operating in the mainstream techniques, such as FL, CL, SSL, and GDL. The originality is
primarily based on the innovative integration of the working mechanisms of FL and CL. As
depicted schematically in Fig. 1, the lighting model of FEEL can be clearly interpreted by
dividing it into three steps: (1) Analogous to FL mechanism, the secondary electron emissions
are generated by ion bombardment at the cathode surface. It is noted that, in FEEL, the gas
pressure used to initiate the production of electrons is significantly lower than that used in FL,
being in the range of 10
-1
to 10
-2
torr. (2) The electrons are subsequently accelerated by the
electric field established between the cathode and the anode. During the course of flight, the
gas excitation and ionization are induced by electron collision with gas. The probability of
electron collision depends on the pressure, electron kinetic energy and gas type. (3) As an
analogy to the anode structure of CL, electrons impact with CL phosphor-coated anode and
transfer the kinetic energy to the emission of visible light. Evidently, over the whole process of
lighting in FEEL, ultraviolet light is unnecessary and therefore there is no need to have
mercury inside the device. Thus, the mechanism meets requirements for developing green
lamps by completely avoiding the pollution issues from the involvement of mercury.

Fig. 1. The lighting mechanism of FEEL is proposed by integrating the mechanisms
operating at the cathode of FL and the anode of CL. Briefly, the electrons are induced by
discharge in the vicinity of cathode, accelerated by electrical field, and finally impacts on the
phosphor-coated anode, which gives rise to the eventual light emission by transferring the
electron kinetic energy into phosphorous radiations. The combination is advantageous in
generating large area planar electron beams for obtaining uniform photon distributions over
the entire CL phosphor screen.

A Lighting Mechanism for Flat Electron Emission Lamp

307
So far FL is the most mature technology for general daily lighting, while CL has been
ubiquitously used in the long history of the display technology. For instance, the fluorescent
tubes and plasma display panels (PDP) belong to the field of FL mechanism, while the cathode
ray tube (CRT), field emission display (FED), and field emission lamp (FEL) are all based on
CL mechanism. It is interesting to note that over the long history of development in the two
respective fields there was essentially no overlap between FL and CL technologies. In fact, they
are working at very different gas pressure conditions. From the required conditions of FL, it is
necessary to infuse a sufficient amount of gas (in the range of 100 torr) in order to sustain the
stable glow discharge; it is also necessary to further produce high enough ultraviolet intensity
during the discharge process to maintain photoluminescence effect for obtaining the visible
light radiated from photoluminescence phosphor. On the contrary, in the case of CL, high
vacuum environment (< 10
-6
torr) is essential for reducing energy dissipation from electron
scattering. The suitable vacuum pressure conditions for FL and CL can be different by eight
orders magnitude or even larger. As will be discussed in more details later, the working
pressure in the current FEEL device is in a more compromised condition, which is in the
middle range between the pressure of FL and CL, and, hence exhibiting some unique features
that were not fully explored previously.
In contrast to the traditional line or point shaped lighting sources, the planar lighting
sources with the uniform light have the important advantage of no glare, thus can create a
comfortable lighting environment. The mercury-free field emission-like (FEL) lighting
technology has been one of the potential candidates for the planar lighting or backlighting
applications. Unfortunately, the random dark points constantly appearing in the lighting
screen of FEL have severely hindered its market development. Usually, an additional
diffuser covering the entire the device surface to obtain uniform light emission is required.
However, the extra diffuser also reduces the effective lighting intensity of the lamp. Thus,
either from the consideration of cost down or from the view point of energy-saving, the
issue of dark points has become one of the technology bottlenecks for FEL. The primary
source of such non-uniform light emission is attributed to the grown-in disorders during the
growth of nanoscale electron emitters. For instance, one of the most favored materials for
electron emitter, the carbon nanotubes (CNTs), turned out to be very difficult to control the
uniformity of nanotube length and alignment over large area by the processes of screen
printing or direct growth. To this respect, the gas discharge and emitter-free characteristic of
FEEL provides an alternative solution for easily generating planar beam-like electrons from
cathode. As an example, the lighting screen shown in Fig. 2 for a typical FEEL device
evidently demonstrates the complete elimination of the dark points. In addition, it also
avoids the unfavorable cost and process time spent in growing the electron emitter material.
Nevertheless, it should be emphasized that making use of gas discharge in unconventionally
low pressure regime is the first attempt for the development of cathodoluminescence
technology and further understandings of the detailed mechanisms are certainly needed.
3. The lighting mechanism
In order to investigate the intrinsic discharge property of FEEL, an experimental setup
was established for simultaneously measuring current-voltage (I-V) curves, Paschen
curves, luminance and optical emission spectroscopy (OES) of FEEL device. As illustrated
in Fig. 3, the experimental FEEL device consists of a cathode glass, a glass spacer, and a

Cathodoluminescence

308
CL phosphor-coated anode glass. The electrode glasses were deposited with a transparent
conducting oxide film (fluorine-doped tin-oxide/FTO). The electrode and spacer were
sealed by glass glue with infused gas as the working gas. The height of glass spacer was
10 mm and size of lighting screen (phosphor surface) was 30 mm x 30 mm. A glass tube
connected the glass spacer and the vacuum system for evacuating and adjusting pressure
inside the device. A DC voltage power (Keithley 248) and a picoammeter (Keithley 6485)
were used for analyzing the discharge and electrical behavior; a Multi-function Color
Analyzer (Ruyico Tech.) and an OES (optical emission spectroscopy) probe (StekkarNet
EPP200V) were positioned on top of device for measuring the optical properties. The OES
spectrum was measured to further reveal the correlated behaviors between glow
excitations and the lighting mechanism of FEEL devices. To obtain the optimal conditions
for FEEL lighting, the gas pressure inside the device was gradually decreased and the
transitions of the lighting mode were simultaneously analyzed by observing the images of
the lighting screen.

Fig. 2. The photo shows the lighting screen of a typical FEEL device demonstrating the
uniform light radiation and absence of dark points ubiquitously encountered in traditional
FED devices.


309

Fig. 3. The schematics of the FEEL device structure and the experimental setup for
measuring the lighting and discharge behaviors. The device consists of cathode, glass
spacer, gas, and CL phosphor-coated anode. The pressure inside the device can be adjusted
via the vacuum system. The experimental setup for simultaneously measuring I-V curves,
Paschen curves, luminance and OES spectrum enabled the revelation of correlations
between discharge and lighting mechanism of FEEL.
Figure 4 depicts the prevailing mechanisms in different gas pressure regimes. With an
applied DC voltage of 5 kV, which ensures the turn-on status for FEEL, the top-view and
side-view images of the device show that the glow discharge features (Fig. 4A) gradually
transform to the FEEL mechanism (Fig. 4C) as the N
2
pressure is approaching to an
optimized condition. To further confirm the predominant light source(s) in each gas
pressure regime the OES spectra are also displayed in Fig. 4. It is clear that only glow
excitation is active and the phosphor coated on the anode does not emit light when the N
2

pressure is higher than 0.2 torr (Fig. 4A), consistent with whats seen in the images. The four
distinguished peaks identified in the OES spectrum with the wavelength respectively
locating at 337, 358, 391.5 and 427.5 nm are originated from the nitrogen gas excitations
belonging to the first negative system (B
2
u
+
X
2
g
+
) (358.2, 391.4 and 427.8 nm) and the
second positive system (C
3
u
B
3
g
) (337.1 nm). In this case, obviously the kinetic energy of
electrons is too low to activate the emission from the phosphor, presumably due to the
frequent electron scatterings by collidings with gas molecules in such high gas pressure
environment. We note that the frequent electron-gas scatterings are exactly responsible for
the discharge glow as what one would expect.

Cathodoluminescence

310

Fig. 4. The lighting mode transitions as a function of gas pressure inside the FEEL devices
with a fixed applied voltage of 5 kV. Both the device images and OES spectra show the close
interplay between the scattering-induced glow discharge and the prevailing of electron
emission lighting (APL, Vol.94, 091501, 2009). Typically, four pressure regimes within which
different dominant lighting mechanisms can be clearly identified:
(A) (P
N2
> 0.2 torr): predominantly nitrogen excitation associated glow discharge.
(B) (0.2 torr> P
N2
>0.11 torr): coexistence of nitrogen glow and phosphor lighting at the
anode observed.
(C) (0.11 torr > P
N2
> 0.1 torr): predominantly phosphor lighting due to the FEEL
mechanism.
(D) (P
N2
< 0.1 torr): no lighting mechanism is active.
As the N
2
pressure is decreased to within the range between 0.2 torr to 0.11 torr (Fig. 4B), it
is clear from the side-view image that phosphor emission starts to emerge at the expense of
the attenuated glow background. The OES spectrum evidently shows the weakened peak
intensity from the N
2
excitation in this pressure range, with an apparent peak center at the
wavelength of 530 nm, which is the characteristic wavelength of the phosphor used in this
study. We believe that, in this lower pressure range, the dissipation of electron kinetic
energy will be less than that in the previous case and there will be some population of
electrons with residual kinetic energy higher than the threshold energy for activating
phosphor emission. In this regime, nonetheless, the gas excitation glow and phosphor
emission are coexistent, indicating that the predominant mechanisms operating in the
system are strongly dependent on the gas pressure and, in fact, controllable. As the N
2

pressure is further decreased to the range of 0.11 torr to 0.1 torr (Fig. 4C), the OES spectrum
displays only a strong peak at 530 nm, reflecting the green light emitted from the phosphor
coated at the anode. Furthermore, the characteristic peaks of the glow excitation of N
2
gas
disappear altogether under this condition. We refer this state as the FEEL state because it


311
appears to be the dominant lighting mechanism. It is noticed that, although the pressure
(0.11 torr) at which the system switches from the glow/FEEL coexistent state to the pure
FEEL state (Fig. 4B and Fig. 4C) appears to be somewhat arbitrary, the existence of even a
slight portion of glow can affect the luminance of the device dramatically. It suggests that
the electron-gas scatterings during the flight journey from cathode to anode can have
significant effect on the luminance efficiency.
Lastly, as the pressure of N
2
is below P = 0.1 torr, neither any lighting is seen in the images
nor any peak is revealed in the OES spectrum. It indicates that the glow excitation of N
2
gas
and lighting from phosphor emission are all not active under this condition (Fig. 4D).
Intuitively, one would expect that at this low pressure condition, the device should be
operating similarly to the CL mode prevailing in FED. However, we note that, for FED
applications, usually the turn-on electrical field of cathode emitter is well-above order of
V/m as indicated by the Fowler-Nordheim (F-N) theory. Consequently, recent
development of the cathode emitter for FED was mainly focused to lowering the turn-on
electrical field. Indeed, the CNT with its advantageous high aspect ratio has successfully
reduced the turn-on field to E = 0.5 ~ 1V/m. In our case, the estimated electrical field is
about 0.5 V/m, near the turn-on field of CNT. It implies that by combining the glow
discharge effect, the present FEEL device has the equivalent effect of CNT field emission.
Nevertheless, without glow discharge, there will be no electron because the applied external
field is not enough to trigger the FED mechanism, leading to virtually no response in the
FEEL device when operated at low gas pressure regime (Fig. 4D).
In order to further clarify the correlations between the characteristics of luminance and glow
excitation in a more quantitative manner, the current density and luminance were analyzed
in the pressure range of FEEL and glow/FEEL states. Figure 5 shows the current density
and the luminance as a function of gas pressure in the FEEL device operated at an applied
dc voltage of 5 kV. It is evident that the maximum luminance is obtained at the same
threshold pressure of 0.11 torr, in accordance with the glow excitation effect on the device
luminance described above. In addition, the current densities of FEEL state are always lower
than those obtained in the glow/FEEL state. It implies that, under the same luminance
condition, the luminance contribution per unit current density in the FEEL state is always
higher than that in the glow/FEEL state. It also indicates that high degree of glow excitation,
which strongly correlates with electron scattering, in fact diminishes luminance
performance of the FEEL device. The significant influence from gas excitation in reducing
the device luminance is believed to arise from the dissipation of electron kinetic energy,
which, in turn, leads to inefficient phosphor excitations.
Another set of systemic measurements that might further help in understanding the intrinsic
properties of the FEEL devices are the current-voltage (I-V) and luminance-current
characteristics of the devices. Figure 6A shows the I-V curves of the FEEL and glow/FEEL
states operated over a wide range of gas pressures. The behaviors display nothing more
than the usual electric breakdown behavior expected from the normal discharge theory.
When breakdown occurs, it indicates that light is turning on and the static voltage drops.
Since the output voltage of the power supply is higher than the breakdown voltage (V
b
), the
sustain voltage (V
sustain
) of the device remains the nearly constant, albeit the power supply
voltage increases continuously to enhance current density. Also, when the FEEL device is in
the lighting state, the system shows features of self-sustain discharge and an apparent

Cathodoluminescence

312
negative resistance. It is evident that FEEL works after breakdown occurs, a behavior akin to
normal glow discharge. The results also indicate a tendency of lower V
b
and V
sustain
for
higher pressure devices. It suggests that the operation window for FEEL and glow/FEEL
states are both located in the left hand side of the Paschen curve, which are sensitively
dependent on parameters such as the gas type, cathode material, pressure condition and
discharge gap width.

Fig. 5. The current density and luminance as a function of gas pressure inside the FEEL
device. A close correlation between luminance efficiency and glow excitation is observed.
The characteristic threshold pressure of 0.11 torr distinguishing the regimes of FEEL and
glow/FEEL states is consistent with the observations shown in Figure 4.


313
Figure 6B shows the characteristics of luminance-current density for FEEL devices operated
over the same gas pressure range. It is noted that, except the apparent saturation behaviors
occurring for the two samples with the lowest gas pressures in FEEL state, there appears to
exist a linear relation between luminance and current density in glow/FEEL state, albeit the
slope is a function of gas pressure. To understand these, we consider the energy budget of
the electrons under discussion. At anode, the maximum arrival kinetic energy for electrons
emitted from the cathode by ion bombarding will be the summation of the initial kinetic
energy and the acceleration energy from electrical field. However, over the course of flight,
the electrons may collide with the gas molecules and loss their kinetic energies. The glow
excitation observed within the pressure region of 0.11-0.25 torr shown in Fig. 4 is primarily
due to these dynamical interactions. Also, as shown in Fig. 5, although the device current
remains essentially constant after the luminance reaches its maximum at P
N2
= 0.11 torr,
further increase in device pressure does cause a reducing device luminance. Combining
with the observation of the linear dependence of luminance on current density under a fixed
device pressure (Fig. 6B), it indicates that, although the photon energy density (luminance)
of the device is linearly related to the electron number density arriving at the anode per unit
time, the internal quantum efficiency of the phosphor is very much dependent on the
effective kinetic energy carried by the electrons. Consequently, a large amount of low kinetic
energy electrons do not guarantee efficient luminance in the glow/FEEL state. On the other
hand, as shown in Fig. 6B, for the low pressure cases even though the number of electrons
(current density) is much lower with their high kinetic energy, high luminance can be
achieved in the FEEL state. It appears that the slopes of luminance versus current density
curves shown in Fig. 6B can be regarded as the effective kinetic energy of the electrons and
the electron kinetic energy is the dominant factor to determine the final luminance in FEEL
mechanism. Finally, the apparent saturation behaviors for the low pressure cases of the
FEEL state might be due to the thermal quench inside the phosphor, which in turn prevents
further increase of device luminance from electron kinetic energy transferring.

Fig. 6. (A) The I-V characteristics and (B) current density-luminance correlations in the FEEL
and glow/FEEL states of the device (APL, Vol.94, 091501, 2009). The I-V curves indicate that
FEEL emits light after electric breakdown. As power supply voltage is higher than breakdown
voltage, the device voltage remains nearly constant although the power supply voltage
continuously increases to enhance current density. (B) Luminance versus current reveals that
luminance of FEEL is linearly related to current density after electric breakdown.

Cathodoluminescence

314
Finally, we discuss the variations of the breakdown voltage V
b
near the operation conditions
of FEEL devices by measuring the Paschen curve of the device. As shown in Fig. 7, the
Paschen curve demonstrates that regimes of FEEL and glow/FEEL are indeed operating in
the lower pressure regime locating on the left hand side of the Paschen curve. As a result,
the steep change in V
b
within a slim region of pressure change is currently a major hurdle to
be dealt with. Improvements by using different working gases as well as cathode materials
with better secondary electron emissions are currently under extensive investigations.

Fig. 7. The Paschen curve of FEEL devices with N
2
as the working gas. The areas highlighted
by the dark-yellow and light-yellow colors respectively indicate the operation regime of the
FEEL and glow/FEEL states in the devices.
4. Energy-saving issues and applicability of FEEL
As described above, the lighting mechanism operating in FEEL devices can be conceptually
regarded as a combination of gas discharge working for FL devices with CL phosphor
lighting. Such combination, nevertheless, has been proven to be feasible in generating large
area planar electron beams which successfully eliminated the dark spots on the phosphor
screen by providing uniformly distributed excitations over the entire anode surface. Other
equally important and extremely crucial issues in general lighting are the considerations of
health and safety, namely how to protect human eyes and skin from long-time exposure to
the lighting environment. To take into account these human factors, the planar light source
which provides the uniform light emission, can naturally support the sufficient amount of
illumination with reduced lighting intensity to avoid of the glare and persistence of vision
from lamp. Thus, the unique planar feature of FEEL nicely meets the requirement of healthy
human factors. To further compare with the traditional planar lamps with diffusers, which


315
are composed of spot or line shaped light sources inside the lamp structure, FEEL doesnt
need extra diffuser to obtain uniform planar lighting from originally concentrated radiation
sources. Hence, without energy dissipation from diffuser, FEEL doesnt need to increase the
power to obtain higher radiated intensity for compensating the energy loss from diffuser. It
means that even total illumination and input power of the FEEL are reduced for energy
saving, it still simultaneously matches the requirements of specification and lighting
environment. Obviously, lower material cost and power consumption become the
production advantages for FEEL. Besides, to consider the heat extraction issue, FEEL is
similar to the planar heat source with low heat density. Without heat concentration, it also
doesnt need special device design for heat extraction.
In addition to planar lighting market, FEEL has high potential to expand its applicability by
modifying the device design. We find that, especially in the FEEL anode, there are several
methods to adjust the styles of light emission through modifying the phosphor pattern. In
the following demonstrations, we illustrate that FEEL not only can be applied to planar
lamp, but also can be designed as many kinds of unique lighting productions. For instance,
the prototypes of transparent lighting glass, gray-scale image lighting, and colorful image
lighting have all been developed.
Figure 8 individually reveals the power turned-off and turned-on status of transparent FEEL
device. It shows that the FEEL is transparent when its power is turned off (Fig. 8A). In Fig.
8B, the phosphor coating on the FEEL anode radiates the visible light when it is turned on
the power. Due to the influence of the light emitted from phosphor, at the turn-on status the
transparency of the device is reduced to a lower level. For transparency purpose, in this case
transparent FTO films were used as the electrodes of the device. The only opaque material,
the phosphor coating on the anode surface, was designed as regular dot-matrix pattern for
light transparency. The area of the dot pattern is around 100 m, so the detailed structure of
the pattern cannot be differentiated by human eyes. The device transparency could be
evaluated by the coverage of phosphor on the whole anode surface. The maximum device
transparency is about 60%, which is limited by the transparency of transparent electrodes
and total coverage of phosphor. It means that FEEL can be a lighting window to receive the
extra natural light in the daytime-indoor, and provides the indoor lighting at night.
Additionally, we found that the isolated glass structure and a very small amount of gas
inside the 4 inch FEEL device naturally give rise to the property of low heat conductivity
(0.07 W/mK), which is about 20% of the heat conductivity of a 4 inch normal glass (1.4
W/mK). The thermal conductivity of the device was measured by thermal conductivity
meter (EKO HC-072). It is interesting to note that FEEL is the only lighting technology to
have the property of heat isolation. It is very useful for the windows of passive house to
prevent the heat from transferring through the FEEL window. In other words, it can
preserve the thermal energy inside the house in winter and avoid the outside heat from
transferring into the indoor in summer.
It is clear from the above description that FEEL does have the potential of being
simultaneously used as the energy-saving thermal insulation window, daylighting system,
and indoor lighting in sustainable passive house for saving the power consumption of air-
condition in summer, indoor heating in winter or daytime-indoor lighting (Fig. 9). In
particular, with the unique features of transparency and heat insulation, it can also meet the
demand of the sustainable building material. It may be anticipated that the buildings

Cathodoluminescence

316

Fig. 8. The photo of a transparent FEEL. (A) It shows that the lamp is transparent when
power is turned off. The phosphor dotted matrix can not be differentiated because of the
size of dotted patterns is only around 100 m. (B) The phosphor coating on the anode of
FEEL radiates the visible light when its power is turned on. It also reveals the partial
transparent effect while seeing through the device.


317

Fig. 9. Schematics demonstrating that it is possible to use FEEL to simultaneously
accomplish the energy-saving thermal insulation glass, daylighting system, and indoor
lighting in sustainable passive house for saving the power consumption of air-condition in
summer, indoor heating in winter or day time indoor lighting.
skylights and elevation windows are the potential applications for green building markets.
Furthermore, the buildings owners are able to independently control the turn-on status for
every FEEL window on the building elevation. If the elevated walls of the building are
analogy to the display screens, then the FEEL glasses can be represented as the display
pixels, which are named as window pixels in this case (Fig. 10). We believe that, the linkage
between lighting and green building material, will change the business model and the users
habits in the lighting market. However, we expect that the installation and material cost for
FEEL will be higher than the normal glass, but the extra cost can be refunded from the
power consumption saving generated from daily operations as described above. In addition,
at night, it is possible to utilize the FEEL windows in commercial mansions to be rewarded
for the advertisement applications such as show of the companys logo and instantaneous
information.

Cathodoluminescence

318

Fig. 10. FEEL can be applied to building skylights and building elevation windows. The
building elevation looks like a large size display when every FEEL glass is independently
controlled the turn-on status.
The rectangle shape is not the limit for FEEL technology. As shown in Fig. 11, the flat FEEL
bulb, which challenges the arbitrary shape capability of FEEL, subverts the normal
impressions of ball shaped incandescent bulb. It also shows the special lighting effect after
phosphor is printed with designed pattern on anode surface.


319

Fig. 11. The flat FEEL bulb subverts the normal impressions of ball shaped incandescent
bulb.
As illustrated in Fig. 12, the radiated gray-scale image on FEEL anode screen is another
unusual demonstration by special screen printing design for phosphor. The way for
revealing the static gray-scale image by FEEL is making a big difference from display panel.
On normal display, the screen is composed of dotted matrix pixels. Every individual pixel
radiates its own correspondence intensity for revealing the gray-scale effect. On FEEL
lighting surface, however, there are no regular pixel matrix and external control circuit.
Instead, it utilizes the area of individual dotted phosphor pattern to form the corresponding
radiated intensity under the same power condition. It means that the radiated intensities are
linear relation to the areas of dotted phosphor patterns. Similar to the case of transparent
device, the areas of the phosphor patterns are too small to differentiate the patterns
structure by human eyes. By this design rule, any digital picture or document content can be
transformed to the phosphor patterns with correspondent areas for reproducing the
radiated image. Thus, FEEL should be able to radiate the high resolution image with gray-
scale. It is interesting to note that the static lighting image from FEEL is always confused
with display image. Although it is only a lamp, it indeed provides unlimited imagination for
display applications.

Cathodoluminescence

320
In Fig. 13, we further extend the gray-scale function to show the possibility of using FEEL
for colorful image demonstrations. Basically, colorful radiation is linear superposition of
three original colors (Red, Green and Blue). It means that any color image can be
decomposed into three individual images with original color. As described before, these
individual RGB images are further transformed into dotted phosphor patterns in
correspondent areas to produce the gray-scale image with single color. At last, RGB
phosphor patterns are overlapped (multi-layers) printing on the same anode of FEEL to
reveal the color image. Once again, its image resolution looks like the static color picture in
display, although there is no display driver and pixel structure inside the device.

Fig. 12. FEEL radiates the precise image with gray-scale. It provides the imagination for
display applications.


321

Fig. 13. FEEL shows the capability of colorful image demonstration by the overlapped
printing of RGB phosphor patterns.

Cathodoluminescence

322
5. Conclusion and remarks
The lighting mechanism for FEEL is based on the innovative integration from mature
theories, such as gas discharge and cathodoluminescence. It has been indicated that the
working range for FEEL may obey the rule of normal glow discharge, with the features of
voltage drop and negative resistance. The measurements of I-V curves, Paschen curve,
luminance and OES show that the electron kinetic energy is the dominant factor to
determine the final luminance efficiency. The first priority to enhance the luminance
efficiency of FEEL is to reduce the electron scattering dissipation by decreasing gas pressure.
However, it conflicts with the requirement for stable discharge. In general, it needs sufficient
amount of gas for maintaining self-sustain discharge, or higher gas pressure means lower
breakdown voltage. The solution to solve such dilemma between low gas pressure and
efficient discharge is to improve the secondary electron emission efficiency of cathode
material under the low gas pressure environment. It is noticed that the suitable gas type
should be carefully selected to match with cathode material for getting higher discharge
efficiency. Besides, the optimized phosphor composition and printing process need to be
further investigated for obtaining higher quantum efficiency of phosphor. So far, FEEL
mechanism shows the high flexibility and potential for applicability in the early stage. It
already presents the potential to expand applicability for normal lighting, power-saving
building material and gray-scale ambiance lighting. We believe that FEEL could be the
candidates of next generation green lighting for providing the comfortable lighting, power-
saving, and ambiance applications.
6. Acknowledgments
The authors would like to acknowledge the support from the Energy Fund of Ministry of
Economics Affairs, Taiwan. The authors would also like to acknowledge the support
from the Energy Foundation. JYJ is supported partially by National Science Council of
Taiwan and by MOE-ATU program operated at National Chiao Tung University
(NCTU).
7. References
Jung-Yu Li, Shih-Pu Chen, Chia-Hung Li, Yi-Ping Lin, Yen-I Chou, Ming-Chung Liu, Po-
Hung Wang, Hui-Kai Zeng, Tai-Chiung Hsieh, and Jenh-Yih Juang (2009). A
lighting mechanism for flat electron emission lamp, APPLIED PHYSICS LETTERS,
Vol.94, pp.091501-1-091501-3, Taiwan
Chia-Hung Li, Ming-Chung Liu, Chang-Lin Chiang, Jung-Yu Li, Shih-Pu Chen, Tai-
Chiung Hsieh, Yen-I Chou, Yi-Ping Lin, Po-Hung Wang, Ming-Shin Chun, Hui-
Kai Zeng, and Jenh-Yih Juang (2011). Discharge and photo-luminance properties
of a parallel plates electron emission lighting device, Optics Express, Vol.19,
pp.A51-A56, Taiwan
T. Jstel, H. Nikol, C. Ronda (1998). New Developments in the Field of Luminescent
Materials for Lighting and Displays, Angew. Chem. Int. Ed, Vol.37, pp. 3084-3103,
Germany


323
M. IImer, R. Lecheler, H. Schweizer, M. Seibold (2000). Hg-free Flat Panel Light Source
PLANON - a Promising Candidate for Future LCD Backlights, SID International
Symposium, Digest of Technical Papers, XXXI, pp. 931-933, Germany
A. A. Talin, K. A. Dean, J. E. Jaskie (2001). Field emission from carbon nanotubes: the first
five years, Solid-State Electronics, Vol.45, pp. 893-914, Switzerland
U. Kogelschatz (2004), Excimer Lamps: History, Discharge Physics, and industrial Applications
Proceedings of SPIE, Vol.5483, pp. 272-286, Switzerland
G. G. Lister, J. E. Lawler, W. P. Lapatovich, V. A. Godyak (2004). The physics of discharge
lamps, Rev. Mod. Phys., Vol. 76, pp. 541-598, USA
J. P. Boeuf (2003). Plasma display panels: physics, recent developments and key issues, J.
Phys. D: Appl. Phys., Vol.36, R53-79, France
M. Itoh, L. Ozawa (2006). Cathodoluminescent phosphors, Annu. Rep. Prog. Chem., Sect. C,
Vol.102, pp. 12-42, Japan
Y. X. Liu, J. H. Liu, C. C. Zhu (2009). Flame synthesis of carbon nanotubes for panel field
emission lamp, Applied surface science, Vol.255, pp.7985-7989, China
R. W. B. Pearse and A. G. Gaydon (1965). The Identification of Molecular Spectra,3rd ed.,
Chapman and Hall, pp. 209220, London
Y. Saito and S. Uemura (2000). Field emission from carbon nanotubes and its application to
electron sources, Carbon, Vol.38, pp.169-182, Japan
J.-M. Bonard, H. Kind, T. Stckli, and L.-O. Nilsson (2001). Field emission from carbon
nanotubes: the first five years, Solid-State Electron, Vol.45, pp.893-914,
Switzerland
M. J. Druyvesteyn, F. M. Penning (1940). The mechanism of electrical discharges in gases of
low pressure , Reviews of Modern Physics, Vol.12, pp.87-174, Holland
M. M Pejovi, Goran S Risti, J. P Karamarkovi (2002). Electrical breakdown in low pressure
gases, J. Phys. D: Appl. Phys., Vol.35, R91-103, Yugoslavia
G. Cho, J. Y. Lee, Dae H. Lee, S. B. Kim, H. S. Song, J. Koo, B. S. Kim, J. G. Kang, E. H. Choi,
U. W. Lee, S. C. Yang, J. P. Verboncoeur (2005). Glow Discharge in the External
Electrode Fluorescent Lamp, IEEE Transactions On Plasma Science, Vol.33, pp.1410-
1415, Korea
J. Reece Roth (1995). Industrial Plasma Engineering-V.1 : Principle, Institute of Physics, p.353,
London
Yuri P. Raizer (1991). Gas discharge physics, Springer-Verlag Berlin Heidelberg, pp.134-135,
New York
A. Miller (1988). Conflguration-independent minimum voltage for the Townsend model of
breakdown in gases, J. Appl. Phys., Vol.63, pp.665-667, New Jersey
Nikolai N. Chubun, Andrei G. Chakhovskoi, Charles E. Hunt (2003). Efficiency of
cathodoluminescent phosphors for a field-emission light source application, J. Vac.
Sci. Technol. B, Vol.21, pp.1618-1621, California
J. J. Rocca, J. D. Meyer, M. R. Farrell, G. J. Collins (1984). Glow-discharge-created electron
beams: Cathode materials, electron gun disigns, and technological applications, J.
Appl. Phys., Vol.56, pp.790-797, Colorado

Cathodoluminescence

324
A.P. Bokhan, P.A. Bokhan and D.E. Zakrevsky (2005). Peculiarities of electron emission
from the cathode in an abnormal glow discharge, Appl. Phys. Lett., Vol.86, pp.1-3,
Russia
P. Hartmann, H. Matsuo, Y. Ohtsuka, M. Fukao, M. Kando, and Z. Donko (2003). Heavy-
particle hybrid simulation of a high-voltage glow discharge in helium, Jpn. J. Appl.
Phys., Vol.42, pp.3633-3640, Hungary

Cathodoluminescence

Uploaded by

Copyright:

Available Formats

Cathodoluminescence

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Cathodoluminescence

Uploaded by

Copyright:

Available Formats

CATHODOLUMINESCENCE

Edited by Naoki Yamamoto

is the concentration of ionized acceptors, EA(0) is the ionization energy at very

were estimated as described above and

[Solomonov et. al., 2011].

taking into account the occurrence of two types

101 additional ones

centers (two oxygen vacancies with three electrons) and

(n = 0, 1, 2, 3, and 4), where

) (Khriachtchev et al., 2002; Liu et al.,

state within t. Because the variation of

You might also like