2nd semester, 2022

Birck Nanotechnology Center

Surface Characterization:
X-ray Photoelectron Spectroscopy,
Auger Electron Spectroscopy

Dmitry Zemlyanov
Senior Scientist
Birck Nanotechnology Center, Purdue University
[email protected]

2nd semester, 2022 ECE 658, Purdue University, Dmitry Zemlyanov 1

Birck Nanotechnology Center

• Introduction: history, terminology, types of samples, etc.

2nd semester, 2022 ECE 658, Purdue University, Dmitry Zemlyanov 2

ECE658, Purdue University, D. Zemlyanov 1

 2nd semester, 2022

Introduction: history, terminology, types of samples

• Hertz (1880) – Spark enhancement

• Hallwachs (1888) – Negatively
charge Zn plate discharged
• J.J. Tomson (1899) – Light
induced electron emission
• Einstein (1905) – Photoelectric
effect explained
(Nobel prize rewarded in 1921)
• Steinhardt and Serfass (1951) –
The process of using photons (light) to Photoemission was applied as
remove electrons from a bulk material analytic tool (ESCA)
is called photoemission. (Nobel prize rewarded in 1981)

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 3

Introduction: history, terminology, types of samples

X-ray photoelectron spectroscopy (XPS)* is a quantitative

spectroscopic technique (typically non-destructive) which can provide:
• the surface elemental composition and empirical formula (top 1–10
nm usually);
• chemical states and electronic states of the elements uniformity of
elemental composition across the surface (or line profiling or
mapping)
• uniformity of elemental composition as a function of ion beam
etching (or depth profiling)
XPS spectra are obtained by irradiating a material with a X-ray beam
while measuring the kinetic energy and number of electrons that are
emitted.

XPS requires ultra high vacuum (UHV) conditions!

* XPS is also called ESCA: Electron Spectroscopy for Chemical Analysis

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 4

ECE658, Purdue University, D. Zemlyanov 2

 2nd semester, 2022

Introduction: history, terminology, types of samples

XPS detects all elements with an atomic number, Z, of 3 (lithium) and

above.
• Neither hydrogen (Z = 1) nor helium (Z = 2) can be detect because
their atomic orditals are too small, resulting in negligible Relative
Sensitivity Factor.
Detection limits for most of the elements are in the parts per thousand
range (in the range of 0.1 at.%). Ppm (10-4 at.%) level detection is
possible, but it is typically required very long collection time
(overnight).

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 5

Introduction: terminology, types of samples

Birck Nanotechnology Center

Type of samples:any vacuum compatible materials (no gas or liquid).*

Materials routinely analyzed by XPS

Inorganic compounds, metal alloys, semiconductors, polymers, pure elements,
catalysts, glasses, ceramics, paints, papers, inks, woods, plant parts, make-up,
teeth, bones, human implants, biomaterials, viscous oils, glues, ion modified
materials
Organic chemicals are not routinely analyzed by XPS because they are readily degraded by
either the energy of the X-rays or the heat from non-monochromatic X-ray sources.

* A specially designed high pressure XPS instruments

can analyze gas and liquid ( see for instance: “ESCA
applied to free molecules” by K. Siegbahn, et.al,
American Elsevier Publishing Company, Inc., New
York, 1967

2nd semester, 2022 ECE 658, Purdue University, Dmitry Zemlyanov 6

ECE658, Purdue University, D. Zemlyanov 3

 2nd semester, 2022

Introduction: terminology, types of samples

Energy of photons and photoelectrons: electron Volt, eV
1 eV – 1.60×10-19 J eV-1
Intensity: count per second, cps

The energy of a photon is given by the Einstein relation : E = hν

h - Planck constant ( 6.62 x 10-34 J s ) ; ν- frequency (Hz) of the radiation.
Frequency of 1 eV photon is 2.4×1015 Hz ; for 1000 eV photon – 2.4×1018 Hz .
Wave length of 1 eV photon is 1.23 μm, for 1000 eV photon – 1.2 nm.

• X-ray Photoelectron Spectroscopy (XPS)

using soft x-ray (50-5000 eV) radiation to examine core-levels.

• Ultraviolet Photoelectron Spectroscopy (UPS)

using vacuum UV (10-50 eV) radiation to examine valence levels.

• Auger Electron Spectroscopy (AES)

using Auger electrons to determine element composition.

2nd semester, 2022 ECE 658, Purdue University, Dmitry Zemlyanov 7

Recommended Literature:

XPS hand books

1. Surface Analysis by Auger and X-ray Photoelectron Spectroscopy, ed. J. T. Grant and D. Briggs,
published by IM Publications, 2003, Chichester, UK
2. Practical Surface Analysis by Auger and X-ray Photoelectron Spectroscopy, 2nd edition, ed. M. P.
Seah and D. Briggs, published by Wiley & Sons, 1992, Chichester, UK
3. Practical Surface Analysis by Auger and X-ray Photoelectron Spectroscopy, ed. M. P. Seah and D.
Briggs, published by Wiley & Sons, 1983, Chichester, UK
4. Handbook of X-ray Photoelectron Spectroscopy, J. F. Moulder, W. F. Stickle, P. E. Sobol, and K.
D. Bomben, published by Perkin-Elmer Corp., 1992, Eden Prairie, MN, USA
5. Handbook of X-ray Photoelectron Spectroscopy, C. D. Wagner, W. M. Riggs, L. E. Davis, J. F.
Moulder, and G. E. Mullenberg, published by Perkin-Elmer Corp., 1979, Eden Prairie, MN, USA
Recommended web resources (UHV and XPS):
1. https://xpssimplified.com
2. https://srdata.nist.gov/xps/
3. http://www.chem.qmul.ac.uk/surfaces/scc/
4. http://www.nottingham.ac.uk/~ppzpjm/amshome.htm
5. http://venables.asu.edu/grad/lectures.html
6. http://www.cem.msu.edu/%7Ecem924sg/LectureNotes.html
7. http://matsci.uah.edu/CourseWare/mts723/lectures
8. http://nanohub.org/members/19703/contributions
2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 8

ECE658, Purdue University, D. Zemlyanov 4

 2nd semester, 2022

Basic Principles of XPS:

photoemission and production of photoelectrons, binding energy

Energy

!w
EB = hν − EK − ϕ
1s 2s 2p EK (The Einstein equation)
φ is a work function of a spectrometer
Vacuum level
j
Fermi edge hn

1s 2s 2p EB = E final ( n − 1) − Einitial ( n )
EB
1s EB ≈ −ε k
(Koopmans’ Theorem)
It will be discussed below

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 9

Basic Principles of XPS: peak labeling

n l j X-Ray Level Electron Level

1 0 1/2 K 1s
2 0 1/2 L1 2s
2 1 1/2 L2 2p1/2
2 1 3/2 L3 2p3/2
3 0 1/2 M1 3s
3 1 1/2 M2 3p1/2
3 1 3/2 M3 3p3/2
3 2 3/2 M4 3d3/2
3 2 5/2 M5 3d5/2

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 10

10

ECE658, Purdue University, D. Zemlyanov 5

 2nd semester, 2022

Basic Principles of XPS: identification of elements

ELEMENT ANALYSIS
Every chemical element has an unique electronic structure, thereby the electrons are emitted
with specific kinetic energies. The emission lines for almost all elements are well tabulated.1

Survey Spectrum covers a wide energy range: poor resolution but high sensitivity.

1NIST X-ray Photoelectron Spectroscopy Database (the National Institute of Standards and Technology,
http://srdata.nist.gov/xps/)
Handbook of X-ray Photoelectron Spectroscopy, C. D. Wagner, W. M. Riggs, L. E. Davis, J. F. Moulder, and G. E.
Mullenberg, published by Perkin-Elmer Corp., 1979, Eden Prairie, MN, USA
2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 11

11

Basic Principles of XPS: identification of elements

Survey Spectrum of a silver foils

Energy of photoelectrons: eV
Intensity: count per second, cps
hν = 1486.6 eV

Acquisition conditions were chosen to have maximum of

sensitivity and a low resolution (pass energy ~160 eV), wide
energy range.
BE= hν-KE-ϕspectrometer
2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 12

12

ECE658, Purdue University, D. Zemlyanov 6

 2nd semester, 2022

Basic Principles of XPS: identification of peak and survey spectrum interpretation

The diagram below shows a real XPS
spectrum obtained from a Ag foil using
Al Kα radiation (1486.6 eV)
The main peaks occur at kinetic
energies of ca. 350, 770, 880, 915,
1110, 1120, 1390, 1430 and 1480 eV.
Since the energy of the radiation is
known it is a trivial matter to transform
the spectrum
BE= hν-KE-ϕspectrometer
so that it is plotted against BE as
opposed to KE.
The most intense peak is now seen to
occur at a binding energy of ca. 370 eV.

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 13

13

Basic Principles of XPS: identification of peak and survey spectrum interpretation

Working downwards from the highest

energy levels (closest to Fermi edge):
• the valence band (4d) emission
occurs at a binding energy of ca. 5
eV.
• the emission from the 4p and 4s
levels gives rise to very weak
peaks at 60 and 97 eV respectively
• the most intense peak at ca. 368 eV
is due to emission from the Ag 3d
levels, whilst the 3p and 3s levels
give rise to the peaks at ca.
572/605 eV and 720 eV
respectively.
• the remaining peak is not an XPS
peak at all ! It is an Auger peak
arising from X-ray induced Auger
emission. It occurs at a kinetic
energy of ca. 350 eV .
2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 14

14

ECE658, Purdue University, D. Zemlyanov 7

 2nd semester, 2022

XPS Instrumentation

Synchrotron-based home-built
ambient pressure XPS system
(BESSY-2, Berlin, Germany)

Kratos Axis XPS spectrometer

(University of Limerick, Ireland)

XPS can be performed using either a commercially

built XPS system, a privately built XPS system or a
synchrotron-based end-station typically with a
custom designed electron analyzer. Home-built XPS/STM system, (Chemical
Engineering, Purdue University)
2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 15

15

XPS Instrumentation
Commercial XPS instruments use either a highly focused 20 to 600
micrometer beam of monochromatic (typically Al Kα) X-rays or a broad
10–30 mm beam of non-monochromatic or (polychromatic) magnesium
X-rays.
Synchrotron-based XPS use a synchrotron radiation (very high energy
range of 100 eV – 20000 eV).
A few, special design, XPS instruments can analyze volatile liquids or
gases, materials at low or high temperatures or materials at roughly 1 torr
vacuum, but there are relatively few of these types of XPS systems.

Palladium single crystal in a home-built ambient pressure Silver foil in a VG(?)/RHK(?) XPS system.
XPS system (BESSY-2, Berlin, Germany). The photo was obtained at www.wikipedie.com.
2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 16

16

ECE658, Purdue University, D. Zemlyanov 8

 2nd semester, 2022

XPS Instrumentation
Birck An
Nanotechnology
electron energy Center
analyzer
A set of stage
manipulators UHV ss chamber

The main components of a commercially

made XPS system include:
• A source of X-rays
• An ultra-high vacuum (UHV)
stainless steel chamber with UHV
pumps
• An electron collection lens
• An electron energy analyzer
• Mu-metal magnetic field shielding
• An electron detector system
• A moderate vacuum sample
introduction chamber
• Sample mounts
• A sample stage
• A set of stage manipulators

A moderate vacuum sample

introduction chamber
A source of X-rays UHV ss chamber

2nd semester, 2022 ECE 658, Purdue University, Dmitry Zemlyanov 17

17

XPS Instrumentation: X-ray anodes

Energy and Widths of some characteristic soft X-ray lines

Line Photon Energy, eV Width, eV
Cu Lα 929.7 3.8
Mg Kα 1253.6 0.7
Al Kα 1486.6 0.85
Si Kα 1739.5 1.0
Y Lα 1922.6 1.5
Zr Lα 2042.4 1.7 Most used anodes
Ag Lα 2984.3 eV 1.3

Can be monochromized

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 18

18

ECE658, Purdue University, D. Zemlyanov 9

 2nd semester, 2022

XPS Instrumentation: Monochromatic radiation

Toroidal quartz
backplane
Monochromatic Al Kα X-rays are normally
Rowland circle produced by diffracting and focusing a beam of
diameter non-monochromatic X-rays off of a thin disc of
natural, crystalline quartz with a <1010>
orientation. The resulting wavelength is 8.3386 Å
which corresponds to a photon energy of 1486.6 eV.

Energy dispersion ΔE ~ Rowland circle diam.

For 500 mm ~ 0.625 eVmm-1

250 mm ~ 1.25 eVmm-1

Electron gun
& x-ray anode
Energy dispersive
direction Fixed mono spot

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 19

19

XPS Instrumentation: Monochromatic radiation

Why do we need Monochromatic Radiation?

Pros: Energy dispersion ΔE for 500 mm Rowland circle

• High resolution is ~ 0.625 eVmm-1, which corresponds to the
resolution of 0.25 eV for 0.4mm diameter spot.
• Small spot
• No thermal damage of a sample The natural Full Width at Half Maximum
(FWHM) is 0.70eV for Mg Kα (1253.6eV) and
• No ghost and satellite peaks 0.85eV for Al Kα (1486.6eV).
• No Bremsstrahlung X-rays

Cons:
• Lower intensity
• Focused radiation might cause
radioactive damage of a sample

The sample is close to a X-ray gun and it is

heated by filament.

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 20

20

ECE658, Purdue University, D. Zemlyanov 10

 2nd semester, 2022

XPS Instrumentation: Ghost and satellite peaks

Satellites arise from less probable electron transition (the
same material).
α stands transition from an adjacent shell.
The main line:
Kα1,2 – from 2p3/2,1/2 (unresolved doblet) → 1s (original hole)

The satellite lines:

Kβ – from valence band → 1s (original hole)
Kα’ – from 2p3/2,1/2 (exited state) → 1s (original hole)
Kα3,4, ..n – the transitions in a multiply ionized atom
Satellite lines appear with higher kinetic
energy values than the dominant Al Kα
induced photoelectric line
Separation from Kα1,2 (eV) and relative intensity (Kα1,2 = 100)
X-ray line Mg Al
Kα’ 4.5 (1.0) 5.6 (1.0)
Kα3 8.4 (9.2) 9.6 (7.8)
Kα4 10.0 (5.1) 11.5 (3.3)
Kα5 17.3 (0.8) 19.8 (0.4)
Kα6 20.5 (0.5) 23.4 (0.3)
Kβ 48.0 (2.0) 70.0 (2.0)
2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 21

21

Instrumentation: Ghost and satellite peaks

X-ray ghosts are due to excitations arising from

impurity elements in the X-ray source.
The most common ghost is Al Kα from Mg Kα
source.
The following table summarizes the main
contamination types that may be encountered

Contamination Anode material: Anode material:

radiation Al Mg
Mg Kα +233 N/A
Al Kα N/A -233
O Kα +961.7 +728.7
Cu Lα +556.9 +323.9

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 22

22

ECE658, Purdue University, D. Zemlyanov 11

 2nd semester, 2022

Instrumentation: Electron Energy Analyzer

ΔE 2W
=
E R0

Resolution 2 times of Width of slits

=
Pass Energy Radius of Analyser

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 23

23

Birck Nanotechnology Center

Instrumentation for XPS: Resolution and Pass energy

PET: Polyethylene terephthalate

The C 1s spectra were obtained at different Pass Energies

2nd semester, 2022 ECE 658, Purdue University, Dmitry Zemlyanov 24

24

ECE658, Purdue University, D. Zemlyanov 12

 2nd semester, 2022

Instrumentation for XPS: Spectrometer resolution

Birck Nanotechnology Center

( FWHM ) + ( Energy Analyzer Resolution )

2 2
Spectrometer resolution = X−ray line

FWHM of X-ray line, V Energy Analyzer Expected Spectrometer

resolution, eV resolution, eV
0.25 (monochromatic) 0.25 0.35
0.7 (Mg Kα) 0.25 0.74
0.85 (Al Kα) 0.25 0.89
0.25 (monochromatic) 0.50 0.56
0.7 (Mg Kα) 0.50 0.86
0.85 (Al Kα) 0.50 0.99

2nd semester, 2022 ECE 658, Purdue University, Dmitry Zemlyanov 25

25

Basic Principles of XPS: Excitation and Relaxation

Excitation (Photoemission)

Auger Process
Relaxation X-ray Fluorescence

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 26

26

ECE658, Purdue University, D. Zemlyanov 13

 2nd semester, 2022

Basic Principles of XPS: Auger process (spectroscopic notation)

n l j X-Ray Level Electron Level

1 0 1/2 K 1s
2 0 1/2 L1 2s
2 1 1/2 L2 2p1/2
2 1 3/2 L3 2p3/2
3 0 1/2 M1 3s
3 1 1/2 M2 3p1/2
3 1 3/2 M3 3p3/2
3 2 3/2 M4 3d3/2
3 2 5/2 M5 3d5/2

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 27

27

Basic Principles of XPS: Auger process

Fermi edge
Auger process involve two electrons and a hole 2p3/2
at a core level: BEL2
2p1/2 L-shell
1. Ionisation occurs by removal of a K-shell BEL1
2s
Energy

electron (1s electron level);

2. One electron “falls” from a higher level L1 BEK
(2s electron level) to fill an initial core hole
in the K-shell;
1s K-shell
3. The energy liberated in the second step is
transferred to a L2 (or L3) electron (2p1/2 or
2p3/2 electron level) and this electron called
Auger electron is emitted with kinetic
energy Auger electron
KEAuger = ( BEK - BEL1 ) - BEL2
= BEK – (BEL1 + BEL2) . Fermi edge
2p3/2 BEL2
The final state is a doubly-ionized atom with 2p1/2 L-shell
BEL1
2s
Energy

core holes in the L1 and L2 (or L3) shells. This

Auger transition is KL1L2.
BEK
Note: the KE of the Auger electron is
independent of the mechanism of initial 1s K-shell
core hole formation.
2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 28

28

ECE658, Purdue University, D. Zemlyanov 14

 2nd semester, 2022

Basic Principles of XPS: Auger process

Note: In the case of X-ray induced Auger emission it is really

meaningless to refer to an associated binding energy.

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 29

29

Basic Principles of XPS: Spin-Orbit Splitting

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 30

30

ECE658, Purdue University, D. Zemlyanov 15

 2nd semester, 2022

Basic Principles of XPS: Spin-Orbit Splitting

Closer inspection of the spectrum shows that emission from some levels
(most obviously 3p and 3d ) does not give rise to a single photoemission peak,
but a closely spaced doublet.

The 3d photoemission is in fact split between two peaks, one at 368.3 eV

and the other at 374.3 eV, with an intensity ratio of 3:2 . This arises from
spin-orbit coupling effects in the initial state.
2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 31

31

Basic Principles of XPS: Spin-Orbit Splitting – j-j coupling

j-j coupling
The total angular momentum of a single isolated ℓ=0→
electron is obtained by summing vectorially the n s
3d5/2 1→p
individual electronic spin and angular momenta.
The total angular momentum, j , is j = l + s . 2→d
j-j coupling is the best for Z > ~75 but in fact this 3→f
nomenclature has been used for the entire Periodic j =l±s
Table.
The total atomic angular momentum is the sum of j,
J= ∑j.

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 32

32

ECE658, Purdue University, D. Zemlyanov 16

 2nd semester, 2022

Basic Principles of XPS: Spin-Orbit Splitting – j-j coupling

Quantum numbers X-ray suffix X-Ray Level Spectroscopic

n l j Level
1 0 1/2 1 K 1s
2 0 1/2 1 L1 2s
2 1 1/2 2 L2 2p1/2
2 1 3/2 3 L3 2p3/2
3 0 1/2 1 M1 3s
3 1 1/2 2 M2 3p1/2
3 1 3/2 3 M3 3p3/2
3 2 3/2 4 M4 3d3/2
3 2 5/2 5 M5 3d5/2
etc etc etc etc
2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 33

33

Basic Principles of XPS: Spin-Orbit Splitting – j-j coupling

Intensity ratio for spin-orbit coupling doublets:

Orbital momentum Doublets Intensity ratio

p p3/2 : p1/2 2:1
d d5/2 : d3/2 3:2
f f7/2 : f5/2 4:3

ℓ=0→
n s
3d5/2 1→p
2→d
3→f
j =l±s
2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 34

34

ECE658, Purdue University, D. Zemlyanov 17

 2nd semester, 2022

Basic Principles of XPS: Spin-Orbit Splitting – L-S coupling

L-S (Russell-Sanders) coupling

The other way of carrying out the vectorial summation is first to sum all the
individual electronic angular momenta (to get total atomic angular momentum)
and then all the individual electronic spin momenta (to get total atomic spin
momentum).
L = ∑l S = ∑s
The total angularNELIN
momentum:
ET AL.
J = L + S or J = L ± S .
L-S coupling should be applied for Z < ~20. to the occupations of the spin-orbit sub-shells, is
lost. In a 40-eV range of energy, there are six XPS
L-S coupling is typically used to calculate Auger
peakstransitions low atomic
with substantial intensity number
as opposed to
two spin–orbit split peaks expected. In fact, the
the

elements. third peak, at E " 20 eV, has comparable inten-

sity with the first peak at E ¼ 0; furthermore,
rel
rel
FIGURE 1. Representative XPS Spectra for the Ne the sixth peak at Erel " 40 eV and the fourth peak
atom. [Color figure can be viewed in the online issue, at Erel " 25 eV also have comparable intensities.
which is available at wileyonlinelibrary.com.] The common assignment made for these six peaks
[7, 8], based on semiempirical Anderson model
this configuration is 2S and, hence, this configura- calculations, is that the first triplet arises from
tion can and does mix with the XPS allowed 2s- ionization of a 3d5/2 electron and that the second
hole configuration, 1s22s12p6, which is also a 2S triplet is from 3d3/2 ionization. In addition, each
multiplet [16]. Thus, the states that would be of the triplets are assigned as arising, as shown in
2nd semester,
obtained by a two configuration CI mixing the Figure 2, from mixing of configurations with dif-
2022 XPS-allowed 2s-hole ECE 658, Purdue University,
configuration Dmitry
and the XPS- Zemlyanov 35
ferent degrees of charge transfer. Specifically,
forbidden configuration would be one that is these are configurations without CT from O to the
35 dominantly the XPS-allowed 2s-hole configuration
and the second, at higher energy, which is domi-
empty Ce 4f shell, denoted CT0 for occupations
3d94f0O(2p6), configurations with single CT from
nantly the XPS-forbidden configuration but with a O to the empty 4f shell, denoted CT1 for occupa-
small weight of the XPS-allowed configuration tions 3d94f1O(2p5), and configurations with dou-
[16]. The second CI state will receive intensity ble CT to the empty 4f shell, denoted CT2 for
proportional to the weight of the allowed XPS occupations 3d94f2O(2p4). Furthermore, the highest
configuration in the CI wavefunction for this BE peaks for each of the triplets, peaks 3 and 6, are
state. Thus, it is appropriate to say that the sec- assigned as arising from CT0 configurations, while
ond state, a dominantly XPS-forbidden state, the lower two BE peaks of each triplet are assigned
steals intensity from the lower state, which is as mixtures of CT1 and CT2 configurations. It is
dominantly the XPS allowed configuration with a worth noting that these semiempirical calculations
small weight of the XPS-forbidden configuration. are for a generic core-hole; they do not distinguish
Basic Principles of XPS: Spin-Orbit Splitting or many-body effect and charge transfer
This interpretation shows us how states domi- the properties of different core levels and the spin–
nated by XPS-forbidden configurations are able to orbit splitting is added with an ad hoc parameter.
gain intensity, possibly considerable intensity, We will demonstrate that our nonempirical calcula-
because they contain contributions from the tions indicate that this analysis neglects important
allowed configurations [9, 17]. This analysis is rig- intra-atomic many-body effects and the assignment
orous if one uses the same set of orbitals for the
initial, N electron, state, and the final, N ! 1 elec-
22 T. Skála et al. / Journal of Electron Spectroscopy and Related Phenomena 169 (2009) 20–25
tron, states. When orbital relaxation to screen the
core-hole [18] is taken into account, satellites may
also gain intensity because of non-zero overlap
integrals between the orbitals of the initial and
final states. These shake satellites somewhat blur
the concept that we have introduced that domi-
nantly XPS-forbidden states steal intensity from
dominantly XPS-allowed states by mixing in CI
wavefunctions. However, the concept of stealing
intensity by CI mixing is useful for qualitatively
understanding the numbers and energy distribu-
tions of reasonably intense satellites.
The Ce 3d XPS of CeO2 shown schematically in
Figure 2 shows how, when there are important
many-body effects, the simple XPS doublet of a FIGURE 2. CeO2 3d XPS with semiempirical Anderson
3d5/2 peak at lower BE and a 3d3/2 peak at higher Model assignments for the charge transfer origin of
CeO 3d XPS with semiempirical Anderson Model
BE with relative intensities of 6:4, corresponding 2 satellites.
assignments for the charge transfer origin of
satellites.
2754 INTERNATIONAL JOURNAL OF QUANTUM CHEMISTRY DOI 10.1002/qua VOL. 110, NO. 15
These are configurations without charge transfer
from O to the empty Ce 4f shell, denoted CT0 for
occupations 3d94f0O(2p6), configurations with single
CT from O to the empty 4f shell, denoted CT1 for
occupations 3d94f1O(2p5), and configurations with
double CT to the empty 4f shell, denoted CT2 for
occupations 3d94f2O(2p4).
Fig. 1. Ce 3d spectra acquired on the CeO2 surface at 120 K before (bottom, fits A–C) and after (top) the deposition of 0.67 nm Sn layer. Photon energy 1486.6 eV, emission
angle 20◦ . The CeO2 spectra were decomposed into three doublets of Voigt profile (fit A). In fit B only the u + v doublet was modelled by a convolution of a Doniach-Šunjić
T. Skála et al. / Journal of Electron Spectroscopy and C. J. Nelin et al./ International Journal of Quantum
with a Gaussian profile and in fit C by a sum of two constrained Voigt peaks as illustrated in inset. The top curve was fitted by three Ce4+ doublets from the fit C (red) and two
Ce3+ related Voigt doublets (blue). The green curve labelled “Sn foil” shows the Sn 3s spectrum measured on clean Sn foil and normalized to tin content in the top spectrum
Related Phenomena 169 (2009) 20–25
using Sn 3d areas. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.) Chemistry 11 (2010) 2752–2764
2nd semester,
3. Results and discussion
2022 The amount of Sn658,
ECE at thisPurdue
saturation point corresponds
University, Dmitryroughly to
Zemlyanov 36
1–2 monolayers. Higher amounts of tin caused only a suppres-
3.1. Low temperature, 120 K sion of Ce 3d intensity without influencing the chemical state of
36
First, the 1.5-nm thick CeO2 (1 1 1) layer was prepared, cooled
cerium.
The top three curves in Fig. 3 of the ceria layer without (a)
to 120 K and carefully checked with LEED, XPS and resonant pho- and with tin (b) show the valence band spectra measured at
toemission. The Ce 3d XPS spectrum is plotted in the bottom part
of Fig. 1. It exhibits a typical structure of stoichiometric CeO2 as
compared to the literature and confirmed by the fitting proce-
dure described in Section 2.3 (Ce3+ states contribution below 0.1%)
and by resonant photoemission (D ratio 0.05), see below. Then, Sn
was deposited on the sample in eight steps up to a thickness of
0.67 nm. After each step, XPS and synchrotron radiation photoe-
mission spectra were recorded. An example of the Ce 3d spectrum
and its decomposition after the last deposition step is plotted as
ECE658, Purdue University, D. Zemlyanov
the top curve in Fig. 1. The evolution of the relative contribution 18
of Ce3+ features in Ce 3d spectrum is plotted in Fig. 2 for both
emission angles. The dependence is qualitatively similar, for the
angle 60◦ , which has higher surface sensitivity (approximately half
of the probing depth at 20◦ ) the absolute values of this ratio are
 2nd semester, 2022

Basic Principles of XPS: Spin-Orbit Splitting or many-body effect and charge transfer
COMPLEX CORE LEVEL SPECTRA AND MATERIALS PROPERTIES

antibonding combinations determined in our or-

bital variations. However, as the same orbitals are
used for all states, the trends in occupations are
meaningful. The main peak at Erel ¼ 0, feature A, is
dominated by the contribution of a single final
ionic state while there are a few states contributing
to each of features B, C, and D. However, the elec-
tronic characters of the different states contributing
to features B to D are reasonably similar. Thus, it is
sufficient to give the data for a single representative
state in each of the features. We have chosen the
state with the largest SA Irel as the representative
state.
Theory for the 3d XPS of CeO2 showing in
A striking feature from Table I is that the main
peak, Feature A, arises from a state dominated,
(I) the hole state with no many-body effects included,
with a weight of 95%, by determinants that do not
involve charge transfer. The satellite peaks, on the
(II) the inclusion of single charge transfer configurations,
other hand, have very little contribution from con-
figurations without CT. For feature B, Wht(CT0) ¼ (III) the inclusion of some double charge transfer
0.012 is the largest weight of configurations with-
out CT for the satellites shown in Table I. Finally,
configurations.
we note that the contribution from double CT,
Wht(CT2), to the XPS, while it is not negligible, is
For the intra-atomic many-body effects, can leads to major
not especially large. These three conclusions,
obtained from rigorous theory, are in direct contra-
changes in the energy separation and in the number of
diction to the results of semiempirical Anderson intense XPS peaks.
model calculations [7, 8]. In the Anderson model
calculations, the intense feature with the highest
Erel, corresponding to our Feature C is found,
incorrectly, to be dominated by configurations
without CT. On the other hand, single and double
CT configurations are claimed to contribute in
comparable amounts to the two lower BE peaks,
our features A and B. Insight into the character of
the 4s-hole states comes from the occupation num-
bers given in Table II. The Ce 4s, 4p, and 4d occu-
pations change from the ideal values of 1, 6, and 10
because of atomic many body effects; i.e., from the
FAC, see Figure 4. For the main peak, feature A,
the atomic many body effects involve moving 0.1
4p electrons into the 4s shell and 0.1 4d electrons
into the 4f shell. There is a CT of another 0.1 elec-
trons from O 2p into the Ce 4f and there is no CT
C. J. Nelin et al./ International Journal of Quantum Chemistry 11 (2010)
into Ce 5d. All of the 4s satellites are dominated by 2752–2764
CT with very little intra-atomic contributions. See FIGURE 6. Theory for the 3d XPS of CeO2 showing in
Ref. [4] for further details. 2nd semester,
I, the hole state with no many-body effects included, in
We now present preliminary results for the 3d II, the inclusion of single charge transfer configurations,
XPS. In Figure 6, we show the XPS for three dif- 2022 and in III, the inclusion of some double charge transfer ECE 658, Purdue University, Dmitry Zemlyanov 37
configurations. Intense individual XPS peaks as well as
ferent CI treatments; here the calculated Irel were
the envelope of all contributions are shown. [Color

37
broadened with a Gaussian of 1.6 eV FWHM. The figure can be viewed in the online issue, which is
top figure is for only the 10 3d-hole configura- available at wileyonlinelibrary.com.]
tions where only spin-orbit splitting is included
and there is no CT. There is a small splitting of

VOL. 110, NO. 15 DOI 10.1002/qua INTERNATIONAL JOURNAL OF QUANTUM CHEMISTRY 2759

Basic Principles of XPS

Photoemission consists of three steps:

• Photon absorption and ionization (initial state effects)

• Response of the atom and neighbors on creation of
photoelectron (final state effects)
• Transport of photoelectron to surface and photoelectron
escape (extrinsic losses)

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 38

38

ECE658, Purdue University, D. Zemlyanov 19

 2nd semester, 2022

Basic Principles of XPS: Initial State Effect - Chemical Shift

EB = E final ( n − 1) − Einitial ( n ) Koopmans’ Theorem or

EB ≈ −ε k the Approximation of the Frozen Orbitals

The binding energy (BE) of an electron is simply the difference between the
initial state in an atom with n electrons and final state, which is a single-
charged atom with n-1 electrons and free photoelectron.

In the approximation of the “frozen” orbitals, when no relaxation followed

photoemission, BE is an eigen energy, which can be calculated using non-
relativistic Hartree-Fock method.
The absence of relaxation means no electronic rearrangement (intra- and
extra-atomic) following photoemission.

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 39

39

Basic Principles of XPS: Initial State Effect - Chemical Shift

If the energy of the atom’s initial state changed, for example by formation of chemical bond
with other atoms, the EB of the electrons in that atom will change. ΔE B = Δε k

Considering an atom as a hollow charged sphere, the effective charge potential model (Siegbahn
et al. 1969) proposed that the orbital shift as

DqB
De iA = Dq A kicore-valence + å
B ¹ A R AB
The charge difference The average interaction Two center–electron approximation: no overlapping
on atom A between between a core electron i between the core electron density on atom A and the
the molecule and and valence electrons on valence electron densities on the other atoms B.
reference system atom A.
+d -d
E Fermi level E
Δε iA = ΔqA kicore−valence + V molecular potential + const
EB(-d)
EB(+d)
Original level

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 40

40

ECE658, Purdue University, D. Zemlyanov 20

 2nd semester, 2022

Basic Principles of XPS: Initial State Effect - Chemical Shift

Dq A kicore-valence + V molecular potential

C-C*F3 +d -d
E Fermi level E
C-O-C*=O EB(-d)
EB(+d)
Original level
C-C*-O-C
C-C*-C
MeC2

292 285 281

Binding Energy, eV

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 41

41

Basic Principles of XPS: Initial State Effect - Chemical Shift

Birck Nanotechnology Center

The O 1s and C 1s spectra of PET: Polyethylene terephthalate

2nd semester, 2022 ECE 658, Purdue University, Dmitry Zemlyanov 42

42

ECE658, Purdue University, D. Zemlyanov 21

 2nd semester, 2022

Basic Principles of XPS: Initial State Effect - Chemical Shift

Birck Nanotechnology Center

The XPS of Polymers Database: Edited by Graham Beamson and David Briggs

2nd semester, 2022 ECE 658, Purdue University, Dmitry Zemlyanov 43

43

Basic Principles of XPS: Initial State Effect - Chemical Shift

Birck Nanotechnology Center

The XPS of Polymers Database: Edited by Graham Beamson and David Briggs

2nd semester, 2022 ECE 658, Purdue University, Dmitry Zemlyanov 44

44

ECE658, Purdue University, D. Zemlyanov 22

 2nd semester, 2022

Basic Principles of XPS: Initial State Effect - Chemical Shift

C*-C C*-N C*-O O=C*-N O=C*-OH C*-N3

Type 1 Type 2 Type 3 (amide) (carboxyl) (arginine)
Type 4 Type 5 Type 6
RGD silane 20 13 7 18 2 1
2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 45

45

Basic Principles of XPS: Initial State Effect - Chemical Shift

C*-N & C-O

*
*
Amide C -C
Residual
Hydrocarbons
Intensity, arb. units

OH-C*=O

*
C Fx C*N3 Experimental data
Curve-fitting result

Peptide constraint

295 290 285 280

Binding Energy, eV

The C 1s spectrum obtained from the powder RGD peptide

C*-C C*-N C*-O O=C*-N O=C*-OH C*-N3

Type 1 Type 2 Type 3 (amide) (carboxyl) Type 6
Type 4 Type 5
RGD silane 20 13 7 18 2 1
2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 46

46

ECE658, Purdue University, D. Zemlyanov 23

 2nd semester, 2022

Basic Principles of XPS: Initial State Effect - Chemical Shift

Birck Nanotechnology Center

2nd semester, 2022 ECE 658, Purdue University, Dmitry Zemlyanov 47

47

Basic Principles of XPS: Initial State Effect - Chemical Shift

PdO
Pdox1

Pd 3d3/2 Pd 3d5/2 Pd
0

249 min
Intensity, arb. units

129 min

Pdox2

37 min

345 340 335

Binding Energy, eV

2nd semester, 2022 ECE 658, Purdue University, Dmitry Zemlyanov 48

48

ECE658, Purdue University, D. Zemlyanov 24

 2nd semester, 2022

Basic Principles of XPS

Photoemission consists of three steps:

• Photon absorption and ionization (initial state effects)

• Response of the atom and neighbors on creation of
photoelectron (final state effects)
• Transport of photoelectron to surface and photoelectron
escape (extrinsic losses)

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 49

49

Basic Principles of XPS: Final State Effects

FINAL STATE EFFECT

Final state effects are those factors that influence the states of the
atom after the photon has hit it or affect on the photoelectron while
it is leaving.

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 50

50

ECE658, Purdue University, D. Zemlyanov 25

 2nd semester, 2022

Basic Principles of XPS: Final State Effects

Localized
core hole
RELAXATION
The photoemission event has left a hole in a
1s 2s 2p
core level. Delocalization of this localized hole
due to inflow (“diffusion”) of charge is called
relaxation.
– Intra-atomic relaxation
The core hole is delocalized due to Consequences
rearrangement of electrons in the orbitals
of the excited atom. – the leaving electron can
– Inter-atomic (extra-atomic) escape at higher kinetic
relaxation energy: the binding energy
The core hole is delocalized due to of the electron to decrease.
movement of electrons from the
surrounding atoms in the material. – Relaxation will NOT cause
an extra peak to appear in
the spectra.

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 51

51

Basic Principles of XPS: Final State Effects

RELAXATION: Silver oxidation

Due to Inter-atomic (extra-atomic) relaxation the photoelectron
escape at higher kinetic energy from Ag2O comparing with Ag
metal.
Ag1+

Ag0

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 52

52

ECE658, Purdue University, D. Zemlyanov 26

 2nd semester, 2022

Basic Principles of XPS: Auger parameter

RELAXATION: Silver oxidation

Modified Auger parameter

(Gaarenstroom & Winograd at 1977):
α’ = KEijj(Auger) + BEi(Photoemission)

Δα’ = ΔKEijj + Δ BEi ≈

(- Δ V + 3Δ Rea) + ( Δ V - Δ Rea)

Δα’ ≈ 2Δ Rea Final state

Evans et al at 1992:

β = KEijj + BEi+ 2BEj

Δβ ≈ 2ΔV ( ΔqA ki
core−valence
) Initial
state

These data: D.Yu. Zemlyanov et al. Applied Surface Science 133(1998) p171
Modified Auger Parameter: C.D. Wagner, A. Joshi, J. Electron Spectrosc. Rel. Phen. 47 (1988) p183
2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 53

53

Basic Principles of XPS: Final State Effects and Satellites

SATELLITES
Shake-up and shake-off satellites arise
when the photoelectron imparts energy to
another electron of the atom.
– Shake-up satellite

– Shake-off satellite

Consequences
– The photoelectron loses kinetic energy
and appears at higher binding energy
in the spectrum.
– Extra peak might appear in the
spectrum.
2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 54

54

ECE658, Purdue University, D. Zemlyanov 27

 2nd semester, 2022

Basic Principles of XPS: Final State Effects and Satellites

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 55

55

Basic Principles of XPS: Final State Effects and Plasmons

Plasmons
Plasmons describe the interaction of the photoelectron with the plasma oscillation of
the outer shell (valence band) electrons. This is an inelastic scattering.
• Plasmons are quantum mechanical pseudoparticles with energy Ep = ħω
• ω= (ne2/εom)1/2/2π n - valence electron density,
e, m - electron charge and mass
εo - dielectric constant of vacuum
• Bulk and surface plasmons may be observed (Esurface plasmon = (1/√2)Ebulk plasmon
• Plasmon intensity reflects how easy to excite plasma oscillation

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 56

56

ECE658, Purdue University, D. Zemlyanov 28

 2nd semester, 2022

Basic Principles of XPS: Final State Effects and Plasmons

Plasmons
Extrinsic satellites occur during transport
of electron to surface. Discrete loss
structure is observed.

Consequences
– The photoelectron loses kinetic energy
and appears at higher binding energy
in the spectrum. Energy loss (plasmon) lines associated with the
– Extra peaks might appear in the 2s line of aluminium (a = 15.3 eV; note surface
plasmon at b)
spectrum.

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 57

57

Basic Principles of XPS: Final State Effects and Plasmons

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 58

58

ECE658, Purdue University, D. Zemlyanov 29

 2nd semester, 2022

Basic Principles of XPS

Photoemission consists of three steps:

• Photon absorption and ionization (initial state effects)

• Response of the atom and neighbors on creation of
photoelectron (final state effects)
• Transport of photoelectron to surface and photoelectron
escape (extrinsic losses)

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 59

59

Basic Principles of XPS

X-ray can readily travel through

solids. Actually, X-rays of 1KeV,
will penetrate 1000 nm (1μm) or
more into material.

Can we say the same about

electrons?

NO !!!
Electrons of 1KeV energy will
only penetrate approximately 5nm
(0.005μm).
2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 60

60

ECE658, Purdue University, D. Zemlyanov 30

 2nd semester, 2022

Basic Principles of XPS: Information depth

In general IMFP depends on

• KE of the photoelectron
• the specific material

"Universal curve" of electron inelastic mean free path λ (IMFP) versus KE

IMFP is average distance between inelastic collisions (Å)

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 61

61

Basic Principles of XPS: Information depth

For normal photoemission, cosθ = 1

When d = λ, 63.3 % of electrons come from
within 1λ of surface.
!w
When d = 2 λ, 86.4 % of electrons come
d 3l
from within 2 λ of surface.
When d = 3 λ, 95.0 % of electrons come
from within 3λ of surface
Typical λ varies from 10 Å to 35 Å.

{
P = exp −d λ cosθ } Probability of electron escaping without loss

= N (1 − exp {−d λ cosθ })

Nelastic is the number of electron escaping
N elastic 0 without loss
N0 is the total number of electron generated

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 62

62

ECE658, Purdue University, D. Zemlyanov 31

 2nd semester, 2022

Basic Principles of XPS: Angle Resolved XPS

!w
d 3l

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 63

63

Basic Principles of XPS: Electron Attenuation Length (EAL)

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 64

64

ECE658, Purdue University, D. Zemlyanov 32

 2nd semester, 2022

XPS Artifacts: Degradation during analysis

Depends on the sensitivity of the material to the wavelength of X-rays used, the total dose of the X-
rays, the temperature of the surface and the level of the vacuum. Metals, alloys, ceramics and most
glasses are not measurably degraded by either non-monochromatic or monochromatic X-rays. Some,
but not all, polymers, catalysts, certain highly oxygenated compounds, various inorganic compounds
and fine organics are degraded by either monochromatic or non-monochromatic X-ray sources.

• Non-monochromatic X-ray sources produce a significant amount of high energy Bremsstrahlung

X-rays (1– 15 keV of energy) which directly degrade the surface chemistry of various materials.
• Non-monochromatic X-ray sources also produce a significant amount of heat (100 to 200 °C) on
the surfaceof the sample because the anode that produces the X-rays is typically only 1 to 5 cm (2
in) away from the sample. This level of heat, when combined with the Bremsstrahlung X-rays,
acts synergistically to increase the amount and rate of degradation for certain materials.
• Monochromatic X-ray sources, because they are far away (50– 100 cm) from the sample, do not
produce any heat effects.
• Monochromatic X-ray sources are monochromatic because the quartz monochromator system
diffracted the Bremsstrahlung X-rays out of the X-ray beam which means the sample only sees
one X-ray energy, for example: 1486.6 eV if Al Kα X-rays are used.
Because the vacuum removes various gases (e.g. O2, CO) and liquids (e.g. water, alcohol, solvents)
that were initially trapped within or on the surface of the sample, the chemistry and morphology of
the surface will continue to change until the surface achieves a steady state. This type of degradation
is sometimes difficult to detect.

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 65

65

XPS Artifacts: Charging

Energy

Analyzer

Ek EKspec Spectrometer
work function
Vacuum level
φ φsp Spectrometer
hν Fermi edge ground

EB Connected if a sample is conducting

1s EB = hn - EK spec - fsp

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 66

66

ECE658, Purdue University, D. Zemlyanov 33

 2nd semester, 2022

XPS Artifacts: Charging

Energy

Analyzer

Spectrometer
Ek EKspec work function

Vacuum level
φsp Spectrometer
φ ground
hν Fermi edge

EB NOT Connected if a sample

is an insulator

1s EB ¹ hn - EK spec - fsp

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 67

67

XPS Artifacts: charging correction and reference lines

• Adventitious carbon: use of the C 1s line of adventitious hydrocarbon on the

specimen surface which is assumed to have a binding energy of 284.8 eV.
• Internal hydrocarbon: use of an internal hydrocarbon group of the specimen
compound, with the same assumed binding energy (284.8 eV).
• Condensed hydrocarbon: a hydrocarbon was co-condensed with the vapor-
deposited specimen and used as the charge reference with an assumed binding
energy of 284.8 eV.
• Valence band minimum: The valence band minimum can be used as a
reference.
• Use of a known line.
• Argon: implanted Ar with the Ar 2p line used as the charge reference. An
average Ar 2p binding energy of 241.82 eV for Ar implanted in graphite has
been recommended [C. J. Powell, Appl. Surf. Science 89, 141 (1995)]. The Ar
2p binding energy, however, will vary depending on the material in which the
ions were implanted [P. H. Citrin and D. R. Hamann, Phys. Rev. B 10, 4948
(1974)].
• Gold: gold vapor deposition method with the Au 4f7/2 binding energy assumed
to be 84.0 eV.
2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 68

68

ECE658, Purdue University, D. Zemlyanov 34

 2nd semester, 2022

XPS Artifacts: Differential (non-uniform) surface charging

Broadening of peak

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 69

69

XPS Artifacts: Methods for charge control

Charge neutralizing system (Kratos)

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 70

70

ECE658, Purdue University, D. Zemlyanov 35

 2nd semester, 2022

XPS Artifacts: Ghost peaks

X-ray ghosts are due to excitations arising from impurity elements in the X-ray source.
The most common ghost is Al Kα from Mg Kα source.

O1s
O KLL
O Auger O2s
Ghost of O1s
Due to Al Kα

Al 2p
F1s C1s
Al 2s

Contamination Anode material: Anode material:

radiation Al Mg
Mg Kα +233 N/A
Al Kα N/A -233
O Kα +961.7 +728.7
Cu Lα +556.9 +323.9
2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 71

71

Birck Nanotechnology Center

• Quantitative Analysis: peak widths, line shapes, background and background

subtraction.
• Quantitative Analysis and Data Processing: sensitivity factors, ionization
cross section, asymmetry parameter, analyzer transmission, intensities, satellite
subtraction, detection limit, effect of thin overlayers, peak area, line shapes,
curve fitting.
• Depth Profiling: non-destructive and destructive methods, angle resolved
XPS, elastic and inelastic scattering, sputtering, sputtering depth calibration.
• Applications: ambient pressure XPS.
• Advance Data Processing : coverage calculation, thin film thickness
calculation.

2nd semester, 2022 ECE 658, Purdue University, Dmitry Zemlyanov 72

72

ECE658, Purdue University, D. Zemlyanov 36

 2nd semester, 2022

XPS Quantitative Analysis: Background

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 73

73

XPS Quantitative Analysis: Background subtraction

Most used backgrounds are Linear, Shirley and universal cross-section Tougaard.

Linear background for Ti 2p Tougaard background for Ti 2p

Shirley background for Ti 2p

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 74

74

ECE658, Purdue University, D. Zemlyanov 37

 2nd semester, 2022

XPS Quantitative Analysis: line shapes

Core level electrons ejected by X-rays appear in spectra with a variety of

shapes. These shapes arise due to a combination of the physics involved in the
ionization process and also distortions due to the measurement mechanism. An
idealization of these influences is the specification of an observed peak as a
convolution of a Gaussian and Lorentzian function;
where in principle the Gaussian described the measurement process
(instrumental response, x-ray line-shape, Doppler and thermal broadening)
while the Lorentzian models the lifetime broadening (natural broadening) due
to the uncertainty principle relating lifetime and energy of the ejected
electrons (ΔE × Δt ≈ h/2π).
The realities of observed spectral features are complex shapes which, in some
cases, are shown theoretically to include asymmetry in the nature of the
photoelectric peaks. Further, the line-shape chosen when fitting a given data
envelope are as much dictated by the uncertainty in the background shape as
any theoretical considerations. For these reasons, when fitting XPS spectra
with synthetic line-shapes a variety of functional forms are required.

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 75

75

Birck Nanotechnology Center

• Quantitative Analysis: peak widths, line shapes, background and background

subtraction.
• Quantitative Analysis and Data Processing: sensitivity factors, ionization
cross section, asymmetry parameter, analyzer transmission, intensities, satellite
subtraction, detection limit, effect of thin overlayers, peak area, line shapes,
curve fitting.
• Depth Profiling: non-destructive and destructive methods, angle resolved
XPS, elastic and inelastic scattering, sputtering, sputtering depth calibration.
• Applications: ambient pressure XPS.
• Advance Data Processing : coverage calculation, thin film thickness
calculation.

2nd semester, 2022 ECE 658, Purdue University, Dmitry Zemlyanov 76

76

ECE658, Purdue University, D. Zemlyanov 38

 2nd semester, 2022

XPS Quantitative Analysis: data quantification

Photoelectron Peak Intensities

Photoelectron peak intensity Nk produced by subshell k
can be calculated within a three-step-like model by
integrating the differential intensities dNk originating in
the various volume elements of the specimen.
⎡ Instrument ⎤
⎡ X − ray flux ⎤ ⎡ Acceptance solid angle of ⎤ ⎢ ⎥
dN k = ⎢ ×
⎥ ⎢ ×
⎥ ⎢ d etection ⎥
⎣ at x, y, z ⎦ ⎣ electron analyser at x, y, z ⎦ ⎢ efficiency ⎥
⎣ ⎦
⎡ Number of atoms (molecules) ⎤ ⎡ Differential cross − ⎤
×⎢ ⎥×⎢ ⎥
⎣ at dx, dy, dz ⎦ ⎣section for k subshell ⎦
⎡ Probability for no − loss ⎤
⎢ ⎥
× ⎢ escape from speciment with ⎥
⎢⎣ neglegible direction change ⎥⎦

C.S. Fadley “Basic Concepts of X-ray Photoelectron Spectroscopy”

From “Electron Spectroscopy, Theory, Techniques, and Applications” C.R. Brundle and A.D. Baker, Eds., Pergamon Press, 1978, Volume 11,
Chapter 1
2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 77

77

XPS Quantitative Analysis: data quantification

Instrument dependent parameters.

This is a number of emitters Under the certain conditions, this
and this allows TO can be treated as CONSTANT.
QUANTIFY XPS data It can be called Instrument Function.

The probability to excite an

electron is different for
⎡ Instrument ⎤ different electron levels
⎡ X − ray flux ⎤ ⎡ Acceptance solid angle of ⎤ ⎢ ⎥
dN k = ⎢ ×
⎥ ⎢ ×
⎥ ⎢ d etection ⎥ and it depends on photon
⎣ at x, y, z ⎦ ⎣ electron analyser at x, y, z ⎦ ⎢ efficiency ⎥ energy as well.
⎣ ⎦

⎡ Number of atoms (molecules) ⎤ ⎡ Differential cross − ⎤

×⎢ ⎥×⎢ ⎥
⎣ at dx, dy, dz ⎦ ⎣section for k subshell ⎦
This is due to travelling through
⎡ Probability for no − loss ⎤ material. This depends on (1)
⎢ ⎥ kinetic energy of
× ⎢ escape from speciment with ⎥ photoelectrons and (2)
⎢⎣ neglegible direction change ⎥⎦ material.
2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 78

78

ECE658, Purdue University, D. Zemlyanov 39

 2nd semester, 2022

XPS Quantitative Analysis: data quantification

The acceptance solid angle Ω0 as well as the effective

specimen A0, over which Ω is not equal to zero, may vary
with electron kinetic energy.
The probability for no-loss escape from the specimen can
be written as T(Ekin, kf, x,y,z), which can be simplified as

æ -z ö
T ( Ekin , k f , x, y, z ) = exp ç ÷
L ( E
è e kin ) cos q ø
The instrument detection efficiency D0 is the probability
that a no-loss electron escaping from the specimen in a
direction encompassed by the acceptance solid angle will
yield a single final count. D0 is supposed to be constant for
the regime of constant Pass Energy.

dσ s ⎛ −z ⎞
dN s = I 0 × ρ × × Ω(Ekin , x, y, z) × D0 × exp ⎜ ⎟ dxdydz
dΩ Λ (E
⎝ e kin )cos θ ⎠

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 79

79

XPS Quantitative Analysis: data quantification

1. Semi-infinite specimen, atomically clean surface, peal k with Ekin=Es

+∞ +∞ 0
dσ s ⎛ −z ⎞
N s (θ ) ≡ ∫ dN s = ∫∫∫I ×ρ× × Ω(Es , x, y, z) × D0 × exp ⎜ ⎟ dx dy dz
−∞ −∞ −∞
0
dΩ Λ (E
⎝ e s )cos θ ⎠
dσ s
= I 0 × Ω0 (Es ) × A0 (Es ) × D0 × ρ × × Λ e (Es )cosθ
dΩ
dσ s
= InstrumentFunction × ρ × × Λ e (Es )cosθ
dΩ

dσ s
N s (θ ) = Instrument Function × ρ × × Λ e (Es )cosθ
dΩ

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 80

80

ECE658, Purdue University, D. Zemlyanov 40

 2nd semester, 2022

XPS Quantitative Analysis: data quantification

N s (θ )
ρi =
dσ k
Λ (E )cosθ × Instrument (Transmission) Function
dΩ e s
Area under Photoemission Peak i
Normalized Area i =
Realative Sensetivity Factori × Transmission Function(Eikinetic ) × IMFP(Eikinetic )

Normalized Area i
Ci (atomic %) = N

∑ Normalized Area j
j

In approximation of homogeneously mixed elements !!!

IMFP(Eikinetic ) ≈ constant × E α
α = 0.5 − 1
2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 81

81

Birck Nanotechnology Center

• Quantitative Analysis: peak widths, line shapes, background.

• Quantitative Analysis and Data Processing: sensitivity factors, ionization
cross section, asymmetry parameter, analyzer transmission, intensities,
background subtraction, satellite subtraction, detection limit, effect of thin
overlayers, peak area, line shapes, curve fitting.
• Depth Profiling: non-destructive and destructive methods, angle resolved
XPS, elastic and inelastic scattering, sputtering, sputtering depth calibration.
• Applications: ambient pressure XPS.
• Advance Data Processing : coverage calculation, thin film thickness
calculation.

2nd semester, 2022 ECE 658, Purdue University, Dmitry Zemlyanov 82

82

ECE658, Purdue University, D. Zemlyanov 41

 2nd semester, 2022

Basic Principles of XPS: Angle Resolved XPS

!w
d 3l

dσ s
N s (θ ) = InstrumentFunction × ρ × × Λ e (Es )cosθ
dΩ

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 83

83

Basic Principles of XPS: Angle Resolved XPS

Native oxide on Al film - A l 2p analysis

-2 Al 2O 3 0 degree
x10
15 degree
100 q
30 degree
!w 45 degree
Intensity(CPS)

80 60 degree
Al 75 degree
60

40

20

84 80 76 72 68
Binding Energy (eV)
2nd semester, 84
2022 ECE 658, Purdue University, Dmitry Zemlyanov

84

ECE658, Purdue University, D. Zemlyanov 42

 2nd semester, 2022

Basic Principles of XPS: Angle Resolved XPS

q Photoemission PEG features q Photoemission S bound to Au

C-O Angle, q S non-bound
Sample C-C 80
0 36.7 53.1 66.4 78.5 Sample to Au
q = 75°
O=C-OH q = 75°

contribution, %%
70
80

Relativecontribution,
70
60
60 S bound to Au
50 S boundStonon-bound to Au
q = 60°
Au
q = 60°
50 S non-bound to Au
Intensity, a.u.

40
40

Relative
30
30
q = 45° q = 45°
20
20

1.0 0.8 0.6 0.4 0.2

-10 0 10 20 30 40 50 60 70 80
q = 30°
cosq
Angle q = 30°

q = 0°
OH
HS O O q = 0°
7
Residual O
OH SH OH OH OH
hydrocarbons O O O O
O
290 285 O O O O 166 164 162 160
7
Binding Energy, eV
O
O O O O Bindimg Energy, eV
7 O 7 7 7

S O S S S

Au

The cleaned gold substrates were then immersed into 0.01 M thiolated polyethylene glycol (PEG) acid (HSC2H4(OC2H4)8COOH) in
ethanol.
2nd semester, 85
2022 ECE 658, Purdue University, Dmitry Zemlyanov

85

Basic Principles of XPS: Angle Resolved XPS

Angle Resolved XPS is non-destructive depth

profiling of 5nm topmost layer!!!

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 86

86

ECE658, Purdue University, D. Zemlyanov 43

 2nd semester, 2022

Basic Principles of XPS: Depth Profiling

Sputtering by
Ar +( He + , Ne + etc) X-
ra
y
gu Photo elec tro ns
n

!w

Sample Sam ple

Depth profiling by ion sputtering is

destructive methods !!!

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 87

87

Basic Principles of XPS: Depth Profiling

SiO SiN
SiO
SiO2 SiN
SiO2
0
Si
PECVD
Intensity, a.u.

st
1 layer of
SiO2
SiOx
interface
Si layer

nd
2 layer of
SiO2

105 100 95 540 535 530 525 405 400 395

Binding Energy, eV

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 88

88

ECE658, Purdue University, D. Zemlyanov 44

 2nd semester, 2022

Basic Principles of XPS: Depth Profiling X-

ra
y
gu Photo elec tro ns
n

!w

Sam ple

Ag foil exposed to NO/O2 mixture

Applied Surface Science 133 (1998) 171–183

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 89

89

Basic Principles of XPS: Depth Profiling

In teraction of S ilver w ith a N O/ O 2 M ixtu re L an gm u ir, Vol. 14, N o. 12, 1998 3246
3245 L an gm u ir, Vol. 14, N o. 12, 1998 Z em lyan ov et a
3244 L an gm u ir, Vol. 14, N o. 12, 1998 Z em lyan ov et al.

du e t o t h e over la p of t h e C K , N K , a n d O K X-r a y em ission

lin es. Th e loca l E E DX a n a lysis per for m ed in t h e a r ea
bet ween t h e AgNO 3 cr yst a ls r evea led a n it r ogen con cen -
t r a t ion sign ifica n t ly lower t h a n t h e st oich iom et r ic va lu e
for AgNO 3 . Th is m a y be expla in ed by a ssu m in g t h e
pr esen ce of a t h in AgNO 3 film : if t h e t h ickn ess of t h e
n it r a t e la yer wa s sm a ller t h a n t h e dept h of a n a lysis of
ou r E E DX, a lso t h e u n der lyin g Ag 2 O wou ld be det ect ed,
r esu lt in g in a n a ppa r en t over a ll n it r ogen deficien cy. Th e
t h ickn ess of t h e film m a y be t h er efor e est im a t ed t o be a
few m on ola yer s. Th e det ect ion of AgNO 3 a s t h e on ly
species in t h e u pper m ost la yer s by E E DX well cor r ela t es
wit h t h e XP S da t a , wh ich h a ve sh own t h e a bsen ce of silver
oxide a n d t h e pr esen ce of on ly AgNO 3 in t h e u pper su r fa ce
la yer s of t h e t r ea t ed silver foil.
Th e m or ph ology of t h e u n der lyin g Ag 2 O la yer is sh own
in F igu r e 1-3, t h e SE M m icr ogr a ph s wer e obt a in ed a ft er
wa sh in g off AgNO 3 wit h dist illed H 2 O. Th e su r fa ce
appears to be considerably rough (Figure 1-3a). The EEDX
a n a lysis r evea led t h e pr esen ce of on ly silver , ca r bon , a n d
oxygen on t h e su r fa ce. No n it r ogen cou ld be det ect ed on
t h is pa r t of t h e sa m ple.
F igu r e 1-2 sh ows t h e cen t r a l a r ea (“bea ch ”) of t h e sa m ple
a t low a n d h igh m a gn ifica t ion , a t t h e bou n da r y wit h t h e
edge wh ich wa s soa ked in wa t er wh er e t h e silver n it r a t e
la yer ca m e on ly sh or t ly in t o con t a ct wit h H 2 O a n d cou ld
n ot be com plet ely dissolved. Th e “bea ch ” a r ea is ch a r -
a ct er ized by t h e developm en t of la r ge AgNO 3 cr yst a ls.
Th e cr yst a ls exh ibit a wide size dist r ibu t ion , r a n gin g fr om
less t h a n 0.1 t o 1.3 µm , a n d a r e ch a r a ct er ized by a n
ir r egu la r sh a pe in con t r a st t o t h e AgNO 3 cr yst a ls for m ed F ig u re 4. Cor e level XP S spect r a of C 1s obt a in ed fr om t h
a ft er t r ea t m en t wit h t h e NO/O 2 m ixt u r e (F igu r e 1-1). As silver foil t r ea t ed wit h a m ixt u r e of NO/O 2 a n d pa r t ia lly soa ke
in dist illed wa t er . Th e spect r a wer e r ecor ded a t t h e posit ion
m a y be expect ed, a con sider a ble r ecr yst a lliza t ion of t h e sh own in F igu r e 1-0. Th e r esu lt of t h e cu r ve fit a n a lysis fo
AgNO 3 cr yst a ls seem s t o h a ve been ca u sed by con t a ct wit h spect r a 1 a n d 9 is pr esen t ed.
wa t er . As ca n be seen fr om t h e m icr ogr a ph a t h igh F ig u re 2. Cor e level XP S spect r a of O 1s obt a in ed fr om Ft higeu re 3. Cor e level XP S spect r a of N 1s obt a in ed fr om t h e
silver foil t r ea t ed wit h a m ixt u r e of NO/O 2 a n d pa r t ia lly soasilver
ked foil t r ea t ed wit h a m ixt u r e of NO/O 2 a n d pa r t ia lly soa ked
magnification in Figure 1-2b, this boundary region exhibits
a differ en t m or ph ology t h a n t h e “n it r a t e” a r ea . Th e la r ge
Ag foil exposed to NO/O mixture
in dist illed wa t er . Th e spect r a wer e r ecor ded a t t h2e posit ion in dist
sh O
s illed wa t er . Th e spect r a wer e r ecor ded a t t h e posit ion s
own in F igu r e 1-0. wa t er a t 300 K on a Ag 2 O sa m ple wa s discu ssed by Wea ve
sh own in F igu r e 1-0. Th e r esu lt of t h e cu r ve fit a n a lysis for
AgNO
F ig u re 1. P h ot ogr a ph 3 cr
of t h e silver yst
foil t r eaat ed
ls inof t hu er e of“bea
a m ixt 2 NO/O ch ” a urnea
(1:100) der aa t mrosph
e discr et er e for
er ic pr essu a n15dm in a t 3001sK aspectnd r a 1 a n d 9 is pr esen t ed.
a n d H oflu n d.27
wa sh ed by dist illed wa t er on t h e r igh t side (1-0). SE M m icr ogr a ph s of (1-1) t h e left “n it r a t e” pa r t , (1-2) t h e “bea ch ” a r ea , wh ich
disper sed on t o a su bst r a t e, wh ich sh ows st r on g m or ph o-
is loca t ed in t h e m iddle bet ween t h e wa sh ed pa r t of t h e t r ea t ed foil a n d t h e u n wa sh ed on e, a n d (1-3) t h e r igh t “oxide” side.
Th e C 1s fea t u r e con sist s of t wo u n r esolved pea ks a
logical similarities to the uncovered Ag2O (see Figure 1-3b). ∼10 -10 m ba r ) a r e bet t er t h a n in t h e SE M (ba se pr essu frrome oxygen in cor por a t ed in t o AgNO 3 . Th is a ssign m en t 284.4 a n d 285.5 eV a n d a n ext en ded t a il t owa r d h igh BE
To ch a r a ct er ize tTh
he m e orEphEology
DX ofa nt hae lysis la yer onoft hteh e r Th egione pa r tsicles
beth aween ve a r egut h la re shAgNO ∼10
a pe a n d 3a r e st a t ist ica lly-6 mbar). The better vacuum partly suppresses carbon is su ppor t ed by t h e sim u lt a n eou s a ppea r a n ce of a N 1s va lu es (F igu r e 4). Th e cu r ve fit a n a lysis of t h is fea t u r
“oxide” a r ea a n d on t h e “n it r a t e” on e, t h e silver foil sh own
in F igu r e 1-0 wa s stcr yst abyl minea ntshofisSEaM.
u died r eaP arr evea
t s 1-1 led pa exclu Be aware of possible artifacts (Ar beam damage)!
dist r ibu t ed on t h e su r fa ce, wit h a m ea n dist a n ce com -
r a ble tsively
o t h eir size. t h eA loca
pr esen
l E E DX cea nofa lysis u n a mdeposit
bigu - ion , wh ich is in deed a m a jor pr oblem of elect pea r on k a t 406.135eV (F igu r e 4), wh ich is ch a r a ct er ist ic of r evea ls t h e pr esen ce of t h r ee pea ks wit h m a xim a a t 284.3
t h r ou gh 1-3 of F igu silver , ca r bon , a n d oxygen .
r e 1 sh ow t h e SE M m icr ogr
obt a in ed fr om t h e “n it r a t e” side of t h e sa m ple, fr om t h e
a ph s
r esu lt su
ou sly
ofppor
r evea
t 3ht bein
led Always double-check!
t h e pr esen ce of n it r ogen in t h
in g t h e XP S da t a , wh ich clea r ly poin t ed ou t t o
e pa m icr oscopes, beca u se t h e a dsor bed h ydr oca r bon s a r e n it r a t e. Th e r igh t a r ea (“n it r a t e”) of t h e sa m ple
r t icles, silver 285.7, a n d 287.6 eV t h a t cou ld be t en t a t ively a ssign ed t
bou n da r y on e a n d fr om Sut hm e m a r izin
“oxide” a r eag, r espect
t h e ively. AgNO e SE g t hM e m aa innsu arlysis,
fa ce com pou t h en d. Th e su con ver t ed in t o gr a ph it e by t h e h igh -en er gy elect r on beaexh
r fa ce m . ibit s a m a in O 1s pea k a t 532.0 eV a n d a N 1s sign a l h ydr oca r bon s, t o O-con t a in in g h ydr oca r bon s su ch a
F igu r e 1-1 clea r ly demfollowin
on st r a t esgt hkey
a t t h efea t u rrees
exposu of t sh
h e ou ldm orbe n ot of
ph ology ed.t h e bou n da r y a r ea is sh own in F igu rSu ch a ca r bon con t a m in a t ion is pr a ct ica lly a bsen t u nader
e 1-1b. t 406.1 eV (F igu r es 2 a n d 3). Th is su ppor t s t h e a lcoh ol, a n d t o ca r bon a t es, r espect ively. Reh r en et a
silver foil t o t h e NO/O 2 m ixt u r e dr a st ica lly ch a n ges t h e Th e or igin a lly sm oot h su r fa ce of t h e silver foil is n ow a ssu m pt ion t h a t t h e m a in species cover in g t h e su r fa ce of r epor t ed 21 a bou t t h e C 1s pea k a t 287.6 eV bein g
m or ph ology of t h e su r faTh ce. eF igu
exposu
r e 1-1a sh r eowsoft hsilver
e for m a - t o au nNO/O
even . At2t em gaptss tm ixta inu qu
o obt r ea nat itt a300
t ive E E DX da tXP a fr omS wor kin g con dit ion s.
t ion of pa r t icles wit hK leavadsr yintgobet t h e for0.1ma nadt0.4
ion µmof . a AgNO
t h e a r ea3slabetyerweenon t h et o t h e su r fa
a lsce. l, e XP S spot of a n a lysis of ou r in st r u m en t is tahp- e t r ea t ed silver is AgNO 3 . pecu lia r it y of ca r bon a t es on t h e silver su r fa ce. Th e r a t i
2ndLangmuir 1998, 14, 3242-3248
size ween AgNO 3 cr yst wer e u n su ccessfu Th
semester, Th is com pou n d ca n be efficien t ly wa sh ed off by H 2 O t o pr oxim a t ely 4 × 7 m m , viz., sign ifica n t ly la r ger t h a n t hTh e e pea k a t 529.5 eV pr esen t ed in F igu r e 2 is ver y close bet ween t h e a r ea of t h e oxygen pea k a t 530.8 eV a n d t h
u n cover a r ou gh , por ou s Ag 2 O la yer , a s sh ownECE in pa r658, t s 1-1 Purdue a r eaUniversity,
a n a lyzed by Dmitry E E DX. Th e “n it r a t e” atroeat h e va lu e of 529.2 90
SE M a n dZemlyanov eV r epor t ed by Tjen g et a l.26 a n d a r ea of t h e ca r bon sign a l a t 287.6 eV is equ a l t o 3.60. Th i
2022 t h r ou gh 1-3a of F igu r e 1. m ea su r ed a bou t 8 × 8 m m a n d t h e “oxide” a r ea wa Wea 27
s ver et a l. for t h e BE of t h e O 1s sign a l of oxygen in r a t io is h igh er t h a n t h e st oich iom et r ic O:C a t om ic r a t i
Th e a s for m ed silver n it r a t e la yer con sist s of sm a ll a ppr oxim a t ely 6 × 8 m m . Sever a l XP spect r a wer Age2 O. F or t h ese r ea son s, t h e pea k a t 529.5 eV m igh t be in ca r bon a t es: t h is is pr oba bly du e t o t h e con t r ibu t ion o

90 AgNO 3 particles, embedded probably in a continuos AgNO 3 r ecor ded a t differ en t posit ion s on t h e cen t er of t h e spot a ssign ed t o t h e oxygen of silver (I) oxide. Th e O 1s pea k ot h er oxygen - a n d ca r bon -con t a in in g species. Th u s, t h
film . Th e sh or t con t a ct of t h is la yer wit h wa t er , a s in t h e of a n a lysis in or der t o obt a in in for m a t ion a bou t t h e spa at ia t 530.8
l eV m a y be du e t o oxygen dissolved in t h e bu lk a ppea r a n ce of a C 1s pea k a t 285.7 eV m a y be a ssign e
“bea ch ” a r ea , r esu lt s in t h e colla pse of t h e st r u ct u r e a n d dist r ibu t ion of t h e elem en t s. Th e cor e level XP spect silver
ra
10,14,21,25,27,38 or t o ca r bon a t es. 16,20-22 E ven t u a lly, a t o oxygen -con t a in in g h ydr oca r bon s, wh ich ca n for m b
t h e for m a t ion of a discr et e, big (u p t o 1.3 µm ) AgNO 3 of t h e O 1s, N 1s, a n d C 1s lin es r ecor ded a t differwea en t k pea k a t 533.2 eV cou ld be dist in gu ish ed by m ea n s con t a ct of t h e t r ea t ed silver foil wit h wa t er a n d a ir . Th
cr yst a l, pa r t ia lly u n cover in g t h e su bst r a t e. Th e a g- posit ion s a r e sh own in F igu r es 2 t h r ou gh 4, r espect ively. of cu r ve fit a n a lysis a ssu m in g a Ga u ssia n pea k sh a pe. A r a t io bet ween t h e a r ea of t h e O 1s pea k a t 530.8 eV a n
glom er a t ion of t h e AgNO 3 pa r t icles in t h e “bea ch ” a r ea Th e est im a t ed posit ion s of t h e cen t er of t h e spot of a n a lysisst u dy of t h e in t er a ct ion of wa t er a n d oxygen wit h Ag(110) t h e su m of t h e a r ea s of t h e C 1s pea ks a t 285.7 a n d 287.
poin t s t o a wea k in t er a ct ion of AgNO 3 wit h t h e u n der lyin g du r in g ea ch m ea su r em en t a r e m a r ked in F igu r e 1-0. Th by eAu et a l.14 r epor t a n O 1s BE of ca . 533.5 eV for m olecu la r eV m ea su r ed a t posit ion 1 (r igh t edge of t h e silver foil) i
com pou n d (su pposedly Ag 2 O). O 1s spect r u m obt a in ed a t t h e r igh t edge of t h e sa mwa plet er . H owever , it is kn own t h a t H 2 O m olecu les on silver a ppr oxim a t ely 0.92.
Th e for m a t ion of AgNO 3 cr yst a ls wa s obser ved a t a (wh er e we a ssu m ed Ag 2 O) sh ows a st r on gly a sym m etoxide r ic pa r t ia lly dissocia t e a n d pa r t ia lly desor b a t 150 K.
14 Th e O 1s, N 1s, C 1s XP spect r a , obt a in ed su bsequ en t l
t em per a t u r e, wh ich is sign ifica n t ly lower t h a n t h e m elt in g fea t u r e (see F igu r e 2-9), wh er ea s t h e O 1s spect rTh u me SE M m icr ogr a ph of t h e a r ea t h a t wa s soa ked in wa t er a ft er sh ift in g t h e sa m ple by st eps of a ppr oxim a t ely 1-1.
poin t of AgNO 3 (T m (AgNO 3 ) ) 485 K).37 Th e pr oba ble m ea su r ed a t t h e left edge displa ys a br oa d, n earrevea ly led a ver y r ou gh a n d por ou s st r u ct u r e. Su ch a n of m m fr om t h e u n wa sh ed “n it r a t e” t o t h e “oxide” a r ea of t h
m ech a n ism of t h e AgNO 3 cr yst a l for m a t ion will be sym m et r ic, pea k a t ∼532.0 eV (F igu r e 2-1). A cu r veoxide fit su r fa ce m igh t sh ow a n en h a n ced a ct ivit y t owa r d foil, a r e sh own in F igu r e 2 t h r ou gh F igu r e 4. Th e cu r v
a n a lyzed in t h e discu ssion . a n a lysis sh owed t h a t bot h spect r a ca n be descr ibed at sh ea a dsor pt ion of H 2 O a t r oom t em per a t u r e, a n d it seem s fit a n a lysis of t h e O 1s sign a l, exa m ples of wh ich a r e sh ow
ECE658, Purdue University, D. Zemlyanov 3.2. X-ra y P h o to e le c tro n S p e c tro s c o p y . XP S wa s
em ployed a s a n a lyt ica l t ool in or der t o det er m in e t h e
lin ea r com bin a t ion s of fou r Ga u ssia n pea ks, h a vin
a ppr oxim a t ely t h e sa m e fwh m of ∼1.7 eV a n d loca t edt oaOH
r eagson a ble t o a ssign t h e O 1s pea k a t 533.2 eV t o H 2 O or
t -gr ou ps. Th e possibilit y of t h e pr esen ce of a dsor bed
in spect r a (1) a n d (9) in F igu r e 4, a llowed a n est im a t io
45
of t h e a t om ic r a t io bet ween t h e differ en t O-con t a in in
n a t u r e of t h e ch em ica l com pou n ds on t h e su r fa ce of t h e 533.2, 532.1, 530.8, a n d 529.5 eV. Com pa r ison of t h ese species a t ea ch poin t of t h e sa m ple. Th e r esu lt is pr esen t e
t r ea t ed silver foil sh own in F igu r e 1-0. Th e va cu u m da t a wit h lit er a t u r e (Ta ble 1) a llows t h e pea ks t o be (38) Bu kh t ia r ov, V. I.; Bor on in , A. I.; P r osvir in , I. P .; Sa vch en ko, V. in t h e in set of F igu r e 2. Th e sa m e pr ocedu r e for t h
con dit ion s in t h e XP S a n a lyzer ch a m ber (ba se pr essu r e a ssign ed a s follows. Th e O 1s pea k a t 532.1 eV a r ises I. J . Catal. 1994, 150, 268. C-con t a in in g species gives t h e r esu lt sh own in t h e in se
 2nd semester, 2022

Basic Principles of XPS: Depth Profiling

Ni – 65 nm
Cr – 60 nm
Ni – 65 nm
Cr – 60 nm
Ni – 65 nm
Cr – 60 nm
Si wafer

1−m
YA ⎛m ⎞ ⎛U ⎞
=α ×⎜ A⎟ ×⎜ B⎟
Sputtering Rate: YB ⎝ mB ⎠ ⎝UA ⎠
•Sputtering rate determines time for depth profile.
•Sputtering rate can be varied with raster size or YA , YB - the sputtering yields,
beam current. mA , mB - atomic masses,
• For sample with layers, sputtering rate can be UA , UB - surface binding energies of A and B,
different for each layer. α a constant and m a parameter of the order of
• For sample with bi-material layers, sputtering rate 0.25.
can be different for each material.
2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 91

91

Basic Principles of XPS: Depth Profiling

Depth profiling by ion sputtering is

destructive methods !!!

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 92

92

ECE658, Purdue University, D. Zemlyanov 46

 2nd semester, 2022

Birck Nanotechnology Center

• Qualitative Analysis: peak widths, line shapes, background.

• Quantitative Analysis and Data Processing: sensitivity factors, ionization
cross section, asymmetry parameter, analyzer transmission, intensities,
background subtraction, satellite subtraction, detection limit, effect of thin
overlayers, peak area, line shapes, curve fitting.
• Depth Profiling: non-destructive and destructive methods, angle resolved
XPS, elastic and inelastic scattering, sputtering, sputtering depth calibration.
• Applications: ambient pressure XPS.
• Advance Data Processing : coverage calculation, thin film thickness
calculation.

2nd semester, 2022 ECE 658, Purdue University, Dmitry Zemlyanov 93

93

Basic Principles of XPS: Ambient pressure XPS

Close-up of sample-first aperture region

Gas phase composition can

be measured by XPS.
gas phase signal:
hν 1 torr·mm ~ a few monolayers
p0

e-
1.0
d mm
d=1
p/p0

0.5

0.0
-2 -1 0 1 2
z [d]
z mm
z=2

gas

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 94

94

ECE658, Purdue University, D. Zemlyanov 47

 2nd semester, 2022

Basic Principles of XPS: Ambient pressure XPS - In situ XPS system at BESSY-2

X-rays enter the cell at Hemisphercal

55° incidence through an electron
SiNx window analyzer
(thickness ~ 1000 Å) (10-9p0)

Analyzer
input lens
mass spectrometer
and additional Focal point
pumping of analyzer
input lens

Third differential
pumping stage (10-8p0)

Experimental cell Second differential

supplied by gas pumping stage (10-6p0)
lines (p0)

First differential
2nd semester,
pumping stage (10-4p0)
2022 ECE 658, Purdue University, Dmitry Zemlyanov 95

95

Basic Principles of XPS: Ambient pressure XPS

O 1s PdO
Pd 3p3/2 O (II)
gas phase
Pd 3d5/2
O(II) PdO
Pdox1
O(III) Pd 3d3/2
Pd 3p3/2 O(I) Pd
0

249 min
249 min
Intensity, arb. units
Intensity, arb. units

129 min
132 min

Pdox2

37 min
40 min

340 335

534 532 530 528 526 Binding Energy, eV

Binding Energy, eV Pd0

Pd(111) in 0.3mbar O2 at 650K

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 96

96

ECE658, Purdue University, D. Zemlyanov 48

 2nd semester, 2022

Birck Nanotechnology Center

• Qualitative Analysis: peak widths, line shapes, background.

• Quantitative Analysis and Data Processing: sensitivity factors, ionization
cross section, asymmetry parameter, analyzer transmission, intensities,
background subtraction, satellite subtraction, detection limit, effect of thin
overlayers, peak area, line shapes, curve fitting.
• Depth Profiling: non-destructive and destructive methods, angle resolved
XPS, elastic and inelastic scattering, sputtering, sputtering depth calibration.
• Applications: angle resolved XPS, ambient pressure XPS.
• Advance Data Processing : coverage calculation, thin film thickness
calculation.

2nd semester, 2022 ECE 658, Purdue University, Dmitry Zemlyanov 97

97

XPS Quantitative Analysis: data quantification – coverage calculation

. 2. Semi-infinite substrate with a non-attenuating

overlayer at fractional monolayer coverage

dσ substraye
sadlayer N adlayer (θ ) × Λ substrate
e
(Esubstrate )cosθ
Coverage ≡ = dΩ
ssubstrate N substrate (θ ) dσ adlayer
×d
dΩ

C.S. Fadley “Basic Concepts of X-ray Photoelectron Spectroscopy”

From “Electron Spectroscopy, Theory, Techniques, and Applications” C.R. Brundle and A.D. Baker, Eds., Pergamon Press, 1978, Volume 11,
Chapter 1
2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 98

98

ECE658, Purdue University, D. Zemlyanov 49

 2nd semester, 2022

XPS Quantitative Analysis: data quantification – film thickness calculation

3. Specimen of thickness t atomically clean

surface, peal k with Ekin=Es:
.
Three different ways to calculate a film thickness:
⎛ ⎛
⎜ 1− exp ⎜ overlayer
−t ⎞⎞
⎟⎟
1) Compare intensity from layer with intensity of infinite
⎜⎝ (Eoverlayer )cosθ ⎠ ⎟⎠
⎝ Λe layer (need two samples);
2) Compare intensity of substrate coated and not coated
(usually distractive);
3) Compare intensities of layer and substrate (preferential).

dσ s ⎛ ⎛ −t ⎞⎞
N overlayer (θ ) = InstrumentFunction × ρ × × Λ overlayer (Eoverlayer )cosθ × ⎜ 1− exp ⎜ overlayer ⎟⎟
dΩ e
⎝ Λ
⎝ e (Esubstatrate
)cos θ ⎠⎠

N overlayer (θ ) ⎛ ⎛ −t ⎞⎞ ⎛ N (θ ) ⎞
∞
= ⎜ 1− exp ⎜ overlayer ⎟⎟ t = −Λ overlayer (Eoverlayer )cosθ × ln ⎜ 1− overlayer ⎟
N overlayer (θ ) ⎝ Λ
⎝ e (Eoverlayer
)cos θ ⎠⎠ e
⎝ N overlayer (θ ) ⎠
∞

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 99

99

XPS Quantitative Analysis: data quantification – film thickness calculation

3. Specimen of thickness t atomically clean

. surface, peal k with Ekin=Es:

Three different ways to calculate a film thickness:

1) Compare intensity from layer with intensity of infinite
layer (need two samples);
2) Compare intensity of substrate coated and not coated
(usually distractive);
3) Compare intensities of layer and substrate (preferential).

⎛ −t ⎞
exp ⎜ overlayer
⎝ Λe (Esubstrate )cosθ ⎠⎟

N substrate (θ ) ⎛ −t ⎞
= exp ⎜ overlayer
without overlayer
N substrate (θ ) ⎝ Λe (Esubstrate )cosθ ⎟⎠
⎛ N substraite (θ ) ⎞
t = −Λ overlayer (Esubstrate ) × cosθ × ln ⎜ without
e overlayer
⎝ N substrate (θ ) ⎟⎠

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 100

100

ECE658, Purdue University, D. Zemlyanov 50

 2nd semester, 2022

XPS Quantitative Analysis: data quantification – film thickness calculation

. 4. Semi-infinite substrate with uniform

overlayer of thickness t

⎛ −t ⎞
dσ l 1 − exp ⎜ overl
ρoverl × × Λ overl (E ) ⎟
N l (θ ) dΩ e l ⎝ Λ e (El )cosθ ⎠
= ×
N s (θ ) dσ s ⎛ −t ⎞
ρsubst × × Λ subst
e
(Es ) exp ⎜ overl ⎟
dΩ ⎝ Λ (E )cosθ ⎠
e s

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 101

101

XPS Quantitative Analysis: data quantification – film thickness calculation

⎛ −t ⎞
dσ l 1 − exp ⎜ overl ⎟
N l (θ ) ρoverl × dΩ × Λ e (El )
overl
⎝ Λ e (El )cosθ ⎠
= ×
N s (θ ) dσ s ⎛ −t ⎞
. ρsubst × × Λ subst
e
(Es ) exp ⎜ overl ⎟
dΩ ⎝ Λ (E )cosθ ⎠
e s

O – for overlayer
S – for substrate
λo is the attenuation length of
photoelectrons within the overlayer that
originated in the overlayer.
so and ss are the relative sensetivity factors
(RSF) for overlayer and substrate

SURFACE AND INTERFACE ANALYSIS Surf. Interface Anal. 29, 403–406 (2000) “The Thickogram: a method for easy
film thickness measurement in XPS” Peter J Cumpson
2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 102

102

ECE658, Purdue University, D. Zemlyanov 51

 2nd semester, 2022

XPS Quantitative Analysis: data quantification – film thickness calculation

⎛ −t ⎞
dσ l 1 − exp ⎜ overl ⎟
N l (θ ) ρoverl × dΩ × Λ e (El )
overl
Λ
⎝ e (E )cos θ ⎠
= × l

N s (θ ) dσ s ⎛ −t ⎞
. ρsubst × × Λ e (Es )
subst
exp ⎜ overl ⎟
dΩ ⎝ Λ (E )cosθ ⎠ e s
−t
Λ overlayer
e
(Eoverlayer )cosθ
dσ overlayer
N overlayer (θ ) ρoverlayer
dΩ
dσ overlayer
dσ substrate N overlayer (θ ) ρoverlayer
N substrate (θ ) ρsubstrate dΩ
dΩ
dσ substrate
N substrate (θ ) ρsubstrate
dΩ

SURFACE AND INTERFACE ANALYSIS Surf. Interface Anal. 29, 403–406 (2000) “The Thickogram: a method for easy
film thickness measurement in XPS” Peter J Cumpson
2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 103

103

where

Birck Nanotechnology Center

Introduction to XPS: film thickness

dσ overlayer
N overlayer (θ ) ρoverlayer
dΩ dσ overlayer
dσ substrate N overlayer (θ ) ρoverlayer
dΩ
. N substrate (θ ) dΩ
ρsubstrate
dσ substrate
N substrate (θ ) ρsubstrate
dΩ
−t
Λ overlayer (Eoverlayer )cosθ
e
Ratio Thickness, Number of
Angle C:Cu nm ML
0 3.75 1.78 5.2
-45 9.39 1.97 5.8
-59.94 15.58 1.71 5.0
Average 1.8 5.4
STDV 0.1 0.4

SURFACE AND INTERFACE ANALYSIS Surf. Interface Anal. 29, 403–406 (2000) “The Thickogram: a method for easy
film thickness measurement in XPS” Peter J Cumpson
2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 104

104

ECE658, Purdue University, D. Zemlyanov 52

 2nd semester, 2022

XPS Quantitative Analysis: data quantification – film thickness calculation

XPS Thickness Solver

Kyle Christopher Smith; David A Saenz; Dmitry Zemlyanov; Andrey A Voevodin (2012), "XPS Thickness Solver,”
http://nanohub.org/resources/xpsts. (DOI: 10.4231/D3N29P603).
2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 105

105

XPS: finally! The END!!!

• Surface sensitive technique (non-destructive)

XPS can provide an element composition of the
topmost ~50Å (5nm) layer with sensitivity ~0.1 at.%.
• Quantitative analysis
XPS can measure the element concentration with
accuracy ~0.1 at.%. Coverage of adlayer can be
calculated and film thickness can be measured.
• Qualitative analysis
Chemical state, hybridization, chemical environment
etc. can be analyzed with XPS.

2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 106

106

ECE658, Purdue University, D. Zemlyanov 53

 2nd semester, 2022

Ultraviolet Photoemission Spectroscopy (UPS)

UV light ( hν = 5 to 100 eV) to excite photoelectron. From an analysis of the kinetic energy and angular
distribution of the photoelectrons, information on the electronic structure (band structure) of the material
under investigation can be extracted with surface sensitivity.

2nd semester,
2022 107
ECE 658, Purdue University, Dmitry Zemlyanov

107

Ultraviolet Photoemission Spectroscopy (UPS)

He I and He II
Excitation: He I and He II

Grotian diagram for He I and He II showing the strongest resonance lines (584 Å,
98% ofGrotian
the emission intensity,
diagram for theIIother
He I and He lines
showing the present
strongest in a He discharge
resonance lines (584 Å,are
98%537
of Å,
the emission
522 Å and 304 Å intensity,
which canthe other
havelines presentof
intensity in the
a Heorder
discharge
of 2are%537
of Å,
the522 Å and
584 304 Å
Å line.
which can have intensity of the order of 2 % of the 584 Å line.
2nd semester,
2022 108
ECE 658, Purdue University, Dmitry Zemlyanov

108

ECE658, Purdue University, D. Zemlyanov 54

 2nd semester, 2022

Ultraviolet Photoemission Spectroscopy (UPS)

Excitation: Gas Discharge Lines
Gas Emission Line Energy (eV)
Wavelength (nm) Relative Intensity (%)

H Lyman α 10.20 121.57 100

Lyman β 12.09 102.57 10

He 1α 21.22 58.43 100

1β 23.09 53.70 ~ 1.5
1γ 23.74 52.22 0.5
2α 40.81 30.38 100
2β 48.37 25.63 <10
2γ 51.02 24.30 negligible

Ne 1α 16.67 74.37 15
1α 16.85 73.62 100
1β 19.69 62.97 <1
1β 19.78 62.68 <1
2α 26.81 46.24 100
2α 26.91 46.07 100
2β 27.69 44.79 20
2β 27.76 44.66 20
2β 27.78 44.63 20
2β 27.86 44.51 20
2γ 30.45 40.71 20
2γ 30.55 40.58 20

Ar 1 11.62 106.70 100

1 11.83 104.80 50
2 13.30 93.22 30
2 13.48 91.84 15 https://en.wikipedia.org/wiki/Ultraviolet_photoelectron_spectroscopy
2nd semester,
2022 109
ECE 658, Purdue University, Dmitry Zemlyanov

109

Ultraviolet Photoemission Spectroscopy (UPS)

Excitation: Synchrotron facility with high brightness of soft radiation

2nd semester,
2022 110
ECE 658, Purdue University, Dmitry Zemlyanov

110

ECE658, Purdue University, D. Zemlyanov 55

 ,- 5 ',, A 1
Ld ,,
t 5 -
~,_
¢"

~
\

L1
ki.,i
l X ,;,; /,"
2 i ~ / f/ //
10J, '~\
t
I/~
'.,..4-.-.~l j---h~
-7--- i \ t c / 2nd semester, 2022
\ " I ,' l,~e '1

L
i
i I
/ /.1'\I I I I
10 I I I I I I
Fig. 3. The calculated bands along FL. The dashed line
corresponds to the free-electron final-state band shifted down F
158 V.M. Tapilin et al. / Surface Science 310 (1994) 155-162 on the photon energy. The dashed-dotted line shows the Fig. 5. As for Fig. 4, but along FM.
position of the Fermi level.
( K l l ~ is t h e s m a l l e s t r e c i p r o c a l l a t t i c e v e c t o r p e r -
p e n d i c u l a r to t h e ( l l l ) - p l a n e a n d E b is t h e e n -
e r g y o f t h e b o t t o m o f t h e l o w e s t b a n d ) is also comparing FM experimental spectra and theoretical spectra taken under the corresponding polar an-
s h o w n in Fig. 3. I n t e r s e c t i o n s o f this e n e r g y curve
h g ~ e- gle.
predictions. For that reason we plotted in Figs. 4
w i t h t h e e n e r g y b a n d s give us t h e e n e r g i e s satisfy- These calculated peak positions, shown as ver-
/ / / / / / / / / / / / /and
/ / / / 5 the dependence of the kinetic energy on
tical bars in Figs. 1 and 2, must be compared with
158
Ultraviolet Photoemission Spectroscopy (UPS)
ing t h e c o n d i t i o n o f e l e c t r o n k - v e c t o r c o n s e r v a -
t i o n d u r i n g Vp h o t o e x c i t a t i o n w h i c h m u s t b e ful-
. M . Tapilin et al. / Surface Science 310 (1994) 155-162
klr given by (11), for a set of emission polar
angles, which Od: were used in the experiment. The the experimental spectra.
intersections of these last curves with the ones First of all, let us consider the normal emis-
( K l l ~ is t h e smallest r e c i p r o c a l lattice v e c t o r p e r - ,~ al,a" b2 representing , the sion. The calculation predicts three peaks at 2.0,
V.M.dependence of calculated
Tapilin et al. / Surface Science 310peak
(1994) 155-162 4.2 and 8.5 eV below the Fermi 159 level. The bars,
p e n d i c u l a r to the ( l l l ) - p l a n e a n d E b iso 8t h e en- FK c
positions on kip give us the calculated position of
hq~ e" ,-; FM 36 °
ergy of the b o t t o m o f the lowest b a n d ) is also the photoemission / peaks for the bulk transition in labeled a, b and c in accordance with Figs. 4 and
shown in Fig. 3. I n t e r s e c t i o n s o f this 7e1n1 1e1r1 7g/ /y/ / / / curve
/ h g ~ e- 2. b2,b3
, / / 0° 5,6 ° show 12 ° the 18 ° calculated
2.4° 3 0 ° 3 6peak
° positions in Figs. 1
with the e n e r g y b a n d s give us t h e e n e r g i e s satisfy-
o b'l,
O---Tk ...... L L3 0 i
and 2. The experimental spectrum has two peaks
///////////////// z
ing the c o n d i t i o n o f e l e c t r o n k-vector conserva- 2 , ,b ~ 0L'2
° 10° 20030040050o60° A II and!o~ B at ~ 1.7 .4", and
Ull 4.2¢ eV which we match with
tion d u r i n g p h o t o e x c i t a t i o n which m u s t b e ful- ~- 0 2&° I I I w a MI ! ~ i ~ 1
the first two calculated peaks. The first peak
I A b'
z al/ja~ b1
a2 ,,Od: ,,, arises mainly from dx2y2 and dz2 states and the
L. ---I--..~"~-'~1 I / ~'1 second one from dxy, dy z and dxz states. The
,~ al,a" b2 , ~ ,,rb 3
,- 5 ',, A 1 .h~Ji / / . local densities of states (LDOS) at the F point of
Ld ,, LLI
o8 FK c
J t 5 - the SBZ are represented for the surface, subsur-
hq~ e" ,-; 36¢"° \ ~512° I ki.,i X ,; ,"
face and bulk layers in Fig. 6. It is shown that the
'5it- i l ,; / the peak at 8.5 eV arises is
7111117/////// 2. b2,b3
,
/
~,_ ~ L1 I
2 LDOS
i '.,..4-.-.~l from
~ j---h~ / f/ // which
J ~ c - 6o
"E
o b'l,
L 10J, '~\ c l ,'X,'i/
very
I / ~ small,
t
-7--- i so \ t cthis/ peak is difficult to observe
above the background of the inelastically scat-
.13 z 2 , ,b j A 0° vl ; , ' I tered electrons unless the transition probability is
>.- ~-
....
2&°5
i .... i .... i
\ ,:i! , " I ,' l,~e '1
I--" z
EF=0

Binding Energy (eV) L

10 15
R i
unusually high.
Peak
/ /.1'\I
AI has I an excrescence on its left which we
z a2 ,, ,,, 10
i I I
butFig. 4. Energies of are
electronic states
in along FK, satisfying the
I..d I I I I I I
I.-- Fig. 2. Fig.As for 3. TheFig. 1, calculated along FM. along
bands Bars FL. labeled
The dashed line assign to the violation of the k j_-conservation
Z
accordance with the curves condition in Fig. 5. of conservation of three-dimensional wave vector
~ ,,rb 3 corresponds to the free-electron during photoexcitation final-state band(solidshifted Dashed lines represent F
lines). down
during photoexcitation at the surface. The sur-
on the photon energy. The dashed-dotted
the dependence of energy line on kshows
N for atheset of emission polar Fig. 5. faceAs for and Fig. 4, subsurface
but along FM. LDOS, shown in Fig. 6,
J position
f i l l e d for t r a nofs i tthei o nFermi
12°s inangles.
when going from P point along F K and F M
level.
t h e bulk. Energies
T h e s e e n eof r g i electronic
es states along ΓK support andthisΓM, satisfying the
assumption.
condition of conservation of three-dimensional
d i r e c t i o n s o f t h e S B Z a r e p l o t t e d in Figs. 4 a n d 5.
spectra wave vector
taken under the corresponding polar an-
T h e comparing
e m i s s i o n experimental
a n g l e 0d, kll spectra
a n d t h eandk i ntheoretical
etic
J ~e n e r gcypredictions. during photoexcitation gle.
o f a n e-m i t6ot For
e d ethat reason
lectro (solid lines). Dashed lines represent
n a r e wer e l aplotted
t e d by tin
h e Figs. 4
"E These calculated peak positions, shown as ver-
.13 j A
e x p r e s sand
0°
the dependence of energy on k|| for
i o n 5 the dependence of the kinetic energy on
tical bars in
a set of emission polar
Figs. 1 and 2, must be compared with
.... i ....
3.841kll12/sin20d,
Eki n = klr given by (11), for a set of emission
i .... i angles. ( 1 1 ) polar the experimental spectra.
>.-
EF=0 angles, which were used in the experiment. The
w5h e r e 10Eki n a1n5 d kll a r e m e a s u r e d in e V a n d A -
o 1
,
I--"
EF=O 5 10 15 First of all, let us consider the normal emis-
Binding tintersections
r e s p e cEnergy
i v e l y . (eV) of these
A s follows f r o mlast curves
(11), o n l y with
for t the
h e ones
z Fig. 1. The P t ( l l l ) photoemission spectra taken along FK of n o r m a lrepresenting
e m i s s i o n d i fthe
f e r edependence
n t e n e r g y r e of
g i o calculated
n s in t h e peak
sion. The calculation predicts three peaks at 2.0,
I..d
I.-- the SBZ. Capital letters label the experimental Fig. 2. As for
peaks. TheFig. 1, sbut
p e c talong h aFM.
v e t hBars
r apositions eons aare
m e labeled
kip I klllus=inthe
give 0. Fcalculated
o r o t h e r e position
mis- of 4.2 and 8.5 eV below the Fermi level. The bars,
Z The Pt(lll) photoemission spectra taken along ΓΚ and ΓΜ of
vertical bars show the energy position where interband
accordancetransi-
tions could contribute according to the calculations. Bars are
with the curves
sionina n Fig.
g l e5.
the s the electrons emitted under the same
photoemission peaks for the bulk transition in labeled a, b and c in accordance with Figs. 4 and
angle with different kinetic energies have differ- 5, show the calculated peak positions in Figs. 1
Fig. 4.
the SBZ. Capital letters label the experimental peaks. The
labeled with small letters in accordance with the curves in
e n t kll a n d m u s t b e t a k e n i n t o a c c o u n t w h e n
and 2. The experimental spectrum has two peaks
vertical bars show thefilled
energy for t rposition
a n s i t i o n s inwhere
t h e bulk. These energies
interband
w h e n going f r o m P p o i n t a l o n g F K 0 ° 10° 20030040050o60° A and B at 1.7 and 4.2 eV which we match with
transitions could contribute 0 a n d F MI I I w
d i r e c t i o n saccording
o f t h e S B Z ato r e the
p l o t t calculations.
e d in Figs. 4 a nadl /5.
ja~ b1 the first two calculated peaks. The first peak
T h e emission angle 0d, kll a n d the kinetic arises mainly from dx2y2 and dz2 states and the
e n e r g y o f an e m i t t e d e l e c t r o n a r e r Le. l a t e d by---I--..~"~-'~1
the I/ ~'1 second one from dxy, dy z and dxz states. The
V.M. Tapilin et al. / Surface eScience
xpression
310 (1994) 155-162 .h~Ji / / . local densities of states (LDOS) at the F point of
2nd semester, LLI
the SBZ are represented for the surface, subsur-
2022 Eki n = 3.841kll12/sin20d, ~5 University,
ECE 658, Purdue (11) Dmitry Zemlyanov I face and bulk layers in 111
Fig. 6. It is shown that the
'5 i I o
w h e r e Eki n a n d kll a r e m e a s u r e d in e V a n d A - ,
it- 1 LDOS from which the peak at 8.5 eV arises is

111spectra taken along FK of

EF=O 5 10 15 respectively. A s follows f r o m (11), only for t h e very small, so this peak is difficult to observe
Fig. 1. The Pt(lll) photoemission n o r m a l emission d i f f e r e n t e n e r g y regions in vl
c l
the ; , ' I ,'X,'i/ above the background of the inelastically scat-
the SBZ. Capital letters label the experimental peaks. The tered electrons unless the transition probability is
vertical bars show the energy position where interband transi-
s p e c t r a have the s a m e I klll = 0. F o r o t h e r emis- ,:i! , unusually high.
sion angles t h e e l e c t r o n s e m i t t e d u n d e r the s a m e R
tions could contribute according to the calculations. Bars are Peak A has an excrescence on its left which we
labeled with small letters in accordance with the curves in angle with d i f f e r e n t kinetic e n e r g i e s have differ-
Fig. 4. Energies of electronic states along FK, satisfying the
e n t kll a n d m u s t b e t a kcondition
e n intoof aconservation
c c o u n t w h eofn three-dimensional wave vector assign to the violation of the k j_-conservation
Fig. 4.
during photoexcitation (solid lines). Dashed lines represent
during photoexcitation at the surface. The sur-
the dependence of energy on k N for a set of emission polar face and subsurface LDOS, shown in Fig. 6,
angles. support this assumption.

Ultraviolet Photoemission Spectroscopy (UPS)

162 V.M. Tapilin et aL / Surface Science 310 (1994) 155-162

three-dimensional wave vector during the elec-

tron excitation. At the same time, the use of
COad grazing incident angles revealed a surface state in
2 +3L CO
y ot300K the spectra in the neighborhood of the K point of
the SBZ which is predicted by the calculations.
The L M T O - T B approximation in the elec-
tronic structure calculations gives us a reasonable
" j ~' clean guide for analysis of angle resolved photoemis-
"~ , ~,aFr ~'~" ./ Pt(111)
sion spectra.
2 , I I f
c -) i ' i

~ o I o 0- Acknowledgement

///////////////////// We are grateful to the Russian Foundation of

J i I I I
Fundamental Research for the financial support
3 6 9 12 of this work under Grant 93-03-4761.
Binding Energy(eV)
Fig. 10. The photoemission spectra taken at grazing incidence
angle for the clean and CO covered P t ( l l l ) surfaces. S, B x References
and B 2 denote the same spectra features as in Fig. 8, s, b 1 and
The photoemission spectra taken at grazing incidence[1] angle
b'1 are the calculated positions of the surface state and the
E.W. Plummer and W. Eberhardt, Adv. Chem. Phys. 49
bulk peaks, respectively. (1982) 533.
for the clean and CO covered Pt(lll) surfaces. S, Bx and B2
[2] R.A. Bartynski and T. Gustafsson, Phys. Rev. B 33 (1986)
denote theto same spectra features
the combination of the 1~- as
andin5trFig. 8, s, b 1and
molecular b'1 are
6588.
[3] M. Lindroos, P. Hofmann and D. Menzel, Phys. Rev. B
the calculated
orbitalspositions of the surface
of CO, in accordance with Ref. state
[16]. and the bulk 33 (1986) 6798.
We assign the photoemission peak S observed [4] S.D. Kevan, Surf. Sci. 178 (1986) 229.
peaks, respectively.
experimentally in the interval 0 a - - 5 5 - 6 5 ° to the [5] H. Kuhlenbeck, H.B. Saalfeld, U. Buskotte, M. Neu-
mann, H.J. Freund and E.W. Plummer, Phys. Rev. B 39
surface state in the band gap C (Figs. 7 and 9) (1989) 3475.
found with the calculation. The predicted posi- [6] M.Yu. Smirnov, V.V. Gorodetskii and A.R. Cholach, Izv.
tion of the peak is shown in Fig. 8 with a bar. The AN SSSR, Ser. Phys. 52 (1988) 1558.
predicted positions of the bulk transitions b I and [7] O.K. Andersen and O. Jepsen, Phys. Rev. Lett. 53 (1984)
2571.
b'1 which, as it has been discussed above, refer to
[8] W.R.L. Lambrecht and O.K. Andersen, Surf. Sci. 178
the B~ experimental peak, are also shown in Fig. (1986) 256.
8. [9] V.M. Tapilin, Surf. Sci. 206 (1988) 405.
V.M. Tapilin et al. / Surface Science 310 (1994) No other spectral features sensitive to the inci-
155-162 [10] B. Wenzien, J. Kudrnovsky, V. Drchal and M. Sob, J.
Phys.: Condensed Matter 1 (1989) 9893.
2nd semester, dence angle of the radiation were observed. The
[11] O.K. Andersen, in: The Electronic Structure of Complex
2022 surface state in gap C at K point mainly consists 112
ECE 658, Purdue University, Dmitry Zemlyanov
of equal parts of dxy, d y z a n d dxz states.
Systems, NATO ASI Series B 113 (1984) 11.
[12] H.L. Skriver, The LMTO Method. Muffin-Tin Orbitals
and Electronic Structure (Springer, New York, 1984).
112 5. Conclusion
[13] H. Krakauer and B.R. Cooper, Phys. Rev. B 16 (1987)
605.
[14] S.M. Cunningham, Phys. Rev. B 10 (1974) 4988.
[15] M. Scheffier, K. Kambe, F. Forstmann, Solid State Com-
The main features of the P t ( l l l ) photoemis- mun. 23 (1977) 789.
sion spectra can be understood in terms of the [16] P. Hofmann, S.R. Bare, N.V. Richardson and D.A. King,
bulk band structure and the conservation of the Solid State Commun. 42 (1982) 645.

ECE658, Purdue University, D. Zemlyanov 56

 2nd semester, 2022

Ultraviolet Photoemission Spectroscopy (UPS)

Work-function of a surface
Secondary ΔE = hν − ϕ surface
electron cut-off Fermi edge

2nd semester,
2022 113
ECE 658, Purdue University, Dmitry Zemlyanov

113

Auger Electron Spectroscopy

Auger Electron Spectroscopy (Auger spectroscopy or AES) was developed in the late 1960’s,
deriving its name from the effect first observed by Pierre Auger, a French Physicist, in the mid-
1920's. AES is a surface specific technique utilising the emission of electrons in the Auger
process and it used to be one of the most commonly employed surface analytical techniques
for determining the composition of the surface layers of a sample.

Auger spectroscopy includes three basic steps:

(1) Atomic ionization (by removal of a core electron);

Usually, this is done by electron beam bombardment. The energy of the beam 3-10 keV.

(2) Electron emission (the Auger process);

This is a relaxation process leaving behind double-charged ion.

(3) Analysis of the emitted Auger electrons.

This is simply a technical problem of detecting charged particles with high sensitivity,
with the additional requirement that the kinetic energies of the emitted electrons must
be determined, which is similar to XPS.

2nd semester,
2022 114
ECE 658, Purdue University, Dmitry Zemlyanov

114

ECE658, Purdue University, D. Zemlyanov 57

 2nd semester, 2022

Auger Electron Spectroscopy

n l j X-Ray Level Electron Level
1 0 1/2 K 1s
2 0 1/2 L1 2s
2 1 1/2 L2 2p1/2
2 1 3/2 L3 2p3/2
3 0 1/2 M1 3s
3 1 1/2 M2 3p1/2
3 1 3/2 M3 3p3/2
3 2 3/2 M4 3d3/2
3 2 5/2 M5 3d5/2

An Auger transition is characterized primarily by :

• the location of the initial hole;
• the location of the final two holes.
The existence of different electronic states (terms) of the final doubly-ionized atom may lead to fine
structure in high resolution spectra.
When describing the transition, the initial hole location (K, for instance) is given first, followed by the
locations of the final two holes (L1 and L2,3) in order of decreasing binding energy: KL1L2,3. transition .
If we just consider these three electronic levels there are clearly several possible Auger transitions:
KL1L1 KL1L2,3 KL2,3L2,3
2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 115

115

Auger Electron Spectroscopy

A hole at a core level can be created by X-ray photon or by

electron. AES uses an electron beam of 3-10 keV.
Fermi edge
Auger process involve two electrons and a hole 2p3/2
at a core level: BEL2
2p1/2 L-shell
1. Ionisation occurs by removal of a K-shell BEL1
2s
Energy

electron (1s electron level);

2. One electron “falls” from a higher level L1 BEK
(2s electron level) to fill an initial core hole
in the K-shell;
1s K-shell
3. The energy liberated in the second step is
transferred to a L2 (or L3) electron (2p1/2 or
2p3/2 electron level) and this electron called
Auger electron is emitted with kinetic
energy Auger electron
KEAuger = ( BEK - BEL1 ) - BEL2
= BEK – (BEL1 + BEL2) . Fermi edge
2p3/2 BEL2
The final state is a doubly-ionized atom with 2p1/2 L-shell
BEL1
2s
Energy

core holes in the L1 and L2 (or L3) shells. This

Auger transition is KL1L2.
BEK
Note: the KE of the Auger electron is
independent of the mechanism of initial 1s K-shell
core hole formation.
2nd semester,
2022 ECE 658, Purdue University, Dmitry Zemlyanov 116

116

ECE658, Purdue University, D. Zemlyanov 58

 2nd semester, 2022

Auger Electron Spectroscopy

This is an Auger spectrum of Pd metal - generated using a 2.5 keV electron beam to
produce the initial core vacancies and hence to stimulate the Auger emission process.
The main peaks for palladium occur between 220 & 340 eV. The peaks are situated
on a high background which arises from the vast number of so-called secondary
electrons generated by a multitude of inelastic scattering processes.

2nd semester,
2022 117
ECE 658, Purdue University, Dmitry Zemlyanov

117

Auger Electron Spectroscopy

Auger spectra are also often shown in a differentiated form : the reasons for this are partly
historical, partly because it is possible to actually measure spectra directly in this form and
by doing so get a better sensitivity for detection. The plot below shows the same spectrum
in such a differentiated form.

2nd semester,
2022 118
ECE 658, Purdue University, Dmitry Zemlyanov

118

ECE658, Purdue University, D. Zemlyanov 59

 2nd semester, 2022

Auger Electron Spectroscopy

Scanning Auger Microscopy ( SAM )

The incident primary electrons resulted in Auger electrons. The electron beam is scanned across
the surface and detection of the Auger electrons is synchronised with scanning.

to electron
energy analyzer
e

Scanning electron
beam
Auger electrons
e e

As with SEM , the attainable resolution is again ultimately limited by the incident beam characteristics.
More significantly, however, the resolution is also limited by the need to acquire sufficient Auger signal
to form a respectable image within a reasonable time period, and for this reason the instrumental
resolution achievable is rarely better than about 15-20 nm.

2nd semester,
2022 119
ECE 658, Purdue University, Dmitry Zemlyanov

119

Auger Electron Spectroscopy a) SEM image of the interface

Scanning Auger Microscopy ( SAM )Imaging – Spectrum Ti (left)-fiber (right)
b) Boron Auger image –
reaction of B from TiB2
coating with the Ti matrix
c) Color composite of Ti Auger
image (blue), B (red), and C
(green). The overlap of Ti
and B gives a magenta
region, showing the
distribution of the a Ti
monoboride reaction phase.
Ti and C overlap gives a sky
blue region, showing the
position of a TiC reaction
phase.
d) Auger spectrum confirms the
SEM image of the interface Ti (left)-fiber (right) presence of TiC
b) Boron Auger image – reaction of B from TiB2 coating with the Ti matrix
c) Color composite of Ti Auger image (blue), B (red), and C (green). The overlap of Ti and B gives a
magenta region, showing the distribution of the a Ti monoboride reaction phase. Ti and C overlap gives a
sky blue region, showing the position of a TiC reaction phase.
d) Auger spectrum confirms the presence of TiC
2nd semester,
2022 120
ECE 658, Purdue University, Dmitry Zemlyanov

120

ECE658, Purdue University, D. Zemlyanov 60

 element present on the surface with very high Ar+ beam and spectroscopic data gathering Characteristics
time. The sputter time can be converted to a Applications with an electron beam (fig. 2). The
depth scale when sputter rates ofdepth
theprofiles
analyzedthrough buried layers or concentration depth profile measured for Information
Sample type depth
materials are known.
interfaces can be obtained by combining SAM
wafers,
with ion beam sputter etching. In such a wires,showing
leadframes, cathodes
the stack of layers. Note that the multilayers etc.)
2nd semester, 2022
depth profile, the intensities of elements Al at the interface with the organic layer with ion beam etching up to several µm
present at the exposed surface of the sample shows some oxidation. Sample size
A typical application: compositional adherence
are monitored as a function of the sputter of layered structures
depth profile of a pixel from an organic
time. The sputter time can be converted to a Applications Lateral resolution
depth scale when sputter rates of the analyzed Information depth
LED display materials are known. (multi)layer structures (e.g.
wafers, wires, oxidation
leadframes, cathodesand/
or contamination of interfaces) < ion
with 50beam
nmetching
under SEM
up to conditions
several µm
A typical application: compositional adherence of layered structures
devices that basically consist of depth
a lightprofile of a pixel from an organic Lateral resolution
emitting organic layer sandwiched LEDbetween
display sition of powders and/or
(multi)layer individual
structures grains
(e.g. oxidation and/
Auger Electron Spectroscopy or contamination of interfaces) < 50 nm under SEM conditions
a cathode and an anode (fig. 2). devices
Indium-tin-
that basically consist of a light
Augerused
oxide (ITO) is typically Depth Profiling
as aemitting
transparent boundaries
organic layer sandwiched between sition of powders and/or individual grains
a cathode and an anode (fig. 2). Indium-tin-
To (e.g.
anode, while a metal obtain
Al) isinformation
used(ITO)
oxide as the about the variation
is typically used as a transparent
of composition with depth below the surface of a
boundaries
sample,
cathode. If a positive it is
voltage necessary
is applied
anode, while to(e.g.
toa metal gradually
Al) is3:used
Fig. remove
SEMas image
the of thematerial
pixel array of(by an ionLEDflux) Fig.
an organic from the surfacedepth
4: Compositional region profilebeing
measured
cathode. If a positive voltage is applied to Fig. 3: SEM image of the pixel array of an organic LED Fig. 4: Compositional depth profile measured
analysed,
the ITO (anode) and whilst
a negative voltage
the
continuing
ITO (anode)
to monitor
is and a negative
display. Indicated
voltage is
and therecord
indisplay. center the
ofthe
Indicated in one Auger
pixel
center thespectra.
of oneispixel area
is the area ona single
on a single
pixel ofpixel of an
an organic LEDorganic
display. LED display.
applied to the metal cathode, a applied
current to thecan
metal cathode, a currentfor
of analysis canthe compositional
of analysis for the compositional depth profile.
depth profile.
flow through the organic layer and light will 100

flow through the organic layer and light will

be generated. The ITO layer is transparent Philips Innovation ServicesC
100AI
3. Auger electron emission AI C
80

acuum be generated. The ITO layer is andtransparent

usually mounted on a glass substrate.

atomic concentration (%)

This offers the possibility to use organic Materials Analysis80 O 60

primary
and usually mounted on a glass substrate.

atomic concentration (%)

tom electron offers a full range of60analytical methods and
In 40
O
This offers the possibility
beam to useFigure 3 shows a SEM image of part of a
organic AI (oxide) 20
display that consists of an array of pixels.
nt L11 Ar ions sputter gun expertise to support 40
both research N
and 0 In
0 5 10 15 20 25 30

ry ©2011 Koninklijke Philips

manufacturing, serving customers AI (oxide)taking
by
energy levels

sputter time (min)

L1 Figure 3 shows a SEM image of part of a metal (cathode)

Electronics N.V.
on All rights reserved.
light emittingSEM image
organic of the pixel array of an
layer
20
2. relaxation display that consists of an array of pixels. organic LED display. an integral,
Indicated solution-oriented
in the approach. N
ITO layer (anode)
center of one pixel is the area of analysis 0
glass substrate
for the compositional depth pro le. 0 5 10 15 20 25 30
©2011 Koninklijke Philips Philips Innovation Services sputter time (min)

K
Electronics N.V.
All rights reserved.
World-class expertise –
Compositional depth pro le measured on a
1. ionisation
working for you
single pixel of an organic LED display.

For more information:

Schematic diagram of the Auger process. In this Fig. 2: Schematic diagram of the set up for
ple, the incident primary electron causes the As with SEMdepth
compositional , theprofiling
attainable resolution
of the is again ultimately limited by the incident beam characteristics.
layered structure Philips Innovation Services
E-mail: [email protected]
More significantly, however, the resolution is also limited by the need to acquire sufficient Auger signal
on of a K-shell electron and finally the emission of a pixel from an organic LED display.
to form a respectable image within a reasonable timewww.innovationservices.philips.com
period, and for this reason the instrumental
Auger electron. resolution achievable is rarely better than about 15-20 nm.
Technical Note 5
e focused electron beam can be scanned 2nd semester,
Each of these pixels is an individual organic November 2011 121
2022 ECE 658, Purdue University, Dmitry Zemlyanov
s the surface, scanning electron
scopy (SEM) images can be provided, 121 elemental depth composition of such an
in combination with SAM analysis can individual pixel, a SAM depth profile can be
lateral distribution of a particular recorded by alternating sputtering with an
ent present on the surface with very high Ar+ beam and spectroscopic data gathering Characteristics
with an electron beam (fig. 2). The
profiles through buried layers or concentration
Auger Electrondepth profile measured for
Spectroscopy Sample type
aces can be obtained by combining SAM
on beam sputter etching. In such a showing the stack of layers. Note that the multilayers etc.)
profile, the intensities of elements Al Summary
at the interface with the organic layer
nt at the exposed surface of the sample shows some oxidation. Sample size
onitored as a function of the sputter Auger Electron Spectroscopy (AES) is a surface-sensitive spectroscopic technique
The sputter time can be converted to a used for elemental analysis of surfaces ; it offers
Applications
scale when sputter rates of the analyzed
• high sensitivity (typically ca. 1%Information
monolayer) for all elements except H
depth
and He.
rials are known. wafers, wires, leadframes, cathodes
• a means of monitoring surface cleanliness of samples
with ion beam etching up to several µm
• quantitative compositional analysis of the surface region of specimens,
pical application: compositional adherence ofbylayered structures
comparison with standard samples of known composition.
h profile of a pixel from an organic Lateral resolution
display (multi)layer
In addition,structures
the basic(e.g. oxidation
technique and/
has also been adapted for use in:
• Auger
or contamination of Depth Profiling : providing<quantitative
interfaces) 50 nm under compositional
SEM conditions
es that basically consist of a light information as a function of depth below the surface
ing organic layer sandwiched between • Scanning
sition of powders and/orAuger Microscopy
individual grains (SAM) : providing spatially-resolved
compositional information on heterogeneous samples
hode and an anode (fig. 2). Indium-tin-
(ITO) is typically used as a transparent boundaries
e, while a metal (e.g. Al) is used as the
de. If a positive voltage is applied to Fig. 3: SEM image of the pixel array of an organic LED Fig. 4: Compositional depth profile measured
TO (anode) and a negative voltage is display. Indicated in the center of one pixel is the area on a single pixel of an organic LED display.
2nd semester,
ed to the metal cathode, a current can 2022 of analysis for the compositional depth ECE
profile.
658, Purdue University, Dmitry Zemlyanov
122

through the organic layer and light will 100

nerated. The ITO layer is transparent

122 AI C
80
sually mounted on a glass substrate.
atomic concentration (%)

O
offers the possibility to use organic 60

40 In
e 3 shows a SEM image of part of a AI (oxide)
ECE658, Purdue University, D. Zemlyanov 20 61
y that consists of an array of pixels. N
0
0 5 10 15 20 25 30