J Mater Sci: Mater Electron

DOI 10.1007/s10854-016-6189-3

Structural and electrical properties of the Al/p-Cu2ZnSnS4 thin

ilm schottky diode
R. Touati1 · I. Trabelsi1 · M. Ben Rabeh1 · M. Kanzari1,2 

Received: 24 September 2016 / Accepted: 5 December 2016

© Springer Science+Business Media New York 2016

Abstract In order to calculate the Schottky barrier 1 Introduction

parameters and to explain the resulting efects, the conduc-
tion mechanisms in a Schottky barrier should be known. In the last few years, polycrystalline thin ilm solar cells
In the present study, we investigated the structural and based on Cu(In,Ga)(S,Se)2 (CIGS) and CdTe have shown
electrical properties of Al/p-Cu2ZnSnS4 (CZTS)/Mo thin signiicant improvement and are quickly maturing toward
ilm Schottky junction. Structural characterization was mass production [1–3]. Under laboratory conditions, CIGS
carried out using X-Ray difraction and Raman Scatter- based solar cells hold the highest record eiciencies of
ing whereas electrical characterization was performed about 20.3% whereas on industrial scale, it has been possi-
by using the current–voltage (I–V) characteristics and by ble to reach ~12–15% for inished solar cell modules [4–6].
recording the AC impedance spectroscopy over a wide Despite the great technical promise of these two technol-
range of temperature up to 558  K in the frequency range ogies, they are hampered by toxicity of cadmium (Cd) and
5 Hz–13 MHz. The complex impedance plots display one scarcity issues related to the constituent elements: tellurium
semicircle with equivalent circuit functions as typical par- (Te), indium (In) and gallium (Ga) [7–9]. These drawbacks
allel RC connected to a serial resistance. The characteris- have drawn attention toward replacing CIGS with a high-
tic parameters such as barrier height, ideality factor and eiciency and lower-cost absorber [3].
series resistance have been calculated from the I–V meas- Recent research trends are advancing towards ind-
urements. At room temperature, this heterostructure has ing alternatives based on earth-abundant and non-toxic
shown non-ideal Schottky behavior with an ideality factor elements. Nowadays, thin ilm photovoltaics community
of 1.56 and 0.829 µA as a saturation current. By the imped- is exploring an alternative material Cu2(Zn,Sn)(S,Se)4
ance spectroscopy technique, we have found that all of the (CZTS) which contains earth-abundant materials such
serial resistance Rs and the parallel resistance Rp decrease as Zn and Sn. CZTS crystallizes in the tetragonal këster-
by increasing temperature whereas the capacitance C0 ite structure (space group I4, a = 5,467  Å, c = 10,923  Å).
increased from 0.76 to 1.07 µF. From the Arrhenius dia- The crystal structure can be derived from copper indium
gram, we estimated activation energy at 0.289  eV which disulide (CuInS2) chalcopyrite structure by substituting
represents the energy diference between the trap level and half of the indium In(III) sites by zinc atoms Zn(II) and
the valence band. the other half by tin Sn(IV) [10]. The resulting bandgap
varies between 1.45 and 1.6  eV which is an ideal match
* R. Touati for the solar spectrum [10, 11]. Copper-Zinc-Tin Sulide
[email protected] (CZTS) also has a high absorption coeicient (104 cm−1)
1
and absorbs nearly all the solar radiation above the band
Laboratoire de Photovoltaïque et Matériaux
Semi-conducteurs, Université de Tunis El Manar, ENIT BP
gap within a few microns [10, 11]. In addition, CZTS ilm
37, le Belvédère, Tunis 1002, Tunisia contains neither rare metals nor toxic materials, and com-
2
Laboratoire de Photovoltaïques et Matériaux
bined with the cadmium-free bufer layer, we can expect
Semi-conducteurs, Université de Tunis El Manar, ENIT- solar cells with complete non-toxicity [12].
IPEITunis Montleury-Université de Tunis, Tunis, Tunisia

13
Vol.:(0123456789)
 J Mater Sci: Mater Electron

Metal–Semiconductor (MS) contacts represent an essen- performed to convert the material into ine powder used
tial component of almost all semiconductor electronic and as raw material for the preparation of thin ilms.
optoelectronic devices. They have been used in various
industrial applications namely as microwave diodes, ield-
efect transistors (FETs) [13], integrated circuits, organic 2.2 Fabrication of Al/p-CZTS/Mo schottky junction
ield-efect transistors (OFETs) [14], light detectors [15]
and as solar cells. CZTS thin ilms, with a 400 nm thick, were processed by
Among the most interesting properties of a MS inter- thermal evaporation of the powder onto glass substrates
face, we may include its Schottky barrier height (SBH).The heated at 60 °C. Glass substrates, serving as back contact,
electronic transport through the MS interface is controlled were previously coated with a 700 nm thick Molybdenum
by the SBH and thus it is of major importance to the suc- sputter layer. In order to form the Al/p-CZTS/Mo-coated
cessful operation of any semiconductor device. There have glass Schottky junction, aluminum (Al) contact was ther-
been numerous textbooks and articles which were pub- mally evaporated onto CZTS thin ilms. Use of Al as
lished since the second half of the twentieth century trying material for Schottky contacts on p-type CZTS yields a
to unravel the SBH mystery [16, 17]. reasonable compromise between a small work function,
Deined application of Schottky diode structures pro- on the one hand, and a low chemical reactivity, on the
vides an elementary method for inspecting semiconduc- other hand [22]. The schematic diagram of the fabricated
tor material parameters. Especially for thin ilm solar cells device (Al/p-CZTS/Mo) is shown in Fig. 1.
based on the kesterite absorber CZTS Schottky diodes con-
sisting of the heterostructure Mo/CZTS/Al are appropriate
to study the intrinsic properties of the absorber layer and, 2.3 Characterization
consequently, support the relatively complex defect spec-
troscopy measurements carried out on the complete cell The structure of CZTS thin ilms has been investigated by
devices [18–21]. X-ray difraction (XRD) and Raman spectroscopy. X-ray
In the present study, we report the fabrication of the difraction analysis was operated in the 2θ range from 10°
Schottky junction Al/ p-CZTS/Mo by thermal evapora- to 90° on a X’Pert PRO PANalytical difractometer with
tion method. The structural characterization of (CZTS)/ CuKα radiation (λ = 1.5406 Å) using a step size of 0.02°
Mo thin ilm was performed using X-ray difraction and and step time of 1 s. Raman scattering experiments were
Raman spectroscopy. The temperature dependence of the performed with a micro-Raman spectrometer (HORIBA
current–voltage characteristics (I–V) of the Schottky diode Jobin Yvon LabRAM HR) at room temperature using an
Al/p-CZTS/Mo diode as well as the impedance spectros- excitation diode laser with a wavelength of 632.81  nm
copy analyzes were estimated to know the response of (Ne/He) and a CCD camera. The current–voltage (I–V)
interface and the thermionic emission mechanism in this characteristic of (Al/p-CZTS/Mo) Schottky junction was
Schottky junction and to determine possible usage in solar performed at room temperature by means of Keithley
cells. 6517 electrometer whereas the electrical measurements
of real and imaginary parts of impedance parameters
(Z′, Z″) were made over a wide range of temperature
2 Experimental procedures (418–558 K) and frequency (5 Hz–13 MHz) by means of
Hewlett–Packard HP 4192 impedance analyzer.
2.1 Synthesis of Cu2ZnSnS4 bulk crystals

The CZTS crystals were grown by the horizontal Bridg-

man method. They were obtained by reacting stoichio-
metric amounts of the elements of 99.999% purity namely
copper (Cu), zinc (Zn), tin (Sn) and sulfur (S) in a quartz
tube sealed under high vacuum of 1.33 × 10−3 Pa. The
tube, arranged in a horizontal position, is inserted into
a programmable furnace (type Nabertherm-Germany)
which maximum temperature reached for this synthe-
sis is 1000 °C maintained for 48  h. The furnace, slowly
cooled to room temperature, allows a good crystallization
of the material. The ingot so obtained is of bluish black
color measuring 20  mm of length. A grinding is then Fig. 1 The structure of the (Al/p-CZTS/Mo) Schottky junction

13
J Mater Sci: Mater Electron

3 Results and discussion

3.1 Structural properties

The structural analysis and phase identiication were per-

formed by mean of XRD and Raman scattering measure-
ments. Given the signiicant overlapping of the main peaks
in the difractograms’ characteristics of CZTS and related
sulides such as ZnS and Cu2SnS3, detection of CZTS
among these phases only by the standard technique X-ray
difraction remains a diicult task. Therefore, Raman spec-
troscopy analysis was performed as a control method for
assessment of crystalline quality and identiication of sec-
ondary phases within CZTS thin ilms [23–25].
Figure 2 illustrates the XRD pattern of CZTS thin ilm
deposited at Mo-coated glass substrate heated at 60 °C. In
addition to the three peaks related to the Mo layer (JCPDS Fig. 3 Raman spectrum of CZTS/Mo thin ilm deposited at 60 °C
card no.42-1120), several relections were observed at
28.34°, 37.56° and 46.97° corresponding respectively to
the (112), (211) and (220) planes of CZTS, which are char- at 138 and 366  cm−1 have been identiied as E symmetry
acteristic of the Kesterite crystal structure (JCPDS card no. mode from the CZTS Kesterite phase [24]. The intense
26–0575). These results agree with previous reports [26, peak around 330  cm−1 could be assigned to the A1 mode
27]. Also, it can be seen that this sample exhibits (112) pre- which is the strongest mode generally observed in the
ferred difraction plane. Raman spectra of CZTS compound. The A1-symmetry
It should be noted that these three XRD peaks overlap vibrations represent pure anion modes which correspond to
with those of Cu2SnS3 and ZnS, so the crystallization of vibrations of sulphur atoms surrounded by motionless clos-
CZTS cannot be conirmed solely by XRD analysis. There- est atoms [33, 34].
fore, Raman spectroscopy has been carried out to obtain According to literature data, the A1 Raman mode peak
further insight into the phase identiication. Raman spec- of CZTS with kesterite structure is observed at 338 cm−1.
troscopy was made over the range 100 to 900 cm−1 and the Nevertheless, in this experiment, the major peak is shifted
Raman spectrum is shown in Fig. 3. toward low wave number direction due to the existence of
The observed peaks at 138  cm−1, 330  cm−1 [28], internal compressive stress and to the disorder in cationic
366 cm−1 [29] and 655 cm−1 are consistent with the Raman sublattice [35–37]. The prevalence of 330  cm−1 peak cor-
peaks of CZTS with Kesterite structure [30–32]. The peaks respond to Cu-poorer conditions meaning that the forma-
tion of high concentration of intrinsic defects in this sample
especially ZnCu antisite and copper vacancy VCu is expected
[38, 39]. In such conditions, all Zn-atoms and half of the
Cu-atoms occupy 2d and 2c sites. As a result, the crystal
structure changes from Kesterite-type (I4) to a disordered
Kesterite phase (I42  m). Therefore, the sample exhibit a
disordered Kesterite phase structure.
The spectrum also shows a peak around 655 cm−1 which
is a second order Raman scattering attributed to CZTS
phase [24]. In addition, the corresponding Raman frequen-
cies of ZnS, SnS and Cu2SnS3 reported in the literature
are respectively 353  cm−1 [40], 192, 218  cm−1 [10] and
295–303  cm−1 [10]. In this sample, the presence of these
secondary phases namely SnS, Cu2SnS3 and ZnS were
detected by peaks observed at 224, 299 and 349  cm−1,
respectively. This is suggesting that secondary phases
mainly Cu2SnS3 and ZnS has not completely reacted during
Fig. 2 X-ray difraction pattern of CZTS/Mo thin ilm deposited at the growth of CZTS phase which explains the low intensity
60 °C of the main Raman peak related to CZTS (330 cm−1).

13
 J Mater Sci: Mater Electron

The crystallite size DW–H was calculated according to the

Williamson-Hall analysis [41]:

𝛽 2 obv = 𝛽 2 crystallite + 𝛽 2 strain + 𝛽 2 istr (1)

( )2
0.9𝜆
× cos2 𝜃 = + (4𝜀)2 sin2 𝜃 (2)
( 2 2
)
𝛽obs − 𝛽istr
DWH

where λ is the wavelength of Cu-Kα radiation (1.5406 Å), θ

is the Bragg angle, βobs is the full width at half-maximum
(FWHM) of the corresponding peak located at θ, βistr is the
instrumental broadening factor (2𝜃 = 0.02◦ ) and βstrain is
the strain correction factor. A plot is drawn with (4ε)² sin²θ
along the x-axis and (β²obs − β²istr)cos²θ along the y-axis for
CZTS thin ilm deposited at 60 °C. From the linear it to the
(data, )the crystallite size was estimated from the y-intercept
2 Fig. 4 The experimental forward and reverse bias current–voltage
0.9𝜆
DWH
, and the strain ε, from the slope of the it (4ε)². characteristics of the Al/p-CZTS/ Mo Schottky junction at room tem-
perature
The crystallite size DW–H as well as the strain ε were found
to be about 20 nm and 15.10−3 respectively. The small grain
size suggests that this ilm has not fully crystallized which is where A is the efective contact area, A* is the Richardson
conirmed by the coexistence of binary (ZnS, SnS), ternary constant (A*= 63.6 A/cm² K² for CZTS [50]) and 𝜙b0 is the
(CTS) and quaternary phases (CZTS) in the ilm. zero–bias barrier height.
By referring to Fig. 4, the rectifying behavior airms that
3.2 Electrical properties this junction acts as a Schottky diode.
Up to the threshold potential, Eq. (3) can be written in log-
3.2.1 Current–voltage (I–V) characteristics arithmic form as:
q
In order to determine Schottky barrier parameters, several LnI = LnI0 +
𝜂KT
(V − Rs I) (5)
methods can be used namely current voltage (I–V) analysis,
capacitance–voltage (C–V) analysis, photoelectron spectros- As illustrated in Fig.  5 and according to the logarithmic
copy and activation energy method [42]. I–V analysis method form of the Eqs. (3, 5), the saturation current I0 is determined
is the simplest of all methods since it involves direct meas- by extrapolating the linear region of the forward-bias semi-
urement of current voltage and provides irst-hand informa- log I–V curves to the zero applied voltage (V = 0). The cal-
tion about the nature of the developed barriers across the culated value of I0 from the intercept of Ln I = f (V) plot at
interface. V = 0 is found to be about 0.829 µA.
Figure  4 illustrates the experimental forward and reverse For the ideality factor η, it’s a measure of conformity of
bias current–voltage (I–V) characteristics of the Al/p-CZTS/ a real diode to pure thermionic emission (TE) theory and if
Mo based Schottky diode at 298  K (room temperature). η is equal to one, pure thermionic emission occurs. How-
These I–V characteristics were analyzed using standard ther- ever, η has usually a value greater than unit. Furthermore, the
mionic emission theory. According to this theory, the current deviation of experimental I–V data from the ideal thermionic
in such a device can be expressed as [43–47]: model is explained by this parameter, which also enables a
[ ( ) ] more understanding of the contribution of other current trans-
q ( ) port mechanisms [51, 52].
I = I0 exp V − Rs I − 1 (3)
𝜂KT It can be determined from the slope of the linear region of
where q is the carrier unit charge, η the diode ideality fac- the forward bias I–V characteristics using the relation [47]:
tor, K is the Boltzmann’s constant, T is the absolute tem- q dV
perature in Kelvin, V is the applied voltage, Rs is the series 𝜂= (6)
KT d(ln(I))
resistance and I0 is the saturation current given by [48, 49]:
( ) For an ideal diode, η should be nearly equal to unit.
2
q𝜙b0 However, when the efects of series resistance and leakage
I0 = AA ∗ T exp − (4)
KT current are involved, it may increase.

13
J Mater Sci: Mater Electron

Fig. 5 The logarithm of the current density Ln (I) versus the forward Fig. 6 The experimental dV/dLn (I) versus I plot obtained from for-
bias voltage (V) of the Al/p-type CZTS/ Mo Schottky junction at ward and reverse bias current–voltage characteristics of the (Al/p-
room temperature type CZTS/ Mo) Schottky structure

3.2.2 Impedance spectroscopy

The high values of η are derived from multiple efects
namely inhomogeneities of ilm thickness, series resistance, AC impedance spectroscopy supplies frequency resolved
interface states, tunneling process and luck of uniformity in information which can separate the contributions of dif-
distribution of the interfacial charges. ferent component regions (e.g. bulk, material/contact and
A high ideality factor is often attributed to defect states interface between the n and p type regions) from the total
in the semiconductor band gap [53, 54]. electrical properties of heterojunction devices through
In the present study, η was found to be about 1.56. diferences in the time constants of each element [52, 57,
Concerning the Schottky barrier height (SBH) 𝜙b0, it can 58].
be evaluated using the following expression: The AC equivalent circuit of a pn heterojunction solar
( ) cell is composed by three elements: the series resistance Rs
KT AA ∗ T 2
𝜙b0 = ln (7) (due to bulk and contact resistances); the parallel resistance
q I0
Rp (due to recombination in the depletion region) and the
The barrier height 𝜙b0 was found to be 0.807 eV. total capacitance C0 (sum of difusion and depletion capaci-
The Schottky diode parameters such as the ideality tances) [57, 59, 60]. The complex impedance of this elec-
factor η and the series resistance Rs were also estimated trical equivalent AC circuit is expressed as:
through the model developed by Cheung and Cheung [55,
56]. 1
Z = Rs + = Z � + jZ ��
Usually, at low forward bias voltages, the current–volt-
1
+ j𝜔Cp (9)
Rp
age characteristics (I–V) are linear on a semi logarithmic
scale. However, it can deviate signiicantly from linearity where Z ̠ is the real part of the complex impedance and Z ̠̠
owing to many efects especially efects of series resistance its imaginary part. Equation (9) can be reorganized to sepa-
R s. rate real and imaginary parts as shown above [57]:
The forward bias current–voltage characteristics due to
thermionic emission of a Schottky contact with the series Rp
Z ̠ = Rs + (10)
resistance can be expressed by Cheung’s function given by: 1 + 𝜔2 C0 2 R2p
dV 𝜂KT
= Rs I + (8)
d(LnI) q 𝜔Cp R2p
Figure 6 illustrates the variation of dV
versus the cur-
Z ̠̠ = (11)
d ln(I) 1 + 𝜔2 C0 2 Rp2

rent density I for the Al/p-CZTS/ Mo Schottky structure. It Since the components of total capacitance are in paral-
has a linear behavior and we can readily obtain both Rs and lel, this equivalent circuit has a single time constant τ = Rp
η which are evaluated to be 21.61 Ω and 1.46 respectively. C0 and the plot of Z′ versus Z″ for a range of frequencies

13
 J Mater Sci: Mater Electron

Fig. 9 The variation of the series resistance and parallel resistance

for diferent temperatures of the Al/p-type CZTS/ Mo Schottky diode
Fig. 7 The Nyquist plots of the (Al/p-CZTS/Mo) Schottky diode at
diferent temperatures

Fig. 8 Equivalent circuit for the (Al/p-type CZTS/ Mo) Schottky

diode

should be a semicircle with diameter Rp, displaced from

origin by Rs [57, 58]. Fig. 10 The variation of the space-charge capacitance versus temper-
ature for (Al/p-CZTS/Mo) Schottky Diode
Figure  7 depicts the Nyquist diagrams (Zε vs. Z ̠)
obtained by plotting the imaginary part with the corre-
sponding real part for the (Al/p-CZTS/Mo) heterostructure conductivity is thermally activated as well as the relaxa-
at zero bias in the temperature range of 418–558 K. tion times distribution [61, 62].
The spectra present one semicircular arc from which Figures 9 and 10 depict the evolution of the equivalent
we can retrieve all characteristics namely relaxation time, circuit parameters as a function of temperature. The val-
resistance and capacitance. These observations present the ues of the serial and the total resistance were extracted
response to excitatory electric ields applied to the sample from low and high frequencies, respectively, by inter-
and correspond to intergranular polarization phenomenon cepting the real axis. The value of capacitance was esti-
occurring at higher frequencies. mated to be rather frequency dependent. It is clear that
Figure 8 shows the modeled equivalent circuit of a given the increase of temperature leads to a decrease in values
sample. It consists of a parallel resistor Rp and a capaci- of both serial and parallel resistances, whereas capaci-
tance Cp connected to a series resistance Rs. tance increases from 7.57107 F at T = 418 K to 1.0710−6
The analysis of experimental data (Fig.  7) show F at T = 558  K. This behavior is probably assigned to
that the semi-circles are slightly depressed and their improvement of the junction structure. Table  1 summa-
maximum shift to higher frequencies as the tempera- rizes the evolution of the equivalent circuit parameters
ture increases. The diameter and the maximum of the ((Rs), (Rp) and capacitance (Cp)) as a function of temper-
semicircles decrease with increasing temperature. This ature of the Al/p-CZTS/ Mo Schottky junction.
observation leads to the conclusion that the electrical

13
J Mater Sci: Mater Electron

To study the relaxation time as a function of tempera-

ture, we plot Zε versus frequency at diferent temperatures
as depicted in Fig.  11. By increasing temperature, it can
be seen that the pic of imaginary part shifts to the higher
frequencies. The frequency ωm (corresponding to the maxi-
mum of imaginary part of the impedance) gives the most
probable relaxation time τm from the condition τm.ωm = 1.
This variation of relaxation frequency at maximum of Zε
as function of temperature according to the Arrhenius law,
is given by [63, 64]:
( )
−Ea
𝜔m = 𝜔0 exp (12)
KB T
where ω0 is a constant, KB is the Boltzmann’s constant and
Ea is the activation( energy.
) As shown in Fig. 12, the expres-
sion Ln 𝜔m = f T 1000
of leads to a linear function, in
( )
Fig. 12 Arrhenius diagram of the cut-of frequency

good agreement with expression (12). The value of activa-

tion energy was estimated to 0.289 eV. substrate temperature heated at 60 °C, crystallizes in disor-
dered Kësterite structure. That’s why eforts are underway
to enhance stoichiometry of CZTS ilm and to inhibit the
4 Conclusion formation of secondary phases, by making diferent anneal-
ing under Sulphur atmosphere.
In this study, Al/p-CZTS/ Mo Schottky diode was fabri- Concerning the basic diode parameters such as barrier
cated by thermal evaporation method and structural char- height, ideality factor and series resistance, they were con-
acterization was investigated using X-Ray difraction and cluded using current–voltage (I–V) and impedance meas-
Raman spectroscopy. The XRD proile reveals the forma- urements in the temperature range (418 K–558 K). The rec-
tion of CZTS Kesterite phase although minor impurity tifying behavior conirms a Schottky junction with a p-type
phases such as SnS, ZnS and Cu2SnS3 were observed by absorber layer. An ideality factor of 1.46 and a serial resist-
mean of Raman scattering measurements. Raman spectros- ance of 21.61 Ω were extracted from the experimental data.
copy conirmed also the presence of internal compressive The barrier height of this heterostructure 𝜙b0 was found to
stress which made the position of Raman peak of CZTS at be 0.807  eV. The complex impedance spectroscopic tech-
336  cm−1 shifting slightly towards 332  cm−1. This obser- nique was performed so as to characterize the relaxation
vation leads us to the conclusion that CZTS, grown at process in (Al/p-CZTS/Mo) Schottky junction diode. The
impedance data show that the observed dielectric responses
can be described by an electrical equivalent circuit and
we have found that all of the serial resistance Rs and the
parallel resistance Rp decrease by increasing temperature
whereas the capacitance C0 increased from 0.76 to 1.07 µF.
An activation energy of 0.289 eV was found. These results

Table 1 Evolution of the equivalent circuit parameters as a function

of temperature of the (Al/ p-CZTS/ Mo) Schottky junction
T (K) Rs (Ω) Rp (Ω) Cp (F)

418 10.00 17.00 7.5752E−7

478 8.12 15.58 8.1834E−7
498 7.84 14.88 8.5684E−7
518 7.70 14.46 8.9059E−7
528 7.53 14.39 9.1299E−7
538 7.50 12.52 1.0286E−6
Fig. 11 Angular frequency dependence of Z″ at various temperatures
558 7.60 12.09 1.0759E−6
of the (Al/p-type CZTS/ Mo) Schottky diode

13
 J Mater Sci: Mater Electron

are very promising for preparing thin ilm solar cells with 31. P.A. Fernandes, P.M.P. Salomé, A.F. da Cunha, Thin Solid Films
p-CZTS as an absorber layer by mean of thermal evapora- 517, 2519 (2009)
32. P.A. Fernandes, P.M.P. Salomé, J. Alloys Compd. 509, 7600 (2011)
tion technique. 33. H. Neumann, Helv. Phys. Acta 58, 337 (1985)
34. M. Grossberg, J. Krustok, J. Raudoja, K. Timmo, M. Altosaar, T.
References Raadik, Thin Solid Films 519, 7403 (2011)
35. M.Y. Valakh, O.F. Kolomys, S.S. Ponomaryov, V.O. Yukhymchuk,
1. N.G. Dhere, Sol. Energy Mater Sol. Cells 91, 1376 (2007) I.S. Babichuk, V. Izquierdo-Roca, E. Saucedo, A. Perez-Rodriguez,
2. N.G. Dhere, Sol. Energy Mater Sol. Cells 95, 277 (2011) J.R. Morante, S. Schorr, and I.V. Bodnar, Phys. Status Solidi RRL
3. E.M. Mkawi, K. Ibrahim, M.K.M. Ali, K.M.A. Saron, M.A. Far- 7(4), 258 (2013)
rukh, N.K. Allam, J. Mater. Sci. 26, 222 (2015) 36. Z. Seboui, Y. Cuminal, N. Kamoun-Turki, J. Renewable Sustain-
4. P. Jackson, D. Hariskos, E. Lotter, S. Paetel, R. Wuerz, R. Men- able. Energy 5, 023113 (2013)
ner, W. Wischmann, M. Powalla, Prog Photovolt 19, 894 (2011) 37. R. Caballero, E. Garcia-Llamas, J.M. Merino, M. León, I.
5. A. Fairbrother, X. Fontané, V. Izquierdo- Roca, M. Espíndola- Babichuk, V. Dzhagan, V. Strelchuk, M. Valakh, Acta Mater. 65,
Rodríguez, S. López- Marino, M. Placidi, L. Calvo-Barrio, A. 412 (2014)
Pérez-Rodríguez, E. Saucedo, Sol. Energy Mater Sol. Cells 112, 38. L. Choubrac, A. Lafond, C. Guillot-Dendon, Y. Moelo, S. Jobic
97 (2013) Inorg. Chem. 51, 3346 (2012)
6. A.U. Sheleg, V.G. Hurtavy, A.V. Mudryi, M. Ya. Valakh, V.O. 39. D. Huang, C. Persson, Thin Solid Films 535, 265 (2013)
Yukhymchuk, I.S. Babichuk, M. Leon, and R. Caballero, Semi- 40. Y. Young-Moon, M.-H. Hyun, S. Nam, D Lee, O. Byungsung,
conductors 48, 1296 (2014) K.-S. Lee, P.Y Yu, Y.D. Choi J. Appl. Phys 91(11), 9429 (2002)
7. M.A. Green, Prog. Photovol. 14, 383 (2006) 41. V.D. Mote, Y. Purushotham, B.N. Dole, J. Theor, Appl. Phys 6, 6
8. B.A. Anderson, C. Azar, J. Holmberg, S. Karlsson, Energy 23, (2012)
407 (1998) 42. A.A.M. Farag, I.S. Yahia, M. Fadel, Int. J. Hydrog. Energy 34,
9. S. Ahmed, K.B. Reuter, O. Gunawan, L. Guo, L.T. Romankiw 4906 (2009)
and H. Deligianni, Adv. Energy Mater. 2(2), 253 (2012) 43. M. Shur, Physics of Semiconductor Devices. (Prentice Hall, Engle-
10. A.-J. Cheng, M. Manno, A. Khare, C. Leighton, S.A. Campbell, wood Clifs, 1990)
E.S. Aydil, J. Vac. Sci. Technol. A 29, 051203 (2011) 44. A. Kefous, M. Siad, S. Mamma, Y. Belkacem, C. Lakhdar
11. H. Katagiri, Thin Solid Films 480–481, 426 (2005) Chaouch, H. Menari, A. Dahmani, W. Chergui, Appl. Surf. Sci
12. P.S. Vasekar, T.P. Dhakal, in Solar Cells – Research And Appli- 218, 337 (2003)
cation Perspectives, ed. by A.M. Acevedo (InTech, Rijeka, 45. M.J. Turner, E.H. Rhoderick, Solid State Electron. 11, 291 (1968)
2013), p. 145 46. S. M. Sze, Physics of Semiconductor Devices, 2nd Edn (John
13. A.R. Brown, C.P. Jarrett, D.M. de Leeuw, M. Matters, Synth. Wiley and Sons, New York, 1981), p. 1–138
Met. 88, 37 (1997) 47. E.H. Rhoderick, R.H. Williams, Metal–Semiconductor Contacts,
14. E. Itoh, K. Miyairi, Thin Solid Films 499, 95 (2006) 2nd Edn (Clarendon Press, Oxford, 1988)
15. T.K. Lin, S.J. Chang, Y.Z. Chiou, C.K. Wang, S.P. Chang, K.T. 48. N. Tugluoglu, S. Karadeniz, M. Sahin, H. Safak, Semicond. Sci.
Lam, Y.S. Sun, B.R. Huang, Solid-State Electron. 50, 750 (2006) Technol. 19, 1092 (2004)
16. W. Leroy, A Fundamental Study of Advanced Metal/Semiconduc- 49. M.M. Patel, K.D. Patel, C.A. Patel, K.K. Patel, V.M. Pathak, R.
tor Contacts (Ghent University, Ghent, 2006), pp. 1–169 Srivastava, ‘PRAJNA’ J. Pure Appl. Sci.,18,119 (2010)
17. A. Tombak, Y.S. Ocak, M.F. Genişel, T. Kilicoglu, Mater. Sci. 50. A.E. Rakhshani, J. Alloys Compd. 675, 387–392 (2016)
Semicond. Process. 28, 98 (2014) 51. F. Jabli, H. Mosbahi, M. Gassoumi, C. Gaquiere, M.A. Zaidi, H.
18. E. Schlenker, V. Mertens, J. Parisi, R. Reineke-Koch, M. Könt- Maaref, J. Appl. Phys 6, 27 (2014)
ges, Phys. Lett. A 362, 229 (2007) 52. S. Alialy, H. Tecimer, H. Uslu, Ş. Altındal, J Nanomed. Nano-
19. M. Igalson, H.W. Schock, J. Appl. Phys 80, 5765 (1996) techol. 4, 167 (2013)
20. A. Jasenek, U. Rau, V. Nadenau, H.W. Schock, J. Appl. Phys 87, 53. J.H. Werner, H.H. Guttler, J. Appl. Phys 69, 1522 (1991)
594 (2000) 54. V. Lakshmi Devi, I. Jyothi, V. Rajagopal Reddy, Chel-Jong Choi,
21. C. Deibel, V. Dyakonov, J. Parisi, Appl. Phys. Lett. 82, 3559 Open Appl. Phys. J. 5, 1 (2012)
(2003) 55. S.K. Cheung, N.W. Cheung, Appl. Phys. Lett 49, 85 (1986)
22. C.L. Chan, I. Shih, J. Appl. Phys 68, 156 (1990) 56. G. Güler, Ö. Güllü, Ş. Karataş, Ö.F Bakkaloğlu J. Phys. 153,
23. P. A. Fernandes, P.M.P. Salomé, A.F. Sartori, J. Malaquias, A.F. 012054 (2009)
da Cunha, B.A. Schubert, J.C. González, G.M. Ribeiro, Sol. 57. H. Bayhan, A.S. Kavasoglu, Turk. J. Phys 27, 529 (2003)
Energy Mater Sol. Cells 115, 157 (2013) 58. A.K. Jonscher, Dielectric Relaxation in Solids, (Chelsea Dielectrics
24. M. Dimitrievska, A. Fairbrother, X. Fontané, T. Jawhari, V. Press, London, 1983)
Izquierdo-Roca, E. Saucedo, A. Pérez-Rodríguez, Appl. Phys. 59. R.A. Kumar, M.S. Suresh, J. Nagaraju, Sol. Energy Mater. Sol.
Lett. 104, 021901 (2014) Cells 60, 155 (2000)
25. B.L. Guo, Y.H. Chen, X.J. Liu, W.C. Liu, A.D. Li, AIP Adv. 4, 60. M.S. Suresh, Sol. Energy Mater.Sol. Cells 43, 21 (1996)
097115 (2014) 61. B. Roy, S. Chakrabarty, O. Mondal, M. Pal, A. Dutta, Mater. Char-
26. C. Chan, H. Lam, C. Surya, Sol. Energy Mater Sol. Cells 94, 207 act. 70, 1 (2012)
(2010) 62. P.A. Fernandes, A.F. Sartori, P.M.P. Salomé, J. Malaquias, A.F. da
27. J.J. Scragg, P.J. Dale, L.M. Peter, Electrochem. Commun. 10, Cunha, M.P.F. Graça, J.C. González, Appl. Phys. Lett. 100, 233504
639 (2008) (2012)
28. A. Shavel, D. Cadavid, M. Ibáñez, A. Carrete, A. Cabot, J. Am. 63. C. Ostos, L. Mestres, M.L. Martinez-Sarrion, J.E. Garcia, A.
Chem. Soc. 134, 1438 (2012) Albareda, R. Perez, Solid State Sci. 11, 1016 (2009)
29. F. Jiang, H. Shen, W. Wang, J. Electron. Mater. 41(8), 2204 64. S. Hamrouni, M.F. Boujmil, K. Ben Saad, Adv. Mater. Phys.
(2012) Chem. 4, 224 (2014)
30. J. He, L. Sun, K. Zhang, W. Wang, J. Jiang, Y. Chen, P. Yang, J.
Chu, Appl. Surf. Sci. 264, 133 (2013)

13