Accepted Manuscript Room temperature deposition of highly crystalline Cu-Zn-S thin films for solar cell applications using SILAR method Edwin Jose, M.C. Santhosh Kumar PII:
S0925-8388(17)31297-5
DOI:
10.1016/j.jallcom.2017.04.097
Reference:
JALCOM 41501
To appear in:
Journal of Alloys and Compounds
Received Date: 25 January 2017 Revised Date:
5 April 2017
Accepted Date: 11 April 2017
Please cite this article as: E. Jose, M.C. Santhosh Kumar, Room temperature deposition of highly crystalline Cu-Zn-S thin films for solar cell applications using SILAR method, Journal of Alloys and Compounds (2017), doi: 10.1016/j.jallcom.2017.04.097. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Room temperature deposition of highly crystalline Cu-Zn-S thin films for solar cell applications using SILAR method Edwin Jose#, M. C. Santhosh Kumar*
RI PT
Optoelectronic materials and Devices lab, Department of Physics, National Institute of Technology, Tiruchirappalli, 620015, Tamil Nadu, India
*Corresponding author.
SC
E-mail addresses: *
[email protected], #
[email protected]
M AN U
Abstract
Cu-Zn-S having inexpensive, non-toxic, and earth-abundant constituent elements, combined with suitable optical and electrical properties is a promising candidate as an absorber material for thin film solar cell fabrication. We report the deposition of Cu-Zn-S thin films using Successive Ionic Layer Adsorption and Reaction (SILAR) method under room temperature and atmospheric conditions. Cu-Zn-S thin films are deposited over glass
TE D
substrates with different Cu/(Cu+Zn) ratio in the precursor solution to investigate the band gap tunability. The as-deposited Cu-Zn-S films are characterized for determining their structural, compositional, morphological, optical and electrical properties. Also studied the stability of electrical properties for a period of six months after the deposition. All the films
EP
exhibited p-type conductivity and a relatively high absorption coefficient value between 104 to 105 cm-1 in the visible and near-IR spectral range. We observed that, under the Zn-rich
AC C
growth conditions, the films formed are having a double band gap structure with lower band gap values in the range 1.6 to 1.7 eV, making them a suitable absorber material for solar cell fabrication. Under the Cu-rich growth conditions, band gap values reached 2.4 to 2.6 eV making them a suitable buffer/window layer in solar cell application. Notably, films with Cu/(Cu+Zn) ratio 0.8 showed an optical transparency of 40 to 70 % in the visible spectrum along with an electrical conductivity of 2900 Scm-1, which is much higher than the other reported p-type transparent conducting materials.
Keywords SILAR deposition, CuZnS, Earth abundant, Thin film solar cells, p-type absorber, p-type transparent conducting film
ACCEPTED MANUSCRIPT
1. Introduction The utilisation of solar energy gives a feasible solution to meet the rapidly increasing energy demands of today's world. The best method to utilise solar energy for the generation of electric power is the use of solar cells or photovoltaic cells. The most common solar cell material is crystalline Silicon. Since the cost of production is very high for Si-based solar
RI PT
cells, attention has been shifted to cost-effective thin film solar cells. To fabricate affordable and efficient thin film solar cells, the constituent elements used in the absorber material
should be inexpensive, non-toxic and earth-abundant. To obtain high energy conversion efficiency, the material should have appropriate optical and electrical properties such as
SC
suitable optical band gap, high optical absorption coefficient, high quantum yield for the
excited carriers, long carrier diffusion length, and low recombination velocity [1]. The current
M AN U
best-developed thin film photovoltaic devices are made from chalcogenide absorbers such as CuInxGa(1-x)Se2 (CIGS) and CdTe with recent world record efficiencies certified at 22.6 % and 22.1 % respectively [2,3]. These absorber materials contain toxic (Cd) and rare and expensive (In and Ga) elements. CuZnS (CZS) with a suitable bandgap of about 1.6 eV and a high absorption coefficient of visible light (above 104 cm-1) is a promising absorber material for solar cell fabrication [4-8]. The less toxic and earth-abundant constituent elements
TE D
Copper, Zinc, and Sulfur makes CuZnS a potential substitute for CIGS and CdTe. So far, only a few research groups have reported successful deposition of ternary CuxZnyS thin films. Innocenti et al. [4] used Electro Chemical Atomic Layer Deposition (ECALD) methodology to synthesise thin films of ternary Cu-Zn sulfides. The work reported
EP
the formation of a sulfide species containing both Cu and Zn in its stoichiometry having a band gap value of 1.64 eV, which is suitable for solar cell applications. Benedetto et al. [5]
AC C
reported deposition of ultra-thin films with chemical composition belonging to the Cu-Zn-S ternary system using layer-by-layer E-ALD electrochemical technique. The reported band gap value of 1.61 eV strongly propose the usefulness of the material for photovoltaics and photochmical applications. Sreejith et al. reported the fabrication of a solar cell using Chemical Spray Pyrolysis deposited CuZnS thin films as the absorber layer [6]. It is reported that the material has a double band gap, and the band gap values vary with the Cu/Zn ratio. A solar cell with efficiency of 1.04 % was fabricated, using p-type CuZnS layer having Cu to Zn ratio 0.4 and band gap value of 1.8 eV as the absorber layer and In2S3 as the buffer layer. Recently, the same group reported an enhancement in the efficiency from 1 % to 1.94 % by an increase of Cu to Zn ratio in the CuZnS absorber [7]. Kitagawa et al. reported the
ACCEPTED MANUSCRIPT fabrication of a Cu-Zn-S (CZS) solar cell with an efficiency of 1.7 % based on CZS thin films deposited by the Spray pyrolysis method [8]. According to the report, there occurs a band gap variation in CZS films in the range of 1.8 to 3.5 eV with the increase in Cu/(Cu+Zn) ratio. The best efficiency was observed for a cell with the CZS absorber having Cu/(Cu+Zn) ratio of 50% and band gap value of 2 eV.
RI PT
There are also reports, regarding the deposition of wide band gap CuxZnyS films that can be used as a p-type transparent conducting material [9–15]. Transparent conducting thin films play an important role as a transparent electrode in many optoelectronic devices due to its high transparency and high electrical conductivity. At present, most of the transparent
SC
conducting materials used in the optoelectronic devices are n-type, such as tin-doped indium oxide (ITO), fluorine-doped tin oxide (FTO), and Al-doped zinc oxide (AZO). In contrast, ptype (hole conducting) transparent conducting materials (p-TCMs) are far less developed [9].
M AN U
With the technological advancements in the fields of optoelectronics, there will be an increased demand for p-type transparent thin films.
Yang et al. reported Electro Chemical Deposition (ECD) of CuxZnyS thin films exhibiting high optical transmission in the visible range, and a bandgap of about 3.2 eV [10]. Mandula et al. [11], Tong et al. [12] and Ichimura et al. [13] reported Photo Chemical
TE D
Deposition (PCD) of p-type CuxZnyS thin films with a bandgap larger than 3 eV. Other reports include deposition of Cu-alloyed ZnS by Diamond et al. [14]. In this report, Cualloyed wurtzite CuxZn1−xS films are prepared by Pulsed Laser Deposition (PLD) technique at elevated temperature (550 °C), and reported p-type nature with the best films exhibiting a
EP
conductivity of 54 Scm-1 and optical transmission of 65% at 550 nm. Later, Woods-Robinson et al. reported the synthesis of CuxZn1−xS films at room temperature by pulsed laser
AC C
deposition (PLD), and achieved a hole conductivity of 42 Scm-1 and a transparency of 50%– 71% averaged over the visible spectrum [9]. Recently, Xu et al. reported Chemical Bath Deposition (CBD) of p-type transparent conducting films of nanocrystalline (CuS)x:(ZnS)1−x structure. The films had an optical transmission above 70% in the visible range of the spectrum, and hole conductivity as high as 1000 Scm−1, which they claim as much greater than that of other state-of-the-art p-type TCMs [15]. In this report, we present the room temperature and atmospheric deposition of Cu-ZnS thin films using Successive Ionic Layer Adsorption and Reaction (SILAR) method. The purpose of this study is to investigate the band gap tunability of Cu-Zn-S with varying Cu to Zn ratio, and hence to establish the feasibility of using this material in thin film solar cells.
ACCEPTED MANUSCRIPT
2. Materials and methods 2.1 Deposition process Cu-Zn-S thin films were deposited on cleaned soda lime glass substrates using SILAR technique under room temperature and atmospheric conditions. The SILAR method is based on the selective adsorption and reaction of the constituent ions from the precursor solutions
RI PT
over the substrate surface. The overall system consists of four beakers; two of which contains cationic and anionic precursor solutions and the other two beakers are filled with deionized (DI) water. The cationic precursor contains a solution of salts of the cation, and the anionic precursor contains a solution of a salt of the anion. A single SILAR deposition cycle starts
SC
with substrate immersion in the cation solution, initiating adsorption of cations over the
substrate surface. It follows a rinsing in deionized water to remove excess and loosely bound
M AN U
cations, ensuring a uniform distribution of the adsorbed cations. The substrate is then immersed in anionic precursor which results in a chemical reaction between the adsorbed cations, and the anions from the precursor, to form the required compound. Subsequent rinsing again with deionized water ensures a uniform film formation well adhered to the substrate surface. A schematic illustration of a single SILAR cycle for the deposition of CuZn-S films is shown in Fig 1. With successive repetition of the same cyclic process, thin films
TE D
of desired thickness can be obtained.
In the present study, CuCl2 and Zinc acetate dissolved in DI water was used as the cationic precursor, while Na2S dissolved in DI water was used as the anionic precursor. Triethanolamine was used as a stabiliser and pH balancer in the cationic precursor. All
EP
reagents were analytical grade and used without further purification. Films were deposited by varying Cu/(Cu+Zn) ratio in the cationic precursor from 0.1 to 0.8 and at the same time
AC C
fixing (Cu+Zn) content always as 0.1 M. Anionic precursor contained 0.05 M Na2S in all the cases. The samples were coded accordingly as CZS 10 to CZS 80. For example, sample CZS 20 contains 0.02 M CuCl2 and 0.08 M Zn acetate (resulting Cu+Zn content as 0.1 M) in the cationic precursor and 0.05 M Na2S in the anionic precursor respectively. The dipping, rinsing, and drying time were kept constant as 15 s, 15 s, and 5 s
respectively. The deposition process was repeated for 50 deposition cycles. The as-deposited films were then ultrasonicated for 5 min in DI water to check film adhesion with the substrate surface as well as to remove surface contaminants. To study the stability and ageing effect, the films were preserved inside a non-vacuum desiccator.
ACCEPTED MANUSCRIPT 2.2 Characterisations and measurements The film thickness was determined by using a combination of stylus profiler (Bruker DektakXT) and non-contact 3D optical profiler (Taylor Hobson CCI MP). The structural characterisation was carried out by grazing incidence X-ray diffraction (Rigaku SmartLab) analysis with fixed grazing angle of incidence 0.5O, scan rate 3 deg/min and step size 0.01
RI PT
degree with Cu Kα radiation (λ=1.54056 Å). The chemical composition of the deposited thin films was determined by X-ray photoelectrons spectroscopy (Kratos – Axis Ultra DLD) using Al Kα radiation as probing source and an argon ion gun for ion milling. The Ultraviolet
photoelectrons spectroscopy (UPS) measurements were carried out using a He I (21.22 eV)
SC
source to determine the work function of the material. Raman spectra were obtained from Horiba LabRAM HR Evolution system coupled with a laser excitation source of wavelength 532 nm. The surface morphology and surface roughness of the films were examined by using
M AN U
field-emission scanning electron microscopy (Zeiss, Ultra-55) and atomic force microscopy (Park-NX10). The optical transmittance (T) and reflectance (R) were recorded at normal incidence mode using a UV-Vis-NIR spectrophotometer (JASCO V670) in the wavelength range of 300 to 3000 nm. The absorption coefficient (α) and the optical band gap (Eg) were calculated from the measured data. The carrier type, carrier concentration, mobility, and
TE D
conductivity were measured at room temperature, according to the van der Pauw configuration with an Ecopia HMS-5000 Hall Measurement System equipped with a 0.59 T magnet. The electrical measurements were repeated as a function of ageing time: as-
EP
deposited, after one month, three months and six months.
AC C
3. Results and discussion 3.1 Surface morphology
Fig 2. shows the FESEM images of the samples CZS 20, CZS 40, CZS 60 and CZS
80. The films are fully covered and well adhered to the substrate surface without any voids. It can be observed that Cu/Zn ratio plays a significant role in the microstructural morphology of the film. When the Cu concentration is higher than Zn concentration (CZS 60 and CZS 80), the films are composed of small petal-like structures of nano-sized dimension with uniform distribution. For higher Zn concentrations (CZS 20 and CZS 40), there happens an aggregation of petal-like structures forming nanoflower kind of patterns. The AFM analysis (Fig 3.) also shows a similar trend. For higher Zn concentration, there occurs coalescing of the grains forming protruding larger aggregates. It can be clearly
ACCEPTED MANUSCRIPT seen that, with the increase in Cu concentration, the surface morphology changes, aggregations are no longer present, and size of the dispersed grains are smaller and almost uniform. Also, the films with higher Cu concentration than Zn, have a columnar structure for the grains, and the grain size in the elongated direction is about the same size as that of the film thickness. The root mean square (RMS) surface roughness values of the films are
RI PT
obtained as 85 nm (CZS 20), 155 nm (CZS 40), 13 nm (CZS 60) and 7 nm (CZS 80). The increase in surface roughness values for the CZS 20 and CZS 40 films is because of agglomerations formed over the surface.
SC
3.2 Elemental composition
The elemental compositions of the samples are investigated using XPS analysis. The top surface layer of the as-deposited films was etched by Ar sputtering before the XPS
M AN U
probing. The obtained spectra are corrected on the C1s peak at 284.5 eV. Fig 4 shows the full range spectra as well as elemental specific high-resolution spectra of as-deposited samples CZS 20, CZS 40, CZS 60 and CZS 80. The full range spectra show photoelectron peaks of constituent elements Cu, Zn and S along with Oxygen and Carbon in their corresponding positions.
TE D
The high-resolution spectrum of Cu showed the typical Cu 2p3/2 and Cu 2p1/2 binding energy values at 932.08 eV and 951.98 eV with a peak splitting of 19.9 eV, indicating the formation of Cu(II) [16, 17]. The peaks for the Zn 2p shows two peaks corresponding to Zn 2p3/2 and 2p1/2 at binding energies of 1022.15 and 1045.06 eV with a peak separation of
EP
22.91 eV, suggesting the presence of Zn(II) [18, 19]. Typical S 2p peaks are observed at binding energy values of 161.56 eV and 162.73 eV consistent with the expected values of S
AC C
2p3/2 and S 2p1/2 in sulfide phases [16-19]. The small intensity peaks corresponding to C1s and O1s noticed in the survey spectra
may be due to atmospheric contaminations over the surface, and no other elements are detected. Hence, it can be concluded that the deposited films are composed of Cu, Zn, and S without any other elemental impurities. Quantitative analysis of the XPS is carried out to find the atomic concentration of constituent elements. From the tabulated results (Table 1), it can be noted that the elemental composition of the films depends weakly on the constituent composition of the precursor solutions.
ACCEPTED MANUSCRIPT 3.3 Structural analysis There are different opinions regarding the structure of Cu-Zn-S material. Reports include deposition of ternary Cu-Zn sulfides [4, 20], non-stoichiometric alloy semiconductor CuxZnyS [10, 13], CuS–ZnS binary composite [7, 21], Cu-alloyed ZnS [9, 14], and Zn-doped CuS [22, 23]. To have a better understanding of the structure, X-ray diffraction patterns are
RI PT
recorded in grazing incidence geometry as it can give stronger signals from the film avoiding unnecessary substrate signals. Since there are no standard XRD reference patterns for ternary CuZnS compound, the obtained XRD patterns (Fig 5) are compared against the standard XRD reference patterns of CuS (JCPDS:06-0464) and ZnS (JCPDS:65-0309).
SC
Interestingly, the obtained XRD patterns show more resemblance with that of standard CuS reference pattern. All the major peaks corresponding to CuS standard are present in every sample with no apparent peak shift. It is also observed that no characteristic peaks of
M AN U
elements such as Cu, Zn, or S are observed. Thus we assume that CuS act as a prominent phase and Zn is incorporated into the host sites of CuS lattice. Since the ionic radius of Zn2+ is similar to that of Cu2+ (0.74 Å and 0.73 Å respectively), there is a high probability for the incorporation of Zn atoms in place of Cu atoms in the CuS lattice [23]. Also, for samples with higher Zn concentration (CZS 10 to CZS 40), there appears an
TE D
extra peak at 2θ value 8.10O, indicating Zn incorporation is affecting the phase or structural properties of the CuS host lattice. For these samples, elemental concentration analysis revealed that Zn concentration is high and comparable with Cu concentration (Table 1). This indicates the possibility towards the formation of a Zn alloyed CuS structure or a non-
EP
stoichiometric CuxZnyS alloy structure (retaining CuS as the host lattice) in addition to the CuS structure. However, for samples with higher Cu concentration (CZS 50 to CZS 80), no
AC C
additional peak or significant difference is observed from that of CuS standard. For these samples, the elemental concentration of Zn is smaller compared with that of Cu, indicating the possibility towards the formation of a Zn-doped CuS structure. Raman spectra (Fig 6) can provide more information regarding the bonding states as
well as crystalline nature of the system concerned. For the pure CuS samples, the Cu-S vibrational mode shows a strong signal around 472-474 cm-1 corresponding to longitudinal optic (LO) mode [23,24]. In the case of as-deposited Cu-Zn-S samples, this peak is shifted to 469 cm-1 consistent with the assumption that, Zn2+ occupy the Cu2+ sites forming a Cu-S-Zn bonding in the CuS host lattice. This frequency shift occurs because of the difference in bond strength of Cu-S-Zn from that of pure Cu-S [23].
ACCEPTED MANUSCRIPT The calculation of full width at half maximum (FWHM) of Cu-S vibrational mode gives values 10.83 cm-1, 10.65 cm-1, 9.12 cm-1 and 8.86 cm-1 for samples CZS 20, CZS 40, CZS 60 and CZS 80 respectively. The increase in FWHM values of samples CZS 20 and CZS 40 in comparison with other two samples suggest a deformation in CuS host lattice due to Zn incorporation. The extra peak observed at 2θ value of 8.10O in the XRD spectra of these
RI PT
samples also support this argument. In addition to this, the Raman spectra also revealed other peaks around 262 cm-1, 134 cm-1, 110 cm-1 and 61 cm-1 in the low-frequency region, with similar frequency shifts
compared with the CuS system [25]. The peak at 262 cm-1 can be assigned to transverse optic
SC
(TO) mode, which is consistent with the reported Raman spectra of CuS [24].
3.4 Optical properties
M AN U
The absorption coefficient (α) and the optical bandgap (Eg) of the films are calculated from the UV-Vis-NIR transmittance measurements. The absorption coefficient (α) is determined based on the formula,
(α) = ln (1/T) / t;
(1)
where (T) is the optical transmittance and (t) is the thickness of the film. All the films exhibit
TE D
relatively high absorption coefficient value between 104 to 105 cm-1 in the visible and near IR spectral range. In general, the absorption coefficient and photon energy are related by the Tauc relation [26],
α hυ = B ( hυ – Eg ) n
(2)
EP
where (B) is a constant and (n) assumes values of (1/2, 2, 3/2 and 3) for allowed direct, allowed indirect, forbidden direct and forbidden indirect transitions respectively. Assuming the transition type is allowed direct, the optical band gaps of the films are estimated from the
AC C
plot of (αhυ)2 against the photon energy hυ, as in Fig 7. The extrapolation of the straight line portion of the curve to the energy axis at (αhυ)2 = 0 gives the band gap. The band gap analysis shows the samples with higher Zn concentration (CZS 10 to
CZS 40) have two band gap energies, in the range of 1.6 to 1.7 eV and 2.1 to 2.3 eV. Sreejith et al. also reported a similar observation, describing CuZnS as having a double band gap nature arising from a mixed structure consisting of CuxS and ZnS phases [6]. In the present case, under the Zn-rich growth condition, the samples are mainly consisting of two phases, a CuS phase and a non-stoichiometric CuxZnyS phase. The higher band gap value ranging between 2.1 to 2.3 eV can be assigned to CuS phase, and the lower band gap value ranging between 1.6 to 1.7 eV can be due to non-stoichiometric CuxZnyS phase. The CuxS system has
ACCEPTED MANUSCRIPT band gap values ranging from 2.1 eV to 2.6 eV for copper deficient CuS phase, matching with the observation [17, 27, 28]. The incorporation of Zn into the CuS lattice forming a nonstoichiometric CuxZnyS alloy structure can significantly alter the band alignment of the host system. Similar band gap values about 1.6 eV are reported for CuZnS system by others [4, 5]. Moreover, the obtained band gap values (1.6 to 1.7 eV) are suitable for an absorber layer in
RI PT
thin film solar cells. The samples with higher Cu concentration (CZS 50 to CZS 80) having a Zn-doped CuS structure exhibit band gap values in the range 2.4 to 2.6 eV, consistent with other
reported values [9,15]. Besides, the average transmittance in the visible region (Fig 7 and
SC
Table 2) for the samples CZS 70 and 80 are significantly higher than the other samples. The higher transparency is directly linked to the smaller grain size, which minimises light absorption and scattering losses [29]. The sample CZS 80 has an average transparency of 63
M AN U
% in the visible region with a peak value of 74 % near 600 nm. This transparency is relatively high and is suitable for application as a p-type TCM.
3.5 Work function
The ultraviolet photoelectron spectroscopy (UPS) technique was applied to determine
TE D
the work function of the as-deposited films. Since films can be categorised into two chemically dissimilar systems, one having nonstoichiometric CuxZnyS alloy structure with band gap ~ 1.6 - 1.7 eV and the other having Zn-doped CuS structure with band gap ~ 2.4 2.6 eV, representative samples are selected for characterisation. He (I) excitation source with
EP
energy 21.2 eV was used as probing source, and the resulted spectra along with enlarged plotting of secondary electron cutoff (Ecutoff) region are represented in Fig 8. The work
AC C
function is calculated by subtracting the secondary cut off energy value from the photon energy of the excitation source [30]. The work function values are 4.77 eV and 4.85 eV for CZS 40 and CZS 60 samples respectively. The measured work function of Zn-doped CuS (CZS 60) is in good agreement with the reported value [23].
3.6 Electrical properties The carrier mobility, carrier concentration, and electrical conductivity of the asdeposited films are measured by the Hall measurement technique performed at room temperature using the Van der Pauw configuration and are summarised in Table 3. The type of conductivity is determined as p-type for all the samples. Overall, the carrier concentration
ACCEPTED MANUSCRIPT is found to vary between 1 to 10 x 1021 cm3, which is in the range of degenerately doped semiconductors. There is not much apparent variation in the mobility for the as-deposited samples, and it appeared between 0.6 to 3.5 cm2 V−1 S−1, comparable to previous reports [9, 15]. In general, the samples exhibit a much higher hole conductivity in comparison with other p-type chalcogenides, and it mainly comes from the relatively high carrier concentration.
RI PT
From the analysis of variation in the electrical properties as a function of time for a duration of six months (Table 3), all the samples are found to retain the p-type conductivity, with not much variation in electrical properties.
It can be seen that the increase in copper content plays a key role in determining the
SC
electrical conductivity and transparency of the films. The higher copper content decreases the grain size achieving a high p-type conductivity along with increased transparency. Compared to other samples, CZS 80 show a higher conductivity as well as transparency values, making
M AN U
the sample suitable for the application as a p-type transparent conducting material. The retention of high conductivity along with carrier concentration and mobility even after a six months duration emphasis the structural and electrical stability of the sample. Along with its high work function, this sample can be considered as a promising candidate as a transparent
4. Conclusions
TE D
hole selective electrode for thin film solar cells.
Cu-Zn-S thin films are deposited at room temperature and atmospheric conditions using Successive Ionic Layer Adsorption and Reaction (SILAR) method. X-ray photoelectron
EP
spectroscopy analysis shows that the as-deposited films are composed of constituent elements Cu, Zn, and S without any other elemental impurities. Based on the X-ray diffraction and
AC C
Raman spectroscopic analyses, it is observed that CuS act as the host lattice and Zn is incorporated into the CuS host lattice. Depending upon Cu to Zn ratio in the precursors, two structurally heterogeneous systems are formed. Under the Zn-rich growth condition, the samples are mainly consisting of two phases, a CuS phase and a non-stoichiometric CuxZnyS phase, exhibiting two band gap energies in the range of 2.1 to 2.3 eV and 1.6 to 1.7 eV respectively. The samples deposited under higher Cu concentration have a Zn-doped CuS structure with band gap values in the range of 2.4 to 2.6 eV. All the films exhibited relatively high absorption coefficient value between 104 to 105 cm-1 in the visible and near - IR spectral range. Hall measurements performed at room temperature confirmed p-type conductivity for all the samples. In general, the samples exhibited a much higher hole conductivity in
ACCEPTED MANUSCRIPT comparison with other p-type chalcogenides resulting from the relatively high carrier concentration in the range of 1 to 10 x 1021 cm3. All the samples are found to retain the p-type conductivity, along with no apparent change in electrical properties even after a period of six months. In conclusion, Cu-Zn-S films having inexpensive, non-toxic, and earth-abundant constituent elements, coupled with suitable optical and electrical properties are deposited
RI PT
through a room temperature and atmospheric deposition process. The obtained band gap values (1.6 to 1.7 eV) are appropriate for absorber layer material in thin film solar cells. The samples having high electrical conductivity and transparency are promising to use as
SC
transparent hole selective electrode in thin film solar cells.
M AN U
Acknowledgements
Part of the reported work (characterization) was carried out at the CeNSE, IISc under INUP at IISc which have been sponsored by DeitY, MCIT, Government of India. The authors acknowledge the MHRD, Government of India for the characterization facilities (Raman, UV-vis and AFM) under the plan fund sanctioned to the Department of Physics,
AC C
EP
TE D
NIT, Tiruchirappalli.
ACCEPTED MANUSCRIPT
References [1]
P. Sinsermsuksakul, J. Heo, W. Noh, A.S. Hock, R.G. Gordon, Atomic layer
deposition of tin monosulfide thin films, Adv. Energy Mater. 1 (2011) 1116–1125. [2]
P. Jackson, R. Wuerz, D. Hariskos, E. Lotter, W. Witte, M. Powalla, Effects of heavy
RI PT
alakali elements in Cu(In,Ga)Se2 solar cells with efficiencies up to 22.6%, Phys. Status Solidi – Rapid Res. Lett. 10 (2016) 583–586. [3]
First Solar, Inc., First solar achieves yet another cell conversion efficiency world
record, http://investor.firstsolar.com/releasedetail.cfm?ReleaseID=956479, February 23,
[4]
SC
2016, (accessed 05.01.17).
M. Innocenti, L. Becucci, I. Bencistà, E. Carretti, S. Cinotti, L. Dei, F. Di Benedetto,
M AN U
a. Lavacchi, F. Marinelli, E. Salvietti, F. Vizza, M.L. Foresti, Electrochemical growth of Cu– Zn sulfides, J. Electroanal. Chem. 710 (2013) 17–21. [5]
F. Di Benedetto, S. Cinotti, F. D’Acapito, F. Vizza, M.L. Foresti, A. Guerri, A.
Lavacchi, G. Montegrossi, M. Romanelli, N. Cioffi, M. Innocenti, Electrodeposited semiconductors at room temperature: an X-ray Absorption Spectroscopy study of Cu-, Zn-, Sbearing thin films, Electrochim. Acta. 179 (2015) 495–503.
M.S. Sreejith, D.R. Deepu, C.S. Kartha, K. Rajeevkumar, K.P. Vijayakumar, Tuning
TE D
[6]
the properties of sprayed CuZnS films for fabrication of solar cell, 202107 (2014) 14–18. [7]
M.S. Sreejith, D.R. Deepu, C. SudhaKartha, K. Rajeevkumar, K.P. Vijayakumar,
Improvement of sprayed CuZnS/In2S3 solar cell efficiency by making multiple band gap [8]
EP
nature more prominent, J. Renew. Sustain. Energy. 8 (2016) 023502. N. Kitagawa, Seigo Ito, Duy-Cuong Nguyen, Hitoshi Nishino, Copper Zinc Sulfur
AC C
Compound Solar Cells Fabricated by Spray Pyrolysis Deposition for Solar Cells, Nat. Resour. 04 (2013) 142–145. [9]
R. Woods-Robinson, J.K. Cooper, X. Xu, L.T. Schelhas, V.L. Pool, A. Faghaninia,
C.S. Lo, M.F. Toney, I.D. Sharp, J.W. Ager, P-Type Transparent Cu-Alloyed ZnS Deposited at Room Temperature, Adv. Electron. Mater. 2 (2016) 1–9. [10]
K. Yang, Y. Nakashima, M. Ichimura, Electrochemical Deposition of CuxS and
CuxZnyS Thin Films with p-Type Conduction and Photosensitivity, J. Electrochem. Soc. 159 (2012) H250. [11]
M. Dula, K. Yang, M. Ichimura, Photochemical deposition of a p-type transparent
alloy semiconductor CuxZnyS, Semicond. Sci. Technol. 27 (2012) 125007.
ACCEPTED MANUSCRIPT [12]
B. Tong, M. Ichimura, Annealing of p-type wide-gap CuxZnyS thin films deposited by
the photochemical deposition method, Jpn. J. Appl. Phys. 55 (2016) 098004. [13]
M. Ichimura, Y. Maeda, Conduction type of nonstoichiometric alloy semiconductor
CuxZnyS deposited by the photochemical deposition method, Thin Solid Films. 594 (2015) 277–281. A.M. Diamond, L. Corbellini, K.R. Balasubramaniam, S. Chen, S. Wang, T.S.
RI PT
[14]
Matthews, L.-W. Wang, R. Ramesh, J.W. Ager, Copper-alloyed ZnS as a p-type transparent conducting material, Phys. Status Solidi. 209 (2012) 2101–2107. [15]
X. Xu, J. Bullock, L.T. Schelhas, E.Z. Stutz, J.J. Fonseca, M. Hettick, V.L. Pool, K.F.
SC
Tai, M.F. Toney, X. Fang, A. Javey, L.H. Wong, J.W. Ager, Chemical Bath Deposition of pType Transparent, Highly Conducting (CuS)x:(ZnS)1-x Nanocomposite Thin Films and Fabrication of Si Heterojunction Solar Cells, Nano Lett. 16 (2016) 1925–1932. Linlin Li, Peining Zhu, Shengjie Peng, Madhavi Srinivasan, Qingyu Yan, A.
M AN U
[16]
Sreekumaran Nair, Bin Liu, and Seeram Samakrishna, Controlled Growth of CuS on Electrospun Carbon Nanofibers as an Efficient Counter Electrode for Quantum DotSensitized Solar Cells, J. Phys. Chem. C 118 (2014) 16526–16535. [17]
E. Güneri, A. Kariper, Optical properties of amorphous CuS thin films deposited
[18]
TE D
chemically at different pH values, J. Alloys Compd. 516 (2012) 20–26. M. Hurtado, S. D. Cruz, R. A. Becerra, C. Calderón, P. Bartolo-Pérez and G. Gordillo,
Aluminum-doped zinc oxide films as transparent conductive electrode for organic lightemitting devices, IEEE 40th Photovoltaic Specialist Conference (2014) 0368–0372 Ding Tang, Qinglong Wang, Fangyang Liu, Lianbo Zhao, Zili Han, Kaiwen Sun,
EP
[19]
Yanqing Lai, Jie Li, Yexiang Liu, An alternative route towards low-cost Cu2ZnSnS4 thin film [20]
AC C
solar cells, Surface and Coatings Technology 232 (2013) 53-59. T.T. T, P. Sheng, J. Li, L. Chen, L. Yuan, C.A. Grimes, Q. Cai, Synthesis and
photocatalytic application of ternary Cu–Zn–S nanoparticle-sensitized TiO2 nanotube arrays, 210 (2012) 425–431. [21]
M. Adelifard, H. Eshghi, M.M. Bagheri Mohagheghi, Synthesis and characterization
of nanostructural CuS–ZnS binary compound thin films prepared by spray pyrolysis, Opt. Commun. 285 (2012) 4400–4404. [22]
P. Wang, Y. Gao, P. Li, X. Zhang, H. Niu, Z. Zheng, Doping Zn2+ in CuS Nanoflowers
into Chemically Homogeneous Zn0.49Cu0.50S1.01 Superlattice Crystal Structure as HighEfficiency n-Type Photoelectric Semiconductors, ACS Appl. Mater. Interfaces. 8 (2016) 15820–15827.
ACCEPTED MANUSCRIPT [23]
S. Bera, A.K. Katiyar, A.K. Sinha, S.P. Mondal, S.K. Ray, Resistive switching
characteristics of a single Zn-doped CuS nanoball anchored with multi-walled carbon nanotubes, Mater. Des. 101 (2016) 197–203. [24]
Z. Zhan, C. Liu, L. Zheng, G. Sun, B. Li, Q. Zhang, Photoresponse of multi-walled
carbon nanotube–copper sulfide (MWNT–CuS) hybrid nanostructures, Phys. Chem. Chem.
[25]
RI PT
Phys. 13 (2011) 20471. N.A. Yeryukov, A.G. Milekhin, L.L. Sveshnikova, T.A. Duda, L.D. Pokrovsky, A.K.
Gutakovskii, S.A. Batsanov, E.E. Rodyakina, A. V Latyshev, D.R.T. Zahn, Synthesis and Characterization of CuxS (x=1-2) Nanocrystals Formed by the Langmuir–Blodgett
SC
Technique, J. Phys. Chem. C. 118 (2014) 23409–23414. [26]
J. Tauc, A. Menth, States in the gap, J. Non. Cryst. Solids. 8-10 (1972) 569–585.
[27]
M. Adelifard, H. Eshghi, M.M.B. Mohagheghi, An investigation on substrate
M AN U
temperature and copper to sulphur molar ratios on optical and electrical properties of nanostructural CuS thin films prepared by spray pyrolysis method, Appl. Surf. Sci. 258 (2012) 5733–5738. [28]
F.A. Sabah, N.M. Ahmed, Z. Hassan, Effects of Concentration and Substrate Type on
Structure and Conductivity of p-Type CuS Thin Films Grown by Spray Pyrolysis Deposition,
[29]
TE D
J. Electron. Mater. 46 (2017) 218-225.
S. Hemathangam, G. Thanapathy, S. Muthukumaran, Tuning of band gap and
photoluminescence properties of Zn doped Cu2S thin films by CBD method, J. Mater. Sci. Mater. Electron. 27 (2016) 2042–2048.
X. Jiang, F.L. Wong, M.K. Fung, S.T. Lee, Aluminum-doped zinc oxide films as
EP
[30]
transparent conductive electrode for organic light-emitting devices, Appl. Phys. Lett. 83
AC C
(2003) 1875–1877.
ACCEPTED MANUSCRIPT
Figure Captions
Fig 1. Schematic illustration of a single SILAR cycle in the formation of CuZnS thin films.
RI PT
Fig 2. FESEM images of CuZnS thin films deposited on glass substrate.
Fig 3. AFM micrographs of CuZnS thin films deposited on glass substrate.
SC
Fig 4. XPS full range and elemental specific high-resolution scans of CuZnS thin films.
Fig 5. GI-XRD patterns of SILAR deposited CuZnS thin films with varying Cu/(Cu+Zn)
M AN U
ratio.
Fig 6. Raman spectra of as-deposited CuZnS thin films.
TE D
Fig 7. Optical properties of CuZnS thin films deposited with varying Cu/(Cu+Zn) ratio.
Fig 8. He (I) UPS spectra of CuZnS thin films along with an expanded view of secondary
AC C
EP
electron cutoff region.
ACCEPTED MANUSCRIPT
Cu
Zn
S
CZS 20
15.94
13.72
70.34
CZS 40
20.35
11.06
68.59
CZS 60
27.18
3.25
69.57
CZS 80
27.89
1.87
70.24
RI PT
Atomic concentration (%)
Thickness (nm)
Bandgap (eV)
Average transmittance (%)
CZS 10
434
1.71, 2.20
33.61
CZS 20
509
1.65, 2.33
44.29
CZS 30
547
1.60, 2.31
35.38
CZS 40
514
1.58, 2.14
46.23
CZS 50
274
2.68
36.45
CZS 60
345
2.65
CZS 70
104
2.51
CZS 80
72
2.46
AC C
EP
TE D
Sample code
27.13 51.14
63.28
Table 2. Summary of optical properties
M AN U
Table 1. Elemental concentration from XPS analysis
SC
Sample code
ACCEPTED MANUSCRIPT
Sample code
Bulk concentration (x 1021 cm3)
Mobility (cm2 V−1 S−1)
Conductivity (S cm−1)
After 1 month
After 3 months
After 6 months
As deposited
After 1 month
After 3 months
After 6 months
As deposited
After 1 month
After 3 months
After 6 months
CZS 10
4.06
3.46
2.26
1.13
1.521
1.402
1.279
0.822
472
386.7
211
205.2
CZS 20
4.80
4.28
2.14
1.99
0.606
0.432
0.362
0.350
195.8
132.4
55.4
40.3
CZS 30
3.19
2.99
1.61
1.35
1.517
1.083
0.745
0.347
438.7
218.7
70.1
46.9
CZS 40
4.45
4.13
2.97
2.33
2.142
1.305
0.645
0.645
596.4
347.7
133
88.9
CZS 50
5.38
3.70
3.56
2.58
2.710
1.464
1.249
0.927
1231
764.5
464.6
377.1
CZS 60
8.01
7.32
5.97
4.06
1.645
0.748
0.533
0.170
882.6
306.9
61.3
30.5
CZS 70
10.1
7.79
6.40
4.23
2.681
2.726
1.514
1.490
2926
1928
1470
1152
CZS 80
6.64
5.69
2.79
1.34
3.519
3.253
2.927
1.745
2964
2827
2673
2495
SC
M AN U
TE D
AC C
EP
Table 3. Summary of electrical properties
RI PT
As deposited
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Research Highlights
Room temperature and atmospheric deposition of Cu-Zn-S thin films.
•
Inexpensive, earth-abundant, and non-toxic constituents for low cost solar cells.
•
Investigation of band gap tunability with varying Cu/(Cu+Zn) ratio.
•
Bandgap values in the range suitable for absorber layer under Zn rich conditions.
•
Optical and electrical properties suitable for p-type TCMs under Cu rich conditions.
AC C
EP
TE D
M AN U
SC
RI PT
•
