Exploration of Cd1−xZnxSe as a window layer for CIGS based solar cell with PEDOT: PSS as back surface field layer

This research investigates the potential of Cd1−xZnxSe thin film for photovoltaic applications. The electrical behavior of CIGS based solar cell is examined with the novel Cd1−xZnxSe as buffer layer material by Solar Cell Capacitance Simulator (SCAPS). The tunability of Cd1−xZnxSe facilities to reduce the defects between absorber and buffer layer by determining the ideal conduction band offset. It is revealed that cross-over occurs between the p-type absorber and the metal back contact if the metal work function is below 4.6 eV. In this research, a thin PEDOT: PSS back surface (BSF) layer was integrated which enhances the device efficiency from 22.5 percent to 28.32% while retaining the metal work function at 5.1 eV. The trade-off between the use of metal having higher work function and inclusion of heavily doped BSF layer is one of the important findings of this research. These findings pave the way for Cd1−xZnxSe to be commercially used as a buffer layer material for CIGS solar cell.


Introduction
Thin film solar cells such as Cadmium Telluride and Copper Indium Gallium (Sulfur, Selenide) (CuIn 1−x Ga x Se 2 ) account for most of the market share due to their lower cost and greater stability (Noufi andZweibel 2006, Sobayel et al 2021).With a current efficiency of 23.35%, CIGS solar cells outperform all other thin film solar cell technologies (Nakamura et al 2019).The key factors for such execution can be attributed to stability, a high conversion rate, a low production cost, and eco-friendliness.In addition, CIGS solar cells are gaining attention due to their capability to perform under lower irradiance.Photovoltaic researchers have been attracted (Hamri et al (2019)) to CIGS material due to its prominent optoelectronic properties like high absorption co-efficient (10 5 cm −1 ), tunability of bandgap, and a smaller amount of material consumption (Romeo et al 2004, Kapur et al 2008, Nakada 2012, Rampino et al 2012, Fischer et al 2014, Reinhard et al 2014).Variating the indium (In) and gallium (Ga) concentrations in CIGS can control the bandgap, which implies that the bandgap is a direct function of the alloy stoichiometry of the CIGS material.In contrast, copper (Cu) vacancies induce the p-type characteristics in the chalcopyrite structure of CIGS.The lattice parameter ratio (c/ a) of the CIGS structure, as shown in figure 1 is near 2 (Srour et al 2016), and the bond between Ga-Se or In-Se, Cu-Se is responsible for any tetragonal distortion.Due to the variation in CIGS composition, conduction band mismatch with the adjacent layer often occurs, which adversely affects the photovoltaic performance.
In copper indium gallium selenide (CIGS) solar cells, the buffer layer is frequently made of cadmium sulfide (CdS).It provides several benefits that boost the device's overall effectiveness and performance.Since CdS is an n-type semiconductor, it has an abundance of electrons and can speed up electron transport.A beneficial electron flow path from the CIGS layer to the TCO (transparent conducting oxide) layer is created when it forms a heterojunction with the p-type CIGS absorber layer.As a result, recombination losses are decreased, and efficient charge carrier collection is increased.Furthermore, In the CIGS absorber layer, CdS can function as a passivation layer to lower the density of surface defects and trap states.The CIGS/CdS interface's recombination losses are decreased, and the carrier lifetime is increased because to this passivation, which boosts device performance.Though CdS is still the most popular buffer layer material for p-type CIGS absorbers, it exhibits a few intrinsic problems: (i) contains a toxic element, Cd.During the production, use and disposal of CdS, there is a risk of environmental contamination if proper handling and waste management protocols are not followed.The toxicity of cadmium raises concerns regarding the long-term sustainability and environmental impact of CdS-based solar cells (ii) CdS has a relatively wide band gap energy (approximately 2.4 eV), which can result in a loss of short-wavelength photons that could be absorbed by the CIGS absorber layer.This limitation reduces the overall light absorption and can lower the efficiency of the solar cell, especially in the blue and ultraviolet regions of the spectrum; and (iii) is deposited using a non-vacuum method (Naghavi et al 2010, Ali et al 2018).When exposed to heat, humidity, and other environmental conditions, CdS is subject to deterioration.The CdS layer may experience chemical reactions and structural alterations over time, resulting in a decline in performance and stability.The longevity and long-term dependability of CdS-based CIGS solar cells may be impacted by this degradation.Cadmium is a less common element than other elements compared to other materials used in solar cells, despite CdS being a comparatively cheap material.The scalability and commercial viability of CdS-based CIGS solar cell technology may be affected by the price and availability of cadmium, which can vary.To achieve higher current density (Jsc), it is practical to keep the CdS thickness at a lower value (<100 nm) to reduce the rate of recombination of minority charge carriers (holes) while crossing the n-type layer.It is also important to optimise the blue response of the spectrum as well.But lower thickness can increase the formation of pinholes during its fabrication on the film, which has a significant impact on the cell performance.Considering all these factors, researchers are exploring the most viable alternatives and an ideal buffer layer that should have the following characteristics: (i) a higher bandgap; (ii) environmentally friendly; (iii) have similar or better optoelectronic properties to CdS; and iv) can be fabricated using low-cost methods.Kieserite solar cells without CdS usually lag CdS-based ones in terms of Voc and Fill Factor (FF) but achieve higher short circuit current (Jsc) due to the higher bandgap (Friedlmeier et al 2015).Besides, many other II-VI polycrystalline semiconducting materials such as CdSe, ZnSe etc have been widely used in a thin-film solar cells as a buffer layer due to their longer absorption spectrum near infra-red (IR) (Munna et al 2021).While both CdS and CdSe have their respective advantages and disadvantages in solar cell applications, CdSe offers a few specific benefits over CdS.Compared to CdS, CdSe has a lower bandgap energy.It has a bandgap of about 1.7 eV (Bagheri et al 2020), which is closer to the ideal range for solar cell effectiveness.Because of its smaller bandgap, CdSe can absorb more of the solar spectrum, including a significant portion of the visible and near-infrared spectra.CdSe has a higher absorption coefficient (Garg 2019), which means that a larger portion of incident light can be successfully absorbed by CdSe, increasing light absorption.It has higher electron and hole mobilities which makes it possible for electrons and holes to move across the solar cell efficiently, lowering recombination losses and enhancing device performance overall.However, there are still difficulties with CdSe, such as the toxicity of cadmium.Efforts are being made to mitigate the environmental and health concerns related to the use of CdSe in solar cells.
Researchers also explore potentiality of promising buffers, especially Zn-based materials by exposing their competitive and benign characteristics rather than an extension of relatively toxic CdS layer's (Moon et al 2023, Rahman et al 2023).ZnSe has a direct bandgap of about 2.7 eV, which is larger than the bandgap of many other p-type solar cell materials.It typically has a reasonably high hole mobility, allowing holes to pass through the material effectively.Moreover, ZnSe is chemically stable, which is an essential characteristic for a material used in the construction of solar cells.Taking this into consideration, we have investigated Cadmium zinc selenide (Cd 1−x Zn x Se), as a potential alternative of CdS buffer layer for photovoltaic applications.It is an n-type semiconductor with intriguing size-dependent characteristics, great stability, and a large band gap that covers the whole electromagnetic spectrum (Gu et al 2005, Pan et al 2013, Wang et al 2016, Yang et al 2015).Cd 1−x Zn x Se is a notable ternary compound material with a wide-ranging potential utilisation in various optoelectronic device applications due to its remarkable optical features (Hankare et al 2006, Vartanian, Asatryan, andKirakosyan 2007).Moreover, film preparation can be performed by diverse fabrication methods (Trager-Cowan et al 1996, Schreder et al 2000, Kale et al 2007).Depending on the composition, the alloyed ternary compound Cd 1−x Zn x Se has tuneable band gaps which can be tuned from 1.74 eV (x = 0 for CdSe) (Ninomiya and Adachi 1995) to 2.72 eV (x = 1 for ZnSe) (Al-Kuhaili et al 2013), which can be derived by the following equation It is commonly known that the highest efficiency for a single band gap semiconductor for the AM1.5 solar spectrum is 33% at a band gap of 1.5 eV.In solar cell applications, the bandgap of a p-type buffer layer is often set to be greater than that of the absorber layer to prevent photo-generated carriers from recombining.Therefore, although the bandgap of CdSe (Eg = 1.72 eV) is ideal for highly efficient CIGS solar cells, the Cd toxicity problem persists.Allowing Zn with Cd can have two effects.First, the bandgap can be tuned to more than 2 eV, which are comparable to the bandgap (Eg = 2.42 eV) of famous CdS.Secondly inclusion of earth abundant and non-toxic Zinc (Zn) lessens the amount of poisonous Cd.Moreover, tunability of bandgap facilities decent battle against photo corrosion (Hankare et al 2006) once being used for optoelectronic devices.It has been reported that Cd 1−x Zn x Se can effectively offset the conduction band mismatch once being used as buffer layer for CIGS solar cell (Sharbati et al 2015).By tuning the gallium/indium ratio in the case of CIGS and sulfur/selenium ratio in CIGSSe compound are two viable options to adjust the conduction band offset between Cd 1−x Zn x Se buffer layer.
Even though Cd 1−x Zn x Se has been used in a variety of optoelectronic devices, there has been little research into photovoltaic applications.Several previous studies (Mahjabin et al 2020, Islam et al 2021, Sobayel et al 2021) used SCAPS software to investigate the underlying properties of CIGS/CdS solar cells.But besides this, no research on the CIGS/ Cd 1−x Zn x Se structure has been published in which the typical buffer layer is replaced by the Cd 1−x Zn x Se layer.As a result, this research focuses primarily on the modelling of CIGS solar cells using Cd 1−x Zn x Se as a buffer layer instead of CdS.Numerical simulations of the novel CIGS/ Cd 1−x Zn x Se structure are performed, and further investigation into the impact of various constituent layer parameters is described.Moreover, the effects of ambient temperature on the proposed structure are being studied.Based on the results, an ideal solar cell model of CIGS/ Cd 1−x Zn x Se (Nurhafiza et al 2021, Hossain et al 2018, 2022) is provided.

Device structure and numerical simulation parameters
To investigate the proposed model, the famous Solar Cell Capacitance Simulator (SCAPS-1D) program developed by the Department of Electronics and Information Systems (ELIS), University of Gent, Belgium was utilized (Burgelman, Nollet, and Degrave 2000, Decock, Khelifi, and Burgelman 2011).Various material parameters for cell structures are manually entered based on different literature reviews and experimental findings.In this simulation work, we have assumed the ambient temperature as 300 K and AM 1.5 G one sun spectrum as illumination.
The schematic structure of the proposed solar cell is shown in figure 2, where the most common and widely used FTO is used as the front contact material and it also acts as the Transparent Conductive Oxide (TCO) layer of the structure.The optimized bandgap of the CIGS is 1.40 eV (based on the higher efficiency from the simulation).Although the typical bandgap of CIGS is around 1.15 eV, tuning material composition, (In and Ga proportion) the bandgap can be varied which is discussed earlier.Typical thicknesses of each layer were obtained from the different literatures (Hall 1952, Hegedus and Shafarman 2004, Alzoubi and Moustafa 2019).Interface defect layers (IDL) of 10 nm thickness were used to provide realistic view and to assess the defect densities subsisting on material interfaces.In addition, the neutral type of defect model is used in the simulation where the density of the defect in the active layer has been considered 1 × 10^14 ^cm −3 .The thermal velocity of electrons and holes of 1×10 7 cm s −1 , the Gaussian energy distribution with a characteristic energy of 0.1 eV, was considered for the model.Tables 1 and 2 display the material parameters of all the constituent layers and the defect properties respectively.

Validation of Cd 1−x Zn x Se
In this work, II-VI compound Cd 1−x Zn x Se is proposed as a substitution of the conventional CdS buffer layer.The performance of two different structures is investigated and their current-voltage (J-V) characteristics are shown in figure 3(a).The efficiency appears better for CIGS/ Cd 1−x Zn x Se heterojunction than that for CIGS/ CdS heterojunction, in which the enhancement is mostly contributed by the increased open-circuit voltage (V OC ).On the other hand, the insignificant change in short-circuit current (J SC ) of both proposed and baseline solar cells is in good agreement with the QE curves as shown in figure 3(b).The QE also confirms the suitability of Cd 1−x Zn x Se based CIGS solar cells.In both cases, the increase of bandgap of Cd 1−x Zn x Se plays the vital rule.
The ratio of the photocurrent produced by a solar cell under monochromatic illumination of a specific wavelength to the value of the spectral irradiance at that same wavelength is known as the spectral response (denoted by SR(λ), with the unit's A/W).Since the number of photons and irradiance are related, the spectral response can be written in terms of the quantum efficiency as following (Laozi. and Hinton 2002) where λ is in micrometers.Depending on the quantum efficiency value utilized, the spectral response in equation (1) may be internal or external.According to the equations given above, incident photons at a specific wavelength directly affect the quantum efficiency of a solar cell.High QE is exhibited by materials with a larger photon absorption capacity although surface roughness, bandgap, the condition of the interfaces between adjacent layers, etc, play an important role.In this study, it has been observed that the device with CdS buffer layer has lower absorption than Cd 1−x Zn x Se alloy due to higher bandgap.Because of its reduced surface recombination and increased responsiveness with longer wavelengths and higher diffusion length, the alloy exhibits highly exceptional promotion red light response.

Conduction band offset behavior analysis
According to Anderson's law, the value of conduction band offset (ΔE c ) can be defined as (Chen et al 2017) Different materials have different electron affinities.The electron affinity is an important parameter for both buffer and absorber layers.Due to the variation of electron affinity, there will be a mismatch of interface levels of both the layers producing spike and cliff type interface, which play a crucial role in determining the efficiency and performance of the solar cell.In spike-like interface, the bottom of the conduction band (Ec) of buffer layer is higher than the corresponding Ec level of absorber layer which refers to a sudden increase in energy level at the junction between CIGS/ Cd 1−x Zn x Se interface as shown in figure 4(a) producing negative Conduction Band Offset (-CBO).In heterostructure like CIGS/Cd 1−x Zn x Se, such type of interface assists in efficiently separating and transporting charge carriers (electrons and holes) generated by sunlight.On the other hand, when Ec of absorber layer is higher than the Ec of buffer layers, cliff-like interface is generated, as shown in figure 4(b) producing positive Conduction Band Offset (+CBO).So, changing of buffer layer material eventually changes the ΔE c value.The increase of open circuit voltage and FF can be ascribed to the proper choice of conduction band offset value.So, the creation of an optimal conduction band offset (CBO) between the buffer layer and the CIGS absorber layer is crucially important (Herberholz et al 1997).In this work, Cd 1−x Zn x Se thin film with a controllable band gap is used to control the CBO between Cd 1−x Zn x Se /CIGS interface.The optimum bandgap of the traditional CdS buffer layer is 2.42 eV, but the bandgap of Cd 1−x Zn x Se film can be varied from 1.74 eV to 2.58 eV, which eventually controls the CBO of Cd 1−x Zn x Se /CIGS layers.From a theoretical standpoint, when the conduction band minimum (CBM) of the buffer layer is lower than that of CIGS, a cliff forms at the buffers/CIGS interface.The difference in electron affinity values of these two layers is responsible for such a cliff.The generated cliff acts as a barrier against injected electrons from the n-type region during forward bias.As a result, electrons gather in numbers at the cliff region promoting the unwanted recombination between majority carriers at the buffer/CIGS interface.Consequently, a leakage conduction path is produced which drops both V oc and FF values.
In this study, the electron affinity of Cd 1−x Zn x Se layer varied from 3.6 eV to 4.8 eV to measure the CBO.The electron affinity denoted as EA (x) for Cd 1−x Zn x Se can be expressed as follows - Where ∆ ( ) EA x represents the change in electron affinity due to the variation in composition x.It is important to note that ∆ ( ) EA x can be either positive or negative depending on the specific material properties and the composition range of x.
From the simulation result shown in figure 5, it has been found that, maximum efficiency of 27% is found when the electron affinity of the buffer layer lies between 4.0 eV to approx.4.2 eV.When it is less than that of the absorber, a cliff type band alignment occurs which triggers the recombination process through Cd 1−x Zn x Se/CIGS interface.So, efficiency reduces drastically to 8.41% at 3.60 eV.On the other hand, when the buffer layer affinity is higher than CIGS, a spike type of band alignment occurs, which acts as barrier for the charge carriers.In this case also, unwanted recombination happens through Cd 1−x Zn x Se/CIGS interface.From figure 4, it is also clear that both open circuit voltage and efficiency maintain their respective higher values approx. in the range between −0.3 eV to −0.1 eV.This is because of the reduction of recombination rate at CIGS and Cd 1−x Zn x Se interface, but with the higher values of CBO, both decrease because of the higher energy barrier for photo-generated carriers under forward bias.So, varying the Zn composition, the electron affinity of Cd 1−x Zn x Se buffer layer can be varied from 4.2 eV to 4.4 eV.
The majority carrier recombination via the Cd 1−x Zn x Se/CIGS interface defects might be significantly reduced with an acceptable conduction band offset.This study shows the importance of Cd 1−x Zn x Se buffer layer for adjusting the conduction band offset between the buffer/absorber interface.

Activation energy analysis
The presence of recombination centers at the buffer/CIGS contact can cause solar cell properties to deteriorate.The temperature dependence of V OC can be used to estimate the activation energy (E a ) at the buffer/absorber  interface.The relationship is (Rau andSchock 1999, Hegedus andShafarman 2004) provided in the following equation, A is the diode ideality factor, kT/q is the thermal voltage, J 00 is the reverse saturation current pre-factor, and J L is the photocurrent, respectively.
From the linear extrapolation data of the range of V OC measurement temperatures (200 K < T < 400 K), Ea is the value of the intersection at 0 K.The Shockley-Read-Hall (SRH) recombination process becomes the main recombination mechanism in the depletion region if both E a and E g match (Dharmadasa 2009, Sobayel et al 2019).However, mismatching of E a and E g resembles that recombination is also occurring at the absorber-buffer interface, the value of E g -E a increases (Wang et al 2010, Todorov et al 2013).
Open circuit voltage at 100% illumination as a function of temperature is shown in figure 6.The activation energy deduced from V OC vs T extrapolated to 0 K is ~1.57eV, which is 0.17 eV higher than the absorber bandgap.(Eg = 1.4 eV at 300 K).It means that, not only the SRH recombination is occurring inside into the cell, there are other recombination processes like Auger and radiative recombination processes are also happening.Because, in a solar cell, when the activation energy differs from the bandgap, the recombination that takes place is called radiative recombination.As a result, lowering the number of recombination centers at the Cd 1−x Zn x Se /CIGS interface is one approach to increase the performance of CIGS solar cells.

Metal work function and cross-over effect analysis
In this work, the efficiency variation of the structure is also explored as a function of back contact material.Mostly, in solar cells, a flat-band model is considered default in the case of defining the metal work function for both back and front contact material.But cell performance is strongly influenced by metal work function values, especially for the back contact.In this simulation, apart from taking the flat band model the work function varied from 4.6 to 5.3 eV and found the linearity between the work function and PCE of the solar cell up to 4.6 eV which is depicted in figure 7.After 5.1 eV and onward the efficiency saturates at approx.25%.
Generally, selecting the proper metal for back contact is crucial for any kind of solar cell.Higher values of metallic work function are appreciable in the case of solar cells to have good Ohmic contact or near Ohmic contact at metal/semiconductor interface.Lower values of metal work function create Schottky type contact and degrade the performance of the cell.But, due to the commercial point of view, it is impractical to use metal having higher work functions.It has been observed in figure 8 that, if the work function of metal back contact is lower than 5.5 eV, cross-over phenomena occur in the current-voltage (J-V) curve which causes drastic loss in the V oc .The intersection of light and dark I-V characteristics due to an increase of the contact saturation current under illumination is known as cross-over effect in solar cell.This phenomenon is attributed to the contribution of electron minority current at the back contact Schottky barrier.The V oc is reduced more when the work function has lower values.This is because secondary junctions are formed between the absorber semiconductor and metal, which is called a Schottky junction.This junction is aligned against the primary PN junction formed between CIGS and Cd 1−x Zn x Se.So, we may conclude that, if light soaked I-V curve intersect with the dark I-V curve, there must be a Schottky barrier in the solar cell structure.In summary, it has been found that higher values of metal work function are suitable for designing high efficiency solar cells to avoid the cross-over effect.From a structural and commercial standpoint, copper-doped carbon (Cu doped C) is utilized as the back contact material in this simulation.Cu doped carbon has a metal work function of 5.0 eV and costs less than comparable modern materials.But in this case the potential diffusion or migration of copper ions or atoms from the Cudoped carbon layer into the PEDOT:PSS layer must be taken into consideration.The PEDOT:PSS layer's electrical characteristics, stability, and even the occurrence of structural flaws could all be impacted by this diffusion or migration.Preventing copper (Cu) migration to the PEDOT:PSS layer in a solar cell is crucial to maintain the stability and performance of the device.Copper (Cu) migration to the PEDOT:PSS layer in solar cells can be blocked by using a thin layer of DMSO-doped PEDOT:PSS (Huang et al 2017).The conventional PEDOT:PSS layer is far from being optimal for the best photovoltaic performance (Huang et al 2017).Highquality PEDOT:PSS films can be readily formed using the PEDOT:PSS aqueous dispersion through conventional solution-processing techniques such as spin casting, slot die coating, doctor blade, spray deposition, screen printing, inkjet printing, etc The PEDOT:PSS film is normally uniform and smooth (roughness < 5 nm) (Xia and Dai 2021).By Incorporating barrier layers such as metal oxides (e.g., TiO2, ZnO)  or organic layers between the Cu-doped Carbon layer and the PEDOT:PSS layer effectively block Cu diffusion.Moreover by combining several approaches like proper encapsulation, material selection, surface treatment, it is possible to effectively block Cu migration into the PEDOT:PSS layer, thus improving the PCE and long term stability of the cell.

Back surface field analysis
For any solar cell, the overallperformance depends on three factors, which are separation, extraction, and transportation of charge carriers.But, due to back surface recombination at the CIGS/metal interface, the efficiency is greatly reduced (Alzoubi and Moustafa 2019).This is because the CIGS absorber material makes a Schottky type contact with the metal electrode which creates a Schottky barrier of the photogenerated carriers reducing the current-voltage (J-V) significantly.From the simulated results, it has been found that high work function is expected to achieve higher efficiency.So, increasing efficiency by using the higher work function values is not a proper solution in the case of thin film solar cells from a structural point of view.Another technical disadvantage is that higher values of metal work function mean high energetic photons will be required to trigger the photoemission process.To overcome this problem, we have introduced a heavily doped, much cheaper and highly conductive poly(3,4-ethylenedioxythiophene) polystyrene sulfonate (PEDOT: PSS) material with a thickness of 0.08 μm as the back surface field (BSF) layer which creates p+, p type junction with the absorber material.Other potential materials for the back surface layer like BaSi 2 , V 2 O 5 , etc, can be used (Moon et al 2023, Rahman et al 2023).PEDOT: PSS is a versatile, electrically conductive polymer.A post-treatment with numerous substances, such as ethylene glycol, dimethyl sulfoxide (DMSO), salts, zwitterions, cosolvents, acids, alcohols, phenol, geminal diols, and amphiphilic fluoro-compounds, can dramatically increase the conductivity of PEDOT: PSS (Bengtsson, Nilsson, and Robinson 2014).The main goal of the preparation of doped PEDOT: PSS utilizing special techniques, such as going through treatment and adding secondary dopants, is to increase its electrical efficiency (Shahrim et al 2021).Other engineering techniques, such as the insertion of nanoparticles, surface modification, and in situ polymerization, can be utilized to boost the conductivity of this polymer.
The inclusion of thin back surface layer is found to be a better and more effective way to boost the performance of the proposed solar cell (Moon et al 2021).Since PEDOT: PSS will be deposited below the absorber layer, there won't be any temperature effects that could affect the cell's functionality.Current-voltage properties of two CIGS solar cell designs with and without the back surface field (BSF) layer are shown in figure 9.A remarkable 6% increase in efficiency has been observed with the inclusion of the BSF layer.This might be attributed due to the reduction of recombination at the back contact.Basically, the back surface field eases the path of photogenerated holes in its migrating towards to the back contact rather than gathering at the CIGS/metal interface which reduces the chances of unwanted recombination.So, the value of both J sc and V oc increases from 30.22 mA cm −2 to 32.37 mA cm −2 and 0.88 V to 0.995 V.In contrast to some researchers who used BaSi2 as the back surface layer and found efficiency of about 26.24% with an absorber thickness of 800 nm, we found efficiency of about 28.32% with an absorber thickness of 1500 nm and 80 nm of PEDOT: PSS layer as the back surface filed layer.These results are summarized in table 3.
The following table shows the results -So, two alternative structures can be designed to increase efficiency of CIGS based solar cell with novel Cd 1−x Zn x Se as buffer layer.The first approach may be the use of a metal electrode with high work function  values.And the second approach is the addition of a back surface layer between absorber and back contact keeping the metal work function at nominal values.

Optimized structure
From the simulation output, it has been found that the addition of Zn to a typical CdSe buffer layer to generate the ternary compound Cd 1−x Zn x Se can have a significant impact on the proposed solar cell's performance.
Figure 10 illustrates the overall optimal structure, where replacing around 20% to 30% toxic Cd with non-toxic, earth abundant Zn allows a lot of freedom in terms of tuning band gap, conduction band offset, and so on.The optimized thickness of CIGS and Cd 1−x Zn x Se layer is 1500 nm and 80 nm respectively.The intrinsic defect values are optimized at 1 × 10 14 /cm 3 for both the layers.Furthermore, using a 80 nm thick PEDOT: PSS as the back surface layer enhanced the device's efficiency from 22.50 percent to 28.32 percent, which is remarkable.Table 4 shows the optimized values of corresponding layers-

Conclusion
The present work deals with the novel Cd 1−x Zn x Se ternary alloys as a potential replacement of traditional CdS buffer layer for CIGS based solar cells and found some noteworthy outcomes.The selection of this novel buffer layer material has two aspects-one is having material with minimum amount of toxic Cd, and the other is tunability properties by varying the x composition.It has been found that, if the metal work function is below 4.6 eV cross-over takes place due to the creation of secondary junction between the p-type absorber and the metal back contact.Primarily it seems that, -higher work function values are appropriate for back contact material to get the Ohmic type behavior, but due to higher costs and the necessity of high energetic photons (for photoelectric effect) it's not a practical approach.In this research work an alternative approach was explored integrating a thin back surface (BSF) layer keeping the metal work function within nominal values.So, it's a practical finding for design engineers to compromise the two properties in order to have better efficiency.The efficiency increases from 22.5 percent without BSF layer to 28.32% with the use of BSF layer keeping the metal work function constant at 5.1 eV.Another remarkable finding is the presence of unwanted recombination other than SRH recombination inside into the structure which is revealed from the activation energy.The outcome of this research work will pave the way of utilizing the Cd 1−x Zn x Se as an alternative buffer layer for CIGS based solar cells instead of widely used, toxic CdS or CdSe.The inclusion of earth abundant Zn not only provides the bandgap tunability but also will be cost effective.

Figure 2 .
Figure 2. Schematic structure of the proposed CIGS/Cd 1−x Zn x Se solar cell.

Figure 3
Figure 3 (a) J-V characteristics, and (b) Quantum Efficiency of the solar cell with CdS and the Cd 1−x Zn x Se buffer layer.

Figure 5 .
Figure 5.Typical solar cell parameter variations as a function of Conduction Band Offset (CBO).

Figure 6 .
Figure 6.Open circuit voltage (V oc ) as a function of temperature (K) extrapolated to 0 K to find the activation energy (Ea).

Figure 7 .
Figure 7. Efficiency variation of the proposed solar cell structure as a function of metal work function.

Figure 8 .
Figure 8. Cross over effect on I-V characteristics as a function of metal work function.

Figure 9 .
Figure 9. Current-Voltage curve with and without the back surface field (BSF) layer.

Figure 10 .
Figure 10.Optimized structure of the proposed cell.

Table 1 .
Material parameters used in this experiment.

Table 3 .
Simulation results of the solar cell structure with and without the Back Surface (BSF) layer.