Open Access
Add to BibSonomy
Abstract
Copper indium telluride (CuInTe2)-based n-CdS/p-CuInTe2/p + -MoS2 double-heterostructure solar cell has been investigated numerically by solar cell capacitance simulator (SCAPS-1D). Initially, an adjusted condition among the most influencing parameters e.g. thickness, carrier doping level, and bulk defects of active materials such as CdS window, CuInTe2 absorber, and p + -MoS2 back surface field (BSF) layers has been obtained by a systematic computation. The proposed solar cell exhibits an improved power conversion efficiency (PCE) of 34.32% with VOC =0.927 V, JSC = 42.50 mA/cm2, and FF = 87.14% under the optimized condition. The PCE can be further enhanced to 38.87% introducing sub-bandgap absorption in the MoS2 (300 nm) BSF with Urbach energy, E0 of 0.4 eV. These detailed simulation results reveal a huge potential of CuInTe2 absorber with MoS2 BSF layer for the manufacture of a cost-effective, high-efficiency double-heterojunction thin film solar cell.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Solar energy is the biggest ample and purest renewable energy source to satisfy the present and future huge global energy crisis. Solar cells are the most promising for directly converting sunlight into electricity, among other potential methods. Thus, advancement in photovoltaic technology has raised exponentially in recent decades. Researchers have concentrated and conducted low-cost, earth-abundant, non-toxic, and high-efficiency photovoltaic devices [1,2]. Thin-film Cu(In,Ga)(Se,S)2 (CIGS(e)) absorber-based solar cells have attained the PCE up to 23.4%, revealing the huge future prospect of chalcopyrite materials in photovoltaic applications [3]. The development of solar devices with multijunction architectures is a probable mean to achieve higher efficiency. Ternary chalcogenide compound semiconductors from group I–III2n + 1–VI3n + 2 (n = 0, 1, and 2) have been studied substantially as absorber materials for designing and fabrication of high efficiency thin film solar devices because of their high optical absorption, tuneable direct bandgap and well physical and chemical stability [4]. The Cu(InGa)Se2 (CIGS) belongs to the I–III–VI2 chalcopyrite ternary compounds, which is widely studied for photovoltaic applications. The highest PCE of CIGS solar cell of 22.2% (on polyimide plastic substrate) and 23.35% has been reported recently by Federal Laboratories for Materials Science and Technology (EMPA) and Solar Frontier, National Institute of Advanced Industrial Science and Technology, respectively [5].
CuInTe2 (CIT) is one of the promising semiconducting compound materials from I–III–VI2 group having the direct bandgap of 0.91–1.10 eV at room temperature (∼300 K) [6,7]. CuInTe2 shows a large absorption coefficient (∼105 cm−1), and enormous non-linear susceptibility in the infrared (IR) spectra with a favourable radiation stability with an exceptional defect tolerance. These distinct properties make it suitable for use in photodetectors [8], photovoltaic cells [6], light emitting diodes [9], and thermoelectrics [10] etc. The low bandgap of CuInTe2 is also suitable for use this material in multi-junction solar cells. In addition, CuInTe2 can readily be synthesised by a number of techniques such as thermal evaporation [11], molecular beam epitaxy (MBE) [6], solvothermal [12], and electrodeposition [13] etc. Therefore, CuInTe2 has huge potential to be applied in photovoltaic devices [4,6,14,15]. Although several studies on CuInTe2 and CuInS2 have been reported, an in-depth study on tellurides compound; particularly CuInTe2 has scarcely performed. To date, the experimental highest PCE of 5.1% of a single-junction CuInTe2 solar cell has been recorded under AM1.5 illumination (100 mW/cm2) [6], although Shockley–Queisser (SQ) detailed balance limit of the compound is ∼31.7% [16]. Therefore, the full potential of CuInTe2 to be used in solar cell has been unexplored yet.
In addition, CdS is a popular n-type semiconducting material that has been studied extensively over the world owing to its tuneable wide optical bandgap, higher carrier density and high transmission with an outstanding stability under constant light exposure [17–19]. Furthermore, the importance of the BSF layer for enhancing the photovoltaic performance has been elucidated in earlier works [20,21]. However, in photovoltaic cells without BSF layer, the short wavelength photons mostly absorbed at absorber layer passing through window layer with a significant part of the longer wavelength portion retains unused. A suitable selection of the BSF layer can offer low energy (sub-bandgap) photon absorptions via the two-step upconversion through Urbach tail-states namely TSA (tail-states-assisted) upconversion [22,23]. Thus, an improved absorption of incident photons originates an enhanced photovoltaic performance. Molybdenum disulphide, MoS2 is a transition metal dichalcogenide semiconductor with a direct tuneable bandgap (1.9 eV for single layer to 1.2 eV for bulk films), a higher absorption coefficient of 106 cm-1 and larger diffusion length of 1 µm [24–26].
In this study, a CuInTe2-based thin film solar cell (TFSC) with p + -MoS2 back surface field (BSF) layer has been investigated in details. The proposed n-CdS/p-CuInTe2/p + -MoS2 double-heterostructure has been numerically computed employing solar cell capacitance simulation (SCAPS-1D) simulator. Simulation results reveal a huge potential of CdS/CuInTe2/MoS2 double-heterostructure for the manufacture of high efficiency TFSC.
2. Methodology and simulation parameters
Figure 1(a) and (b) show the architecture and illumined energy diagram of Al/n-CdS/p-CuInTe2/p + -MoS2/Ni heterostructure configuration. The n-type CdS and p + -MoS2 layer with bandgaps of 2.4 and 1.62 eV were utilized as the window and BSF layers, respectively. Aluminum (Al) with work function (WF) of 4.2 eV and nickel (Ni) with WF of 5.35 eV were chosen as front and metal back contact layers, respectively, considering band alignment for the efficient carrier transportation from active layers to metal electrodes. The quantitative electronic parameter’s values of CIT, MoS2 and metal contact interfaces such as back contact barrier height respect to Ec (ΦBn~ 1.62 eV) and Ev (ΦBp ~ 0.0 eV) of MoS2, conduction band offset, CBO between CIT and MoS2 (ΔEc~ 0.48 eV) and valence band offset VBO (ΔEv~ 0.04 eV) at thermal equilibrium condition were determined from the following literature to ensure the development of a favorable band alignment with ohmic contact [27].
Fig. 1. (a) Proposed Al/n-CdS/p-CuInTe2/p + -MoS2/Ni heterojunction solar PV cell and (b) corresponding energy diagram of the device.
A computer-based window application of SCAPS-1D simulator was used to investigate the impacts of each active layer properties like thickness, doping, and defect density with an Al/n-CdS/p-CuInTe2/p + -MoS2/Ni heterostructure configuration. A detailed information on potential of SCAPS-1D for I–III2n + 1–VI3n + 2 material-based solar cell simulation was described elsewhere [28].
SCAPS–1D was embarked for solving equation of Poisson related to electrons and holes for obtaining solar cell characteristics; J-V, C-V, C-f and quantum efficiency (QE) [29–31].
The solar PV device was illumined under solar spectrum of AM 1.5 G having power density of 100 mW/cm2. The electron and hole thermal velocity of 1.0 × 107 cm/s was considered in the calculation. The shunt and series resistances were considered with equitable values of 1000 Ω.cm2 and 2.5 Ω.cm2, respectively, considering practical existence of unavoidable carrier recombination or leakage in solar cells to perceive the actual device. In addition, 300 K was presumed the working device temperature. The electronic parameters of each active material are summarized in Table 1. The optical model from SCAPS was used for each layer for optical data. The interface defects also play vital part in the efficiency of solar cells. In the present simulation, adequate amount of defects of 1010 and 1010 cm-2 were taken as the CdS/CuInTe2 and CuInTe2/MoS2 interface defects, respectively.
Table 1. Simulation parameters utilized for different active layers in CuInTe2 solar cell.
3. Results and discussion
3.1 CuInTe2 absorber impression on PV performance
Figure 2(a) displays the influence of thickness of CuInTe2 absorber on the PV outcomes of the Al/n-CdS/p-CuInTe2/p + -MoS2/Ni thin film solar cells. The values of thickness, carrier and bulk defects densities were set to 800 nm, 1017 cm−3, and 1014 cm−3, respectively. The JSC increases linearly from 32.05 to 42.0 mA/cm2 with the expanding of thickness of CuInTe2 absorber from 200 to 700 nm and it reaches and retains at constant value of 42.6 mA/cm2 at absorber thickness beyond 700 nm. The FF also increases slightly from 86.9 to 87.3 while VOC decreases sub-linearly from 0.96 to 0.90 V with increasing the breadth of the absorber from 200 to 1400 nm. Thus, the resultant PCE increases from 28.0 to 34.4%. For a thick p-CIT absorber (above 1000 nm), PV parameters show a trend to be almost constant. This behavior results from the enhanced absorption of the incident photons at wide area of p-CIT absorber layer and thereby, a markedly improved electron-hole generation, resulting in the higher JSC. The back metal contact reaches close to the depletion region at a thinner absorber layer, which causes insufficient effective area for the efficient absorption of incident photon [37]. This similar characteristic also observed in CIGS-based several previous studies [38–40]. Herein, the thickness of absorber layer of 800 nm was chosen considering adjusted conditions among cell parameters.
Fig. 2. Variation of performance of n-CdS/p-CuInTe2/p + -MoS2 PV cell at a different CIT layer (a) thickness (b) acceptor concentration, NA and (c) defect density, Nt.
Figure 2(b) exhibits the consequence of carrier of CuInTe2 absorber on the device performance for a varied acceptor concentration, NA of 1014–1017 cm-3 at a constant width of 800 nm. A marked enhancement in VOC from 0.89 to 1.05 V and FF from 80.0 to 86.0% with slight improved in JSC is observed and consequently the efficiency rises from 29.0 to 33.5% when the acceptor density raises from 1014 to 1016 cm−3. The reverse saturation current is suppressed with the rise in NA owing to the rise in built-in potential, thereby increases the VOC. Beyond the absorber carrier density, NA of 1017 cm-3, the FF reaches to a constant value of 87.5% while the VOC showing trend to further increase. At higher carrier concentrations (>1017 cm−3), the JSC would be reduced due to the smaller width of the depletion region and parasitic absorption due to free carrier. A higher value of acceptor carrier concentration also originates higher carrier recombination in the bulk region, which is also observed in previous report [41]. Thus, the device with acceptor density, NA of 1017 cm−3 with a layer thickness of 800 nm is found as an optimum value with the maximum PCE of ~35% keeping a trade-off among solar parameters.
Figure 2(c) depicts the impression of bulk defects, Nt of CuInTe2 absorber material on the device performance for a varied Nt of 1011–1018 cm-3 at a constant thickness of 800 nm and acceptor density NA of 1017 cm-3. The VOC reduces drastically from 1.04 to 0.55 V, while the FF from 81.0 to 70.0%, and consequently PCE decreases from 39.4 to 19.6% when the Nt increases from 1011 to 1018 cm-3. As observed in the figure, the JSC retains a constant value of 42.0 mA/cm2 for an Nt of 1015–1016 cm-3. However, FF is lower at the defects <1014 cm-3 and it increases to 87.14% up to the defect density of 1014 cm-3. This may happened due to the increase in diode ideality factor, n. The FF, VOC and n are related as $\textrm{FF} = \frac{{{v_{oc - \ln ({v_{oc}} - 0.72)}}}}{{{v_{oc}}}}$, with ${v_{oc}} = \frac{{{V_{oc}}}}{{n{K_B}T/q}}$, where, KB is the Boltzmann constant, T is the absolute temperature, q is the electronic charge [42]. The ideality factor is an indicator of the recombination mechanisms ruling within the diode. If the recombination takes places in the neutral regions of the diode, then n = 1 whereas if it dominantly occurs in the space charge region due to defects, n = 2. For diodes whose recombination at the hetero-face is dominant n can be higher than 2 [42–44]. The ideality factor affects the fill factor of the solar cell and it is that when n increases the fill factor decreases. It is observed in this simulation that recombination in the space charge region dominates at a defect of 1 × 1011 cm3, which is reduced in the case of higher defects (not shown in the figure). Therefore, it can be concluded that at lower defect the diode ideality factor is higher which results lower FF. When the defects further increase above 1 × 1014 cm-3, the FF again decreases due to increase in ideality factor. Both JSC and FF start to decrease linearly beyond the Nt value of 1016 cm-3. The highest efficiency of ~40.0% with JSC of ~42.0 mA/cm2, VOC of ~1.00 V and FF of ~85.0% is obtained at a minimum Nt of 1012–1013 cm-3. However, considering the unavoidable bulk defect in an absorber material, the defect value of 1014 cm-3 was selected for further analysis to obtain a realistic outcome. The PCE of 34.32% is gained with JSC of 42.50 mA/cm2, VOC of 0.927 V, and FF of 87.14% at Nt of 1014 cm-3. At a higher value of defects, the recombination due to Shockley-Read-Hall (SRH) (${\mathrm{\Re }_{SRH}}$) is prevailing, thereby a drastic reduction in performance parameters is observed. The impact of defect density may be demonstrated by following Eq. (8)-10 [45,46].
Carrier life time:
Diffusion length:
At higher Nt value (over 1016 cm-3), the SRH recombination is predominant, thereby a drastic reduction in performance parameters is observed in the CIT solar cell as delineated in the figure.
3.2 CuInTe2 absorber layer thickness effect on quantum efficiency
Fig. 3. (a) The quantum efficiency (QE) and corresponding (b) J-V characteristics at a different layer thickness of CIT absorber with n-CdS/p-CuInTe2/p + -MoS2 heterostructure and (c) QE at a different layer thicknesses and carrier concentration of CIT absorber at a particular wavelength of 1100 nm.
Figure 3(a) and (b) show the quantum efficiency (QE) and corresponding J-V characteristics of n-CdS/p-CuInTe2/p + -MoS2 dual-heterostructure, respectively at a different CIT thickness of 200–1100 nm. The QE increases systematically at wavelength range of 400–1120 nm with increasing CIT absorber layer thickness from 200 to 1100 nm thereby results in noticeable improvement in photocurrent JSC, consequently an enhanced PCE of over 34.0% is obtained (Fig. 3(b)). A significant improvement of JSC is observed at 800-1100 nm region of solar spectrum particularly. The JSC escalates from 33.5 to 43.2 mA/cm2 with a slight decrease in VOC from 0.95 to 0.92 V corresponding to a rise in QE from 55 to 97% at a wavelength of 800–1100 nm region. However, it is also visualized from the figure that the value of QE slightly exceeds 100% with a highest value of 100.13% in CIT solar cell in the wavelength range of 500-700 nm. In SCAPS simulation, the QE is computed by comparing the currents at the working point conditions and that generated with adding an extra number of monochromatic photons. The QE may become negative or larger than 100% leading to an error if the difference between the currents is smaller compared to them.
Figure 3(c) shows the increment of QE for a particular wavelength of 1100 nm at different dopings and thicknesses of CIT layer. The QE systematically increases from 30 to a 76% with increasing the CIT breadth from 200 to 1000 nm for a certain carrier density, NA between 1014–1017 cm-3. On the other hand, the QE changes insignificantly with increasing the NA at a specific CIT layer thickness which indicates that CIT absorber with an adjusted layer width of 700-900 nm and specific charge carriers of 1015–1018 cm-3 can absorb sufficient incident photons to convert in electrical energy directly. However, the improvement of photon absorption in n-CdS/p-CuInTe2/p + -MoS2 dual-heterostructure can be described by the decrease of dark current due to the reduction in surface recombination velocity and enhanced collection photocarriers by BSF layer [36].
3.3 Impact of CdS window on performance
Figure 4(a) shows the undulation of solar cell output parameters of JSC, VOC, FF and PCE of CIT PV cells with CdS window width in the range of 50–350 nm taking other parameters identical as summarized in Table 1. The value of JSC decreases from 42.55 to 42.12 mA/cm2 and thereby the PCE downfalls from 34.4 to 34.0% with an almost constant VOC and FF with rising the CdS thickness of the device from 50 to 350 nm. This trend of decrement in photocurrent might be owing to an increment of carrier recombination at a longer diffusion length, and higher doping concentration (1018 cm−3) compared with the CIT (1017 cm−3) at a larger CdS layer thickness. Since the thickness of the CdS film improves, additional photons get absorbed outside of the space charge width, and the minority charge current enhances, consequently, the charge recombination rate increases, and therefore the efficiency drops [47]. In addition, this reduction in efficiency might be illustrated by scattering between the absorber and window layer [48]. The outcomes are similar with earlier works [49–51]. Thickness of CdS window of ∼100 nm is needed to limit photon absorption losses since a thick buffer layer reduces the solar cell efficiency. Additionally, it is also found challenging to synthesis high-quality thin CdS films in practice, therefore, 100 nm-CdS window layer was used in this study to achieve optimized condition.
Fig. 4. Variation of PV performance with CdS window layer (a) thickness, (b) doping and (c) bulk defects of n-CdS/p-CuInTe2/p + -MoS2 Solar cell.
Figure 4(b) depicts the variation of solar device parameters of JSC, VOC, FF and PCE for a varied donor density, ND of 1015 to 1021 cm-3 at a constant 100 nm thickness of CdS layer. The PCE retains at almost constant of 34.35% up to the ND of 1018 and further increase in ND gives rise to a noticeable reduction in JSC and consequently in the PCE. The JSC declines from 42.57 to 39.94 mA/cm2, the FF from 87.07 to 86.60%, therefore, the PCE decreases from 34.35 to 31.95% with an insignificant change in VOC from 0.927 to 0.924 V when ND increments from 1018 to 1021 cm-3. When the donor density, ND of CdS rises from 1015 to 1018 cm-3, the n-type quasi-Fermi level close to the CdS/CIT junction moves toward the band of conduction, in contrary, the p -type quasi-Fermi level shifts away from the valence band and as a consequence the bending of valence band is increased [52]. As a consequence, the density of holes at the CdS/CIT junction reduces, leading to the insignificant interfacial recombination. However, severe recombination of photogenerated carriers into CdS, CIT and CdS/CIT interface at a ND of ≥1018 cm-3 causes marked decrease in photocurrent and consequently the cell’s efficiency. Thus, the CdS donor density of 1018 cm-3 has been chosen as optimum value of ND with suppressed carrier recombination.
Figure 4(c) demonstrates the variation of cell parameters of JSC, VOC, FF and PCE at a varied bulk defect density, Nt of 1011–1018 cm-3 at a constant 100 nm layer thickness and donor density, ND of 1018 cm-3 of CdS window. The defects or trap states in the material are responsible for the non-radiative recombination that reduces the charge carries life time (τ), which causes faster recombination reducing the diffusion length (L) and makes hurdle in the transportation of photogenerated charge carriers as seen in Eq. (8),9. A favorable absorber layer thickness is corresponded a unique specific value of Nt. The cell parameters of JSC, VOC, FF decreases noticeably when defect density, Nt increases from 1011–1018 cm-3 and consequently the PCE decreases from 34.4 to 32.7%. A proper adjustment (Nt of 1014 cm-3 for CdS thickness of 100 nm) is required for obtaining high efficiency CIT based solar cells.
3.4 MoS2 BSF layer impression on PV cell
Figure 5(a) illustrates the PV performance of CuInTe2-based PV device with and without p + -MoS2 back surface layer. The VOC increments markedly from 0.701 to 0.907 V and JSC from 42.6 to 44.05 mA/cm2 with the insertion of a 200 nm-thick p + -MoS2 BSF layer for an acceptor density NA of 1019 cm-3 between CIT absorber and back metal contact of Ni in pristine n-CdS/p-CuInTe2 heterostructure. The improvement in absorption of higher wavelength photons in the range of 600–1120 nm originates an additional photocurrent as can be visualized in QE characteristics depicted in Fig. 5(b). The electric field produced at the interface of CIT/ MoS2 as p-p+ heterojunction acts as an important barrier for flowing of electrons to the back surface and this reflection of electrons guides the suppression of saturation current, resulting an improvement in both VOC and JSC simultaneously and thus an improved PCE of 34.05%, which was 24.21% without BSF layer.
Fig. 5. (a) J-V and (b) QE characteristic curves of CuInTe2-based double-heterojunction solar device without and with MoS2 BSF layer.
However, now we investigate the impact of fluctuation in different parameters of MoS2 BSF layer on the PV outcome of the n-CdS/p-CuInTe2/p + -MoS2 device. Figure 6(a) shows the performance of the cell for a varied MoS2 layer thickness in the range of 50–350 nm at a constant NA of 1019 cm-3 and Nt of 1014 cm-3. A slight change in PV parameters are observed for both upper and lower values of tentative selected thickness of 200 nm. The MoS2 is a chalcogenide material showing a tunable bandgap of 1.2–2.0 eV with a higher absorption coefficient of 106 cm−1 in the UV-visible range and free carrier density of 1 × 1019 cm-3, therefore, an extreme advancement in electrical and optical properties appears by insertion of BSF layer between CIT and Ni. However, a small change appears when the BSF layer thickness is changed in the range of 50–350 nm [53–55]. The device efficiency of 34.32% with JSC of 42.50 mA/cm2, VOC of 0.927 V and FF 87.14% is obtained at an optimized thickness value of 200 nm with NA of 1019 cm-3, which is used for further investigation.
Fig. 6. The undulation in device performance with (a) thickness, (b) doping and (c) defect density of MoS2 back surface layer of n-CdS/p-CuInTe2/p + -MoS2 solar cell.
Figure 6(b) delineates the performances of the designed n-CdS/p-CuInTe2/p + -MoS2 solar cell with the undulation of MoS2 acceptor density, NA from 1016 to 1022 cm−3. The JSC is almost constant at a value of 42.50 mA/cm2 in the entire range of investigation. Further, the VOC and FF and therefore, PCE are also almost unchanged in the entire range of investigation as delimitated in the figure.
Figure 6(c) demonstrates the influence of defect density of MoS2 BSF layer on performance of CIT thin film solar cell. Herein, neutral/donor defects were considered and varied from 1010 to 1018 cm−3 keeping of all rest parameters unchanged. The VOC shows nearly constant characteristics up to an Nt of 1015 cm−3 and it begins to decline sub-learnedly with higher defects level that make hurdle to generate significant electron-hole pairs (EHPs) and transport to the metal contact. The VOC falls from 1.04 to 0.53 V with defect density increasing from 1012 to 1018 cm−3. The JSC is almost constant in the defect range of 1010-1016 cm-3 whereas FF is initially lower and attains a higher value of 87.14% at a defect of 1014 cm-3 and then starts to decrease again at higher defects. The change in FF is related to the increase diode ideality factor at low and very high defects [42–44]. Thus, the use of MoS2 as BSF layer with CuInTe2 absorber layer under optimized values of electrical and optical parameters including the reasonable defects level and cell operating temperature offers a high PCE of over 34% for the constructed n-CdS/p-CuInTe2/p + -MoS2 double-heterojunction TFSC.
3.4.1 Effect of Sub-bandgap absorption in MoS2 on performance
The sub-bandgap photons may get absorbed by the Urbach energy states and these lower energy sub-bandgap photons can participate in two-step photon upconversion whereby the absorption of two photons in series produces one electron-hole pair and thereby make photocurrent [56,57]. This Tail-States-Assisted (TSA) upconversion can take place in a material when it has adequate doping, preferred bandgap and lofty absorption coefficient in the sub-bandgap energy spectra [22,23,36]. In SCAPS simulator, the sub-band gap absorption is defined as ${\alpha _{sub - band\; gap}}({h\upsilon } )\approx \textrm{exp}\left( { - \frac{{{E_{g - h\upsilon }}}}{{{E_0}}}} \right)$, $h\upsilon < {E_g}$, where ${E_0}$ is the Urbach tail energy [58]. The up-conversion and degree of enhancement of the cell current will depend on Urbach energy, E0. The higher Urbach energy can highly enhance the quantum efficiency (QE) in the longer wavelength. Moreover, several reports on sub-bandgap absorption owing to tail-states are already found for organic solar cells [59,60].
MoS2 has an absorption coefficient, α of ∼1.77 × 105 cm− 1 at the wavelength of 1800nm (found from data extrapolated in SCAPS) deposited by hydrothermal method [61] that corresponds to an E0 of 0.55 eV with initial α of 106 cm-1 in the visible range although the Urbach energy will be different for different film deposition methods. Therefore, MoS2 BSF layer will significantly enhance QE and hence JSC when it will be deposited with high Urbach energy.
Figure 7 depicts the change in JSC and PCE with respect to Urbach energy, E0 and thickness of MoS2 BSF layer keeping other parameters unchanged obtained by aforementioned systematical studies. In Fig. 7(a), the JSC is enhanced systematically with an increase of Urbach energy from E0 from 0.1 to 0.5 eV for a certain value of MoS2 layer thickness from 0 to 0.40 µm. The variations in VOC and FF were not so significant and therefore not shown here. In general, when photons with energy of Eph ≥ Eg (bandgap) enter into the semiconductor layer (i.e. MoS2), they will be absorbed immediately and create e-h pairs and contributes to promote the photocurrent of the solar cell. And also any two photons with energy (Eph1 + Eph2) ≥ Eg will produce JSC by TSA process. As a results, an improvement of PCE from 34.32 to 39.4% is observed when Urbach energy E0 of MoS2 increased from 0.1 to 0.5 eV for a layer thickness from 0 to 0.40 µm as shown in Fig. 7(b). It is noted that, the Urbach energy characteristically rely on the level of free carrier and crystallinity of the photoactive materials [22,62].
Fig. 7. The fluctuation in (a) JSC and (b) PCE in terms of the Urbach energy and thickness of MoS2 layer.
3.5 Temperature dependent PV performance
The operating temperature significantly affects the cell performance. The temperature raises the atomic vibration in the active layer of the device as well the bandgap of the active layer decreases. Therefore, the performance of the solar device decreases.
The calculation of the efficiency of n-CdS/p-CuInTe2/p + -MoS2 solar cell was performed at temperature of 300 K. However, the device performance was also computed varying the operating temperature of the solar device in the range of 250 to 350 K and is depicted in Fig. 8. The PCE of the device is 36.74% at 250 K that decreases to 31.81% when the temperature rises to 350 K. As shown in the figure, the VOC and FF of the device reduce with temperature that eventually decrease the device performance.
The temperature coefficient of the PCE, TC can be calculated using the following equation [63]:
Where, ${\eta _{STC}}$ is the PCE of the device at standard test conditions (STC) at 298 K, η is the PCE of the device at temperature, T K. The TC of the CuInTe2-based heterojunction is calculated to be -0.0493% K-1 indicating the stability of the cell against temperature.3.6 Overall cell performance
The PV parameters of p-CuInTe2 solar device without and with p + -MoS2 BSF layer shown in Table 2. The PCE of 24.21% obtained in a single-junction n-CdS/p-CuInTe2 solar cell while the efficiency of 34.32% having VOC of 0.927 V, JSC of 42.50 mA/cm2, and FF of 87.14% achieved with p + -MoS2 BSF layer for n-CdS/p-CuInTe2/p + -MoS2 double-heterostructure. The PCE can be further enhance close to ∼42% for a double-heterojunction thin film solar cell (DHJSC) introducing TSA upconversion by Urbach states in the MoS2 BSF layer which is congruent with the detailed-balance limit revealed by Shockley-Queisser (SQ).
Table 2. The PV parameters of CuInTe2 solar cell without and with p + -MoS2 BSF layer.
4. Conclusions
The potential of CuInTe2-based solar cell with Al/n-CdS/p-CuInTe2/p + -MoS2/Ni double-heterostructure design has been unveiled systematically by SCAPS-1D simulator. The most influencing parameters e.g. thickness, doping level, and bulk defect density of active layers of CdS window, CuInTe2absorber, and MoS2 BSF have been optimized. The layer thickness of 100, 800 and 200 nm, and carrier concentration of 1.0 × 1018, 1.0 × 1017, and 1.0 × 1019 cm-3 have been recorded as the optimized values for CdS, CuInTe2, MoS2 layers, respectively. The PCE of 24.21% has been achieved for the pristine i.e. n-CdS/p-CuInTe2 single-junction solar cell. The PCE is enhanced to 34.32% including a VOC of 0.927 V JSC of 42.50 mA/cm2, and FF of 87.14% with the use of MoS2 BSF layer. The efficiency the solar device can be further improved approaching to Shockley-Queisser (SQ) detailed balance limit of double-heterojunction solar cell of ∼42% introducing TSA upconversion through Urbach energy in MoS2 BSF layer. This detailed study reveals a huge potential of CuInTe2-based double-heterojunction solar cells and provides a guideline for fabricating cost-effective, highly efficient thin film solar cells.
Acknowledgements
The authors highly acknowledge Dr. Marc Burgelman, University of Gent, Belgium, for providing SCAPS simulation software.
Disclosures
The authors declare no competing financial interest.
Data availability
Simulation details and associated data are available free of charge from authors upon reasonable request.
References
1. C. S. Durganjali, S. Bethanabhotla, S. Kasina, and S. Radhika, “Recent developments and future advancements in solar panels technology,” J. Phys.: Conf. Series 1495, 012018 (2020). [CrossRef]
2. R. M. France, J. F. Geisz, T. Song, W. Olavarria, M. Young, A. Kibbler, and M. A. Steiner, “Triple-junction solar cells with 39.5% terrestrial and 34.2% space efficiency enabled by thick quantum well superlattices,” Joule 6(5), 1121–1135 (2022). [CrossRef]
3. J. K. Larsen, O. Donzel-Gargand, K. V. Sopiha, J. Keller, K. Lindgren, C. Platzer-Björkman, and M. Edoff, “Investigation of AgGaSe2 as a Wide Gap Solar Cell Absorber,” ACS Appl. Energy Mater. 4(2), 1805–1814 (2021). [CrossRef]
4. M. Lakhe and N. B. Chaure, “Characterization of electrochemically deposited CuInTe2 thin films for solar cell applications,” Sol. Energy Mater. Sol. Cells 123, 122–129 (2014). [CrossRef]
6. T. Mise and T. Nakada, “Low temperature growth and properties of Cu–In–Te based thin films for narrow bandgap solar cells,” Thin Solid Films 518(19), 5604–5609 (2010). [CrossRef]
7. K. Liu, M. Ji, H. Li, and C. Xu, “Calculation of energy band and optical properties of CuInTe2 with doping,” Integr. Ferroelectr. 197(1), 33–42 (2019). [CrossRef]
8. S. Wagner, J.L. Shay, and P. Migliorato, “CuInSe2/CdS heterojunction photovoltaic detectors,” Appl. Phys. Lett. 25(8), 434–435 (1974). [CrossRef]
9. P. W. Yu, Y. S. Park, S. P. Faile, and J. E. Ehret, “Electroluminescence and photovoltaic detection in Cd-implanted CuInSe2 p-n junction diodes,” Appl. Phys. Lett. 26(12), 717–719 (1975). [CrossRef]
10. R. Liu, L. Xi, H. Liu, X. Shi, W. Zhang, and L. Chen, “Ternary compound CuInTe2: a promising thermoelectric material with diamond-like structure,” Chem. Commun. 48(32), 3818 (2012). [CrossRef]
11. A. M. Abo El Soud, T. A. Hendia, L. I. Soliman, H. A. Zayed, and M. A. Kenawy, “Effect of heat treatment on the structural and optical properties of CuInTe2 thin films,” J. Mater. Sci. 28, 1182–1188 (1993). [CrossRef]
12. C. Kim, H. K. Dong, Y. S. Son, H. Kim, J. Y. Bae, and Y. S. Han, “Solvothermal synthesis and characterization of a CuInTe2 absorber for thin-film photovoltaics,” Mater. Res. Bull. 47(12), 4054–4058 (2012). [CrossRef]
13. M. Lakhe, S. K. Mahapatra, and N. B. Chaure, “Development of CuInTe2 thin film solar cells by electrochemical route with low temperature (80 °C) heat treatment procedure,” Mater. Sci. Eng. B 204, 20–26 (2016). [CrossRef]
14. S. Kim, M. Kang, S. Kim, J.-H. Heo, J. H. Noh, S. H. Im, S. I. Seok, and S.-W. Kim, “Fabrication of CuInTe2 and CuInTe2−xSex Ternary Gradient Quantum Dots and Their Application to Solar Cells,” ACS Nano 7(6), 4756–4763 (2013). [CrossRef]
15. N. Ntholeng, B. Mojela, S. Gqoba, M. Airo, S. Govindraju, M. J. Moloto, J. Van Wyk, and N. Moloto, “Colloidal synthesis of pure CuInTe2 crystallites based on the HSAB theory,” New J. Chem. 40(12), 10259–10266 (2016). [CrossRef]
16. M. Bercx, N Sarmadian, R. Saniz, B. Partoens, and D. Lamoen, “First-principles analysis of the spectroscopic limited maximum efficiency of photovoltaic absorber layers for CuAu-like chalcogenides and silicon,” Phys. Chem. Chem. Phys. 18(30), 20542–20549 (2016). [CrossRef]
17. T. Sinha, L. Verma, and A. Khare, “Variations in photovoltaic parameters of CdTe/CdS thin film solar cells by changing the substrate for the deposition of CdS window layer,” Appl. Phys. A 126(11), 867 (2020). [CrossRef]
18. T. Sinha, D. Lilhare, and A. Khare, “Temperature-dependent structural, morphological and optical properties of chemical bath-deposited CdS films,” Rare Met. 40(3), 701–707 (2021). [CrossRef]
19. J.-F. Han, G.-H. Fu, V. Krishnakumar, H.-J. Schimper, C. Liao, W. Jaegermann, and M. P. Besland, “Studies of CdS/CdTe interface: Comparison of CdS films deposited by close space sublimation and chemical bath deposition techniques,” Thin Solid Films 582, 290–294 (2015). [CrossRef]
20. J. Hossain, “Design and simulation of double-heterojunction solar cells based on Si and GaAs wafers,” J. Phys. Commun. 5(8), 085008 (2021). [CrossRef]
21. B. K. Mondal, S. K. Mostaque, and J. Hossain, “Theoretical insights into a high-efficiency Sb2Se3-based dual-heterojunction solar cell,” Heliyon 8(3), e09120 (2022). [CrossRef]
22. S. K. Mostaque, B. K. Mondal, and J. Hossain, “Simulation approach to reach the SQ limit in CIGS-based dual-heterojunction solar cell,” Optik 249, 168278 (2022). [CrossRef]
23. J. Hossain, B. K. Mondal, and S. K. Mostaque, “Computational investigation on the photovoltaic performance of an efficient GeSe-based dual-heterojunction thin film solar cell,” Semicond. Sci. Technol. 37(1), 015008 (2022). [CrossRef]
24. A. Kuddus, A. Rajib, K. Yokoyama, T. Shida, K. Ueno, and H. Shirai, “Mist chemical vapor deposition of crystalline MoS2 atomic layer films using sequential mist supply mode and its application in field-effect transistors,” Nanotechnology 33(4), 045601 (2022). [CrossRef]
25. H.-L. Liu, T. Yang, J.-H. Chen, H.-W. Chen, H. Guo, R. Saito, M.-Y. Li, and L.-J. Li, “Temperature-dependent optical constants of monolayer MoS2, MoSe2, WS2, and WSe2: spectroscopic ellipsometry and first-principles calculations,” Sci. Rep. 10(1), 15282 (2020). [CrossRef]
26. A. Kuddus, K. Yokoyama, and H. Shirai, “Direct synthesis of submillimeter-sized few-layer WS2 and WS0.3Se1.7 by mist chemical vapor deposition and its application to complementary MOS inverter,” Semicond. Sci. Technol. 37(9), 095020 (2022). [CrossRef]
27. H. Luo, Y. Zhang, and H. Li, “Effect of MoS2 interlayer on performances of copper-barium-tin-sulfur thin film solar cells via theoretical simulation,” Sol. Energy 223, 384–397 (2021). [CrossRef]
28. M. A. Ashraf and I. Alam, “Numerical simulation of CIGS, CISSe and CZTS-based solar cells with In2S3 as buffer layer and Au as back contact using SCAPS 1D,” Eng. Res. Express 2(3), 035015 (2020). [CrossRef]
29. S. Abbas, S. Bajgai, S. Chowdhury, A. S. Najm, M. S. Jamal, K. Techato, S. Channumsin, S. Sreesawet, M. Channumsin, A. Laref, K. S. Rahman, and A. M. Holi, “Numerical Simulation of the Performance of Sb2Se3 Solar Cell via Optimizing the Optoelectronic Properties Based SCAPS-1D,” Materials 15(18), 6272 (2022). [CrossRef]
30. M. F. Rahman, M. M. A. Moon, M. K. Hossain, M. H. Ali, M. D. Haque, A. Kuddus, J. Hossain, and A. B. M. Ismail, “Concurrent investigation of antimony chalcogenide (Sb2Se3 and Sb2S3)-based solar cells with a potential WS2 electron transport layer,” Heliyon 8(12), e12034 (2022). [CrossRef]
31. J. Hossain, B. K. Mondal, and S. K. Mostaque, “Design of a highly efficient FeS2-based dualheterojunction thin film solar cell,” Int. J. Green Energy 19(14), 1531–1542 (2022). [CrossRef]
32. M. A. Green, “Intrinsic concentration, effective densities of states, and effective mass in silicon,” J. Appl. Phys. 67(6), 2944–2954 (1990). [CrossRef]
33. J. M. Adamczyk, L. C. Gomes, J. Qu, G. A. Rome, S. M. Baumann, E. Ertekin, and E. S. Toberer, “Native Defect Engineering in CuInTe2,” Chem. Mater. 33, 359–369 (2021). [CrossRef]
34. S. R. Islam Biplab, M. H. Ali, M. M. A. Moon, M. F. Pervez, M. F. Rahman, and J. Hossain, “Performance enhancement of CIGS-based solar cells by incorporating an ultrathin BaSi2 BSF layer,” J. Computational Electronics 19, 342–352 (2020). [CrossRef]
35. J. Hwang, M. Lee, B.-K. Yu, M.-K. Han, W. Kim, J. Kim, R. A. R. A. Orabi, H. Wang, S. Acharya, J. Kim, Y. Jin, H. Park, S. Kim, S.-H. Yang, and S.-J. Kim, “Enhancement of thermoelectric performance in a non-toxic CuInTe2/SnTe coated grain nanocomposite,” J. Mater. Chem. A 9(26), 14851–14858 (2021). [CrossRef]
36. A. Kuddus, S. K. Mostaque, and J. Hossain, “Simulating the performance of a high-efficiency SnS-based dual-heterojunction thin film solar cell,” Opt. Mater. Express 11(11), 3812–3826 (2021). [CrossRef]
37. P. Chelvanathan, M. I. Hossain, and N. Amin, “Performance analysis of copper–indium–gallium–diselenide (CIGS) solar cells with various buffer layers by SCAPS,” Curr. Appl. Phys. 10(3), S387–S391 (2010). [CrossRef]
38. A. Benmir and M. S. Aida, “Analytical Modeling and Simulation of CIGS Solar Cells,” Energy Procedia 10, 618–627 (2013). [CrossRef]
39. H. Heriche, Z. Rouabah, and N. Bouarissa, “High-efficiency CIGS solar cells with optimization of layers thickness and doping,” Optik 127(24), 11751–11757 (2016). [CrossRef]
40. Z. Rafik, I. Bouchama, L. djedoui, and E. Zaidi, “Improving the efficiency of CIGS solar cells by inserting a BSF µc-Si:H layer,” (Preprint).https://doi.org/10.21203/rs.3.rs-1177456/v1.
41. A. Cuevas, “The recombination parameter J0,” Energy Procedia 55, 53–62 (2014). [CrossRef]
42. M. Leilaeioun and Z. C. Holman, “Accuracy of expressions for the fill factor of a solar cell in terms of open-circuit voltage and ideality factor,” J. Appl. Phys. 120(12), 123111 (2016). [CrossRef]
43. G. A. H. Wetzelaer, M. Kuik, M. Lenes, and P. W. M. Blom, “Origin of the dark-current ideality factor in polymer:fullerene bulk heterojunction solar cells,” Appl. Phys. Lett. 99(15), 153506 (2011). [CrossRef]
44. P. Calado, D. Burkitt, J. Yao, J. Troughton, T. M. Watson, M. J. Carnie, A. M. Telford, B. C. O’Regan, J. Nelson, and P. R. F. Barnes, “Identifying Dominant Recombination Mechanisms in Perovskite Solar Cells by Measuring the Transient Ideality Factor,” Phys. Rev. Appl. 11(4), 044005 (2019). [CrossRef]
45. S. Kim, J. A. Márquez, T. Unold, and A. Walsh, “Upper limit to the photovoltaic efficiency of imperfect crystals from first principles,” Energy Environ. Sci. 13(5), 1481–1491 (2020). [CrossRef]
46. S. T. Jan and M. Noman, “Influence of layer thickness, defect density, doping concentration, interface defects, work function, working temperature and reflecting coating on lead-free perovskite solar cell,” Sol. Energy 237, 29–43 (2022). [CrossRef]
47. M. I. Hossain, P. Chelvanathan, M. M. Alam, M. Akhtaruzzaman, K. Sopian, and N. Amin, “Potential buffer layers for Cu2ZnSnS4 (CZTS) solar cells from numerical analysis,” IEEE Conference on Clean Energy and Technology (CEAT), 18-20 November 2013.
48. C. S. Solanki, Solar photovoltaics: fundamentals, technologies and applications, (Phi learning pvt. Ltd., 2015).
49. K. Mukhopadhyay, P. F. H. Inbaraj, and J. J. Prince, “Thickness optimization of CdS/ZnO hybrid buffer layer in CZTSe thin film solar cells using SCAPS simulation program,” Mater. Res. Innovations 23(6), 319–329 (2019). [CrossRef]
50. O. K. Simya, A. Mahaboobbatcha, and K. Balachander, “A comparative study on the performance of Kesterite based thin film solar cells using SCAPS simulation program,” Superlattices Microstruct. 82, 248–261 (2015). [CrossRef]
51. A. A. Abdelkadir, E. Oublal, M. Sahal, and A. Gibaud, “Numerical simulation and optimization of n-Al-ZnO/n-CdS/p-CZTSe/p-NiO (HTL)/Mo solar cell system using SCAPS-1D,” Results in Optics 8, 100257 (2022). [CrossRef]
52. T. Hayakawa, T. Nishimura, H. Sugiura, N. Suyama, K. Nakada, and A. Yamada, “Control of Donor Concentration in n-Type Buffer Layer for High-Efficiency Cu(In,Ga)Se 2 Solar Cells,” IEEE J. Photovoltaics 8(6), 1841–1846 (2018). [CrossRef]
53. A. Kuddus, A. B. M. Ismail, and J. Hossain, “Design of a highly efficient CdTe-based dual-heterojunction solar cell with 44% predicted efficiency,” Sol. Energy 221, 488–501 (2021). [CrossRef]
54. J. Y. Kwak, “Absorption coefficient estimation of thin MoS2 film using attenuation of silicon substrate Raman signal,” Results Phys. 13, 102202 (2019). [CrossRef]
55. K. Ali and Z. Ali, “Analytical study of electrical performance of SiGe-based n+-p-p+ solar cells with BaSi2 BSF structure,” Sol. Energy 225, 91–96 (2021). [CrossRef]
56. E. Yablonovitch, T. Tiedje, and H. Witzke, “Meaning of the photovoltaic band gap for amorphous semiconductors,” Appl. Phys. Lett. 41(10), 953–955 (1982). [CrossRef]
57. C. H. Wu and R. Williams, “Limiting efficiencies for multiple energy-gap quantum devices,” J. Appl. Phys. 54(11), 6721–6724 (1983). [CrossRef]
58. SCAPS: model for optical absorption. https://users.elis.ugent.be/ELISgroups/solar/projects/scaps/SCAPS%20Application%20Note%20Absorption%20Models.pdf.
59. M. Presselt, M. Bärenklau, R. Rösch, W. J. D. Beenken, E. Runge, S. Shokhovets, H. Hoppe, and G. Gobsch, “Subbandgap absorption in polymer-fullerene solar cells,” Appl. Phys. Lett. 97(25), 253302 (2010). [CrossRef]
60. W. J. D. Beenken, F. Herrmann, M. Presselt, H. Hoppe, S. Shokhovets, G. Gobsch, and E. Runge, “Sub-bandgap absorption in organic solar cells: experiment and theory,” Phys. Chem. Chem. Phys. 15(39), 16494–16502 (2013). [CrossRef]
61. S. R. Tulsani, A. K. Rath, and D. J. Late, “2D-MoS2 nanosheets as effective hole transport materials for colloidal PbS quantum dot solar cells,” Nanoscale Adv. 1(4), 1387–1394 (2019). [CrossRef]
62. M. V. Limaye, S. C. Chen, C. Y. Lee, L.Y. Chen, S. B. Singh, Y. C. Shao, Y. F. Wang, S. H. Hsieh, H. C. Hsueh, J. W. Chiou, C. H. Chen, L.Y. Jang, C. L. Cheng, W. F. Pong, and Y. F. Hu, “Understanding of sub-band gap absorption of femtosecond-laser sulfur hyperdoped silicon using synchrotron-based techniques,” Sci. Rep. 5(1), 11466 (2015). [CrossRef]
63. M. M. A. Moon, M. H. Ali, M. F. Rahman, J. Hossain, and A. B. M. Ismail, “Design and Simulation of FeSi2-Based Novel Heterojunction Solar Cells for Harnessing Visible and Near-Infrared Light,” Phys. Status Solidi A 217(6), 1900921 (2020). [CrossRef]
