Open Access
Add to BibSonomy
Abstract
This article demonstrates a novel high efficiency ZnS/SnS/MoS2 dual-heterojunction thin film solar cell. The device has been optimized with respect to the thickness, doping concentration, and defect density of each constituent layer including working temperature and back contact metal work function using SCAPS-1D simulator. The MoS2 plays a promising role to serve as a back surface field (BSF) layer with commendatory band alignment, which provides an opportunity for higher absorption of longer wavelength photons utilizing the tail-states-assisted (TSA) two-step photon upconversion approach. The insertion of MoS2 in the ZnS/SnS pristine structure offers a significant improvement of the power conversion efficiency (PCE) within the detailed-balance limit with a rise from 20.1 to 41.4% with VOC of 0.91 V, JSC of 53.4 mA/cm2 and FF of 84.9%, respectively. This result reveals MoS2 as an effective BSF for low cost, highly efficient dual-heterojunction structure for future fabrication.
© 2021 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
The recent challenge in photovoltaics is the eco-friendly fabrication of highly efficient and cost-effective solar cells. For large scale production, thin film solar cells have proved their potential to meet these expectations. Besides, the thin film technology offers the opportunity to bring alternative materials overcoming the limit of conventional Si technology. To achieve an economical goal, the earth abundant, non-toxic and inexpensive materials have been attracted as a priority to the researchers [1]. When such materials bear convenient electrical and optical properties such as an appropriate optical band gap, high optical absorption coefficient, high quantum yield for excited carriers, long diffusion length and life time for minority carriers, a significant conversion efficiency can be desired [2].
For choosing an appropriate material for absorber layer, the ability to absorb light as much as possible in order to stimulate electrons to higher energy states and the ability to pass those excited electrons from the solar cell to the outer circuit are importantly taken into consideration [3]. Solar cells face a significant loss if a portion of approaching light spectra is not properly absorbed. For example, a primary challenge comes in reduction of the reflectance by using a highly transparent material at the front surface [4]. In addition, if a highly absorbent material is used in absorber layer, there is a possibility of photon absorption close to the surface. Generally, a wide band gap window layer is useful in avoiding surface recombination at the front. However, regulation of front surface and absorber properties is hard to manage to reduce significant loss. Therefore, an emphasis is put on back surface passivation and back reflection in solar cell structure to improve the photo conversion efficiency from its usual range [5].
In recent years, the non-toxic, orthorhombic tin mono-sulphide (SnS) has been considered as an alternative absorber of Si to grow the thin-film solar cells [6,7]. The band gap of ∼1.30 eV is close to the value of Si, but absorption coefficient comes with much higher values in the order of 105 cm-1, which allow SnS absorber to have thinner structure than crystalline Si layer [3]. Such a thin SnS layer is able to absorb the majority of solar spectrum above the band gap [8]. Besides, both Sn and S are earth abandoned materials and capable of supplying huge amount of solar energy on daily basis. Use of these materials can also be an alternative to popular but rare elements like Cd, Te, Ga, Se or CZTS [8]. Furthermore, there are reports on synthesis of SnS with low-cost methods like spin coating, dip coating and spray pyrolysis [9–11]. Hence, the environmentally benign solar cells with SnS absorber layer can be fabricated with a cost-effective technique. While, the theory foresees that a single-junction SnS-based solar cells would have a maximum energy conversion efficiency of 32%, some SnS-based cells exhibit only 2% of photo conversion efficiency [8]. A theoretical investigation with ZnS/SnS structure with buffer layer predicts the theoretical efficiency to be 12.08% [12]. The theoretical efficiency of ZnS/SnS heterojunction can be reached to 16.26%, while the addition of NiO2 as BSF and a suitable metal back contact, Mo enhances the efficiency up to 25.1% [7,13]. Therefore, the SnS was found as a promising absorber material in solar cells, but the SnS has never been applied yet with a material showing the Tail-States-Assisted (TSA) two-step upconversion in the double-heterojunction solar cells with a preferred band gap, high doping concentration and high absorption coefficient at longer wavelength. Additionally, in solar cell fabrication, it is crucial to minimize unwanted Fresnel surface reflection losses over the whole range of the solar spectrum at the interface between air and the top layer of solar cells to improve light harvesting in solar cells. Thus, a better cell performance can be obtained using wider band gap materials like ZnS which provide easy passage of photons to window-absorber junction. Moreover, the properties of ZnS like high refractive index (2.25 at 632 nm), high effective dielectric constant (9 at 1 MHz), wide wavelength pass band and an opportunity to employ suitable metal contact for SnS based cells invigorate the ZnS to be manufactured as more favorable window layer rather than using common window materials [14]. Also, the thin film fabrication of ZnS has already been proven to be an economical and eco-friendly one for the fabrication of heterojunction thin film solar cells [15]. In addition, the majority of the fabricated SnS heterojunction used CdS or doped CdS as the window materials. But CdS is carcinogenic to humans, and inhalation of dust of the CdS can have harmful effects on the human body like kidneys (impairment) and bone (bone weakness) [16]. Thus, n-ZnS window rather than the common window materials with SnS absorber can lead to an attractive result in heterojunction cells if proper BSF material can be chosen to build up a novel structure.
The significance of the BSF layer for the improvement of solar cell performance have already been demonstrated in previous reports [17–20]. However, in solar cells, photons of different wavelengths get immersed in different absorption thickness. In the usual case, short wavelength photons get involved at window and absorber layer which are nearer than the surface while longer wavelength photons are trapped at deeper depth [21]. Thus, a significant portion of falling spectrum gets involved within the absorber but the longer wavelength portion remains unused in the conventional structures. Some of previous studies involve introducing a suitable bottom layer that can absorb longer wavelength photons through tail-states-assisted (TSA) two-step photon upconversion process that plays a vital role in enhancing photovoltaic performances [22–24].
The BSF layer between the rear contact metal and base absorber lessens the recombination loss. Incorporation of heavily doped precise material moderates the conduction band barrier height [25]. Molybdenum disulfide, MoS2 is a promising low-cost, sunlight harvester with band gap of 1.5 eV, a high extinction coefficient in the order of 104 cm-1 and long diffusion length of 1 µm [26]. Studies reveal its ability to form hetero-structure with SnS enhancing the separation of photogenerated free carriers [27,28]. In these works, MoS2 was synthesized by chemical vapor deposition. Other techniques like atomic layer deposition, electrodeposition and pulsed laser deposition can also be used to fabricate the thin layer of MoS2 [29]. However, SnS-MoS2 junction can facilitates the separation of photogenerated free carriers and satisfies photoluminescence intensity. Studies also reveal that the nonlinear absorption of the SnS-MoS2 heterostructure is superior to that of the MoS2 monolayer and SnS nanosheet [28].
In this work, we introduce MoS2 in the back surface layer between a formed heterojunction with SnS and rear contact material Mo (4.95 eV). Mo was selected primarily because it is one of the constituent materials of MoS2, which has effect on reducing the tunnel barrier and eliminating the Schottky barrier which results in the improvement of contact properties [30,31].
Herein, we design and investigate the PV performance of a novel heterojunction (ITO/n-ZnS/p-SnS/p+-MoS2/Mo) solar cell with SCAPS-1D simulator. The influence of several physical parameters of absorber layer along with window and back surface layer has been studied in details.
2. Device modeling and simulation
2.1 Device modeling
The construction of proposed ITO/n-ZnS/p-SnS/p+-MoS2/Mo structure is presented in Fig. 1(a). Because of its excellent transparency, ITO is a preferred material for a substrate in heterojunction cells [22]. This portion is illuminated to sunlight with a thin n-ZnS window layer to form a pn junction with p-SnS absorber. As the band gaps of ITO and ZnS are high compared to SnS absorber and MoS2 back surface layers, they are expected to provide a high optical throughput in dual-heterojunction structure. Wider band gap of ZnS with a suitable electron affinity helps forming a proper junction with SnS and band gap between p-SnS and p+-MoS2 enables suitable formation of second heterojunction between absorber and back surface layer. ITO and Mo were used as front and back contact materials, respectively.
Fig. 1. The schematic (a) block and (b) illuminated energy diagram of n-ZnS/p-SnS/p+-MoS2 dual-heterojunction solar cell (Fn, Fp: quasi-Fermi levels of electrons and holes, respectively; Ec, Ev: conduction and valance band edges, respectively).
The simulated energy band diagram of the designed heterojunction solar cells in Fig. 1(b) further clarifies the modeling of dual-heterojunction structure. The conduction band (CB) of the SnS absorber layer is lower than that of the ZnS window layer, with a very small conduction band offset (CBO: 0.05 eV) between them. It allows the electrons in SnS absorber to transport through ZnS to ITO where holes become blocked due to very high valance band offset (VBO: 2.24 eV) between the ZnS window and SnS absorber layer. It has the advantage of reduction in reverse saturation current and uprising in open circuit voltage. Additionally, the higher band gap than the SnS absorber prevents photon loss in the window and also creates slight improvement in short circuit current density [13]. On the other hand, the valance band of the thin film TMDC (transition metal dichalcogenide) material MoS2 is higher (CBO: 0.40 eV) than that of the SnS absorber layer, and the valance band offset (VBO:0.04 eV) between them is substantially smaller. Thus, the CBO between the SnS absorber and MoS2 BSF layers restricts the flow of electrons and transport holes from SnS to the Mo back electrode.
2.2 Device simulation
The numerical simulation and analysis were performed using SCAPS-1D software, which solves the Poisson equation for electrostatic potential and continuity equation for free electrons and holes in each layer of the proposed structure. Based on the solutions of these two equations, electrical properties such as the energy band diagram, current-voltage characteristics, recombination profile, and quantum efficiency have been extensively studied. Default parameters like global air mass of AM1.5G and the illumination of 1000 W/m2 under single sun with the working temperature of 300 K have been chosen to perform the simulation while initially keeping the series resistance set at ideal value. The Gaussian energy distribution have been used to consider both single-acceptor and single-donor with similar bulk defects in each of the bulk layers with significant values. Table 1 represents the optoelectronic parameters of active materials used as the input parameters during simulations. The optical absorption coefficient data of ZnS, SnS and MoS2 as well as the surface work function (WF) of Mo were collected from the reported literatures [32–35]. It is noted that SCAPS simulator automatically calculated the sub-band gap effect in the MoS2 layer [22–24]. The ranges of each parameter of ZnS, SnS and MoS2 active layers were chosen same as observed in the previous literatures and/or experimentally observed outcomes.
Table 1. Simulation Parameters for n-ZnS/p-SnS/p+-MoS2 dual-heterojunction Solar Cell a
3. Results and discussion
3.1 Photovoltaic performance of the SnS solar cell without a BSF layer
3.1.1 Impact of thickness, doping, and defect density of the SnS absorber layer
The effects of thickness, carrier concentration and defect density of SnS absorber layer on the performance of n-ZnS/p-SnS solar cell have been discussed in this section. The absorber layer thickness and carrier concentration are critical influencing parameters of the solar cell structure. It is quite essential to optimize these two parameters while incorporating bulk defect density in the absorber layer to get indication about real time optimum cell performance which will in turn be helpful for further enhancement and device fabrication.
Figure 2(a) indicates that the thickness of SnS absorber layer has a significant influence on the cell performance and the PCE can be fine-tuned by changing the absorber layer thickness. The cell current JSC exhibits a significant rise from 7 to 30 mA/cm2 exponentially up to a thickness of 750 nm of SnS layer and then becomes almost constant while VOC increases from 0.61 to 0.75 V with increase of the absorber layer thickness and trends to retain increasing at a flat rate for thickness of 750 to 1100 nm. On the other hand, the fill factor increases from 81.0 to 82.2% maintaining exponential nature corresponding to the SnS layer thickness from 50 to 750 nm and then its suddenly rise from 82.2 to 83.25% (∼1.05%). This abrupt change of FF though small in magnitude (1.05%) may appeared due to the existence of critical length of carrier diffusion, afterward the recombination will start to dominate. Figure 2(b) shows the quantum efficiency (QE) spectra for the change in SnS absorber layer thickness. It is viewed from the figure that QE significantly increases with the thickness of SnS layer up to 700 nm, then it almost saturates for the thickness >700 nm due to the saturation of absorption of incident light. Since cell current slightly increases after SnS layer of 750 nm the QE spectra also slightly rises with thickness up to 1100 nm of SnS layer. Consequently, the optimum cell efficiency of ∼21% corresponding to the SnS layer thickness of 750- 800 nm was achieved. Therefore, the SnS layer thickness of 700 nm was chosen to investigate the further improvement of the designed cell with hole-transport layer of MoS2.
Fig. 2. The SnS layer thickness dependent variation of (a) output parameters and (b) QE of n-ZnS/p-SnS heterojunction solar cell.
Figure 3(a), depicts the effect of carrier concentration of SnS absorber layer in the range of 1012 to 1017 cm−3 to achieve optimum cell performance with a thickness of 700 nm. It is observed that, both the Voc and FF increases from 0.51 to 0.83 V and 79 to 83%, respectively with the increase of carrier concentration up to the order of 1016 cm−3 where the PCE of the of designed n-ZnS/p-SnS cell reaches its saturation at ∼22%. The Jsc retains nearly at 33.5 mA/cm2 up to 1015 cm−3 and decreases a little to a value of 32 mA/cm2 at a doping of 1017 cm−3. This is due to the growth of the degree of domination of the recombination among the abundant carriers throughout the cell with the enhancement of SnS carrier concentration. Furthermore, the impurity scattering as well as carrier recombination, affects the resultant cell performance at higher carrier concentrations of the SnS absorber layer. The change of the total equilibrium carrier density affects the carrier lifetime, therefore the performance of the cell. The observation of the change of carrier lifetimes directly τn and τp as a function of the excess carrier concentration is an obvious study to investigate the excess carrier effect. However, previous study reports that the carrier life time of τn and τp decreases around 1 ms while the excess carrier concentration of cell active layers exceeds 1015 cm-3 for Si based solar cell [36]. Considering a certain donor concentration, for both Nd <Na and Nd >Na, the lifetimes are approximately constant at low injection levels and decreases at high injection levels. The reason is that when Δn, Δp < Nd or Na, the lifetimes τn and τp are defined by the equilibrium carrier concentrations. At high injection levels the trap levels are filled by minority carriers results in the increase of carrier recombination rate and the decrease of carrier lifetimes. When the semiconductor is compensated (Nd ≈ Na) the lifetime values decrease with increasing Δn, Δp for all injection levels (vice versa for a certain acceptor concentration). This is related to the smaller value of the total equilibrium carrier concentration, which is in the order of intrinsic carrier concentration. The simulation results observed for different carrier concentration apparently reveals this phenomenon.
Fig. 3. The variation of PV parameters of n-ZnS/p-SnS heterojunction solar cell with (a) doping density and (b) bulk defect concentration of SnS absorber layer.
Figure 3(b) reveals that the photovoltaic parameters face a negligible change in the defect density range from 1010 cm−3 to 1014 cm−3 and afterward this value, a drastic degradation of cell parameters is appeared at defect density > 5×1014 cm−3 due to the domination of recombination loss. Consequently, the recombination rate gets modulated with the SRH recombination at a higher defect density of SnS absorber. Therefore, the layer thickness of 700 nm, carrier concentration of 1016 cm−3 with defect density of 1014 cm−3 were considered as optimum parameters for SnS layer for further investigation. Herein, the defect density of 1014 cm-3 was chosen for further study though the cell performance starts to deteriorate at 1012 cm-3, considering the optimum as well as feasible defect density level of 1012-1015 cm-3 that is observed in metal chalcogenide-based thin films [37,38].
3.1.2 ZnS window layer impact on PV performance
The influences of thickness, carrier concentration and defect density of ZnS window layer on n-ZnS/p-SnS heterojunction solar cells have been investigated in this section. Figure 4(a), b and c show the PV performance of n-ZnS/p-SnS solar cells with respect to thickness, carrier concentration and defect density, respectively of ZnS window layer. In Fig. 4(a), the VOC and FF lessen slightly with the thickness of window layer while JSC gradually decreases above the thickness of 100 nm. Consequently, the PCE of the proposed structure follows the same trend of decrement. The drop-off of the JSC with thicker window layer is due to obstruction of lower wavelength photons to reach the absorber layer by parasitic absorption. As a remedy, the use of an anti-reflection coating (ARC) layer on the top of ZnS can effectively reduce the front reflection of light; therefore, the cell current as well as efficiency of the solar cells can be improved [39,40]. Figure 4(b) shows the impact of ZnS carrier concentration on the cell parameters in the range of 1013 to 1018 cm− 3. It is observed that the VOC starts to decrease from 0.744 to 0.727 V corresponding to the increase in carrier concentration from 1013 to1018 cm− 3 and the JSC also declines though small in extent from 33.4 to 32.8 mA/cm2. On the contrary, the FF gradually increases from 80.8 to 82.4% corresponding to the concentration of 1013 to1017 cm− 3 and retains on this saturated value at >1017 cm− 3. Therefore, the overall cell performance is negligibly affected by ZnS carrier concentration up to 1017 cm−3. With further higher order of carrier concentration, the PCE devaluates from 20.1 to 19.8% for doping concentration of 1018 cm−3 due to the higher recombination for the presence of surplus carriers throughout ZnS layer. Since the ZnS layer thickness is very small compare with SnS absorber layer, carrier density in ZnS region cannot play noticeable role on cell performance. Figure 4(c) depicts that the VOC and JSC negligibly affected by defect density of the thin window layer. The VOC is almost independent at the value of ∼0.73 V and the JSC is also almost independent with defect density up to 1014 cm− 3. Afterwards, JSC as well as FF follows a slight decrement from 33.16 to 32.83 mA/cm2 and 82.25 to 82.16%, respectively with inclusion of defect level from 1015 to 1016 cm− 3. Herein, a trivialize decrement of PCE from 20.13 to 19.95% has been observed for the defects of 1016 cm−3 of ZnS window layer. In summary, the optimum cell efficiency of ∼20.1% corresponding to the ZnS layer thickness of 50-100 nm was achieved. However, the ZnS layer thickness of 100 nm with donor density of 1017 cm− 3 and defect density of 1015 cm− 3 were chosen to investigate the further improvement of the designed cell. These optimized parameters of ZnS window layer are in good agreement with results found in ZnS based similar type of heterostructure in previous studies [41–43]. It is noted that the above-mentioned study for the optimization of pristine ZnS/SnS cell parameters was performed under ideal condition of series and shunt resistances to explore the effect of each active layer parameters solely and thereafter, the effect of series and shunt resistance was studied utilizing the optimized parameters values.
Fig. 4. The variation of PV parameters of n-ZnS/p-SnS heterojunction solar cell with (a) thickness, (b) doping density and (c) bulk defect concentration of ZnS window layer.
3.2 Photovoltaic performance of a SnS solar cell with a MoS2 BSF layer
3.2.1 Impact of thickness and doping density of the BSF layer on PV performance
The BSF plays an important role for the collection of photo generated charge carriers in the absorber layer. The effect of the thickness and carrier concentration of MoS2 BSF layer on the performance of the designed device has been evaluated in the ranges of 0-300 nm and 1013-1018 cm−3, respectively. As depicted in Fig. 5, the JSC exhibits a remarkable rising from 33 to ∼60 mA/cm2 with incorporation of BSF layer thickness up to 200 nm and then tends to be constant at a value of ∼62 mA/cm2. While decrement of JSC from ∼62 mA/cm2 to ∼40 mA/cm2 is observed for the increase in the carrier concentration from 1013 to 1018 cm−3. The insertion of MoS2 BSF layer enhances the VOC by 0.182 V due to the higher built-in potential developed at ZnS/SnS and SnS/MoS2 heterointerfaces [22–24]. However, the Voc and FF decrease from ∼0.92 to ∼0.88 V and 82 to 57% with the increase of BSF layer thickness from 50 to 300 nm. The Voc and FF exhibit increment from 0.86 to 0.92 V and 70 to 86% respectively with the increase of MoS2 doping concentration from 1013- 1018 cm− 3 only. Further doping of MoS2 causes the reduction of cell current thereby the cell efficiency.
Fig. 5. The variations of PV performance parameters depending on the acceptor density and thickness of the MoS2 BSF layer of n-ZnS/p-SnS/p+-MoS2 dual-heterojunction solar cell.
The cell performance primarily enhances with the inclusion of MoS2 layer thickness up to 200 nm due to accession of higher wavelength absorption of solar spectrum by TSA photon upconversion where two sub-band gap photons are absorbed in order by an Urbach energy state that produces an electron-hole pair [22–24] and easy pathway for photogenerated carriers to the metal electrode. A detail of the TSA process can be found in other work [22]. Beyond the thickness of 200 nm, the cell current tends to decrease due to shorter diffusion length and carrier lifetime. Higher recombination rate owing to excessive carrier concentration (>1018 cm− 3) also induces reduction of the cell current although it improves the FF of the designed n-ZnS/p-SnS/p+-MoS2 dual-heterojunction device. Moreover, high carrier concentration produces a strong electric field, at the SnS/MoS2 interface, which blocks the flow of minority electrons towards the interface, thus reducing the interface recombination. The Fermi level shifts of MoS2 BSF towards the valence band and makes an efficient collection of holes at the back-contact forming a nearly ohmic contact with the metal electrode Mo [44]. From analysis, the optimum performance of n-ZnS/p-SnS/p+-MoS2 dual-heterojunction solar cell with MoS2 BSF was determined at MoS2 carrier concentration and thickness of 1×1017 cm−3 and 200 nm, respectively.
3.2.2 Impact of thickness and defect density of BSF layer on PV performance
Figure 6 shows the dependency of the designed cell parameters on bulk defects and thickness of MoS2 BSF layer over the ranges of 1011-1016 cm−3 and 0-300 nm, respectively. The photovoltaic parameters VOC, JSC and FF have been considerably affected over a certain level of defect density (1015 cm− 3). The PCE of n-ZnS/p-SnS/p+-MoS2 dual-heterojunction solar cell falls down from ∼44.0 to ∼36.0% due to SRH recombination over the defect density of 1015 cm−3. The effect of the bulk defect density of MoS2 is negligible up to 1015 cm− 3 as majority carrier dominates throughout the BSF layer. In Fig. 6, the photocurrent (JSC) drops from ∼57.0 to ∼54.0 mA/cm2 with a small decrease of FF from 85.0 to 84.0% while the VOC retains at constant value of 0.915 V for corresponding intensification in the defect density of 1011-1016 cm−3. The presence of higher defects density causes a higher SRH recombination and inevitably affects the cell performance. The optimum cell efficiency of ∼41.4% has been obtained with VOC of 0.913 V, JSC of 53.0 mA/cm2 and FF of 85% considering the trade of among all the cell parameters with the MoS2 BSF thickness of 200 nm, the carrier concentration of ∼1×1017 cm−3 at defect density of 1014 cm−3.
Fig. 6. The variations of PV performance parameters depending on the defect concentration and thickness of the MoS2 BSF layer of n-ZnS/p-SnS/p+-MoS2 dual-heterojunction solar cell.
3.2.3 Impact of interface defect density on PV performance
Figure 7 shows the impact of the ZnS/SnS and SnS/MoS2 interface defect density on the cell performance corresponding to the range of 1011 to 1016 cm−2. The existence of lattice (or thermal) mismatch and different crystallographic structures of the absorber and buffer layer creates a network of dislocations between the active layers which act as electronic defects and promote any unwanted recombination of photogenerated carriers. In addition, inter-diffusion of metal cations among adjacent active layers induces the structural defects at the interfaces during the fabrication process [45]. Therefore, interface engineering is a crucial part in designing and fabrication of the solar cell. It has been found from the figure that the solar cell performance is negligibly influenced by both ZnS/SnS and SnS/MoS2 interface defect density up to 1012 cm−2. The interface defect density over 1012 cm−2 distorts the cell parameters VOC, JSC and FF to noticeable extents and consequently overall cell performance. The defect presents at both interfaces of ZnS/SnS and SnS/MoS2 increases the carrier trapping probability as well as the series resistance which is responsible for the reduction of the cell current and noticeable drops of FF. In Fig. 7(a), the increase of carrier recombination rate causes a gradual declination in both JSC and VOC from 53.4 to 52.1 mA/cm2 and 0.913 to 0.898 V, respectively for the corresponding change in ZnS/SnS interface defect density from 1011 to 1016 cm− 2. Here, the FF marginally decreases from 84.97 to 83.30% due to the increase of series resistance of the cell. In consequence, the PCE gets degraded from 41.4 to 39.1%, with the interface defect density up to 1016 cm−2. On the other hand, a similar impact of defect density for SnS/MoS2 is observed as shown in Fig. 7(b). The JSC, VOC and FF partly decrease from 53.3 to 52.5 mA/cm2, 0.925 to 0.814 V and 84.85 to 84.35%, respectively and thereupon the PCE from 42.0 to 35.7% for the corresponding change in defect density from 1014 to 1016 cm−2 at SnS/MoS2 interface. This study suggests that the interface defect density over a certain value has a considerable impact on the overall performance of the SnS-based cells. The suitable choice of the active materials with as minimum as possible lattice mismatch including the proper steps to achieve optimum defects which are unavoidably added during the fabrication process, may offer most favorable cell performance. The interface defects of 1×1012 cm-2 were considered for both ZnS/SnS and SnS/MoS2 interfaces during the simulation.
Fig. 7. The variations of PV performance parameters with (a) ZnS/SnS and (b) SnS/MoS2 interface defect concentration.
3.3 Effect of series and shunt resistance
Figure 8 represents the effect of series (RS) and shunt (RSH) resistance on n-ZnS/p-SnS/p+-MoS2 dual-heterojunction solar cell performance in the ranges of 0-7 Ω-cm2 and 100-1500 Ω-cm2, respectively. The Rs consists of various resistances such as the resistance at interface between the semiconductors and metal contacts with their own bulk resistances as well those are the sources of RS in solar cell while the reverse saturation liable for RSH [46]. To design and fabricate the high efficiency PV devices, it is required to achieve low series and high shunt resistances of the solar cells. Figure 8(a) illustrates negligible impact of series resistance on VOC and JSC. However, the FF is reduced retaining the linear-like nature from 79.5 to 51.0% at RS varied from 0–7Ω cm2, followed by decreases in the PCE from 41.4 to 24.5%. On the contrary, the VOC and JSC are found nearly constant corresponding to the change in RSH from 1500 to 300 Ω-cm2, immediately after which FF significantly drops from 82 to 72%, resulting the PCE decrement from 40 to 34% corresponding to further decreased value of RSH from 300-100 Ω-cm2 as shown in Fig. 8(b). It is revealed from these simulation results that the performance of the designed dual-heterojunction solar cell are markedly influenced by the RS and RSH.
Fig. 8. The impacts of (a) series and (b) shunt resistance on the PV performance of n-ZnS/p-SnS/p+-MoS2 dual-heterojunction solar cell.
3.4 Effect of work function and working temperature
Figure 9(a) shows the influence of back contact metal work function (WF) on cell parameters for WF of 3.3 to 5.3 eV. Here, both VOC and FF rise significantly from 0.35 to 0.91 V and 61 to 82% along with minimal increase in JSC from 49.2 to 51.5 mA/cm2. Consequently, the PCE escalates from 11.0 to 41.0% with increasing the metal WF till a certain value of 5- 5.1 eV and thereafter it reaches and retains on saturation value. This refers the decrement of barrier height for transporting the majority carrier throughout the cell to metal electrode with higher metal WF. Therefore, WF of back metal contact has a significant impact on cell performance that can be compensated by the use of metal contact having WF with optimum barrier height.
Fig. 9. The effects of (a) back contact metal work function and (b) working temperature on the PV performance of n-ZnS/p-SnS/p+-MoS2 dual-heterojunction solar cell.
Figure 9(b) depicts the impact of the working temperature (WT) on output parameters of PV cell in the range of 280 to 400 K. Overall, the PCE decreases from 41.4 to 31.8% with increasing the working temperature, WT. An increase in WT modifies the density of intrinsic carriers in photoactive semiconductor materials to retain its balanced relationship with the absorption coefficient and shoots up the velocity of charged carriers that acts as a reason of reduction of the bond energy [39]. Therefore, the recombination rate of photogenerated carriers: electrons and holes boost up resulting a reduced number of free carriers in the cell, which in turn drops off the overall cell performance at higher WT.
3.5 Quantum efficiency of the n-ZnS/p-SnS/p+-MoS2 dual-heterojunction solar cell
Figure 10 shows the effect of varying MoS2 BSF layer thickness on the quantum efficiency (QE) of designed dual-heterojunction cell in the range of 0-300 nm. The QE as a function of light wavelength (λ) is defined as the ratio of current passing through the load circuit i.e., the number of charge carriers generated and getting out at load circuit from the solar cell to the number of incident photons. The entire collection of photogenerated electron-hole pairs (EHPs) indicates 100% external QE (EQE) for corresponding wavelength incident on the active layers of the cell. As observed in the figure, the quantum efficiency of the designed SnS-based dual-heterojunction solar cells improves significantly at wavelength of λ>900 nm after the incorporation of MoS2 BSF layer. The rate of increment of spectral absorption quite higher up to the BSF layer thickness of 200-250 nm and absorption rate starts to decrease and trend to reach a saturation value over 250 nm thickness of BSF layer. The sub-band gap absorption of longer wavelength occurs in the BSF layers because of the band tail-states assisted (TSA) two-step photon up-conversion with the combined effect of absorption coefficient and doping density as well as higher carrier concentration in the order of ∼1017 cm− 3 that offers an easy conductive path for transporting the photogenerated carriers to output terminal [22–24,47–51].
Fig. 10. The change in quantum efficiency of n-ZnS/p-SnS/p+-MoS2 dual-heterojunction solar cell with the thickness of MoS2 BSF layer.
3.6 Optimized device structure of the n-ZnS/p-SnS/p+-MoS2 dual-heterojunction solar cell
Figure 11 represents the J–V characteristic and quantum efficiency (QE) of the ITO/n-ZnS/p-SnS/Mo heterojunction and ITO/n-ZnS/p-SnS/p+-MoS2/Mo dual-heterojunction solar cells. The optimized values of thickness of ZnS, SnS and MoS2 layers were 100, 700 and 200 nm, respectively. The doping density of the corresponding layers were 1×1017, 1×1015 and 1×1017 cm-3, respectively and defect density of 1×1014 cm-3 was considered for each layer. The optimized interface defect density of 1×1012 cm-2 were taken for both ZnS/SnS and SnS/MoS2 interfaces. The ideal values of series (0 Ω-cm2) and shunt (∞ Ω-cm2) resistances were considered. The pristine ITO/n-ZnS/p-SnS/Mo solar cell shows the optimized PCE of 20.1% with the VOC, JSC, and FF of 0.731 V, 33.4 mA/cm2 and 84.95%, respectively. On other hand, the PCE of 41.4% with VOC, JSC, and FF of 0.913 V, 53.3 mA/cm2 and 84.9%, correspondingly is achieved from the optimized ITO/n-ZnS/p-SnS/p+-MoS2/Mo dual-heterostructure. The insertion of MoS2 as BSF offers the massive enhancement of PCE of ∼20% as shown in Fig. 11(a). As depicted in Fig. 11(b), the quantum efficiency of the designed SnS-based dual-heterojunction solar cells gets improved at wavelength of λ>900 nm with the incorporation of MoS2 BSF layer into ITO/ZnS/SnS/Mo pristine cell. The insertion of MoS2 BSF having suitable band gap and favorable band alignment with adjacent SnS absorber layer offers a drastic improvement of light absorption at higher wavelength (λ>900 nm) that is the origin of the huge improvement of cell performance from 20.11 to 41.40%. The optimized output parameters of SnS-based dual-heterojunction solar cells with and without MoS2 BSF layer are summarized in Table 2.
Fig. 11. The (a) simulated J-V characteristics and (b) QE of the optimized structure of n-ZnS/p-SnS/p+-MoS2 dual heterojunction solar cell.
Table 2. The optimized output parameters for ZnS/SnS solar cell with and without MoS2 BSF layer
4. Conclusion
A novel SnS-based n-ZnS/p-SnS/p+-MoS2 dual-heterojunction thin film solar cell has numerically been analyzed and optimized employing SCAPS-1D software. This study reveals that the incorporation of MoS2 as BSF layer significantly improves the PCE of the ZnS/SnS heterojunction solar cell. Different parameters that influenced the PV performance significantly were studied and elucidated in details in this study. The layer thickness, carrier concentration, bulk and interfacial defect density of each active materials, working temperature and back contact metal work function are appeared as the major influencing factors for optimizing the PV performance. The optimized PCE of the ZnS/SnS solar cell without BSF layer is about 20.1%, whereas the insertion of MoS2 BSF layer, acted also as bottom absorber layer enhances the PCE of the designed ZnS/SnS based dual-heterojunction solar cell to 41.4% with VOC of 0.91 V, JSC of 53.3 mA/cm2 and FF of 84.9%, respectively. These results would be helpful to perform feasible fabrication of an economical and environmentally benign SnS-based dual-heterojunction solar cell.
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. J. A. Andrade-Arvizu, M. Courel-Piedrahita, and O. Vigil-Galán, “SnS-based thin film solar cells: perspectives over the last 25 years,” J Mater Sci: Mater Electron 26(7), 4541–4556 (2015). [CrossRef]
2. Photovoltaic Materials, 1st ed. (Imperial College Press, 1998), https://www.osti.gov/etdeweb/biblio/631275.
3. M. Minbashi, A. Ghobadi, M. H. Ehsani, H. Rezagholipour Dizaji, and N. Memarian, “Simulation of high efficiency SnS-based solar cells with SCAPS,” Sol. Energy 176(August), 520–525 (2018). [CrossRef]
4. R. Sharma, A. Gupta, and A. Virdi, “Effect of single and double layer antireflection coating to enhance photovoltaic efficiency of silicon solar,” J. Nano- Electron. Phys. 9(2), 02001–02004 (2017). [CrossRef]
5. J. N. Duenow and W. K. Metzger, “Back-surface recombination, electron reflectors, and paths to 28% efficiency for thin-film photovoltaics: a CdTe case study,” J. Appl. Phys. 125(5), 053101 (2019). [CrossRef]
6. V. R. Minnam Reddy, H. Cho, S. Gedi, K. T. R. Reddy, W. K. Kim, and C. Park, “Effect of sulfurization temperature on the efficiency of SnS solar cells fabricated by sulfurization of sputtered tin precursor layers using effusion cell evaporation,” J. Alloys Compd. 806, 410–417 (2019). [CrossRef]
7. S. Ahmmed, A. Aktar, J. Hossain, and A. B. M. Ismail, “Enhancing the open circuit voltage of the SnS based heterojunction solar cell using NiO HTL,” Sol. Energy 207, 693–702 (2020). [CrossRef]
8. P. Sinsermsuksakul, L. Sun, S. W. Lee, H. H. Park, S. B. Kim, C. Yang, and R. G. Gordon, “Overcoming efficiency limitations of SnS-Based solar cells,” Adv. Energy Mater. 4(15), 1400496 (2014). [CrossRef]
9. B. P. Reddy, M. C. Sekhar, S. V. P. Vattikuti, Y. Suh, and S. H. Park, “Solution-based spin-coated tin sulfide thin films for photovoltaic and supercapacitor applications,” Mater. Res. Bull. 103, 13–18 (2018). [CrossRef]
10. S. H. Chaki, M. D. Chaudhary, and M. P. Deshpande, “SnS thin films deposited by chemical bath deposition, dip coating and SILAR techniques,” J. Semicond. 37(5), 053001 (2016). [CrossRef]
11. K. T. Ramakrishna Reddy, N. Koteswara Reddy, and R. W. Miles, “Photovoltaic properties of SnS based solar cells,” Sol. Energy Mater. Sol. Cells 90(18-19), 3041–3046 (2006). [CrossRef]
12. S. Pandey and D. K. Dwivedi, “Theoretical investigation for enhancement of performance of SnS based solar cell,” 3rd Int. Conf. Condens. Matter Appl. Phys. 2220, 140010 (2020).
13. J. Xu and Y. Yang, “Study on the performances of SnS heterojunctions by numerical analysis,” Energy Conversion and Management 78, 260–265 (2014). [CrossRef]
14. J. Hasanzadeh, A. Taherkhani, and M. Ghorbani, “Luminescence and Structural Properties of ZnS: Cu Nanocrystals Prepared Using a Wet Chemical Technique,” Chinese journal of physics 51, 540–550 (2013). [CrossRef]
15. M. A. Jafarov, E. F. Nasirov, and S. A. Jahangirova, “Nano-ZnS thin films for solar cell,” Nanosist.: fiz. him. mat. 6, 644–649 (2015). [CrossRef]
16. L. Slewa, Preparation and characterization of SnS and ZnS thin films for fabrication of solar cells, Dissertation, Salahaddin University, Erbil (2014), [CrossRef] .
17. O. Von Roos, “A simple theory of back surface field (BSF) solar cells,” J. Appl. Phys. 49(6), 3503–3511 (1978). [CrossRef]
18. S. Sepeai, M. Y. Sulaiman, K. Sopian, and S. H. Zaidi, “Investigation of back surface fields effect on bifacial solar cells,” AIP Conf. Proc. 1502(1), 322–335 (2012). [CrossRef]
19. J. Lee, C. Park, V. A. Dao, Y. J. Lee, K. Ryu, G. Choi, B. Kim, M. Ju, C. Jeong, and J. Yi, “Direct metallization local Al-back surface field for high efficiency screen printed crystalline silicon solar cells,” J. Nanosci. Nanotechnol. 13(11), 7551–7555 (2013). [CrossRef]
20. X. Yan, E. C. Wang, N. Chen, L. Zhang, X. Gong, X. Zhang, and S. Duttagupta, “Investigation of phosphorus diffused back surface field (BSF) in bifacial nFAB solar cells,” Sol. Energy 179, 335–342 (2019). [CrossRef]
21. R. A. Z. Razera, H. Boudinov, I. Zanesco, and A. Moehlecke, “Passivation analysis of the Emitter and selective back surface field of silicon solar cells,” SBMicro 2017 - 32nd Symp. Microelectron. Technol. Devices Chip Sands, co-located Symp. 30th SBCCI - Circuits Syst. Des. 2nd INSCIT - Electron. Instrumentation, 7th WCAS - IC Des. Cases 17th SForum - Undergrad. 0–3 (2017).
22. S. K. Mostaque, B. K. Mondal, and J. Hossain, “Simulation approach to reach the SQ limit in CIGS-based dual-heterojunction solar cell,” (2021). arXiv:2103.08841 [physics.app-ph]
23. B. K. Mondal, S. K. Mostaque, M. A. Rashid, A. Kuddus, H. Shirai, and J. Hossain, “Effect of CdS and In3Se4 BSF layers on the photovoltaic performance of PEDOT:PSS/n-Si solar cells: Simulation based on experimental data,” Superlattices Microstruct. 152, 106853 (2021). [CrossRef]
24. 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]
25. M. Dey, M. F. Shahriar, A. Ali, M. Dey, and N. K. Das, “Design and optimization of an efficient molybdenum disulfide (MoS2) solar cell with tin sulfide BSF,” 2nd Int. Conf. Electr. Comput. Commun. Eng. ECCE 2019, 1–5 (2019).
26. S. Kohnehpoushi, P. Nazari, and B. A. Nejand, “MoS2: a two-dimensional hole-transporting material for high-efficiency, low-cost perovskite solar cells,” (2018).
27. X. Li, S. Zhang, X. J. Wang, G. F. Huang, L. X. Xia, W. Hu, and W. Q. Huang, “A two-dimensional MoS2/SnS heterostructure for promising photocatalytic performance: first-principles investigations,” Physica E: Low-dimensional Systems and Nanostructures 126, 114453 (2021). [CrossRef]
28. M. Diao, H. Li, R. Hou, Y. Liang, J. Wang, Z. Luo, Z. Huang, and C. Zhang, “Vertical heterostructure of SnS-MoS2 synthesized by sulfur-preloaded chemical vapor deposition,” ACS Appl. Mater. Interfaces 12(6), 7423–7431 (2020). [CrossRef]
29. H. Rashid, K. S. Rahman, M. I. Hossain, N. Tabet, F. H. Alharbi, and N. Amin, “Prospects of molybdenum disulfide (MoS2) as an alternative absorber layer material in thin film solar cells from numerical modeling,” Chalcogenide Lett. 11(8), 397–403 (2014).
30. G. Yoo, S. Lee, B. Yoo, C. Han, S. Kim, and M. S. Oh, “Electrical contact analysis of multilayer MoS2 transistor with molybdenum source/drain electrodes,” IEEE Electron Device Lett. 36(11), 1215–1218 (2015). [CrossRef]
31. J. Kang, W. Liu, and K. Banerjee, “High-performance MoS2 transistors with low-resistance molybdenum contacts,” Appl. Phys. Lett. 104(9), 093106–7 (2014). [CrossRef]
32. S. Adachi, Optical Constants of Crystalline and Amorphous Semiconductors (Springer, 1999).
33. S. N. Nwankwo, S. Campbell, R. K. T. Reddy, N. S. Beattie, V. Barrioz, and G. Zoppi, “Temperature controlled properties of sub-micron thin SnS films,” Semicond. Sci. Technol. 33(6), 065002 (2018). [CrossRef]
34. J. Q. Hu, X. H. Shi, S. Q. Wu, K. M. Ho, and Z. Z. Zhu, “Dependence of electronic and optical properties of MoS2 multilayers on the interlayer coupling and van hove singularity,” Nanoscale Res. Lett. 14(1), 1–13 (2019). [CrossRef]
35. M. Bernardi, M. Palummo, and J. C. Grossman, “Extraordinary sunlight absorption and one nanometer thick photovoltaics using two-dimensional monolayer materials,” Nano Lett. 13(8), 3664–3670 (2013). [CrossRef]
36. S. Z. Karazhanov, “Impurity photovoltaic effect in indium-doped silicon solar cells,” J. Appl. Phys. 89(7), 4030–4036 (2001). [CrossRef]
37. R. Kondrotas, C. Chen, X. X. Liu, B. Yang, and J. Tang, “Low-dimensional materials for photovoltaic application,” J. Semicond. 42(3), 031701 (2021). [CrossRef]
38. F. Y. Ran, Z. Xiao, Y. Toda, H. Hiramatsu, H. Hosono, and Toshio Kamiya, “n-type conversion of SnS by isovalent ion substitution: geometrical doping as a new doping route,” Sci. Rep. 5(1), 10428 (2015). [CrossRef]
39. A. Kuddus, M. F. Rahman, J. Hossain, and A. Bakar Md Ismail, “Enhancement the performance of CdS/CdTe heterojunction solar cell using TiO2/ZnO Bi-layer ARC and V2O5 BSF layers: a simulation approach,” Eur. Phys. J. Appl. Phys. 92, 20901 (2020). [CrossRef]
40. B. Swatowska, T. Stapinski, K. Drabczyk, and P. Panek, “The role of antireflective coatings in silicon solar cells - the influence on their electrical parameters,” Opt. Appl. 41(2), 487–492 (2011).
41. K. H. Maria, P. Sultana, and M. B. Asfia, “Chemical bath deposition of aluminum doped zinc sulfide thin films using non-toxic complexing agent: effect of aluminum doping on optical and electrical properties,” AIP Adv. 10(6), 065315 (2020). [CrossRef]
42. S. Yamaga, A. Yoshikawa, and H. Kasai, “Growth and properties of iodine-doped ZnS films grown by low-pressure MOCVD using ethyliodine as a dopant source,” J. Cryst. Growth 106(4), 683–689 (1990). [CrossRef]
43. S. Tobbeche, S. Kalache, M. Elbar, M. N. Kateb, and M. R. Serdouk, “Improvement of the CIGS solar cell performance: structure based on a ZnS buffer layer,” Opt. Quant Electron 51(8), 284 (2019). [CrossRef]
44. A. Rolland, L. Pedesseau, M. Kepenekian, C. Katan, Y. Huang, S. Wang, C. Cornet, O. Durand, and J. Even, “Computational analysis of hybrid perovskite on silicon 2-T tandem solar cells based on a Si tunnel junction,” Opt. Quantum Electron. 50(1), 21 (2018). [CrossRef]
45. S. G. Kumar and K. S. R. K. Rao, “Physics and chemistry of CdTe/CdS thin film heterojunction photovoltaic devices: Fundamental and critical aspects,” Energy Environ. Sci. 7(1), 45–102 (2014). [CrossRef]
46. H. Heriche, Z. Rouabah, and N. Bouarissa, “New ultra thin CIGS structure solar cells using SCAPS simulation program,” Int. J. Hydrogen Energy 42(15), 9524–9532 (2017). [CrossRef]
47. J. Hossain, M. M. A. Moon, B. K. Mondal, and M. A. Halim, “Design guidelines for a highly efficient high-purity Germanium (HPGe)-based double-heterojunction solar cell,” Optics & Laser Technology 143, 107306 (2021). [CrossRef]
48. A. E. H. Benzetta, M. Abderrezek, and M. E. Djeghlal, “Contribution to improve the performances of Cu2ZnSnS4 thin-film solar cell via a back surface field layer,” Optik 181, 220–230 (2019). [CrossRef]
49. L. Cai, W. Wang, L. Jin, Z. Yao, W. Lin, L. Meng, B. Ai, Z. Liang, Y. Huang, F. Zhang, P. P. Altermatt, and H. Shen, “12.29% Low temperature–processed dopant-free CdS/p-Si heterojunction solar cells,” Adv. Mater. Interfaces 6(12), 1–7 (2019). [CrossRef]
50. 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]
51. 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), 253302–4 (2010). [CrossRef]
