Concurrent investigation of antimony chalcogenide (Sb₂Se₃ and Sb₂S₃)-based solar cells with a potential WS₂ electron transport layer

Abstract

Antimony (Sb) chalcogenides such as antimony selenide (Sb₂Se₃) and antimony sulfide (Sb₂S₃) have distinct properties to be used as absorber semiconductors for harnessing solar energy including high absorption coefficient, tunable bandgap, low toxicity, phase stability. The potentiality of Sb₂Se₃ and Sb₂S₃ as absorber material in Al/FTO/Sb₂Se₃(or Sb₂S₃)/Au heterojunction solar cells (HJSCs) with 2D tungsten disulfide (WS₂) electron transport layer (ETL) layer has been investigated numerically using SCAPS-1D solar simulator. A systematic investigation of the impact of physical properties of each active material of Sb₂Se₃, Sb₂S_3, and WS₂ on photovoltaic parameters including layer thickness, carrier doping concentration, bulk defect density, interface defect density, carrier generation, and recombination. This study emphasizes the exploration of causes of low performance of actual devices and demonstrates the individual variation in the open-circuit voltage (V_OC), short-circuit current density (J_SC), fill factor (FF), power conversion efficiency (PCE) and quantum efficiency (QE). Thereby, highly potential heterostructures of Al/FTO/WS₂/absorber (Sb₂Se₃ or Sb₂S₃)/Au proposed, in which, the PCE over 28.20 and 26.60% obtained with V_OC of 850 and 1230 mV, J_sc of 38.0 and 24.0 mA/cm², and FF of 86.0 and 89.0% for Sb₂Se₃ and Sb₂S₃ absorber, respectively. These detailed findings revealed that the Sb-chalcogenide heterostructure with potential WS₂ ETL can be used to realize the fabrication of feasible thin film solar cells and thus the design of high-efficiency high-current (HEHC) and high-efficiency high-voltage (HEHV) solar panels.

Graphical abstract

Highlights

• •Antimony chalcogenide (Sb₂Se₃ and Sb₂Se₃)-based TFSCs with WS₂ electron transport layer were studied by SCAPS-1D simulator.
• •Systematic investigation on the impacts of thickness, doping, bulk, and interface defect densities on the PV performance.
• •PCE of 28.20% (26.60%) was found in a 1280 nm thick n⁺/n/p junction Sb₂Se₃ (Sb₂S₃) solar cell under adjusted condition.
• •The simulation was verified with the Shockley–Queisser (SQ) limit including experimental as well as simulation works.

Heterojunction solar cell; Sb₂Se₃; Sb₂S₃; WS₂ electron transport layer; SCAPS-1D.

1. Introduction

Photovoltaic technologies provide a clean, eco-friendly, cost-effective, and long-term solution to meet rising worldwide energy demand by directly converting solar energy into electricity [[1], [2], [3], [4], [5]]. Different types of solar cells have been technologically advanced to meet increasing energy demands [[6], [7], [8], [9]], including silicon (Si) [[10], [11], [12], [13], [14], [15]], cadmium telluride (CdTe), copper indium gallium selenide (CIGS), copper zinc tin sulfide (CZTS), polymer, inorganic metal chalcogenide, dye-sensitized solar cell (DSSC) [[16], [17], [18]], quantum dot (QD), and perovskite-based solar cells [[19], [20], [21], [22]].

A single-junction solar cell with a bandgap of around 1.14 eV of the absorber layer can provide the highest power conversion efficiency (PCE) of 33.3%, according to the Shockley-Queisser (SQ) model [23]. Thin-film solar cells (TFSCs) based on silicon, CIGS, and CdTe, on the other hand, have attained PCEs of over 20% [6,24,25]. Despite the abundance of Si on earth, the rarity of In, Ga, and Te, as well as the toxicity of Cd, limit their use on a broad scale, which has been one of the key challenges for CIGS and CdTe solar cells [26,27]. As a result, in the last decade, various novel earth-abundant absorber materials, such as Cu₂SnS₃ [28], Cu₂GeS₃ [29], Cu₂(Sn,Ge)S₃ [30], GeSe [31], CuSbS₂ [32], CuSbSe₂ [33], Cu₂ZnSnSe₄ [34], FeS₂ [35], FeSi₂ [36], SnSe [37], Sb₂Se₃ [38], and Sb₂S₃ [39], have been suggested as an alternative for the eco-friendly, and golden triangle requirements: low cost, high PCE, and long-term durability of TFSCs [1].

Among them, Sb₂Se₃ and Sb₂S₃ have attracted considerably and advanced hurriedly [[39], [40], [41], [42], [43], [44]] due to their excellent photovoltaic properties, such as a suitable bandgap (1.1–1.7 eV) for single-junction solar cells [45], a high absorption coefficient (>10⁵ cm⁻¹ at visible light) [46], a low melting point (550 °C) [47], low-cost constituents [48], and low structural complexity with only one crystallographic phase [49].

To fabricate Sb₂Se solar cells, several major approaches [40,48,50] with different buffer layers [40,51,52] have been used. It's worth noting that while Sb₂Se has been investigated for more than 40 years, its usage as absorber material in solar cells has only recently become popular, with significant progress occurring since 2009. A comparative study of experimentally fabricated Sb₂Se₃-and Sb₂S₃-based heterojunction solar cells (HJSCs) are summarized in Tables S1 and S2 in the supplementary materials. Overall, the first experimental efficiency of the Sb₂Se₃ and Sb₂S₃ TFSC in 2009 was 0.66% [53] and 3.7% [54] which improved to 9.2% in 2019 [55] and 6.53% in 2020 [56], respectively. Lin et al. fabricated glass/Mo/Sb₂Se₃/CdS/ITO/Ag solar cells using the sputtering technique and delivered PCE of 7.43% [57]. In recent times, Tang et al. have achieved PCE 8.64% by sputtering technique with record V_OC of 0.52 V [58].

However, using the SCAPS-1D simulator, multiple Sb₂Se [[59], [60], [61]] and Sb₂Se₃ [62] based solar cell architectures with varying electron transport layers (ETLs) [63,64] and hole transport layers (HTLs) [65,66] have recently been explored, with reasonably good photovoltaic (PV) performance. Basak et al. have recently explored the Sb₂S₃ (or Sb₂Se₃)/CdS heterojunction solar cell (HJSC) in SCAPS-1D and found PCEs of 9.51% and 12.62%, respectively, for Sb₂S₃ and Sb₂Se₃ HJSC [67]. However, according to the review of simulation-based studies on these solar cells, the maximum efficiency is 23.18% [65]. Yet their theoretical PCE is inferior to that of comparable commercial TFSCs [24,68]. Furthermore, while CdS is widely used as a window or buffer layer in PV devices due to its appropriate bandgap [69], the toxicity of Cd [70] leads researchers to continue looking for other environmentally friendly and wide bandgap materials to use as window or buffer layer in antimony (Sb) chalcogenide-based solar cells [70]. In this perspective, tungsten disulfide (WS₂) offers relaxed carrier transport by acceptable band alignment with Sb₂Se₃ and Sb₂S₃-based solar cells, while also being earth-abundant, economical, adhesive, and non-toxic [70]. Consequently, performing systematic numerical analysis in the SCAPS-1D simulator, we have designed, simulated, and proposed simplified designs for the efficiency enhancement of Sb₂Se₃ and Sb₂S₃-based solar cells employing WS₂ as ETL. Our proposed devices, with their optimized parameters, provide superior results to the devices proposed earlier [[65], [66], [67]].

2. Modeling and simulation

Figure 1 shows the schematic structure and band alignment of proposed Al/FTO/WS₂/Sb₂Se₃ (or Sb₂S₃)/Au HJSC structure, where FTO, WS₂, and Sb₂Se₃ (or Sb₂S₃) are used as the n⁺-type window layer, n-type electron transport, or buffer layer, and the p-type active absorber layer, respectively.

Figure 1.

(a) Proposed Al/FTO/WS₂/Sb₂Se₃(or Sb₂S₃)/Au HJSC, and (b) corresponding band diagram.

In this work, a one-dimensional solar cell capacitance simulator (SCAPS-1D) of version 3.3.07 [71,72] has been used to conduct the whole investigation. Though the SCAPS simulator allows researchers to explore devices with up to seven layers [69], the current structure of the designed solar cell with only three layers has a substantial impact on PV performance, as will be described in subsequent section 3. Light travels through the FTO-coated glass substrate and WS₂ ETL and finally reaches the absorber layer in this approach. FTO is a common material for substrate in HJSCs due to its chemical and mechanical durability, thermal stability, low toxicity, and inexpensive raw material and processing costs [73,74] which is why we used this substrate to achieve a higher value for the design cell's performance to cost ratio. However, considering the practical constraints we assumed 90% transmission (visible absorption of FTO is 0.04) of light incident onto the surface of the FTO layer [69].

Since, Au is a noble metal having two major advantages: (a) chemically inert and (b) high work function metal (5.1–5.3 eV), which is required for hole extraction efficiently at the back interface in p-type absorbers (VBM of 5.36 eV) forming ohmic contact, Au was chosen as back metal contact. J. Zhang et al reported that the forward current in Sb₂Se₃ solar cells with Al(WF of 4.0–4.2 eV) back contact is very low even under large forward bias (0.2 mA/cm² at 1 V, owing to a large barrier at the back interface for hole extraction and consequently blocking current across the solar cell significantly [75]. Although they proposed NiO_x/Ni as a promising one alike Au as back contact, Au was preferred to investigate the efficacy of the absorber and ETL layer in detail with no use of HTL or any HTL like NiO_x as recommended. On the other hand, Al (WF of 4.0–4.2 eV) was used as a front metal electrode with FTO (4.0 eV) considering both the work function for obtaining ohmic contact and its availability. Since, Aluminum is the most abundant metal on Earth, as well as one of the cheapest metals, therefore, Al was chosen as the front metal contact with FTO TCO.

Nevertheless, SCAPS-1D software was developed based on the semiconductor principal equations and is largely used to model semiconductor solar cells [71]. More particularly, the Poisson, hole, and electron continuity equations are employed to determine the numerical assessments of the modeled solar cell structure, as shown in Eqs. (1), (2), and (3), respectively [65,76].

$$\frac{{\partial }^2\Psi }{\partial x^2}+\frac{q}{\epsilon }\left[p\left(x\right)-n\left(x\right)+N_D+N_A+{\rho }_p-{\rho }_n\right]=0$$ (1)

$$\frac{1}{q}\frac{{\partial J}_p}{\partial x}=G_{op}-R\left(x\right)$$ (2)

$$\frac{1}{q}\frac{{\partial J}_n}{\partial x}={-G}_{op}+R\left(x\right)$$ (3)

where, Ψ, q, ε, N_A (N_D), and ρ_p (ρ_n) denote electrostatic potential, the charge of an electron, dielectric permeability, hole (electron) concentration, and hole (electron) distribution, respectively. In addition, J_P (J_n), G_OP, R, and p (n) present current density contributed by hole (electron), optical carrier generation rate, recombination rate, and free acceptor (donor) concentration, respectively.

The drift-diffusion formulas of Eqs. (4) and (5) have been used to calculate the transport properties of the donor and acceptor of semiconducting material [65,76].

$$J_p=\frac{{\mu }_{p}p}{q}\frac{{\partial E}_{FP}}{\partial x}$$ (4)

$$J_n=\frac{{\mu }_{n}n}{q}\frac{{\partial E}_{Fn}}{\partial x}$$ (5)

where, μ_P (μ_n) and E_FP (E_Fn) specify the acceptor (donor) mobility and acceptors’ (donors’) fermi level, respectively.

The designed solar cell has been simulated under global air mass (AM) 1.5 spectrum at a one sun illumination (100 mW/cm²) at a temperature T = 300 K. The thermal velocity of electron and hole was set to 10⁷ cm/s and kept constant throughout the entire simulation. Tables 1 and 2 show all other necessary layer properties of Sb₂Se₃-and Sb₂S₃-based HJSCs that are necessary for conducting the whole simulation [[60], [61], [62], [63], [64], [65], [66], [67],69,70,77,78].

Table 1.

Layer properties for simulation 60, 61, 62, 63, 64, 65, 66, 67, 69, 70, 77, 78, 79].

Parameters (unit)	FTO	WS2	Sb2Se3	Sb2S3
Thickness (nm)	50	30	1200	1200
Bandgap (eV)	3.60	2.20	1.20	1.62
Electron affinity (eV)	4	3.95	4.16	3.70
Dielectric permittivity	9	13.60	14.50	7.08
CB effective DOS (cm−3)	2.2×1018	2.2×1018	2×1018	2×1019
VB effective DOS (cm−3)	1.80×1019	1.80×1019	1019	1019
Electron mobility (cm2V−1s−1)	100	100	16.70	9.80
Hole mobility (cm2V−1s−1)	25	100	16.70	10
Donor density, ND (cm−3)	5×1018	1017 (Sb2Se3), 1018 (Sb2S3)	0	0
Acceptor density, NA (cm−3)	0	0	1013	1015
Defect type	acceptor	acceptor	donor	donor
Bulk defect density, Nt (cm−3)	1014	1018	1012	1012
Electron capture cross-section, σe (cm2)	10–15	10–15	10–15	10–15
Hole capture cross-section, σp (cm2)	10–15	10–15	10–15	10–15
Defect position above EV (eV)	0.6	0.6	0.6	0.6

Table 2.

Interface parameters used in the ETL/absorber interface of the designed HJSCs.

Parameters (unit)	WS2/Sb2Se3 Interface	WS2/Sb2S3 Interface
Defect type	Neutral	Neutral
Electron capture cross- section, σe (cm2)	10–19	10–19
Hole capture cross-section, σp (cm2)	10–19	10–19
Defect position above the highest EV (eV)	0.06	0.06
Defect density (cm−2)	1015	1010

3. Result and discussions

3.1. Impact of the absorber layer on the PV performance

3.1.1. Thickness effect

Figure 2 shows the influence of thickness variation on the solar cell performance parameters in the range of 0.3–3.0 μm at constant values of the rest all other parameters as declared in Tables 1 and 2 at 300 K. The cell's PCE increased from 23.9 to 28.5% and 20.9 to 26.6 with the increase of Sb₂Se₃ and Sb₂S₃ absorber thickness from 0.3‒ 3.0 μm, respectively. The J_sc increased linearly first in both devices corresponding to the absorber layer thickness of ≤0.75 μm and then it reached a saturated value of 38 and 24 mA/cm², in contrast, the FF tends to decrease almost linearly for Sb₂Se₃ and Sb₂S₃ absorber, respectively. This increment of J_sc is owing to the marked improvement of photon absorption at thicker Sb₂Se₃ or Sb₂S₃ absorber layer. This is reasonable because a thicker absorber has a greater number of photons to produce higher electron-hole pairs (EHPs) [69,80].

Figure 2.

Effect of absorber layer thickness (a) Sb₂Se₃ and (b) Sb₂S₃ on the PV parameters.

In Figure 3, the quantum efficiency (QE) response of the corresponding solar cells at different absorber thicknesses also reveals similar consequences as observed in J_sc. Further, an increase of cell resistance, as well as diffusion length at thicker absorber layer, causes the decrease of FF [69,81]. In addition, a negligible change in V_oc from 0.845‒ 0.848 V and 1.22–1.23 V for Sb₂Se₃ and Sb₂S₃ absorbers are observed, respectively. Thus, the highest PCE of 28.5 and 26.6% were observed at a Sb₂Se₃ and Sb₂S₃ absorber thickness of 1.2 μm at adjusted photovoltaic parameters values, which is chosen for further investigation.

Figure 3.

Impact of the absorber layer thickness of (a, b) Sb₂Se₃, and (c, d) Sb₂S₃ solar cells on the JV and QE, respectively.

3.1.2. Doping concentration effect

Figure 4 illustrates the impact of doping concentration N_A of Sb₂Se₃ and Sb₂S₃ absorber layer varied in the range of 10¹²–10¹⁹ cm⁻³ at 300 K on the solar cell parameters at a constant value of the rest others parameters as summarized in Tables 1 and 2. The change of each output parameter was affected negligibly by varied N_A to a value of 10¹⁷ cm⁻³. The J_sc dropped significantly from ∼39.0 to 31.0 mA/cm² and ∼25.0 to 20.0 mA/cm² and FF reduced from 85.0 to 59.0% and 85.0 to 59.0%, in contrast, V_oc increased from 840 to 930 mV and 1230–1295 mV for Sb₂Se₃ and Sb₂S₃ absorber respectively. The decrease of photocurrent at higher acceptor concentration owing to the domination of recombination of photogenerated hole pairs at higher carrier density. The increase of V_oc originated from the improved build-in-potential by further lowering fermi-level at higher acceptor concentration and thereby larger band offset compare with lower doping level at the absorber/buffer interface. Therefore, a sharp decrease in J_SC and therefore the PCE is observed. By adjusting the PV parameters value, the optimum N_A was found at 10¹³ and 10¹⁵ cm⁻³ for Sb₂Se₃ and Sb₂S₃ absorbers, respectively. Thus, the PCE of 28.20 and 26.6%, V_OC of 840 and 1230 mV, J_SC of 38.40 and 24.11 mA/cm², FF of 86.55 and 89.6% at an optimum N_A value for Sb₂Se₃ and Sb₂S₃ absorbers, respectively.

Figure 4.

Effect of the doping concentration of (a) Sb₂Se₃, and (b) Sb₂S₃ absorber layer on the solar cell performance parameters.

Figure 5 shows the impact of the doping concentration of Sb₂Se₃ and Sb₂S₃ absorber layers on the JV and QE spectra of the corresponding solar cells, respectively. The response of QE reduces markedly in the case of both absorber configurations when the hole doping level exceeds 10¹⁷ cm⁻³ which further reveals the recombination of the photogenerated carrier as expected from an increase in the recombination of free carrier charges within the bulk. The lower energy (longer wavelength) photons are absorbed noticeably in the absorber layer. Consequently, a dramatic impact of the doping concentration on the collected conversion efficiency was observed [82].

Figure 5.

Impact of the absorber layer doping concentration of (a, b) Sb₂Se₃, and (c, d) Sb₂S₃ solar cells on the JV and QE, respectively.

3.1.3. Effect of the carrier concentration of the absorber layer on the G-R profile

Figure 6 demonstrates the influence of carrier (electron and hole) concentration and total generation-recombination (G-R) profiles as a function of Sb₂Se₃ and Sb₂S₃ absorber layer thickness. The effective density of states (DOS) in the valence bands, the hole concentration of Sb₂S₃ is slightly larger than that of Sb₂Se₃ owing to the difference in acceptor concentration in the absorbers, and. In contrast, the electron concentration of the Sb₂S₃ absorber is lower than that of the Sb₂Se₃ absorber. The carrier generation and recombination profiles obtained by a systematic study revealed the potentiality of each Sb-chalcogenides-based absorber of Sb₂S₃ and Sb₂Se₃ compare with reported inorganic, organic, and compound semiconductor materials used as an absorber at a specified and adjusted carrier concentration and defect density [[83], [84], [85], [86], [87]]. Thus, Sb-chalcogenides-based Sb₂S₃ and Sb₂Se₃ appear of much more promising and efficient absorber materials, which can be used for the fabrication of high-performance thin film solar cells with lower e-h recombination, in contrast, higher carrier generation.

Figure 6.

Influence of (a) hole and (b) electron carrier concentration, and total (c) generation and (d) recombination concerning the absorber layer thicknesses.

3.2. Impact of WS₂ ETL layer on the photovoltaic performance

Figure 7 shows the effect of thickness and doping concentration of WS₂ ETL on the performance of Sb₂Se₃-based solar cell devices in the range of 0.01–0.21 μm and 10¹⁴–10²¹ cm⁻³ respectively, keeping unchanged all other associated parameters as summarized in Tables 1 and 2. The V_OC increases from 0.8483 to 0.9293 V, while the J_SC decreases from 38.40 to 31.63 mA/cm² with an increase of donor concentration from 10¹² to 10¹⁹ cm⁻³. The increment of V_OC originated from the improved electric field with higher resultant V_bi developed at higher donor concentration in WS₂ ETL. On the other hand, the availability of a higher density of carriers causes a larger recombination rate of photogenerated carriers, consequently reducing J_SC [88,89]. In addition, an insignificant impact of donor concentration on FF was observed. However, fewer photons may reach the absorber layer when the ETL thickness increases, resulting in a reduction in EHP generation due to parasitic absorption by the ETL itself. Consequently, the JSC and the PCE of both the solar cells tend to decrease [36,69,81]. Thus, the highest PCE of ∼29.0% was observed at WS₂ ETL thickness of ∼0.03 μm and carrier concentration of ≤ 10¹⁷ cm⁻³.

Figure 7.

Impact of thickness and doping concentration of WS₂ ETL on the photovoltaic performance (a) J_sc, (b) V_oc, (c) FF, and (d) PCE of Sb₂Se₃-based solar cell.

Figure 8 demonstrates the impact of thickness and doping concentration of WS₂ ETL on the performance of Sb₂S₃-based solar cells in the range of 0.01–0.21 μm and 10¹⁴–10²¹ cm⁻³ respectively, at a constant of all other parameters as summarized in Tables 1 and 2. The V_OC increases from 1.21 to 1.23 V, in contrast, the J_SC decreases from 24.11 to 20.52 mA/cm² with the increase of donor concentration from 10¹² to 10¹⁹ cm⁻³. The increment of V_OC originated by the improved electric field with enhanced resultant V_bi developed at higher donor concentration in WS₂ ETL as observed in the case of Sb₂S₃-based cells. On the other hand, higher carrier density causes increased recombination of photogenerated carriers, consequently, reduced J_SC observed. The photons of longer wavelengths are deeply absorbed in the absorber layer at higher doping concentration owing to a smaller mean free path at a higher carrier density state.

Figure 8.

Impact of thickness and doping concentration of WS₂ ETL on the photovoltaic performance (a) J_sc, (b) V_oc, (c) FF, and (d) PCE of Sb₂S₃-based solar cell.

Thus, the optimum thickness of WS₂ ETL was found of 0.03 μm with a donor concentration of 10¹⁷ and 10¹⁸ cm⁻³ for both Sb₂Se₃ and Sb₂S₃ absorber-based heterostructures, respectively, considering the tradeoff condition among solar cell parameters.

3.3. Influence of the bulk and interface defect density on the PV performance

Defects are mainly originated from dislocations and grain boundaries (GBs) in the synthesized layers. The defect density reduces the carrier lifetime and mobility markedly, which in turn acts as a carrier trapping center or carrier recombination [90]. Thus, cell performance is significantly affected. A detailed study on the impact of bulk and interface defect density has been performed in this section.

3.3.1. Effect of the absorber's bulk defect density

The influence of the bulk defect density of the Sb₂Se₃ and Sb₂S₃ absorber layers on the PV parameters was explored in this section, and the findings are presented in Figure 9.

Figure 9.

Impact of the bulk defect density of (a) Sb₂Se₃, and (b) Sb₂S₃ absorber layer on the photovoltaic performance.

The carrier recombination rate in the p-type materials can be defined by Eq. (6) [90],

$$\frac{dn}{dt}=-k_3{n}^3-k_2{n}^2-k_{1}n$$ (6)

where K_3, K₂, and K₁ are the 1^st, 2^nd, and 3^rd order decay constant. Herein, k_i (i = 1,2,3) refers to an interaction between i carriers (electrons/holes/excitons, depending on the species corresponding to the feature). The k₃ represents Auger recombination whereby a third carrier absorbs the energy released from a recombination event and so on, k₂ represents bimolecular recombination with exciton-exciton annihilation, k₁ corresponds either recombination of excitons or recombination between a free carrier and a trapped (localized) carrier [91]. The maximum power loss in the solar cell is due to the non-radiative SRH recombination process [90,92,93] and the defects in the absorber layer mostly cause the SRH recombination, which is generally defined by Eq. (7),

$$R_{SRH}=\frac{np-n_i^2}{\tau \left(p+n-2n_i\cosh \left(\frac{E_i-E_t}{kT}\right)\right)}$$ (7)

where τ, E_t, k, and T represent the charge carrier lifetime, defect energy level within the bandgap, Boltzmann constant, and solar cell operating temperature, respectively. However, carrier lifetime (τ) has been calculated using the following Eq. (8) [76],

$$\tau =\frac{1}{\sigma \times N_t\times V_{th}}$$ (8)

where σ, N_t, and V_th represent the capture cross-section area of the charge carrier, defect concentration, and charge carrier thermal velocity, respectively.

In Figure 9, the PCE is almost unchanged with a defect density of 10¹³ cm⁻³ and 10¹² cm⁻³ for the Sb₂Se₃ and Sb₂S₃-based devices. The FF, V_OC, and J_SC decreased noticeably, and consequently, the PCE decreased from 28 to 18% and from 27 to 19% in Sb₂Se₃ and Sb₂S₃-based solar cells corresponding to defect density of 10¹⁰–10¹⁶ cm⁻³. Higher SRH-recombination rate is dominant at a defect density of ≥ 10¹³ cm⁻³ which deteriorates the cell performance markedly [67]. Thus, a defect density level of < 10¹³ cm⁻³ is required to achieve the highest cell performance.

3.3.2. Effect of the WS₂ ETL's bulk defect density

Figure 10 shows the impacts of ETL's bulk defect density on the solar cell's parameters in the range of 10¹⁰‒ 10²⁰ cm⁻³, while all other parameters remained unchanged as summarized in Tables 1 and 2. It's worth noting that the PV performance metrics of both solar cells degrade noticeably with increased ETL defects. Even though the values of the designed two solar cells' cell parameters are different in magnitude, the trend of lowering the cell performance is very similar to an increase in defects level. Although the PCE retains constant at the highest value up to 10¹⁸ cm⁻³, it decreases linearly from ∼28 to ∼18% and from ∼27.0 to ∼19% when the defect increases from 10¹⁸ to 10²⁰ cm⁻³. Thus, the defect density is required to be 10¹⁸ cm⁻³ for achieving the highest cell performance.

Figure 10.

Impact of the WS₂ ETL bulk defect density on the photovoltaic performance of (a) Sb₂Se₃, and (b) Sb₂S₃ solar cells.

3.3.3. Effect of the ETL/absorber interface defect density

Figure 11 demonstrate the influence of the interface defect density of the ETL/absorber interface on device performance at the defect density of 10¹⁰–10¹⁸ cm⁻² at a WT of 300 K, at an unchanged other parameters value as described in Tables 1 and 2.

Figure 11.

Influence of the ETL/absorber interface defect density on the device performance of (a) Sb₂Se₃, and (b) Sb₂S₃ solar cells.

The impact of defect density on cell performance is almost negligible until the interface defect density reaches 10¹⁵ cm⁻² for the WS₂/Sb₂Se₃ interface and 10¹¹ cm⁻³ for the WS₂/Sb₂S₃ interface. As perceived from Figure 11, the interface of the Sb₂S₃ solar cell has a more critical impact on the PV characteristics than the Sb₂Se₃ solar cell beyond these marginal values in both devices. All of the metrics drop dramatically as their charge carriers recombine notably with the opposing charge carriers before reaching the junction and at a lower rate of creating EHPs [6,81].

In addition, the effect of temperature variation of Sb₂Se₃ and Sb₂S₃ based heterojunction solar cell is shown in Figure S1. It has been revealed that at high temperatures, the efficiency of Sb₂Se₃ and Sb₂S₃ based heterojunction solar cell reduces, which is consistent with the reported works as well [94].

3.4. The J-V and QE spectra of the optimized solar cells

The current density-voltage (J- V) curve and the corresponding QE spectrum as a function of the light wavelength of the optimized HJSCs are shown in Figure 12. In Figure 12(a), the simulation results of voltage (V_OC), current (J_SC), fill factor (FF) and efficiency (ɳ) of the optimized Sb₂Se₃ (Sb₂S₃) proposed solar cells is 850 mV (1230 mV), 38.40 mA/cm² (26.60 mA/cm²), 86.56% (89.60%) and 28.20% (24.11%), respectively. Figure 12(b) reveals that the QE of the Sb₂Se₃ (Sb₂S₃) structure falls to almost 0% at the wavelength of 800 nm (1050 nm), while it is over 97% at a lower value of these wavelength values for Sb₂Se₃ and Sb₂S₃ HJSCs. The degree of current improvement depends on the band gap of the absorber material, reversely, the value of V_oc enhanced due to the generation of high built-in potential at the absorber interface, therefore, having an almost ideal band gap for absorbing visible spectra significantly, Sb₂Se₃ and Sb₂S₃ appeal great attention for the fabrication of high-efficiency HJSCs [59,65,67].

Figure 12.

The (a) J-V characteristics, and (b) QE spectra of the proposed Sb₂Se₃ and Sb₂S₃ heterojunction solar cells.

Table 3 compares PV performance to previously report experimental and simulation studies. The simulation may be easily verified because the results are consistent with prior studies and are within the SQ limit for single-junction solar cells. Both solar cells proposed here, however, outperform the devices proposed previously [24,61,[64], [65], [66], [67]].

Table 3.

Optimized PV parameters of the proposed HJSCs with comparison to similar HJSCs in the literature. [Theo. = Theoretical SQ limit, Exp. = Experimental, Sim. = Simulation].

Absorber material	Types	Junction Formed	JSC (mA/cm2)	VOC (mV)	FF (%)	PCE (%)	Ref.
Sb2Se3	Theo.	Homo	39.99	935	87.70	32.74	[95]
	Exp.	Hetero	30.80	423	58.10	7.50	[96]
	Exp.	Hetero	32.58	400	70.30	9.20	[97]
	Exp.	Hetero	30.86	488	67.19	10.12	[98]
	Sim.	Hetero	38.15	410	74.08	11.52	[65]
	Sim.	Hetero	38.40	850	86.56	28.20	∗
Sb2S3	Theo.	Homo	25.47	1309	90.50	30.14	[95]
	Exp.	Hetero	14.73	645	65.69	6.27	[61]
	Exp.	Hetero	15.29	748	57.07	6.53	[64]
	Sim.	Hetero	23.73	970	72.32	16.65	[24]
	Sim.	Hetero	24.00	1230	89.60	26.60	∗

^∗This work.

4. Conclusions

The sb-chalcogenides of Sb₂Se₃ and Sb₂S₃-based high-efficiency HJSCs with WS₂ ETL were investigated numerically using SCAPS-1D solar cell simulator. While comprehensive optimization research was carried out, their photovoltaic performance was compared to that found in the literature. The highest PCE of 28.20% with V_OC of 850 mV, J_SC of 38.40 mA/cm², and FF of 86.56% was obtained from Sb₂Se₃ absorber-based heterostructure, while the highest PCE of 26.60% with V_OC of 1230 mV, J_SC of 26.60 mA/cm², and FF of 89.60% was obtained from Sb₂S₃ absorber based heterostructure together with WS₂ ETL layer. Simulation results indicate that WS₂ could be a competitive ETL for fabricating low-toxicity, cost-effective, and highly efficient Sb₂Se₃ and Sb₂S₃-based HJSCs for harnessing solar energy. Furthermore, these devices' higher performance compared with previous reports reveals the high potentiality of Sb-chalcogenide as an absorber material that could be efficient one to realize both high-efficiency high-current (HEHC) and high-efficiency high-voltage (HEHV) solar panels.

Declarations

Author contribution statement

Md. Ferdous Rahman: Conceived and designed the experiments; Performed the experiments; contributed reagents, materials, analysis tools or data; analyzed and interpreted the data; wrote the paper.

Md. Mahabub Alam Moon, Md. Hasan Ali, Md. Dulal Haque, Abdul Kuddus, Jaker Hossain and Abu Bakar Md. Ismail: analyzed and interpreted the data; wrote the paper.

M. Khalid Hossain: Conceived and designed the experiments; analyzed and interpreted the data; wrote the paper.

Funding statement

Prof. Dr. Md. Ferdous Rahman was partially supported by an innovation fund from the ICT Division, Govt. of Bangladesh [Grant No. 22IF15426 (2021–2022)].

Data availability statement

Data will be made available on request.

Declaration of interest's statement

The authors declare no conflict of interest.

Additional information

No additional information is available for this paper.

Appendix A. Supplementary data

The following is the supplementary data related to this article: