Swapnil
Barthwal
a,
Siddhant
Singh
a,
Kumar
Haunsbhavi
b,
Ramashanker
Gupta
c,
V. Vadhana
Sharon
d,
Shivaraj
Maidur
d,
Abhishek K.
Chauhan
a,
David E.
Motaung
b,
Rahul
Kumar
e and
Ramesh
Karuppannan
*a
aDepartment of Physics, Indian Institute of Science, Bengaluru 560012, India. E-mail: kramesh@iisc.ac.in
bDepartment of Physics, University of the Free State, P.O. Box 339, Bloemfontein ZA9300, South Africa
cBCMaterials, BasqueCenter for Materials, Applications and Nanostructures, Martina Casiano, UPV/EHU Science Park, 48950 Leioa, Spain
dDepartment of Physics, Kristu Jayanti (deemed to be) University, Bengaluru 560064, India
eDepartment of Physics and Material Science, Thapar Institute of Engineering and Technology, Patiala 140074, India
First published on 13th November 2025
Antimony selenosulfide [Sb2(S,Se)3] is a scientifically interesting, and technologically intriguing photovoltaic (PV) material for the next generation of solar cells. Recently, power conversion efficiencies (PCEs) of 10.92% and 20.86% have been achieved in single-junction Sb2(S,Se)3 cells, under standard (AM1.5G) and indoor illumination (1000 lux), respectively. Prototype Si/Sb2(S,Se)3 and Sb2Se3/Sb2(S,Se)3 tandem solar cells have demonstrated PCEs exceeding 10%. However, various intractable factors, mainly the anisotropic carrier transport, anion-vacancy (VS/Se) and cation anti-site (SbS/Se) defects, and non-optimized interfaces cumulate to notable photocurrent and photovoltage losses in Sb2(S,Se)3 solar cells. A comprehensive understanding of these performance-limiting factors can be instrumental in amplifying the PCE of Sb2(S,Se)3 solar cells, beyond state-of-the-art. In this context, this review provides a comprehensive discussion on the device engineering strategies and establishes a robust framework for the fabrication of high PCE (>15%) Sb2(S,Se)3 solar cells.
In this context, antimony selenosulfide [Sb2(S,Se)3] has emerged as a promising photon-harvesting material for next-generation PV technology. This quasi one-dimensional (Q-1D) semiconductor offers the advantages of non-toxic and earth-abundant constituents, exceptional optoelectronic properties and high physicochemical stability. It exhibits a unique combination of strong optical absorption (α > 105 cm−1), tunable bandgap (1.1–1.7 eV, depending on the S/Se atomic ratio), low Urbach and exciton binding energy values, ultra-flexibility, and solution-processibility. These merits endorse it as a compelling alternative to the conventional PV materials.9–14 Recent advancements have demonstrated significant improvements in power conversion efficiency (PCE) for Sb2(S,Se)3-based solar cells, achieving notable results in both single-junction and tandem configurations. Sb2(S,Se)3 solar cells are also gaining traction in niche applications such as indoor photovoltaics (IPV), building integrated photovoltaics (BIPV), and photo-integrated rechargeable batteries/supercapacitors, showcasing their versatility across diverse energy-harvesting environments.15–23
Sb2(S,Se)3 films are readily obtained in the desired stoichiometry via various synthesis routes, such as vacuum-based processes (pulsed laser deposition,24,25 sputtering,26 thermal evaporation,27 close space sublimation,28 vapor transport deposition29) and solution-processing techniques (hydrothermal,6–16 spin coating,30,31 chemical bath deposition32). Despite the diversity in synthesis routes, all the high PCE (>10%) Sb2(S,Se)3 solar cells (summarized in Table 1) exclusively consist of hydrothermally synthesised, Sb-rich, n-type Sb2(S,Se)3 thin-films, sandwiched between CdS and Spiro-OMeTAD, serving as an electron and hole transport layer (ETL and HTL), respectively. The lowest open-circuit voltage (VOC)-deficit and fill factor (FF) values achieved in Sb2(S,Se)3 solar cells are 0.48 V33 and 73.14%,34 respectively. These values disclose the availability of significant scope for further device optimization and efficiency enhancement.
| Device architecture | Deposition technique | Novelty | Absorber bandgap (eV) | Device parameters | V OC-deficit (V) | Ref. | |||
|---|---|---|---|---|---|---|---|---|---|
| V OC (V) | J SC (mA cm−2) | FF (%) | PCE (%) | ||||||
| FTO/SnO2/CdS/Sb2(S,Se)3/Spiro-OMeTAD/Au | Hydrothermal | Texturization of FTO substrate to enhance light scattering and maximize charge-carrier generation | 1.30 | 0.569 | 27.38 | 70.04 | 10.92 (certified as 10.70) | 0.48 | 33 |
| FTO/CdS/Sb2(S,Se)3/Spiro-OMeTAD/Au | Incorporation of sodium selenosulfate for in situ defect passivation of deep-level cation antisite (SbSe) defect | 1.44 | 0.633 | 24.91 | 68.60 | 10.81 | 0.54 | 10 | |
| FTO/CdS/Sb2(S,Se)3/Spiro-OMeTAD/Au | LiF doping in CdS to improve its morphology and conductivity, Li-diffusion into Sb2(S,Se)3 passivates VSe and SbS defects | 1.53 | 0.673 | 24.45 | 65.39 | 10.76 | 0.59 | 35 | |
| FTO/CdS/Sb2(S,Se)3/Spiro-OMeTAD/Au | Ethanol assisted additive hydrothermal route, to synthesize Sb2(S,Se)3 thin-films with mitigated defect-density and reformed morphology | 1.43 | 0.630 | 25.27 | 67.35 | 10.75 | 0.54 | 11 | |
| FTO/Zn(O,S)/CdS/Sb2(S,Se)3/Spiro-OMeTAD/Au | Alkali metal fluoride (NaF, KF, RbF, CsF) assisted solution post treatment of Sb2(S,Se)3 thin-films to enhance the crystallinity, conductivity, and energy-level alignment | 1.51 | 0.673 | 23.7 | 66.8 | 10.70 | 0.57 | 12 | |
| FTO/CdS/Sb2(S,Se)3/Spiro-OMeTAD/Au | Na2HPO4 additive to promote [hk1]-oriented growth of Sb2(S,Se)3 films, along with mitigation of SbS defects | 1.48 | 0.666 | 25.19 | 63.60 | 10.67 | 0.55 | 36 | |
| FTO/CdS/Sb2(S,Se)3/Spiro-OMeTAD/Au | Alkaline sodium borohydride induced reaction kinetics regulation to mitigate oxide impurities and passivate deep-level traps | 1.48 | 0.662 | 25.07 | 64.0 | 10.62 | 0.55 | 37 | |
| FTO/SnO2/CdS/Sb2(S,Se)3/Spiro-OMeTAD/Au | Ultrathin SnO2 bridge layer to improve poor electrical and chemical contact at FTO/CdS interface, reduction in S-vacancy defect in CdS film | 1.29 | 0.540 | 26.94 | 72.99 | 10.58 | 0.50 | 34 | |
| 0.552 | 26.31 | 73.14 | 10.48 | 0.49 | |||||
| FTO/SnO2/CdS/Sb2(S,Se)3/Spiro-OMeTAD/Au | Post deposition treatment of Sb2(S,Se)3 films using MgCl2, passivation of anion vacancy defects via (Cl-doping) | 1.53 | 0.671 | 25.44 | 65.21 | 10.55 | 0.59 | 38 | |
| FTO/SnO2/CdS/Sb2(S,Se)3/Spiro-OMeTAD/Au | Temperature-gradient solution deposition to manipulate S/Se distribution and amend unfavorable band structure | 1.46 | 0.675 | 22.92 | 68.13 | 10.55 | 0.52 | 39 | |
| FTO/CdS/Sb2(S,Se)3/Spiro-OMeTAD/Au | Synthesis of high-quality Sb2(S,Se)3 films by utilizing ethylenediaminetetraacetic acid (EDTA) as a strong coordination additive, to regulate the nucleation and growth process | 1.57 | 0.664 | 23.8 | 66.3 | 10.5 | 0.63 | 9 | |
| FTO/CdS/Sb2(S,Se)3/Spiro-OMeTAD/Au | Post surface-treatment using ammonium sulfide [(NH4)2S], mitigation of SbS antisite defects | 1.46 | 0.686 | 22.04 | 68.8 | 10.41 | 0.50 | 15 | |
| FTO/CdS/Sb2(S,Se)3/Spiro-OMeTAD/Au | Hydrazine hydrate (N2H4) solution assisted post-treatment to modify CdS ETL, removal of residual Cd oxychlorides, improving CdS/Sb2(S,Se)3 interfacial band alignment | 1.52 | 0.678 | 23.63 | 66.07 | 10.30 | 0.57 | 16 | |
| FTO/CdS/Sb2(S,Se)3/Spiro-OMeTAD/Au | NH4F additive assisted regulation of the band-gap gradient in the Sb2(S,Se)3 layer and modification of the CdS/Sb2(S,Se)3 interface | 1.44 | 0.630 | 24.76 | 65.95 | 10.28 | 0.55 | 17 | |
| FTO/CdS/Sb2(S,Se)3/Spiro-OMeTAD/Au | Exploring thermally evaporated, up-scalable, stable and inorganic MnS as the HTL, to potentially substitute Spiro-OMeTAD | 1.49 | 0.646 | 24.29 | 64.5 | 10.14 | 0.58 | 18 | |
| FTO/CdS/Sb2(S,Se)3/Spiro-OMeTAD/Au | Regulation of heterogeneous nucleation via BaBr2 additive, and simultaneous mitigation of bulk and interface defects | 1.51 | 0.668 | 23.90 | 63.0 | 10.12 | 0.57 | 40 | |
| FTO/CdS/Sb2(S,Se)3/Spiro-OMeTAD/Au | Novel deposition strategy to regulate the morphology, Se/S elemental ratio, and defects in Sb2(S,Se)3 films | 1.49 | 0.630 | 23.7 | 67.7 | 10.10 | 0.59 | 19 | |
| FTO/CdS/Sb2(S,Se)3/Spiro-OMeTAD/Au | Ethylenediamine (EDA) solvent annealing of the CdS buffer layer to improve the CdS/Sb2(S,Se)3 interfacial bonding | 1.54 | 0.651 | 23.71 | 65.23 | 10.10 | 0.62 | 41 | |
| FTO/CdS/Sb2(S,Se)3/Spiro-OMeTAD/Au | Surface treatment of the CdS underlayer by KI, modulating band-alignment at CdS/Sb2(S,Se)3 from cliff-like to spike-like | 1.53 | 0.663 | 23.75 | 63.91 | 10.06 | 0.60 | 42 | |
| FTO/SnO2/CdS/Sb2(S,Se)3/Spiro-OMeTAD/Au | Employing a low-cost precursor (Na2SeSO3) to regulate the S/Se elemental ratio, tuning band gap, and suppressing defect density in Sb2(S,Se)3 films | 1.35 | 0.551 | 26.01 | 70.1 | 10.05 | 0.54 | 20 | |
| FTO/SnO2/CdS/Sb2(S,Se)3/Spiro-OMeTAD/Au | CsI doping to modulate the energy level, carrier transport, and defect-density in Sb2(S,Se)3 films | 1.55 | 0.671 | 22.93 | 65.3 | 10.05 | 0.60 | 21 | |
Existing review works on Sb2(S,Se)3 solar cells primarily emphasize the material and electrical properties of Sb2(S,Se)3, summarizing notable developments, strategies for suppressing open-circuit voltage (VOC) deficits, and defect engineering.43–50 This review is unique in providing a comprehensive and critical analysis of the latest scientific advancements in Sb2(S,Se)3 solar cells, with a particular focus on material engineering, device design, interface optimization and sustainability. It aims to bridge the knowledge gap and comprehensively address the key challenges that currently limit the device performance of Sb2(S,Se)3 solar cells, such as carrier transport anisotropy, defect management, and interface losses. Furthermore, it explores the potential strategies for surpassing PCE benchmarks for Sb2(S,Se)3 solar cells, making this review an essential reference for researchers aiming to advance the field of Sb2(S,Se)3 solar cells.
Advanced research trends in Sb2(S,Se)3 solar cells can be broadly grouped into three primary categories. First, the material optimization, which involves fine-tuning of stoichiometry, regulating bandgap gradients, controlling defect densities, and optimizing crystallographic orientation in Sb2(S,Se)3 thin films, which are critical to the performance of Sb2(S,Se)3 solar cells. Second, the focus on sustainability, where efforts are being made to develop CdS- and Spiro-OMeTAD-free Sb2(S,Se)3 solar cells, with the aim to address environmental and economic concerns, without any compromise with device performance. Third, significant progress is being made in the device engineering aspects, with innovative strategies like bulk/3D heterojunctions, bifacial solar cells, and tandem architectures. The following subsections detail these advancements in a comprehensive manner.
![]() | ||
| Fig. 1 Schematic of the energy-level alignment in the corresponding solar devices, fabricated via different routes: (a) HTD, (b) HTD + VTD (Se), and (c) HTD + VTD (Sb2Se3 + Se). (d) J–V characteristics, (e) EQE spectra and (f) dV/dJ versus (J + JSC–GV) plots for the corresponding cells. Adapted with permission from ref. 51 Copyright 2023, Royal Society of Chemistry. | ||
![]() | ||
| Fig. 2 (a) Schematic diagram of energy level alignment in Sb2(S,Se)3 devices with or without NaF–SPT. (b) SEM micrographs of the corresponding Sb2(S,Se)3 thin-films. (c) Statistical boxplots for PCE in Sb2(S,Se)3 solar cells. (d) J–V characteristics, (e) EQE spectra, and (f) SCLC spectra of Sb2(S,Se)3 solar cells treated with or without NaF–SPT process. (g) I–V characteristics of Sb2(S,Se)3 films. The schematics of the devices are shown in the insets. (h) Light intensity dependence of VOC for control and NaF–SPT Sb2(S,Se)3 devices. Adapted with permission from ref. 12 Copyright 2022, Wiley-VCH. | ||
Wang et al.51 demonstrated a hybrid growth method to regulate the bandgap gradient in Sb2(S,Se)3 thin-films. The novel strategy consisted of two steps: the first step involved hydrothermal deposition (HTD), followed by vapor transport deposition (VTD). The devices were fabricated in the superstrate architecture FTO/CdS/Sb2(S,Se)3/PTAA/Au. Leveraging the isomorphic structures of Sb2S3 and Sb2Se3, a favourable bandgap gradient in Sb2(S,Se)3 solar cells was attained by carefully engineering the S/Se elemental ratio. As a result of the benign bandgap gradient (Fig. 1b and c), devices fabricated via the hybrid methods (HTD+VTD) delivered better photovoltaic performance than the HTD devices (Fig. 1d–f). Li et al.52 and Liu et al.53 achieved a V-shaped bandgap grading of Sb2(S,Se)3 films through dual-source vapor transport deposition and (Sb2S3 and Sb2Se3) co-sublimation techniques, respectively. The bandgap gradient strategy has been found to be strategic in the simultaneous improvement of short circuit current density and photovoltage, leading to a record PCE (of 9.02%) in CdS-free Sb2(S,Se)3 solar cells.53
Zhao et al.12 demonstrated a novel solution post-treatment technique (SPT) to regulate the S/Se atomic ratio, and to induce a shallow bandgap gradient within the hydrothermally deposited Sb2(S,Se)3 films. Alkaline metal fluoride (NaF, KF, RbF, and CsF) were used as additives, and a notable improvement was obtained in the morphology, crystallinity, and conductivity of the Sb2(S,Se)3 films. In particular, Sb2(S,Se)3 films with the NaF additive exhibited favorable energy level alignment for charge transportation (Fig. 2a), compact morphology with distinct grain boundaries (Fig. 2b), lower trap-density (Fig. 2f), and higher conductivity (Fig. 2g). As a result, the NaF–SPT synthesized champion device (FTO/Zn(O,S)/CdS/Sb2(S,Se)3/Spiro-OMeTAD/Au) exhibited a remarkable PCE of 10.70%. The low value of diode ideality factor (1.28) in the ameliorated device indicated the improved diode characteristics, and suppression of non-radiative recombination pathways on adopting the NaF–SPT technique for Sb2(S,Se)3 thin-film synthesis.
S/Se elemental ratio is not only decisive in dictating the charge carrier transport in Sb2(S,Se)3 films, but also creates a tradeoff in the photocurrent and photovoltage in Sb2(S,Se)3 solar cells.54 The effect of Se-content on the absorption spectra, energy-level alignment, band-gap, and carrier-lifetimes can be visualized in Fig. 3a–d.55 Although higher Se-content narrows the bandgap, it is acknowledged to mitigate the deep-level defects, lengthen the carrier lifetimes, and promote the desired [hk1]-orientation. This vertical alignment of grain boundaries is particularly beneficial for carrier transport and extraction in Sb2(S,Se)3 films, as it presents fewer barriers to the transport of charge carriers.
![]() | ||
| Fig. 3 (a) UV-Visible absorption spectra, (b) energy-level alignment (with respect to CdS and Spiro-OMeTAD), (c) Tauc plots, and (d) transient decay kinetics (scatter) and fitting curves (solid lines) monitored at 600 nm, for the Sb2(S,Se)3 films deposited on the glass/FTO/CdS substrates. Adapted with permission from ref. 55 Copyright 2024, Wiley-VCH. | ||
Guo et al.56 demonstrated an efficient strategy for regulating crystallographic orientation, optimizing energy band alignment, and mitigating of defects in Sb2(S,Se)3 thin film solar cells (Fig. 4a–f). A monoatomic Al2O3 layer was used to engineer the CdS/Sb2(S,Se)3 interface. It was revealed that in the absence of the Al2O3 interface layer, the Sb2(S,Se)3 films predominantly display the [hk0]-orientation. In contrast, following the introduction of a monoatomic Al2O3 layer, the Sb2(S,Se)3 films exhibit a highly preferred orientation, primarily dominated by the (2 1 1) and (2 2 1) crystallographic planes, along with compact, dense and pin-hole free grains. Benefitting from the favorable [hk1]-orientation, Sb2(S,Se)3 films deposited over Al2O3 exhibited higher tunneling current (210 pA), than the control sample (150 pA). Thicker Al2O3 films were found to be detrimental for the device, particularly deteriorating the photocurrent due to their insulating properties. Furthermore, the modified films exhibited intrinsic semiconducting properties, more conducive to suppression of recombination losses and improvement in carrier collection. Consequently, the Al2O3 engineered device exhibited a PCE of 9.39% (compared to 7.24% in the control counterpart), along with significant improvement in photocurrent (Fig. 4a and b). The device based on mono-atomic Al2O3 exhibited a lower diode ideality factor (1.65), lower dark-reverse current (3.61 × 10−6 mA cm−2), higher built-in voltage (0.94 V) and lower series resistance than the control sample (Fig. 4c–f). DLTS measurements revealed two types of hole traps (SbS1 and SbS3) in the control films, while the Al2O3 engineered sample exhibited only one hole trap (SbS3) along with mitigated defect-density. In a parallel study, the Al2O3 interface layer was also found to improve CdS/Sb2(S,Se)3 interface quality and promote [hk1]-orientation, leading to improve photovoltage and PCE (of 6.25%) in the FTO/CdS/Al2O3/Sb2Se3/carbon solar cell.57
![]() | ||
| Fig. 4 (a) J–V characteristics, (b) EQE spectra, (c) dark J–V characteristics, (d) C–V plots, (e) NC–V, and (f) EIS curves of the Sb2(S,Se)3 solar cells, with and without a monoatomic Al2O3 layer. Adapted with permission from ref. 56 Copyright 2024, Elsevier. | ||
Recently, Ren et al.36 proposed a chelate engineering strategy to deposit high-quality Sb2(S,Se)3 films with preferred [hk1]-orientation, and a low density of deep-level defects. Tetrahedral (PO4)3− introduced via dibasic sodium phosphate (Na2HPO4, DSP) additive adsorbed on the polar planes (101) of the underneath CdS layer, and promoted heterogeneous nucleation and vertical [h,k,1]-oriented growth of Sb2(S,Se)3 ribbons. (PO4)3− introduction also passivated the detrimental SbS1 antisite defects. The CdS (101) surface is known to be hostile for the [hk1]-oriented growth of Sb2(S,Se)3 ribbons, owing to poor bonding at the interface. Considering the adsorption of [(SbO)3(PO4)] on the CdS (101) planes, the bonding principle at CdS/Sb2(S,Se)3 is likely broken. Fig. 5a illustrates a schematic of the mechanism for growth of Sb2(S,Se)3 film. Without DSP addition, the Cd and S atoms on the CdS (101) facet form a connection with the S(e) and Sb atoms of the Sb2(S,Se)3 film, partly forming the [hk1]-oriented Sb2(S,Se)3 ribbons. With DSP introduction, the [(SbO)3(PO4)] is adsorbed on the CdS (101) surface, in which the Cd and S atoms on the CdS (101) surface preferentially bind with the O and Sb atoms of [(SbO)3(PO4)]. As the (PO4)3− forms a tetrahedron, the three SbO+ sets on the corners of the tetrahedron are distributed at ∼109.5° to each other. Subsequently, the S(e) and Sb atoms of the Sb2(S,Se)3 layer connect with the Sb and O atoms of the [(SbO)3(PO4)], generating the [hk1]-oriented Sb2(S,Se)3 because of the spatial orientation. Sb2(S,Se)3 films formed with DSP addition (W-DSP) exhibit higher crystallinity (lower FWHM), higher texture coefficients (Fig. 5b), and lower Urbach energy (Fig. 5f). As a result of improved crystallinity, [hk1]-oriented growth, and lower density of trap states, the W-DSP device exhibits improved PCE of 10.67% (Fig. 5c), outperforming the control counterpart (8.59%).
![]() | ||
| Fig. 5 (a) Schematic illustration of the growth mechanism of the Sb2(S,Se)3 film with DSP assistance during the hydrothermal deposition process. XRD patterns (b1) and amplified XRD patterns (b2) of Sb2(S,Se)3 films with and without DSP introduction. (b3) FWHM values extracted from the XRD patterns for the (120), (211), and (221) diffraction peaks. (b4) Texture coefficients for Sb2(S,Se)3 films with and without DSP introduction. (c) J–V profile, (d) EQE, (e) plots of d(EQE)/dE versus photon energy, and (f) Urbach energy calculation. Adapted with permission from ref. 36 Copyright 2025, Wiley-VCH. | ||
It is imperative to develop novel reaction-kinetics regulation strategies to improve the electronic quality of the solution-processed Sb2(S,Se)3 films, and to address the persistent challenge of carrier recombination losses in Sb2(S,Se)3 solar cells. In this context, Fu et al.37 introduced sodium borohydride (SB) into the precursor solution (schematic Fig. 6a). SB was found to be highly effective in accelerating the decomposition of selenourea (SU) and promoting Sb2Se3 phase formation through an alkaline effect. It simultaneously reduced SbO+ to Sb3+via a chemical reduction pathway. This dual action not only suppressed Sb2O3 impurity formation but also passivated deep-level defects such as sulfur-vacancies (VS), and anti-sites SbS1, leading to improved crystallinity and a preferred [hk1]-orientation. Deep-level transient spectroscopy (DLTS) measurements revealed only one electron trap (VS, density of 1.26 × 1013 cm−3) in the case of SB-added films; in contrast the control film exhibited hole traps (SbS1, density of 5.89 × 1013 cm−3) in addition to VS (density of 9.46 × 1013 cm−3). Consequently, the regulated reaction kinetics significantly reduced recombination losses and enhanced carrier transport through better band alignment and defect control. As a result, the optimized devices achieved a power conversion efficiency of 10.62% (0.0684 cm2), comparable to the best reported for this material system (Fig. 6b and c).
![]() | ||
| Fig. 6 (a) Schematic representation of the reaction kinetics and the growth-orientation for the Sb2(S,Se)3 films without and with sodium borohydride (SB) addition. SU, STS and APT refer to selenourea (Se source), sodium thiosulfate (S source) and antimony potassium tartrate (Sb source), respectively. (b) The J–V characteristics, and (c) EQE profile of the control and optimized (SB additive based) devices. Adapted with permission from ref. 37 Copyright 2025, Wiley-VCH. | ||
![]() | ||
| Fig. 7 Schematic of the commonly adopted deposition strategies for Sb2(S,Se)3 thin film deposition. (a) Spin coating, adapted with permission from ref. 58 Copyright 2024, Wiley-VCH. (b) Chemical bath deposition (CBD), adapted with permission from ref. 59 Copyright 2023, Elsevier. (c) Hydrothermal deposition and (d) thermal evaporation, adapted with permission from ref. 56 Copyright 2024, Elsevier. (e) Solid state diffusion, adapted with permission from ref. 60 Copyright 2017, Wiley-VCH. (f) Rapid thermal evaporation (RTE) and (g) vapor transport deposition (VTD) adapted with permission from ref. 61 Copyright 2020, Wiley-VCH. (h) Closed space sublimation (CSS), adapted with permission from ref. 62 Copyright 2023, Wiley-VCH. | ||
Solution processed Sb2(S,Se)3 thin films in general exhibit unfavourable [hk0]-orientation, compared to those fabricated using vapor deposition techniques. Moreover, the solution processed Sb2(S,Se)3 thin films contain secondary (oxide) phases in trace quantities, which are detrimental to device performance. In contrast, vapor deposition enables better control of phase purity, and stoichiometry. Key vapor deposition techniques developed for Sb2(S,Se)3 thin-film growth include Co-evaporation, rapid thermal evaporation, close surface sublimation, vapor transport deposition (VTD), magnetron sputtering, and pulsed laser deposition. It has been revealed that the increase in kinetic energy of vapor particles promotes the [hk1]-orientation in Sb2(S,Se)3 films. High deposition rates, and high adhesion coefficient for (hk1) planes leads to high growth rate along the [hk1]-directions. The vapor-pressure (and kinetic energy of vapor particles) can easily be regulated via controlling the (evaporation) source temperature, source-substrate distance, and vapor-density. Co-evaporation of Sb, S, and Se precursors, has demonstrated high-quality films with minimal defects.26,31,63
As hydrothermal deposition remains the most-used method for Sb2(S,Se)3 thin film deposition, researchers have made meticulous effort to regulate the nucleation and growth of Sb2(S,Se)6 ribbons. Rapid hydrothermal deposition is kinetically more favourable to the [hk1]-oriented growth of Sb2(S,Se)6 ribbons. Tang et al.19 modified the hydrothermal recipe to regulate the growth kinetics and achieved >10% PCE in Sb2(S,Se)3 solar cells for the first time. Increasing the selenourea ratio in the precursor solution promoted the [hk1]-oriented growth of Sb2(S,Se)3 films. A recent study by Li et al.64 demonstrates that by controlling the size and number of added presynthesized Sb2S3 particles in the hydrothermal precursor solution, the thickness, crystallinity and defects in the Sb2S3 can be effectively regulated. This strategy can be potentially investigated for hydrothermal deposition of electronic grade Sb2(S,Se)3 films.
Lu et al.65 deposited Sb2(S,Se)3 thin films using VTD, employing two independent sources of Sb2S3 and Sb2S3. By optimizing the evaporation conditions, Sb2(S,Se)3 films with a bandgap of 1.33 eV were achieved, which matches closely with the optimal (detailed balance) bandgap (of 1.3 eV) for harvesting solar insolation (AM1.5G, 100 mW cm−2). The HTL free device (ITO/CdS/Sb2(S,Se)3/Au) delivered a PCE of 7.03%, along with a VOC of 524 mV, JSC of 25.2 mA cm−2, and FF of 53.2%. Pan et al.66 adopted the same HTL-free architecture to fabricate the champion device with a PCE of 7.1%. Sb2(S,Se)3 powders (with different S/Se elemental ratios) were synthesized via the ball milling method, and Sb2(S,Se)3 films with tuneable bandgaps were obtained by VTD.
| Material | Deposition technique | Defect characterization technique | Trap | Type | E T (eV) | σ (cm2) | N T (cm−3) | Ref. |
|---|---|---|---|---|---|---|---|---|
| Sb2(S,Se)3 | CBD | O-DLTS | H1 | SbS1 | E V + 0.616 | 6.03 × 10−17 | 5.89 × 1013 | 69 |
| E1 | VS | E C − 0.764 | 1.4 × 10−15 | 9.46 × 1013 | ||||
| Sb2(S,Se)3 | Hydrothermal | H1 | VS1 | E V + 0.416 | 4.62 × 10−16 | 3.94 × 1013 | 38 | |
| H2 | VS2 | E V + 0.695 | 3.13 × 10−15 | 8.08 × 1013 | ||||
| MgCl2 treated Sb2(S,Se)3 | H1 | VS1 | E V + 0.414 | 4.42 × 10−17 | 3.50 × 1013 | |||
| H2 | VS2 | E V + 0.694 | 3.06 × 10−15 | 3.22 × 1013 | ||||
| Li-doped Sb2(S,Se)3 | O-DLTS | E1 | VSe | E C − 0.452 | 6.60 × 10−16 | 1.25 × 1014 | 35 | |
| H1 | SbS | E V + 0.716 | 9.01 × 10−15 | 1.97 × 1014 | ||||
| Sb2(S,Se)3 | E1 | VSe | E C − 0.400 | 1.54 × 10−15 | 4.19 × 1014 | |||
| H1 | SbS | E V + 0.726 | 8.57 × 10−15 | 9.86 × 1014 | ||||
| H1 | SbSe | E V + 0.779 | 3.32 × 10−15 | 9.56 × 1013 | 1 | |||
| Sb2(S,Se)3 | H1 | SbS2 | E V + 0.502 | 1.05 × 10−17 | 2.24 × 1014 | 9 | ||
| H3 | SbS3 | E V + 0.766 | 1.28 × 10−15 | 1.21 × 1015 | ||||
| DLTS | E1 | SbSe2 | E C − 0.57 | 2.48 × 10−12 | 1.45 × 1015 | 68 | ||
| E2 | SbSe1 | E C − 0.71 | 4.39 × 10−12 | 4.78 × 1015 | ||||
| H1 | SbS | E V + 0.761 | 3.97 × 10−15 | 6.45 × 1015 | ||||
| KI treated Sb2(S,Se)3 | E1 | SbSe2 | E C − 0.584 | 1.99 × 10−13 | 6.49 × 1014 | |||
| H1 | SbS | E V + 0.784 | 1.6 × 10−14 | 2.97 × 1015 | ||||
| Sb2(S,Se)3 | O-DLTS | H1 | SbS1 | E V + 0.500 | 1.99 × 10−17 | 6.28 × 1012 | 11 | |
| H2 | SbS2 | E V + 0.671 | 5.91 × 10−17 | 1.88 × 1013 | ||||
| E1 | VS2 | E C − 0.747 | 1.60 × 10−16 | 2.47 × 1013 | 67 | |||
| H1 | SeSb | E V + 0.290 | 6.61 × 10−17 | 6.55 × 1012 | ||||
| H2 | SbS1 | E V + 0.638 | 3.18 × 10−15 | 4.15 × 1012 | ||||
| H3 | SbS2 | E V + 0.483 | 1.48 × 10−16 | 4.64 × 1014 | ||||
| H4 | SbSe2 | E V + 0.682 | 3.67 × 10−14 | 8.63 × 1014 | ||||
| H5 | SbS3 | E V + 0.790 | 4.87 × 10−14 | 9.75 × 1014 | ||||
| DLTS | H1 | SbS | E V + 0.560 | 5.98 × 10−18 | 4.97 × 1015 | 53 | ||
| H2 | E V + 0.770 | 2.68 × 10−16 | 6.40 × 1015 | |||||
| E1 | SSb | E C − 0.563 | 1.40 × 10−16 | 1.97 × 1015 | ||||
| Vertical transport deposition (VTD) | Temperature-dependent admittance spectroscopy | H1 | VSe | E V + 0.203 | 2.43 × 10−19 | 5.37 × 1014 | 62 | |
| H2 | SbSe | E V + 0.364 | 3.85 × 10−16 | 1.27 × 1015 |
Zhao et al.10 proposed a novel in situ passivation (ISP) technique to passivate the SbSe defects and mitigate the non-radiative recombination in the Sb2(S,Se)3 solar cells. It was revealed that the additive (sodium selenosulfate, Na2SeSO3) induced in situ selinization in the Sb2(S,Se)3 films. This strategy produced Sb2(S,Se)3 films with improved crystallinity, morphology, orientation, and carrier lifetimes, along with significantly mitigated density of SbSe defects (Fig. 8a–h). Application of the ISP strategy improved the conductivity in the Sb2(S,Se)3 films from 2.93 × 10−5 to 6.61 × 10−5 S cm−1 (Fig. 8f). O-DLTS, SCLC, steady state PL, and TAS studies echo the high quality (and lower defect density) of the passivated films. Moreover, this technique was found to be instrumental in fine-tuning the energy level alignment, and to promote efficient transport of photogenerated carriers. Benefiting from the efficient defect passivation and improved band-alignment, the champion device delivered a PCE of 10.81% (Fig. 8i), which is among the highest recorded PCEs for single-junction Sb2(S,Se)3 solar cells.
![]() | ||
| Fig. 8 The energy levels and defect levels of (a) the control-Sb2(S,Se)3 film and (b) the target-Sb2(S,Se)3 film. (c) The values of NT (left) and (σ·NT)−1 (right) for Sb2(S,Se)3 devices. (d) The SCLC spectra (the device structural diagram is shown in the inset), (e) PL spectra and (f) I–V characteristics of Sb2(S,Se)3 films. TAS decay kinetics at 642 nm for (g) control-Sb2(S,Se)3 and (h) target-Sb2(S,Se)3 films. (i) The J–V characteristics of the Sb2(S,Se)3 devices. Adapted with permission from ref. 10 Copyright 2024, Wiley-VCH. | ||
Recently, Su et al.38 demonstrated a novel post-deposition, MgCl2-assisted activation process to mitigate deep-level anion-vacancy defects, facilitating the formation of large grains and improving the band alignment at the Sb2(S,Se)3/HTL interface. As illustrated in Fig. 9a and b, two-hole traps H1 and H2 were identified in the pristine and MgCl2 treated Sb2(S,Se)3 films, which were attributed to S-vacancies (VS1 and VS2, respectively). The MgCl2 treatment of Sb2(S,Se)3 films were found to dramatically mitigate the H1 and H2 defects (Table 2), extend the (average) carrier lifetime (Fig. 9g and h), and improve photogenerated carrier transport, leading to an impressive PCE of 10.55%.
![]() | ||
| Fig. 9 (a) DLTS signals from the control and MgCl2-treated Sb2(S,Se)3 devices. (b) Arrhenius plots derived from the DLTS signals. (c) The statistical histogram of calculated σ × NT for different hole traps in the control and MgCl2 treated Sb2(S,Se)3 devices. Schematic of (d) position of H1 (VS1) and H2 (VS2) defects in the Sb2(S,Se)3 lattice, (e) and (f) band edge positions and defect levels of the control and MgCl2-treated Sb2(S,Se)3, respectively, including the Fermi level (EF), and defect energy levels (H1, H2), relative to the vacuum level. (g) and (h) Transient kinetic decay (scatter) and corresponding biexponential curve fittings (solid line) monitored at 687 nm of the control and MgCl2-treated Sb2(S,Se)3 films. Adapted with permission from ref. 38 Copyright 2025, Wiley-VCH. | ||
Li et al.68 investigated a novel KI-assisted post-deposition surface treatment for Sb2(S,Se)3 films. The strategy manipulated the crystallization process to form compact films with larger grain size, forming better band alignment and inhibiting the formation of SbSe defects in Sb2(S,Se)3 films. As a result, the heterojunction quality is significantly improved in the Sb2(S,Se)3 solar cells, leading to a PCE of 9.22%. Zhu et al.70 investigated the effect of triethanolamine (TEA) additive in the hydrothermal deposition of Sb2(S,Se)3 films. TEA was found to be instrumental in regulating the S/Se elemental ratio, facilitating [021] and [061] crystallite orientation, and mitigating the detrimental VSe defects, resulting in a PCE of 9.94%.
Recent advancements have highlighted the importance of interface engineering in improving the efficiency and stability of Sb2(S,Se)3 solar cells. One of the key challenges in Sb2(S,Se)3 devices is the high density of surface defects at the Sb2(S,Se)3/HTL interface, which can lead to charge recombination. To mitigate this, self-assembled monolayers (SAMs) and passivating interlayers have been introduced at the interfaces. Recently, Wu et al.71 for the first time developed two self-assembled monolayers (SAMs) of TPA-2Th and TPA-2Py, to mitigate surface defects and ameliorate band alignment at the Sb2(S,Se)3/HTL interface. The ameliorated device delivered a PCE of 8.21%. Other engineering strategies to improve the Sb2(S,Se)3/HTL interface include postdeposition treatment with thioacetamide,72 SbCl3, CuCl2,73 organic chloride,74 and metal fluorides.12 In particular, organic chloride salt [diethylamine hydrochloride, DEA(Cl)]74 modified Sb2(S,Se)3 films were found to have better morphology, phase-purity, mitigated surface defects and improved band-alignment, leading to a PCE of 9.17% in the Sb2(S,Se)3 solar cell. Alkali metal fluoride-assisted post-deposition treatment is also found to be instrumental in regulating the S/Se elemental ratio, improving the morphology, crystallinity, and conductivity of the Sb2(S,Se)3 thin film, leading to a PCE of 10.7%.12 CdS layer post-treatment via CdCl2,12 KCl,75 and LiF35 was found to be instrumental in defect-passivation, and improving band alignment at the CdS/Sb2(S,Se)3 interface.
The CdS layer employed in the Sb2(S,Se)3 solar cell is usually deposited by CBD and air-annealed (at ∼400 °C), and exhibits high roughness (due to the high roughness of the underlying FTO), the presence of impurity phases (e.g., CdO, Cd(OH)2) and moderate conductivity. Additionally, these CdS films have a high concentration of S-vacancies, poor crystallinity and low (free) electron-density, which is not conducive to efficient transport of photogenerated electrons from the Sb2(S,Se)3 layer to the FTO electrode. To counter these limitations, post-treatment of CdS with chloride salts, such as CdCl2,76 SbCl3,77 and ZnCl2,78 has been widely used. CdCl2 treatment of CdS, however, leaves traces of Cd-oxychlorides, which can be eliminated by immersing it into hydrazine hydrate (N2H4) solution. This strategy has enabled regulation of the CdS/Sb2(S,Se)3 interface band alignment and recombination losses, leading to a PCE of 10.30%.16 Liu et al.75 demonstrated KCl-assisted post-treatment of the CdS layer in Sb2(S,Se)3 solar cells. The surface engineering strategy improved the morphological and electrical properties of CdS films, leading to a PCE of 9.98%, along with a significant improvement in JSC and FF in Sb2(S,Se)3 solar cells. KOH79 incorporation was found to be effective in improving the transmittance and crystallinity of CdS films. Ishaq et al.42 employed a thin, spin-coated layer of KI at the CdS/Sb2(S,Se)3 interface and observed significant mitigation of interface and bulk defects. On thermal annealing, K majorly diffused into CdS, while I diffused preferentially into the Sb2(S,Se)3 layer. KI-treatment modulated the band-alignment at the CdS/Sb2(S,Se)3 interface from cliff-like to a more conducive spike-like; and mitigated the detrimental VS/Se and SbS/Se defects, leading to a PCE of 10.06%.
Recently, Liu et al.35 investigated LiF doping in CdS thin films, and observed (thermally induced) in situ diffusion of high mobility Li+ into the Sb2(S,Se)3 absorber. Small ionic radii of Li+ assisted its migration through the grain boundaries, entering the crystal structure, and settling at the gap between [Sb4(S,Se)6]n ribbons, thus forming a Li+ gradient in the Sb2(S,Se)3 layer. Li+ effectively passivated the VSe and SbS defects, via the formation of strongly coordinated Li–S and Li–Se bonds. This strategy not only improved the opto-electronic properties of the CdS and Sb2(S,Se)3 layers, but also created a favorable energy level alignment for the transportation of photogenerated charge carriers towards the respective electrodes. This strategy also modulated the band-offset at the CdS/Sb2(S,Se)3 interface from the normal cliff-like band-offset to a moderately spike-like structure, and suppressed the interfacial defect density (∼3.48 × 1012 cm−2), inhibiting the interface and SCR recombination. The mitigation of defects and improvement in heterojunction quality culminated in improved PV performance (PCE of 10.76%) on adopting the lithiation strategy.
Al-doping was introduced to the CdS buffer layer, which facilitated tuning of the band alignment at the CdS/Sb2Se3 interface from the unfavorable cliff-like to (benign) spike-like.80 Ethylenediamine (EDA) solvent treatment was developed to chemically engineer the underlying CdS film, and found to be instrumental in improving the bonding at the Sb2(S,Se)3/CdS interface.41 Ag81 and Cu82 doping in CdS layer strengthened its n-type semiconductivity, and mitigated the interface defect density at Sb2Se3/CdS. Recently, Peng et al.34 employed an ultrathin layer at the FTO/CdS interface to mitigate S-vacancies in the CdS films, and a drastic improvement was observed in the crystallinity and electrical properties of the CdS thin films. As a result of interface engineering, the Sb2(S,Se)3/CdS interface defect density was reduced by more than 50%, and a PCE of 10.58% was achieved.
In contrast to post-treatment strategies, Zhang et al.21 demonstrated a CsI doping strategy to efficiently dope Sb2(S,Se)3 films, and simultaneously improve their morphology and mitigate the deep-level defects. Cs and I were found to co-ordinate with the Se/S and Sb-ions, respectively. The CsI-doped Sb2(S,Se)3 thin film solar cells exhibited a conducive band-alignment, suppressed carrier recombination, and improved photovoltaic performance (PCE of 10.05%). Recently, Ren et al.40 demonstrated simultaneous passivation of bulk and interface defects, on employing BaBr2 additive during hydrothermal deposition of Sb2(S,Se)3 films. Br− was found to adsorb on the polar planes of CdS, selectively exposing the non-polar planes of CdS, which are conducive in facilitating [hk1]-oriented growth of Sb2(S,Se)3 grains. Ba2+ on the other hand, was found to co-ordinate with (S/Se)2−, and passivate the grain-boundaries. As a result of regulated nucleation kinetics, improved morphology and crystallinity of Sb2(S,Se)3 films, a PCE of 10.12% was achieved in the Sb2(S,Se)3 solar cell. Large cation (Bi3+,83 Cs+,21 and Ce3+ (ref. 84)) doping is found to tune the high tensile strain in Sb2(S,Se)3 films to lower compressive strain, in addition to improve the [hk1]-orientation and mitigation of bulk, and grain boundary defects in Sb2(S,Se)3 films. (NH4)2S-induced hydrothermal sulfurization initially developed for reducing the trap density in Sb2S3 thin films,85 was found to be equally effective for trap-passivation in Sb2(S,Se)3 films.15
A suitable choice of dopant can introduce p- or n-type semiconductivity in Sb2S3 and Sb2Se3 thin-films, enabling the fabrication of homojunction solar cells. Cu,86 Sn,87 Pb,88 and Pt89 serve as p-type dopants, while Bi,90,91 I,92,93 Cl,94 Br,40 and Te95 serve as n-type dopants in Sb2S3 and Sb2Se3 thin-films. Recently, Chen's group demonstrated a homojunction Sb2Se3 solar cell with PCE of 10.15%. The p- and n-type semiconductivity was regulated on either side of the junction by regulating the Sb:Se stoichiometry.96 Thus, the versatility in doping enables different conductivity types, making homojunctions based on Sb2(S,Se)3 promising pathways to highly efficient Sb2(S,Se)3 solar cells.
Nanorod textured Sb2S3 and Sb2Se3 thin film solar cells have been fabricated, with the aim of improving the light trapping and carrier-extraction.99–102 Nanotexturization is a well-known engineering strategy to elongate the optical path for enhanced photon-trapping, and enlarged interface-area to facilitate charge extraction. Shi et al.84 for the first time fabricated a nanorod-textured Sb2(S,Se)3 solar cell prepared using a hydrothermal method. As schematically shown in Fig. 10a, the Sb2(S,Se)3 layer was composed of vertically grown Sb2(S,Se)3 nanorods over a compact layer of Sb2(S,Se)3. CsF additive was employed to promote the crystallinity and [221]-orientation in the Sb2(S,Se)3 nanorod array. Compared to the control device, the nanorod-textured (CsF-40) solar cell demonstrated better carrier management (higher EQE) and diode-characteristics (lower η value of 1.51), leading to PCE improvement from 6.83% to 9.37% (Fig. 10b–g). The nanotexturization strategy was also found to reduce the band offsets at the CdS/Sb2(S,Se)3 and Sb2(S,Se)3/Spiro-OMeTAD interface, benefitting the photogenerated carrier transport (Fig. 10h and i).
![]() | ||
| Fig. 10 (a) Schematic illustration of an Sb2(S,Se)3 solar cell based on a nanorod-textured Sb2(S,Se)3 bilayer absorber. (b) J–V curves of the best devices with and without CsF additive. (c) Certified J–V and power density plots. (d) EQE spectra and integrated JSC of the best devices with and without CsF additive. (e) Dependence of VOC on different light intensities for the control and CsF-processed devices. (f) Histogram of PCE for 25 devices based on the control and CsF-processed Sb2(S,Se)3 films. (g) Nyquist plots of the control and CsF-40 Sb2(S,Se)3 solar cells obtained in the dark at 0.5 V bias voltage (inset: equivalent circuit used for Nyquist fitting). (h) and (i) Schematics showing the band alignments of the solar cells based on the control and CsF-40 Sb2(S,Se)3 absorbers. Adapted with permission from ref. 84 Copyright 2022, Elsevier B.V. | ||
Recently, Dong et al.33 demonstrated a novel charge-carrier management strategy to maximize photon absorption and charge-carrier collection. A textured FTO (T-FTO) substrate was employed as the front electrode, and a thin SnO2 layer (deposited via ALD, spin-coating, and CBD) was used at the FTO/CdS interface to mitigate the shunt paths arising due to textured morphology. The insertion of the SnO2 layer facilitated the growth of a conformal CdS layer and an optimal bandgap profile in the Sb2(S,Se)3 absorber layer. The T-FTO substrate exhibited higher roughness (Fig. 11a–d), and haze factor (Fig. 11e) than the FTO substrate with a micro-scale smooth surface (S-FTO). Higher haze factor indicates better photon harvesting ability and contributes to enhanced photon absorption by the absorber layer. In particular, the devices consisting of a T-FTO electrode, and ALD deposited SnO2 layer were found to outperform those with S-FTO, and the SnO2 layer deposited via spin coating (SC) and CBD (Fig. 11f). The small area (0.1225 cm2) device demonstrated a record PCE of 10.92% (certified as 10.70%, Fig. 11g), while the large area device (1 cm2) exhibited a PCE of 9.27% (Fig. 11h). Carrier loss analysis is illustrated in Fig. 11i. The carrier transport loss decreased from 13.39% in the reference cell (cell-ref) to 8.25% in the cell-target device, whereas non-radiative recombination loss decreased from 15.49% to 11.25%.
![]() | ||
| Fig. 11 Top-view SEM images of (a) S-FTO, and (b) T-FTO electrodes. AFM micrographs of (c) S-FTO, and (d) T-FTO substrates. (e) Haze factor of the S-FTO and T-FTO substrates. Inset: The average haze factor. (f) Statistical distribution of PCE in cells consisting of SnO2 inter-layer deposited with different techniques. (g) J–V curve of the champion cell-target device. Inset: Cell configuration. (h) J–V curve of large-area (1.0 cm2) Sb2(S,Se)3 solar cell. Inset: Cell image. (i) The FF loss analysis of cell-ref and cell-target. Adapted with permission from ref. 33 Copyright 2025, Nature Springer. | ||
Attempts to directly replace CdS with oxide-based ETLs have shown a drop in FF (and PCE), due to the chemical incompatibility and lattice-mismatch existing at the oxide/Sb2(S,Se)3 interface. Hu et al.69 optimized the growth of Sb2(S,Se)3 films on TiO2 film using additive engineering, to achieve a PCE of 8.52%. Dong et al.106 demonstrated that in contrast to the crystalline/compact TiO2 films, nanoparticle/amorphous TiO2 films provide more anchoring sites, and are more conducive to the nucleation and growth of [hk1]-oriented Sb2(S,Se)3 films. Li et al.34 optimized the co-sublimation method to achieve [hk1]-orientation, and a V-shaped double graded bandgap profile in Sb2(S,Se)3 films deposited over TiO2 films. The conducive band-alignment and suppressed deep-level defects lead to an impressive PCE of 9.02%, which is the highest for CdS-free Sb2(S,Se)3 solar cells. Recently, Qin et al.107 employed NaCl solution treatment to enlarge the bandgap and improve the quality of CBD grown TiO2 films. The Na-ions diffused into the Sb2(S,Se)3 layer, passivated the Se-vacancies, and were found to play a key role in optimization of the TiO2/Sb2(S,Se)3 interface. Wang et al.108 investigated TiCl4-passivated TiO2 films, while Chen et al.109 utilized atomic-layer deposited grown zinc tin oxide (ZTO) to engineer the band-alignment and defects at the ETL/Sb2Se3 interface; conceptually similar interfacial-engineering could be extended to Sb2(S,Se)3 solar cells.
Zhao et al.110 replaced the pristine CdS layer with a Zn(O,S)/CdS bilayer ETL. The novel device FTO/Zn(O,S)/CdS/Sb2(S,Se)3/Spiro-OMeTAD/Au demonstrated a PCE of 9.62%. This strategy enabled improvement in JSC by 1.5 mA cm−2, and 75% reduction in the usage of toxic CdS. In contrast, the device employing sole Zn(O,S) as the ETL exhibited high (dark) leakage current density, poor diode characteristics, and severe non-radiative recombination, resulting in a mediocre PCE of 4.20%. The SnO2/CdS bilayer ETL has recently gained tremendous attention and has been frequently employed to achieve PCE values exceeding 10% in Sb2(S,Se)3 solar cells.
As illustrated in Table 3, attempts to substitute CdS layer has led to a drop in efficiency in Sb2(S,Se)3 solar cells. Another strategy is to replace the CdS layer (thickness ≈ 60 nm) with bilayer ETL such as Zn(O,S)/CdS and SnO2/CdS.114 This strategy allows us to reduce the thickness of CdS to low levels (10–20 nm) without compromising the device performance. In particular, the SnO2 nanoparticle dispersion is found to be successful in settling at the valleys at the topography of the FTO substrate, mitigating its roughness and the possibility of direct contact of FTO with the Sb2(S,Se)3 absorber.112 NH4F treatment has been applied to the dispersion of SnO2 nanoparticles and increases the n-type conductivity of the SnO2 film through fluorine doping (leading to PCE of 8.26% in Sb2S3 solar cells).115 Qi et al.116 demonstrated a PCE of 8.9% in the ZTO/CdS bilayer ETL-based Sb2(S,Se)3 solar cells, where an ultrathin and insulating polymer layer was employed at the HTL/Ag interface to mitigated the shunt paths.
| Device architecture | Sb2(S,Se)3 deposition technique | ETL deposition technique | Device parameters | Ref. | |||
|---|---|---|---|---|---|---|---|
| V OC (V) | J SC (mA cm−2) | FF (%) | PCE (%) | ||||
| FTO/TiO2/Sb2(S,Se)3/Spiro-OMeTAD/Au | CBD | Spin-coating | 0.487 | 26.63 | 65.71 | 8.52 | 69 |
| FTO/TiO2/Sb2(S,Se)3/Spiro-OMeTAD/Au | Hydrothermal | Spin-coating | 0.600 | 21.43 | 54.79 | 7.08 | 106 |
| FTO/Zn(O,S)/CdS/Sb2(S,Se)3/Spiro-OMeTAD/Au | CBD | 0.656 | 22.74 | 64.54 | 9.62 | 110 | |
| FTO/Zn(O,S)/Sb2(S,Se)3/Spiro-OMeTAD/Au | 0.418 | 20.15 | 49.85 | 4.20 | |||
| FTO/TiO2/Sb2(S,Se)3/Spiro-OMeTAD/Au | Co-sublimation | Spin-coating | 0.506 | 27.81 | 64.1 | 9.02 | 53 |
FTO/TiO2/BiI3 doped Sb2(S,Se)3/PEDOT : PSS/Au |
Spin-coating | Spray pyrolysis + screen printing | 0.520 | 21.5 | 63.0 | 7.05 | 83 |
| FTO/TiO2/Sb2(S,Se)3/Spiro-OMeTAD/Au | Spin-coating | Spin-coating | 0.478 | 25.8 | 58.1 | 7.15 | 111 |
| FTO/TiO2/Sb2(S,Se)3/Spiro-OMeTAD/Au | CBD + spin coating | Spin-coating | 0.560 | 19.48 | 52.34 | 5.71 | 60 |
| FTO/SnO2/CdS/Sb2(S,Se)3/Spiro-OMeTAD/Au | Hydrothermal | Spin coating + CBD | 0.72 | 21.69 | 55.35 | 8.67 | 112 |
| FTO/SnO2/CdS/Sb2(S,Se)3/carbon/Ag | Hydrothermal + selenization | Chemical vapor deposition (CVD) + CBD | 0.74 | 12.5 | 55.8 | 5.2 | 113 |
| Mo/Sb2Se3/ZTO/ITO/Ag | Sputtering of Sb + selenization | ALD | 0.440 | 32.84 | 60.77 | 8.63 | 109 |
| ITO/TiO2/Sb2Se3/Au | VTD | Spin coating | 0.394 | 25.5 | 52.9 | 5.33 | 108 |
Recently, Dong et al.33 comprehensively investigated the SnO2/CdS bilayer ETL in Sb2(S,Se)3 solar cells, employing a SnO2-layer deposited by ALD, CBD and spin-coating. The ALD grown SnO2-layer was found to be the optimal buffer layer, enabling a record PCE of 10.92% (certified as 10.70%) in Sb2(S,Se)3 solar cells. This study elucidates that the underneath conformal SnO2-coating promotes a conformal deposition of CdS. This strategy regulates the defects and creates an optimal bandgap profile in the Sb2(S,Se)3 absorber, highly conducive to the transport of photogenerated charge carriers.
The oxide/CdS bilayer ETL architecture has started gaining significant attention in Sb2S3 and Sb2Se3 thin-film solar cells as a strategy to overcome the limitations of a standalone CdS layer (∼70 nm thick). In this design, a thin oxide layer (typically SnO2 or TiO2) is first deposited on the FTO substrate—commonly by spin-coating a diluted colloidal nanoparticle solution- followed by chemical bath deposition (CBD) of the CdS layer. Liu et al.76 employed a SnO2/CdS bilayer ETL treated with Ce2S3, achieving an efficiency of 7.66%, while Wang et al.117 investigated a TiO2/CdS nanoarray ETL treated with PbSe, which yielded 8.06% efficiency in Sb2S3 solar cells. The incorporation of ultrathin Ce2S3 and PbSe interfacial layers facilitated preferential crystallization of Sb2S3 along the [hk1] orientation and effectively passivated interface defects at the CdS/Sb2S3 junction. Su et al.118 revealed that the SnO2/CdS underlayer promotes the growth of large grains, smoother surfaces, and reduced defect density in Sb2S3 films compared to those grown directly on CdS. SnO2/CdS has shown promising performance in Sb2Se3 solar cells with PCE values approaching the 10% benchmark (9.50%, Che et al.;119 9.37%, Sheng et al.;77 and 9.28%, Zhao et al.120).
| Device architecture | Sb2(S,Se)3 deposition technique | HTL deposition technique | Device parameters | Ref. | |||
|---|---|---|---|---|---|---|---|
| V OC (V) | J SC (mA cm−2) | FF (%) | PCE (%) | ||||
| FTO/CdS/Sb2(S,Se)3/DTPThMe–ThTPA/Au | Hydrothermal | Spin-coating | 0.638 | 23.18 | 65.50 | 9.69 | 121 |
| FTO/CdS/Sb2(S,Se)3/MnS/Au | Thermal evaporation + air annealing | 0.664 | 21.26 | 65.48 | 9.24 | 122 | |
| FTO/CdS/Sb2(S,Se)3/MnS (crystalline)/Au | Thermal evaporation + annealing in N2 environments | 0.655 | 22.98 | 64.2 | 9.67 | 18 | |
| FTO/CdS/Sb2(S,Se)3/MnS (amorphous)/Au | Thermal evaporation | 0.604 | 18.67 | 59.7 | 6.73 | ||
| FTO/CdS/Sb2(S,Se)3/MnS/Au | Thermal evaporation + air annealing | 0.664 | 21.26 | 65.48 | 9.24 | 122 | |
| FTO/CdS/Sb2(S,Se)3/MAPbBr3/Au | Spin coating | 0.580 | 19.05 | 58.24 | 6.42 | 123 | |
| FTO/CdS/Sb2(S,Se)3/CsPbBr3/Au | 0.620 | 21.50 | 58.55 | 7.82 | |||
| FTO/CdS/Sb2(S,Se)3/PbS/C/Ag | Hydrothermal | Hydrothermal | 0.650 | 18.8 | 65.06 | 8.0 | 124 |
| FTO/CdS/Sb2(S,Se)3/Spiro-OMeTAD:TMT-TTF/Au | Spin-coating | 0.650 | 20.63 | 67.66 | 9.02 | 125 | |
| FTO/CdS/Sb2(S,Se)3/CuPc doped P3HT/Au | Thermal evaporation | Spin-coating | 0.491 | 27.7 | 60.9 | 8.25 | 126 |
| FTO/CdS/Sb2(S,Se)3/NiO/Au | Hydrothermal | Spin-coating | 0.644 | 19.12 | 60.14 | 7.40 | 127 |
| ITO/CdS/Sb2(S,Se)3/Au | VTD | — | 0.524 | 25.2 | 53.2 | 7.03 | 65 |
| FTO/CdS/Sb2(S,Se)3/MoO3/Au | Hydrothermal | E-bean evaporation | 0.684 | 16.53 | 63.75 | 7.20 | 128 |
| FTO/CdS/Sb2(S,Se)3/MnS/Au | Thermal evaporation | 0.623 | 23.7 | 64.5 | 9.50 | 129 | |
| FTO/CdS/Sb2(S,Se)3/MnS/ITO | 0.541 | 23.4 | 58.3 | 7.41 | |||
| FTO/CdS/Sb2(S,Se)3/MXene | Hydrothermal | Thermal spraying | 0.64 | 20.9 | 62.7 | 8.29 | 130 |
| FTO/CdS:O/Sb2(S,Se)3/PbS/C/Ag | Hydrothermal | Painting with carbon paste | 0.518 | 25.98 | 67.21 | 9.05 | 131 |
MnS emerges as a low-cost ($4 per g), environment-friendly, stable, and scalable HTL in Sb2(S,Se)3 solar cells. Qian et al.18 deposited (80 nm thick) an MnS layer over the Sb2(S,Se)3 films using thermal evaporation, followed by annealing to improve their crystallinity, mobility and conductivity. In addition, the annealing also improved the work-function of MnS, which contributed to improved built-in electric field and higher photovoltage. Although the MnS-based champion device (PCE of 9.67%) slightly underperformed compared to the device based on Spiro-OMeTAD (10.14%, Fig. 12a–c), it outperformed the latter in terms of reproducibility, longevity/stability (Fig. 12d) and cost-competitiveness. The unencapsulated MnS-based device exhibited PCE retention of >95% after 60 days of storage in ambient air (relative humidity of 15–40%), while the spiro- counterpart could retain only 58% of the initial PCE. Wang et al.122 proposed air annealing of thermally evaporated MnS, to improve its p-type conductivity and band-alignment with the Sb2(S,Se)3 layer. CuSbS2 and CuSbSe2 can be a potential choice for the HTL in Sb2(S,Se)3 solar cells, owing to their high p-type conductivity, matching the valence-band edges and low-lattice mismatch with Sb2(S,Se)3.132
![]() | ||
| Fig. 12 (a) PCE distribution histograms of Sb2(S,Se)3 solar cells with different HTLs: as-deposited MnS HTL, post-annealed MnS HTL and Spiro-OMeTAD. (b) J–V curves; (c) EQE curves and the integrated current density of the Sb2(S,Se)3 devices with different HTLs. (d) Decay of normalized PCE over time of Sb2(S,Se)3 devices with different HTLs. Adapted with permission from ref. 18 Copyright 2022, Royal Society of Chemistry. | ||
Among the inorganic HTL candidates, NiO stands out owing to its wide-bandgap (∼3.8 eV), deep valence-band edge (∼−5.3 V), thermal-stability, benign-synthesis, and low-cost.133,134 Huang et al.127 investigated NiO as the HTL in Sb2(S,Se)3 solar cells, and as a result of conducive band alignment, Sb2(S,Se)3/NiO achieved a PCE of 7.40%. The NiO-based device exhibited substantially higher stability than its Spiro-OMeTAD-based counterpart, by retaining 92% of the initial PCE after 700 h. Xing et al.128 employed a non-toxic, high work-function and wide band-gap oxide MoO3 as the HTL in Sb2(S,Se)3 solar cells. High purity MoO3 thin films were obtained by e-beam evaporation and despite their n-type semiconductivity, the films were found to selectively extract holes and alleviate interface recombination. As a result, an improvement in built-in voltage and decrement in back contact resistance was obtained, leading to a PCE of 7.2%.
Li et al.130 synthesized pure, uniform and highly conducting 2D MXene (Ti3C2Tx) nano-flakes and then coated it on the surface of Sb2(S,Se)3 thin film by thermal spraying. The negatively charged Tx group of MXene tightly attached and healed the sulfur-vacancy defects on the Sb2(S,Se)3 surface. This effect effectively reduced the recombination and enhanced the carrier transport at the back interface. As a result of ameliorated charge transport, the PCE of 8.29% was obtained in the MXene-based device, outperforming those with Au (4.0%) and carbon (2.8%) electrodes. A blend of Spiro-OMeTAD and lanthanide (Ln)-doped upconversion nanoparticles was found to serve as an extra light harvester and HTL, leading to a nominal PCE of 9.17%.135
Carbon is known to serve a binary function (of HTL and back electrode),136 and has been employed to fabricate noble metal-free Sb2(S,Se)3 solar cells. However, the devices yielded poor PCE (<3%), owing to high photovoltage loss at the Sb2(S,Se)3/C interface. Insertion of PbS124 or P3HT137 layers improved the PCE of the carbon-based device to 8.0% and 4.15%, respectively. A study by Ren et al.138 suggests that the controlled oxidation of antimony chalcogenide films can promote amorphous phases (such as Sb2O3, Se, SeOx) at the surface, and tune the otherwise Ohmic contact to a Schottky contact. Comprehensive studies on metal electrodes with Sb2(S,Se)3 are lacking and need to be performed for attaining the commercialization potential of Sb2(S,Se)3-based devices.
![]() | ||
| Fig. 13 (a) J–V curves of the Sb2(S,Se)3 semitransparent top cell and Si bottom cell. (b) EQE curves of the Sb2(S,Se)3 semitransparent top cell, Si bottom cell covered by the Sb2(S,Se)3 semitransparent top cell, and Si cell alone. Adapted with permission from ref. 129 Copyright 2023, Wiley-VCH. | ||
Recently, Yang et al.140 attempted to address the issue of severe sputtering damage during ITO deposition, and formation of a hole-extraction (Schottky) barrier at the MnS/ITO interface. An amorphous indium-zinc oxide (IZO) layer prepared by the soft sputtering method as the passivation layer to construct high quality ST-SSCs with minimal sputtering damage and excellent carrier transporting ability. This structure eliminates the hole extraction barrier at the MnS/ITO interface while protecting the underlying MnS from sputtering damage. In addition to alleviating the damage to the MnS layer, the IZO layer assists formation of the desired ohmic contact at the MnS/ITO interface. As a result, the semitransparent solar cell (FTO/CdS/Sb2(S,Se)3/MnS/IZO/ITO) delivered a record PCE of 8.26%. On coupling with the Sb2Se3 bottom cell, a PCE of 10.69% was achieved in the 4T tandem solar cell.
Chen et al.142 demonstrated Sb2S3 IPV minimodules (area of 5 cm2, PCE of 12.82%) to power IoT wireless sensors and realized the long-term continuous recording of environmental parameters under WLED illumination in an office. The minimodule delivered competitive performance to the commercial a-Si
:
H IPV devices. Gao et al.55 performed a systematic study on the effect of the optical bandgap of Sb2(S,Se)3 films on the comparative PV performance of the devices under AM1.5G and indoor light conditions. Sb2(S,Se)3 films with different bandgaps were synthesized by regulating the concentrations of S and Se source precursors in the hydrothermal deposition. The solar cells based on Sb2(S,Se)3 with 34.3% Se-content (and 1.50 eV bandgap) enabled an optimal PCE of 9.04% under standard AM1.5G illumination (Fig. 14b), while the device based on Sb2S3 with 1.74 eV bandgap achieved an optimal PCE of 20.34% under 1000 lux indoor illumination (Fig. 14e). The study concludes that though an appropriate Se/S atomic ratio is beneficial for improving the crystallinity of the Sb2(S,Se)3 film and passivating the trap states, the band gap remains a key factor in determining the suitability of this material for IPVs. Interestingly, PCE values exceeding 16% were obtained within a wide bandgap range from 1.5 to 1.7 eV, demonstrating the great prospects for Sb2(S,Se)3 as a light-harvesting material for indoor solar cells (Fig. 14d).
![]() | ||
| Fig. 14 (a) Cross-sectional SEM image of the Sb2(S,Se)3 solar cell. (b) J–V characteristics, and (c) EQE spectra of the Sb2(S,Se)3 solar cells under AM1.5G. (d) PCE distribution histograms of the Sb2(S,Se)3-based indoor solar cells. (e) J–V characteristics of the devices at the light intensity of 1000 lux under LED. (f) The relationship between output power and bandgap under AM1.5G and LED 1000 lux light sources. Adapted with permission from ref. 55 Copyright 2024, Wiley-VCH. | ||
Cao et al.143 investigated low Se-content (Sb/S/Se = 1
:
1.42
:
0.06) Sb2(S,Se)3 as an absorber material in indoor solar cells, fabricated in the architecture FTO/SnO2/CdS/Sb2(S,Se)3/Spiro-OMeTAD:TMT-TTF/Au. Tetrakis(methylthio)tetrathiafulvalene (TMF-TTF) was introduced to prepare the Spiro-OMeTAD:TMT-TTF (36.6/1.0 mg) hybrid HTL. The composite HTL exhibited higher mobility than the pristine Spiro-OMeTAD. The Se-content in the Sb2(S,Se)3 films was found to be instrumental in regulating the favourable [hk1]-orientation, and in promoting charge transport properties. The champion device exhibited PCE values of 18.53, 17.62, 17.07, 17.30, 16.24, and 15.38%, under white LED illuminances of 1000, 500, and 200 lx with color temperatures of 3347 and 6103 K, respectively. Recently, Yang et al.140 achieved a record PCE of 20.86% in a semitransparent Sb2(S,Se)3 solar cell, under 1000 lux indoor illumination.
Cao et al.147 performed a meticulous theoretical investigation to optimize triple-junction tandem solar cells, consisting of Sb2S3 (top cell)/Sb2(S,Se)3 (mid cell)/Sb2Se3 (bottom cell) stacking. A PCE of 32.98% was proposed via a band engineering strategy. Recently, Salem et al.63 demonstrated a PCE of 22.00% (Fig. 14a–c) in a Se/Sb2(S,Se)3 dual-junction, monolithic, two-terminal (2T) tandem solar cell, using the Silvaco TCAD simulations. The proposed Sb2(S,Se)3 bottom was designed HTL-free, while ITO served as the interlayer. Dahmardeh et al.148 demonstrated a PCE of 22.19% in Sb2Se3/Sb2(S,Se)3 tandem solar cells, employing the SCAPS simulation tool. The thickness of both absorber layers and ETL parameters were optimized to achieve current matching between the (top and bottom) subcells and facilitating efficient transport of photogenerated carriers (Fig. 15).
![]() | ||
| Fig. 15 Proposed monolithic HTL-free Se/Sb2(S,Se)3 tandem SC: (a) schematic of the HTL-free tandem SC design, emphasizing the removal of hole-transport layers from both sub-cells, (b) J–V characteristics of the HTL-free tandem configuration, comparing the top, bottom, and overall tandem cell performance, and (c) EQE curves of the HTL-free tandem. Adapted with permission from ref. 63 Copyright 2025, Elsevier. | ||
Single junction Sb2(S,Se)3 solar cells have been extensively investigated using SCAPS simulation, and the key results are enlisted in Table 5. The simulation works prophesise attaining PCE values exceeding 20% in single junction Sb2(S,Se)3 solar cells, on adopting high levels of interface and bulk-defect passivation, and carefully tuning the band alignment at the ETL/Sb2(S,Se)3 and HTL/Sb2(S,Se)3 interfaces. Our previous computational work on Cd-free Sb2(S,Se)3 solar cells endorses ZnSe as a potential alternative to the toxic-CdS based ETL in Sb2(S,Se)3 solar cells.144 On the other hand, CuSbS2 has emerged as the most appropriate substitute to the Spiro-OMeTAD layer. The optimized devices based on ZnSe (ETL) and CuSbS2 (HTL) exhibited a theoretical PCE of 20.02%, paving the way for boosting the PCE of Sb2(S,Se)3 solar cells beyond the current record of 10.92%.
| Device architecture | Absorber thickness (µm) | V OC (V) | J SC (mA cm−2) | FF (%) | PCE (%) | Ref. |
|---|---|---|---|---|---|---|
| FTO/CdS/Sb2(S,Se)3/Spiro-OMeTAD/Au | 0.60 | 1.310 | 24.05 | 58.56 | 18.43 | 149 |
| 0.30 | 0.654 | 24.04 | 63.87 | 10.04 | ||
| ITO/Cds/Sb2(S,Se)3/MoS2/Mo | 0.80 | 0.950 | 35.32 | 75.96 | 25.67 | 150 |
| FTO/Cd0.6Zn0.4S/Sb2(S,Se)3/Spiro-OMeTAD/Au | 0.50 | 0.881 | 26.67 | 74.22 | 17.43 | 151 |
| FTO/ZnO/Sb2(S,Se)3/Cu2O/Au | 0.55 | 0.951 | 27.70 | 68.00 | 18.00 | 152 |
| FTO/ZnO/Sb2(S,Se)3/NiO/Au | 0.50 | 0.940 | 27.60 | 67.00 | 17.30 | |
| FTO/Cd0.6Zn0.4S/Sb2(S,Se)3/Cu2O/Au | 0.90 | 0.889 | 27.40 | 71.00 | 17.30 | |
| ITO/CdS/Sb2(S,Se)3/MnS/Au | 0.30 | 0.722 | 25.29 | 69.54 | 12.70 | 153 |
| Al:ZnO/Zn0.4S0.6/TiO2/Sb2(S,Se)3/MoSe2/Mo | 2.00 | 0.638 | 32.34 | 75.75 | 15.65 | 154 |
| FTO/ZnSe/Sb2(S,Se)3/CuSbS2/Au | 0.50 | 0.930 | 28.64 | 74.54 | 20.01 | 144 |
| FTO/CdS/Sb2(S,Se)3/Spiro-OMeTAD/Au | 0.60 | 0.990 | 30.93 | 87.09 | 26.77 | 155 |
| FTO/SnO2/CdS/Sb2(S,Se)3/CuI/Au | 0.30 | 0.902 | 27.51 | 85.39 | 21.19 | 156 |
| FTO/SnO2/CdS/Sb2(S,Se)3/CuI/C | 0.30 | 0.897 | 23.20 | 85.55 | 17.82 |
MnS has emerged as the top choice to substitute Spiro-OMeTAD in Sb2(S,Se)3 solar cells, while CdS remains an indispensable choice for the ETL. As of now, adopting a bilayer ETL, consisting of a SnO2 or TiO2 layer decorated with an ultrathin (∼10 nm) CdS layer is a favored strategy to reduce the usage of CdS, without compromising the device performance. Various doping (particularly large cations and anions), post-deposition treatment, and additive engineering techniques (especially metal fluorides) have been found to be instrumental in alleviating the sluggish charge carrier dynamics in Sb2(S,Se)3 thin films, by regulating the defects and tuning the strain in the films. As the high temperature annealing induces the formation of SbS/Se anti-site defects, the development of low-temperature synthesis protocols can be crucial in mitigating the formation of scavenging deep-level defects. Prototype Sb2(S,Se)3 solar cells have shown PCE > 20% under indoor light conditions, suggesting great potential for these devices in indoor PV and building integrated PV. Future research should prioritize long-term stability, large-area manufacturing, and integration with tandem structures to unlock their full potential in the renewable energy landscape. The synergy of defect-engineering and band-alignment optimization is anticipated to achieve >20% PCE in Sb2(S,Se)3 solar cells, pushing the technology toward commercialization.
| This journal is © The Royal Society of Chemistry 2026 |