Thickness‑Dependent Photocurrent and Optoelectronic Dynamics in TiO₂/Sb₂S₃/P3HT Planar Hybrid Solar Cells

Introduction

Sb₂S₃ has emerged as a promising absorber for solid‑state thin‑film solar cells, thanks to its moderate bandgap of 1.7 eV and high absorption coefficient of 1.8 × 10⁵ cm⁻¹ [1, 2]. Thin films can be fabricated via spray pyrolysis, electrodeposition, chemical deposition, or thermal vacuum evaporation [3–6]. Recent device optimizations have achieved photo‑conversion efficiencies (PCE) of 5.7–7.5 % in Sb₂S₃ systems [1–10], yet these remain below the performance of dye‑sensitized and perovskite cells [11, 12]. Most efforts focus on device engineering, but a fundamental understanding of the interplay among absorption, internal field, transport paths, and kinetics is essential for further improvement. In this work, we employ the conventional TiO₂/Sb₂S₃/P3HT n‑i‑p architecture to dissect the charge generation and dissociation dynamics across a range of Sb₂S₃ thicknesses.

Varying the Sb₂S₃ thickness in TiO₂/Sb₂S₃/P3HT cells modifies (i) photon harvesting and carrier concentrations; (ii) internal electric field magnitude; (iii) electron/hole transport distances; and (iv) recombination rates [16–21]. While these effects have been qualitatively attributed to bulk resistance, photon absorption, and field strength, a quantitative analysis of thickness‑dependent photovoltaic parameters is lacking. We fabricated TiO₂/Sb₂S₃/P3HT n‑i‑p cells (Fig. 1) with Sb₂S₃ layers of 96–373 nm, examined the associated charge‑carrier transport processes, and identified the key factors limiting photocurrent and open‑circuit voltage.

Illustration of TiO₂/Sb₂S₃/P3HT n‑i‑p solar cell architecture. h⁺ denotes the hole and e⁻ the electron.

Methods

Reagents

Etched FTO‑coated glass substrates were sourced from Huanan Xiangcheng Co., Ltd., China. SbCl₃ (99 %), Na₂S₂O₃ (99 %), and titanium diisopropoxide (75 % in isopropyl alcohol) were purchased from Adamas‑beta. P3HT was obtained from Xi’an Polymer Company, China, and Ag (99.999 %) from Alfa.

Device Fabrication

Substrates were cleaned by ultrasonication in soap water, acetone, and isopropanol for 60 min each, followed by 30 min UV‑ozone treatment. A 0.15 M TiO₂ precursor solution was spin‑coated at 4500 rpm for 60 s, then annealed at 125 °C for 5 min and 450 °C for 30 min. Sb₂S₃ was deposited by chemical bath deposition (CBD) [5, 10, 22]: an acetone solution of SbCl₃ (0.3 M) was added dropwise to Na₂S₂O₃ (0.28 M) in an ice bath (~5 °C). The TiO₂‑coated FTO was inverted into the aqueous solution; once the solution turned orange, deposition proceeded for 1, 1.5, 2, or 3 h. The resulting Sb₂S₃ layers were rinsed with de‑ionized water, dried under N₂, and annealed for 30 min in a glovebox (O₂: 0.1 ppm, H₂O: 0.1 ppm) under N₂. Finally, P3HT (15 mg mL⁻¹) was spin‑cast at 1500 rpm for 60 s inside the glovebox, and MoO₃ (10 nm) followed by Ag (100 nm) electrodes were evaporated through a shadow mask.

Instruments and Characterization

XRD patterns were recorded on an MXP18AHF diffractometer (Cu Kα, λ = 1.54056 Å). SEM imaging employed a ZEISS GeminiSEM 300. Absorption spectra were obtained with a Shimadzu UV‑2600. J–V characteristics were measured under AM 1.5 illumination (100 mW cm⁻²) with a 94023A Oriel Sol3A simulator (Newport Stratford, Inc.). EQE spectra were measured from 300–900 nm using a Zolix QE/IPCE kit. IMPS/IMVS were recorded with an IviumStat.H workstation under ambient conditions, using a 28.8 mW cm⁻² white LED and a 10 % sinusoidal perturbation. KPFM mapping was performed with an Agilent SPM 5500 (MAC III controller) to probe surface potential (SP).

Results and Discussion

Deposition and Characterization of Sb₂S₃/TiO₂ Film

FE‑SEM images (Fig. 2a) confirm uniform Sb₂S₃ layers on TiO₂, with thicknesses tuned by CBD time. The average thickness d increases linearly with deposition time t (Fig. 2b), ranging from 96 to 373 nm. XRD patterns (Fig. 3) match orthorhombic Sb₂S₃ (JCPDS #42‑1393) and show no secondary phases. UV‑vis spectra (Fig. 4) reveal TiO₂ absorption onset at 386 nm (3.21 eV) and a TiO₂/Sb₂S₃ absorption edge near 750 nm. Absorption intensity rises with longer CBD time, consistent with increasing Sb₂S₃ thickness.

a Cross‑sectional FE‑SEM images of Sb₂S₃ films on TiO₂. b Plot of Sb₂S₃ thickness d versus CBD time t.

XRD patterns of Sb₂S₃ films deposited for 1–3 h.

UV‑vis absorption of TiO₂ and TiO₂/Sb₂S₃ films (1–3 h).

Solar Cells

J–V curves (Fig. 5a) and performance metrics (Table 1) demonstrate that device efficiency peaks when Sb₂S₃ thickness is 175 nm (t = 1.5 h). The optimal cell achieves a PCE of 1.65 %, J_sc = 6.64 mA cm⁻², V_oc = 0.61 V, and FF = 40.81 %. This result aligns with prior studies [16, 26] that report optimal Sb₂S₃ thicknesses between 100–120 nm for TiO₂/Sb₂S₃/P3HT cells. Excessively thick Sb₂S₃ (> 280 nm) leads to marked performance degradation due to charge transport limitations.

a J–V curves and b EQE spectra for varying Sb₂S₃ thicknesses.

Charge Transport

J_sc rises sharply as Sb₂S₃ thickness increases from 96 to 175 nm, then declines for thicker layers. This trend reflects a balance between enhanced photon absorption (Fig. 6) and reduced internal electric field strength. Using the Beer–Lambert law, the absorbed photon fraction N_a/N_i rises from 61 % (96 nm) to 93 % (373 nm), yet J_sc drops from 6.64 to 2.64 mA cm⁻² when d = 373 nm, indicating that absorption alone cannot explain the behavior.

Illustration of Sb₂S₃ thickness d‑dependent illumination depth x and E_in.

a Semilogarithmic dark J–V plots. b Dependence of V_in, J_sc, N_a/N_i, E_ke, and E_kh on Sb₂S₃ thickness d.

The built‑in voltage V_in decreases with thickness, leading to a reduced internal field E_in = V_in/d. Calculated kinetic energies (E_ke, E_kh) drop below thermal energy (E_kt = 26 meV) for d ≥ 175 nm, indicating that field‑driven drift becomes negligible for holes. Consequently, J_sc declines despite higher absorption. When d exceeds the hole diffusion length (~180 nm), an “inferior” region emerges where holes cannot reach P3HT, further reducing photocurrent and V_oc.

KPFM measurements (Fig. 8) show that surface potential SP on Sb₂S₃ decreases with thickness, reflecting reduced electron diffusion to the surface. P3HT SP behaves oppositely: for thin Sb₂S₃, P3HT is photo‑excited and SP decreases with d; for d > 280 nm, SP rises again due to diminished hole collection. These observations confirm that thicker Sb₂S₃ hampers hole transport into P3HT, lowering both J_sc and V_oc.

Illustration of SP measurement of Sb₂S₃/P3HT interface by KPFM.

IMPS/IMVS analyses (Fig. 10) reveal that electron transit time τ_IMPS increases with Sb₂S₃ thickness, while recombination lifetime τ_IMVS remains constant, indicating that charge collection efficiency η_c = 1 – τ_IMPS/τ_IMVS decreases for thicker layers. This supports the conclusion that transport path, not recombination, limits performance when d > 175 nm. Electron drift is confined near the TiO₂ interface, whereas holes must traverse the full Sb₂S₃ thickness; beyond the hole diffusion length, the collection efficiency drops sharply.

a IMPS and b IMVS spectra. c Illustration of electron and hole diffusion for short/long wavelengths. d τ_IMPS and τ_IMVS versus CBD time.

Although the current PCE of TiO₂/Sb₂S₃/P3HT cells remains modest, the insights gained point to clear pathways for improvement: enhancing the built‑in field via alternative transport layers, increasing hole diffusion length (e.g., by adding conductive dopants), and optimizing interfacial engineering to promote efficient charge transfer.

Conclusion

This work elucidates the mechanisms governing photocurrent in TiO₂/Sb₂S₃/P3HT n‑i‑p solar cells across a wide range of Sb₂S₃ thicknesses. When the absorber is thinner than the hole diffusion length, absorption and internal electric field primarily control photocurrent; when thicker, a “dead” region forms that limits hole collection and reduces both J_sc and V_oc. KPFM and IMPS/IMVS measurements confirm that thicker Sb₂S₃ layers suffer from diminished electron surface potential and prolonged charge transport, leading to lower collection efficiency. These findings highlight the critical balance between photon harvesting and charge transport and provide a roadmap for optimizing Sb₂S₃‑based hybrid cells.

Abbreviations

CBD:

Chemical bath deposition

E_in:

Internal electric field

E_ke:

Kinetic energy of the electron

E_kh:

Kinetic energy of the hole

E_kt:

Thermal energy at ambient temperature

EQE:

External quantum efficiency

FF:

Fill factor

IMPS:

Intensity‑modulated photocurrent spectra

IMVS:

Intensity‑modulated photovoltage spectra

J–V:

Current density–voltage

J_sc:

Short‑circuit current

KPFM:

Kelvin probe force microscope

N_a:

Absorbed photon

N_i:

Incident photons

P3HT:

Poly(3‑hexylthiophene‑2,5‑diyl)

PCE:

Photoelectric conversion efficiency

SEM:

Scanning electron microscopy

SP:

Surface potential

UV‑vis:

Ultraviolet‑visible spectroscopy

v_e:

Drift velocity of the electron

v_h:

Drift velocity of the hole

V_in:

Built‑in voltage

V_oc:

Open‑circuit voltage

XRD:

X‑ray diffraction