Enhancing Solid‑State CuInS₂ Quantum‑Dot Solar Cells: Optimizing Charge Recombination via Controlled SILAR Deposition and Annealing
Abstract
CuInS₂ quantum dots (QDs) were integrated into a mesoporous TiO₂ film through a spin‑coating‑assisted successive ionic layer adsorption and reaction (SILAR) protocol to fabricate solid‑state QD‑sensitized photoanodes. The photovoltaic output of the resulting QD‑sensitized solar cells (QDSSCs) depends strongly on the number of SILAR cycles, which governs QD coverage and surface defect density on TiO₂. Subsequent high‑temperature annealing further reduces charge recombination and accelerates electron transport. Annealing at 400 °C delivers a record 3.13 % power‑conversion efficiency (PCE), with an open‑circuit voltage of 0.68 V, a short‑circuit photocurrent density of 11.33 mA cm⁻² and a fill factor of 0.41, attributed to suppressed recombination and enhanced electron injection.
Background
Quantum‑dot‑sensitized solar cells (QDSSCs) combine multiexciton generation with tunable band gaps, making them strong candidates for next‑generation photovoltaics. Selecting a sensitizer with an optimal band gap is essential for high efficiency. CuInS₂ (CIS) is a direct‑gap I‑III‑VI₂ semiconductor with a bulk band gap of ~1.5 eV, an absorption coefficient approaching 10⁵ cm⁻¹, non‑toxicity and excellent stability, which together render it an attractive photosensitizer. Earlier reports of CIS‑based QDSSCs have reached efficiencies up to 8.5 % using post‑synthetic assembly; however, this approach limits QD loading and electron coupling with TiO₂. Direct deposition via SILAR can densely load QDs and improve interfacial electronic coupling. Solid‑state QDSSC architectures avoid the long‑term stability issues of liquid electrolytes, yet reported efficiencies remain modest. Here we push the efficiency of solid‑state CIS‑QDSSCs beyond 3 % by optimizing SILAR deposition and annealing.
Methods
Materials
Indium acetate, copper(II) chloride dihydrate, sodium sulfide nonahydrate, titanium isopropoxide, hydrochloric acid, spiro‑OMeTAD, chlorobenzene, tert‑butylpyridine, lithium bis(trifluoromethanesulfonyl)imide, and acetonitrile were sourced from Alfa Aesar and Sigma‑Aldrich. TiO₂ paste (DSL 18NR‑T) was obtained from Dyesol. All chemicals were used without further purification. Ultrapure deionized water was employed for aqueous solutions.
Preparation
A 70 nm compact TiO₂ layer was spin‑coated onto cleaned FTO glass (4000 rpm, 30 s) from a mixture of 350 µL titanium isopropoxide and 35 µL HCl in 5 mL ethanol, followed by a multi‑step air anneal (incrementally to 500 °C, 1 h at 500 °C). A 2‑µm mesoporous TiO₂ layer was deposited by spin‑coating diluted 18NR‑T paste (800 rpm, 10 s) and heat‑treated. CIS QDs were deposited onto the mesoporous layer by spin‑coating‑assisted SILAR: 80 µL of 25 mM CuCl₂ + 50 mM In(OAc)₃ solution was applied and spun (800 rpm, 20 s), followed by 80 µL of 100 mM Na₂S, also spun at 800 rpm. One SILAR cycle consists of these two steps. The film was rinsed with deionized water and dried with N₂ between cycles. After the desired number of cycles, the photoanodes were annealed under N₂ at temperatures between 200–500 °C for 30 min to improve CIS crystallinity. A hole‑transport material (HTM) solution containing 300 mg spiro‑OMeTAD, 2.91 µL chlorobenzene, 28.77 µL tBP and 126 µL Li‑TFSI was spin‑coated under N₂. Gold counter electrodes (0.09 cm² active area) were thermally evaporated.
Characterization
UV‑vis absorption spectra were recorded on a Perkin Elmer Lambda 950. Cross‑sectional SEM was performed on an FEI nova nano SEM450, with energy‑dispersive spectroscopy (EDS) mapping. Current–voltage (J‑V) characteristics were measured under AM1.5 (100 mW cm⁻²) illumination using a 300 W xenon lamp (XES‑100S1). Incident photon‑to‑current efficiency (IPCE) spectra were obtained with an Enlitech QER3011 system (150 W xenon). Electrochemical impedance spectroscopy (EIS) was conducted on a Zahner Zennium workstation (−0.1 to −0.5 V bias, 20 mV AC, 0.1–1 Hz). Time‑resolved photoluminescence (TRPL) was measured with an Edinburgh Instruments FLS 900, excited at 543 nm.
Results and Discussion
The device architecture consists of a TiO₂ photoanode sensitized with CIS QDs, a spiro‑OMeTAD HTM and a gold counter electrode (Figure 1). SEM cross‑sections reveal uniform QD distribution and intimate contact between layers. EDS mapping confirms homogeneous dispersion of Cu, In and S across the mesoporous TiO₂ matrix.
Figure 2 illustrates the spin‑coating‑assisted SILAR procedure. UV‑vis spectra (Figure 3a) show that the absorbance increases with the number of cycles, with a slight red shift indicating growth of the CIS layer. The film color evolves from pale yellow (4 cycles) to deep black (20 cycles). J‑V curves (Figure 3b) demonstrate that both J_SC and PCE rise from 2.49 mA cm⁻² / 0.14 % (4 cycles) to 4.21 mA cm⁻² / 0.75 % (20 cycles), then slightly decline at 24 cycles, reflecting an optimal QD loading that maximizes light harvesting while minimizing recombination in the thicker CIS layer.
Annealing studies (Figure 5) reveal that absorbance peaks at 400 °C, beyond which aggregation and oxidation reduce the optical response. J‑V measurements (Figure 6a) show the highest PCE of 3.13 % at 400 °C, with V_OC = 0.68 V, J_SC = 11.33 mA cm⁻² and FF = 0.41. IPCE spectra (Figure 6b) confirm a strong response (≈66 %) in the 400–550 nm range, a 20 % improvement over the 200 °C device, indicating enhanced light harvesting and electron injection.
EIS (Figure 7a) indicates a larger recombination resistance (R_r) and longer electron lifetime (τ_n ≈ 117 ms) for the 400 °C annealed device, consistent with reduced interfacial recombination. TRPL (Figure 8) shows a shorter PL lifetime with increasing annealing temperature, signifying faster electron transfer from CIS to TiO₂. The electron‑transfer rate (k_et) reaches 1.17 × 10⁷ s⁻¹ at 400 °C, supporting the superior charge‑transport dynamics.
Conclusions
Spin‑coating‑assisted SILAR enables precise control over CIS QD loading on TiO₂, while high‑temperature annealing (400 °C) optimizes the CIS/TiO₂ interface, suppresses recombination and accelerates electron injection. The resulting solid‑state CIS‑QDSSCs achieve a PCE of 3.13 %, V_OC of 0.68 V, J_SC of 11.33 mA cm⁻², FF of 0.41, IPCE of 66 % (400–550 nm) and an electron‑transfer rate of 1.17 × 10⁷ s⁻¹. This work demonstrates a viable route to high‑performance solid‑state QDSSCs and provides guidance for further stability improvements.
Nanomaterials
- High‑Efficiency Graphene Solar Cells: 9% Power Conversion with TFSA Doping
- Enhancing Photovoltaic Efficiency by Detecting Spatially Localized Excitons in Self‑Organized InAs/InGaAs Quantum Dot Superlattices
- Optimizing Gold Nanoparticle Placement in TiO₂ Enhances Dye‑Sensitized Solar Cell Efficiency
- Boosting Perovskite Solar Cell Efficiency Using Er3+-Yb3+-Li+ Tri‑Doped TiO₂ Up‑Conversion Layer
- Enhanced Power Conversion in Flexible Fibrous Dye‑Sensitized Solar Cells via Multilayer TiO₂ Photoanodes and Composite Pt Counter Electrodes
- Dye‑Sensitized Solar Cells: Fundamentals, Advances, and Commercial Outlook
- Significant Efficiency Boost in CdS/CdSe Quantum Dot-Sensitized Solar Cells Using (001)-Oriented TiO2 Nanosheet Photoanodes
- Optimizing Niobium-Doped Titanium Oxide Layers for Enhanced Silicon Quantum Dot Solar Cells
- Enhancing Quantum Dot Solar Cell Efficiency with Concentrating Photovoltaics and Dual Photoanodes
- Climate Change Threatens Solar Panel Efficiency: MIT Study Reveals Impact