The study, conducted entirely with the SCAPS-1D simulation software developed at Ghent University, systematically explored how every major design parameter of a solar cell affects its output. The researchers modeled an n-i-p architecture of the form FTO/TiO2/Cs2BiAgI6/Cu2O/Ag, in which a titanium dioxide electron transport layer extracts electrons from the Cs2BiAgI6 absorber, while a copper oxide hole transport layer ferries holes to a silver back contact. Under the standard AM 1.5G solar spectrum at 1000 W/m2 and a temperature of 300 K, the optimized structure produced an open-circuit voltage of 1.2998 V, a short-circuit current density of 26.68 mA/cm2, a fill factor of 83.80 percent, and the headline efficiency of 29.06 percent.

One of the most striking findings concerns the choice of back-contact metal. The team tested silver, gold, and platinum electrodes and found remarkably similar performance across all three. The platinum variant actually edged ahead with a simulated efficiency of 30.03 percent, an open-circuit voltage of 1.2816 V, a current density of 26.69 mA/cm2, and a fill factor of 84.94 percent, while the gold contact yielded 29.00 percent. Because the differences were marginal, the researchers selected silver for their reference design, a pragmatic choice given silver’s cost advantage over platinum and its favorable work function for hole collection.

Layer thickness proved to be a delicate balancing act. Varying the electron transport layer revealed that 120 nm of TiO2 delivered the optimal 29.06 percent efficiency, with efficiency rising only slowly as the layer thickened before the fill factor and power conversion efficiency began to decline. The hole transport layer told a similar story: at 140 nm of Cu2O, the device reached its peak, and the relationship between thickness and efficiency was found to be directly proportional up to that point. These results echo earlier work on organic semiconductor devices, where transport layers must be thick enough to resist field-induced degradation, typically between 10 and 20 nm at minimum, yet thin enough to avoid parasitic light absorption and excessive series resistance.

The absorber layer itself, the heart of the device, was swept from 650 nm to 1050 nm. Efficiency rose steadily with thickness, confirming a positive correlation, although the fill factor moved in the opposite direction. The optimum landed at 750 nm, where the simulated device achieved its maximum performance. This thickness optimization matters because the absorber must be thick enough to capture incoming photons and generate electron-hole pairs, but not so thick that photogenerated carriers recombine before reaching the transport layers.

Perhaps the most consequential parameter was the bulk defect density of the Cs2BiAgI6 absorber. The researchers varied defect concentrations from 10^11 to 10^15 cm−3 and found that efficiency peaked at the lowest density, 10^11 cm−3, remaining nearly constant below that threshold. Simulations at even lower densities suffered from convergence failures in the iterative solver, a common numerical limitation when defect populations become vanishingly small. The physics here is governed by Shockley-Read-Hall recombination, the dominant loss pathway in perovskite devices: defects act as traps that capture charge carriers and dissipate their energy non-radiatively, shortening carrier lifetimes and eroding both quantum efficiency and open-circuit voltage. The recombination rate depends on the carrier capture cross-section, thermal velocity, and trap density, so every order-of-magnitude reduction in defects translates directly into longer diffusion lengths and better charge collection.

Temperature emerged as a persistent adversary. Sweeping the operating temperature from 300 K to 380 K, the team observed that open-circuit voltage, fill factor, and efficiency all declined as temperature rose, while short-circuit current density increased. The mechanism is well understood: additional thermal energy generates more charge carriers and slightly narrows the bandgap, boosting current, but it simultaneously raises series resistance, lowers shunt resistance, and accelerates recombination, dragging down voltage and fill factor. The open-circuit voltage follows a logarithmic dependence on the ratio of short-circuit current to the dark saturation current, so any increase in recombination-driven dark current exacts a heavy toll on voltage. The best performance was therefore obtained at the coolest simulated condition, 300 K, a reminder that thermal management will be essential in real-world deployments.

Bandgap engineering also proved decisive. The Cs2BiAgI6 absorber’s native bandgap of 1.5 eV sits close to the ideal value for single-junction solar energy conversion, and the simulations confirmed that this is where the device shines, delivering roughly 29 percent efficiency. As the bandgap was widened, the material absorbed fewer photons from the solar spectrum, reducing current density, fill factor, and overall efficiency, with the lowest recorded efficiency of 26.08 percent at a wider gap. Beyond about 1.6 eV the simulation became numerically infeasible. The authors note that alloying strategies, such as substituting small amounts of tin into related double perovskites like Cs2AgBiBr6, can push bandgaps as low as 0.5 eV, underscoring how compositional tuning could open further design space without introducing toxicity.

Electron affinity, the energy difference between the vacuum level and the conduction band, was varied from 3.8 to 4.3 eV. Efficiency fell as the affinity increased, with the optimum at 4.00 eV yielding the familiar 29.06 percent benchmark and the worst case at 4.3 eV producing 28.65 percent. The band alignment between the absorber and its transport layers, characterized by conduction band offsets, directly shapes how easily electrons flow out of the absorber and how much voltage is lost at the interfaces. The study highlights that conduction band offset and absorber thickness are among the most critical levers for maximizing performance in double perovskite devices.

Beyond the headline efficiency, the team examined quantum efficiency across the solar spectrum. The simulated devices achieved approximately 98.68 percent quantum efficiency at a wavelength of 360 nm, holding strong through much of the visible range before declining toward 820 nm, while the photon energy of the absorbed light fell from 4.13 eV at 300 nm as wavelength increased. Defects and interface imperfections can devastate quantum efficiency, with prior studies reporting drops from over 90 percent to as little as 20 percent when carrier capture cross-sections grow, which is why precise external quantum efficiency measurements, accurate to within 5 percent for estimating short-circuit current, remain essential for validating any future experimental device. The authors caution that translating these simulated results into the laboratory will be challenging: environmental factors such as moisture and temperature sensitivity, long-term stability, and the economics of large-scale manufacturing all stand between a 29 percent numerical model and a commercial product. Still, the message of the study is unambiguous. With careful control of absorber thickness, defect density, doping, band alignment, and contact materials, lead-free Cs2BiAgI6 double perovskites can theoretically rival the efficiency of lead-based champions while eliminating one of the technology’s most serious environmental liabilities, offering a credible path toward sustainable, eco-friendly photovoltaics.

Subject of Research: Numerical optimization of lead-free Cs2BiAgI6 double perovskite solar cells using SCAPS-1D simulation

Article Title: A performance analysis of inorganic Cs 2 BiAgI 6 perovskite solar cell: A comprehensive study using SCAPS -1D

Article References: Shawon, M., Fatema, K., Akter, R., Hossain, S., & Shalauddin (2026). A performance analysis of inorganic Cs2BiAgI6 perovskite solar cell: A comprehensive study using SCAPS -1D. Results in Optics, Article 101187. https://doi.org/10.1016/j.rio.2026.101187

Image Credits: AI Generated

DOI: 10.1016/j.rio.2026.101187

Keywords: perovskite solar cells, Cs2BiAgI6, lead-free photovoltaics, SCAPS-1D, double perovskite, power conversion efficiency, Shockley-Read-Hall recombination, defect density, bandgap engineering, electron transport layer, hole transport layer, quantum efficiency