Machine Learning-Assisted Optimization of Iodide Electrolytes for Efficient Indoor Dye-Sensitized Solar Cells with...

Area of Science:

• Materials Science
• Renewable Energy
• Photovoltaics

Background:

• Dye-sensitized solar cells (DSSCs) show promise for indoor photovoltaics (IPVs) and Internet of Things (IoT) devices due to their performance under low light.
• Copper-based redox shuttles offer high photovoltage in DSSCs but lack long-term stability.
• Iodide-based electrolytes are stable but typically yield lower photovoltage under indoor lighting.

Purpose of the Study:

• To optimize iodide-based electrolytes for high-performance indoor DSSCs.
• To enhance photovoltage and power conversion efficiency for IPVs.
• To develop a stable and efficient DSSC technology for indoor applications.

Main Methods:

• A novel photoanode architecture was designed to minimize charge recombination.
• A design of experiments (DoE) approach was integrated with machine learning (ML) for electrolyte optimization.
• Support vector machines and Bayesian optimization were employed to accelerate the development process.

Main Results:

• Achieved power conversion efficiencies exceeding 23% at 1000 Lux under indoor lighting.
• Optimized ML targeting photovoltage, reaching 683 mV at 500 Lux with an iodide/triiodide redox couple.
• Explored electrolyte formulations via Bayesian optimization, yielding 713 mV at 1000 Lux.

Conclusions:

• The developed strategy successfully optimizes iodide-based electrolytes for high-performance IPVs.
• The results demonstrate competitive efficiencies and high photovoltages for iodide/triiodide DSSCs under indoor conditions.
• This work offers a stable and efficient alternative for powering IoT devices and other indoor applications.