(488f) Model-Guided Optimization of Polymer-Electrolyte Dye Sensitized Solar Cells

Dye sensitized solar cells (DSSCs) have several appealing attributes. Their components and fabrication are inexpensive, are easy to fabricate, have tunable optical properties such as color and transparency, and can be integrated in unique ways into buildings and apparels. A DSSC is composed of three main components: (1) a photosensitive dye which converts solar light into electrons, (2) a mesoporous TiO₂ photoanode which provides a large surface area for dye adsorption and a pathway for electron transport, and (3) a liquid electrolyte comprised of a I^-/I^-₃ redox couple which is used to regenerate the dye and acts as a hole transport medium. These components have been studied over the past two decades but the conversion efficiency of the cells has increased only by ~2% to 13%. Most of the research to design and optimize DSSCs during this timeframe has been based on scientific intuition and trial-and-error experimentation, a slow and inefficient process. However, there have been a few computational/theoretical optimization studies. First-principles-model-guided design and optimization can lead to improved design and operation of DSSCs through systematically calculating optimal design specification and operating conditions.

First-principles modeling combined with experimental knowledge can provides a deeper understanding into DSSC material properties and processes than experimental knowledge alone. For example, Mathew et al. [1] used quantum chemical calculations to design a dye molecule that led to a record DSSC efficiency. This work showed that a rational, model-guided design of DSSC materials can improve the DSSC performance substantially. In our previous work, we used first-principles macroscopic mathematical models to theoretically and experimentally investigate the effects of polymer-electrolyte chemistry on the performance of polymer-electrolyte DSSCs [2]. We determined how the polymer chemistry affects DSSC interfacial processes and subsequent affected device performance. Polymer electrolytes can reduce interfacial recombination and allow for addressing the disadvantages of liquid electrolytes such as being prone to leakage and evaporation which hinder DSSC applications.

In this work, we first conduct DSSC optimization based on a first-principles DSSC mathematical model to calculate optimal design specifications for a polymer-electrolyte DSSC and then fabricate and test the optimal cell to validate the optimality of the cell. Optimal values of DSSC design variables such as photoanode thickness, dye absorption coefficient, iodide and triiodide diffusion coefficient, irradiation intensity, and polymer chemistry are will be calculated. The mathematical model includes Bulter-Volmer kinetics along with time-dependent continuity and transport equations for all charged speciesâ??including electrons, iodide, triiodide, and lithium (the cation)â?? to describe the cell and to predict cell conversion efficiency and current-voltage behavior. The optimal design specifications that maximize the conversion efficiency are calculated by solving a constrained nonlinear optimization problem using a Nelder-Mead algorithm. Optimization validation will be conducted with two sets of DSSC experiments: (1) a DSSC is fabricated based on the predicted optimal design specifications, and the conversion efficiency of the fabricated cell is compared to the conversion prediction predicted by the model. (2) Cells with design specifications that are perturbed values of the theoretically-calculated optimal design specifications are fabricated and tested for conversion efficiency to investigate the true optimality of the design specifications.

[1] S. Mathew, A. Yella, P. Gao, R. Humphry-Baker, F. E. CurchodBasile, N. Ashari-Astani, I. Tavernelli, U. Rothlisberger, K. NazeeruddinMd, M. Grätzel, Nat Chem 2014, 6, 242.

[2] Y. Y. Smolin, S. Nejati, M. Bavarian, D. Lee, K. K. S. Lau, M. Soroush, Journal of Power Sources 2015, 274, 156.