The widescale deployment of renewable fuels requires cost effective energy conversion technologies. Photoelectrochemical (PEC) hydrogen production is one promising approach that has demonstrated a high solar-to-conversion (STH) efficiency, as high as 15% using GaInP based tandem cell[1], albeit further improvements of STH, durability, and device cost are still necessary. Chalcopyrite based photoelectrodes are very attractive options for the PEC approach in that they offer a great flexibility in choice of component materials, which offers ways to reduce the cost and to improve the performance. However, the greater flexibility simultaneously introduces additional complexity in understanding the correlation between basic material properties and the device performance, which is a major challenge for developing a synthesis procedure to obtain the material with intended property.

Based on first-principles DFT simulations coupled with the state-of-art experimental characterization, we intend to comprehensively address the relationships between the expected properties of materials and the measured performance in a real device. Particularly with the chalcopyrites, the material property often deviates significantly from the ideal due to the constraint (or choices) in the synthesis procedure, which tends to obscure the factors that may influence the device performance.

We will first review the basic properties (band alignments, alloy phase diagram etc.) of candidate chalcopyrite materials under thermodynamic condition, and discuss how the calculated alloy phase diagram is used to estimate the range of usable synthesis condition.[2-4] The calculated defect energy diagrams as a function of chemical potential of constituent elements can be then used to predict types and concentrations of defects that are likely to present for a given synthesis condition, and assess about likely scenario about the performance of device based on charge carrier lifetime, the Fermi level pinning, and doping characteristics (propensity of being p-type or n-type). Here, let us emphasize that combining theoretical information with experimental characterization can be used to calibrate the theoretical predictions that may not capture some relevant factors that may be present in the real device materials. Such an integrated approach will allow us to provide more reliable and effective feedbacks for development of optimal synthesis and processing steps.

With this in mind, we will talk about the current knowledge on the relevant factors for chalcopyrite photoelectrodes such as alkaline impurities, grain boundaries, composition gradient (ex. copper deficiency), diffuse hetero-junction profile, and discuss our strategy for addressing these issues.

This work performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344 and funded by the Department of Energy office of Energy Efficiency & Renewable Energy (EERE).

References:

1. Todd Duetsch et al., 2016 DOE Annual Merit Review (https://www.hydrogen.energy.gov/pdfs/review16/pd115_deutsch_2016_o.pdf)

2. V. Stevanovic et al. PCCP 16, 3706 (2014)

3. J. Varley, V. Lordi, J. App. Phys. 116, 063505 (2014)

4. J. Varley, V. Lordi, N. Gaillard, T. Ogitsu, in preparation.