Modeling Enhanced Performances by Optical Nanostructures in Water-Splitting Photoelectrodes

Abstract

Material nanostructuring and optical phenomena on a nanoscale such as plasmonic effects and light scattering have been widely studied for improving the solar-to-hydrogen efficiency of photoelectrochemical (PEC) water-splitting electrodes. In this work, we report a method for analyzing the contributions of optical effects from nanostructures for enhancing the PEC performances. Electromagnetic simulations are performed for the precise calculation of generated power density in a semiconductor material. In addition, the transport and transfer of photogenerated charges to the electrolyte are modeled by using the conservation of minority carriers. The surface loss parameter, diffusion length, and doping density of the semiconductor material are determined by fitting the model to an incident photon to current efficiency (IPCE) curve experimentally measured on the bare reference photoelectrode. These parameters are then used to compute the IPCE spectra of the photoelectrode for which an optical enhancement strategy is used, such as nanostructuring or plasmonics. The method is validated using published experimental data. The calculated IPCE enhancement ratio originating from optical effects is in quantitative agreement with experimental observations for both periodic and random optical structures. The model can be used to study in detail the key enhancement mechanisms for the IPCE from optical nanostructures and, in particular, discriminate between optical and nonoptical (e.g., catalytic) enhancement.

High Refractive Index Dielectric Nanoparticles for Optically‐Enhanced Activity of Water‐Splitting Photoanodes

“…We finally compared the effect of TiO 2 nanoparticles with respect to enhancement strategies using plasmonic nanoparticles. Mie scattering theory was applied to TiO 2 nanoparticles as well as 70 nm Ag@SiO 2 nanoparticles that were reported to enhance optically the PEC performances of BiVO 4 under back illumination thanks to their plasmonic properties [25,32] . Figure 6a shows the differential scattering efficiency with respect to the scattering angle at 410 nm for a 300 nm TiO 2 nanoparticle, a 70 nm Ag@SiO 2 nanoparticle (10 nm shell thickness) and a TiO 2 nanoparticle of the same size.…”

“…[32]. The light intensity distribution in periodic geometries describing realistically fabricated Mo:BiVO 4 /Co−Pi/TiO 2 and Mo:BiVO 4 /Co−Pi samples was first calculated with electromagnetic simulations (we showed in our recent work that periodic boundary conditions can be used to describe precisely randomly distributed plasmonic nanoparticles on BiVO 4 [32] ). The model was then simultaneously fitted to the measured EQEs for the bare Mo:BiVO 4 /Co−Pi samples under front and back illumination to determine the hole diffusion length L , the donor density N D and the charge transfer parameter R S .…”

“…The calculated EQEs under front and back illumination was first fitted to the experimentally measured spectra for Mo:BiVO 4 /Co−Pi to determine the hole diffusion length L , the donor density N D and the hole transfer efficiency R S [32] . The optimized values are 62.3 nm, 3.97×10 20 cm −3 and 0.91, respectively, and the coefficient of determination is R 2 =0.95.…”

“…Covering BiVO 4 with such layers cannot be used as a beneficial strategy for water-splitting if there is no proper energy band alignment enabling efficient extraction of the photogenerated charges. Our method for modelling optical enhancement in water splitting photoelectrodes [32] assumes that the semiconductor/electrolyte charge transfer parameter is equal for both the bare sample and the sample including an optical enhancement strategy, which is a wrong assumption in this case.However, the model can be used to study the optical contribution. We simulated, therefore, a geometry including a 300 nm layer of refractive index n = 1.8 on top of BiVO 4 , which yields almost no reflectance in water in the range 430-470 nm, similar to BiVO 4 / TiO 2 (Figure 4a).…”

“…[28] When measurements with a sacrificial hole scavenger in the electrolyte were used to isolate the optical contributions, [29,30] plasmonic effects originating from metallic nanoparticles deposited on the surface were found to be more effective when the light was incident from the substrate side (back illumination). [25,31,32] However, materials such as hematite are known to exhibit higher performances when light is incident from the electrolyte side (front illumination), because of the limiting hole transport with respect to electron transport. [33,34] In addition, numerous reported devices performing bias-free water-splitting include a BiVO 4 [8,[35][36][37][38][39][40][41] or α-Fe 2 O 3 [40,[42][43][44] photoanode working under front illumination.…”