Determination of Electron Extraction in Semiconductor Photoanodes
Photoelectrochemical (PEC) water oxidation using semiconductor photoanodes is increasingly attracting attention due to its potential as an environmentally-friendly method for converting solar energy into hydrogen. Further development in the efficiency of operation of photoanodes requires a fundamental understanding of the physical mechanisms occurring within them – in particular, the charge carrier transport, recombination, and transfer at the photoanode/electrolyte interface or the collecting electrode. While existing optoelectronic models capture the effect of specific mechanisms occurring in the bulk or at the photoanode/electrolyte interface, they cannot be applied to consistently reproduce both the steady-state and transient optoelectronic response (time and frequency domain) of the photoanode. Furthermore, these models do not account for a fundamental feature in all photovoltaic devices – the difference between the internal (average quasi-Fermi level splitting within the photoanode) and external voltage that drives the photocurrent and the excess recombination current arising from non-ideal charge carrier extraction.
This work develops an analytical model to consistently interpret the steady-state and small-perturbation response (both in the time and frequency domain) of photoanodes for solar water-splitting. In addition to accounting for the fundamental mechanisms of charge-carrier generation, recombination, and slow hole transfer at the photoanode/electrolyte interface, the model overcomes the key shortcomings of existing models in the literature. These include consistency across measurement/bias conditions and the non-consideration of imperfect electron extraction at the collecting contact and its corresponding effect on the recombination rate in the bulk. The model is applied to analyse the time constants obtained from intensity-modulated photocurrent spectroscopy (IMPS) and intensity-modulated photovoltage spectroscopy (IMVS) measurements of a hematite photoanode, obtaining an electron extraction velocity of 100 cm s−1 close to the 1 sun open-circuit potential, that corresponds to an electron mobility of 0.022 cm2 V−1 s−1. The model further predicts a linear dependence of the photocurrent versus anodic voltage, an observation whose origin is strongly debated in the literature in the case of hematite photoanodes. The generality of the model allows its extension to other photoanodes and photovoltaic systems, by the addition or removal of specific physical mechanisms.
Further informations ca be found here