Metal oxides have emerged as attractive candidates for photoelectrochemical water splitting, mainly due to their good stability in aqueous solutions, easy synthesis, and low cost. One of the most promising metal oxide photoanodes is bismuth vanadate (BiVO4). This metal oxide evolves oxygen under illumination with visible light, and is stable in aqueous solutions with pH between 3 and 11. However, in the past, BiVO4 suffered from low charge injection efficiency. This problem has been solved by functionalizing its surface with water oxidation catalysts (e.g. CoPi, RhO2, FeOOH, NiFeOx, MnOx)1-5, resulting in significant performance improvements. However, the true nature of semiconductor-catalyst interactions and the improvement mechanisms are still unclear, and requires further attention. In this study, we used intensity modulated photocurrent spectroscopy (IMPS) to examine the surface charge carrier dynamics of spray-deposited bismuth vanadate (BiVO4) photoanodes and the effect of depositing different co-catalysts on its surface. An LED was used to illuminate the sample with a modulated intensity, and the real and imaginary parts of the opto-electrical impedance were recorded. To interpret the resulting spectra, we used a model developed by Peter et al. that allows one to distinguish the rate constants for surface recombination and charge injection into the electrolyte. We found that the photocurrent of BiVO4 is mainly limited by surface recombination. This is evident by the significant reduction of the surface recombination, when CoPi and NixMnyOz were deposited on the BiVO4 photoanodes. Surprisingly, we did not observe any enhancement of charge transfer rate constant into the electrolyte. This result suggests that BiVO4 is already a good oxidation catalyst by itself, which is consistent with the relatively positive potential of the valence band edge, resulting in a high thermodynamic driving force. This observation contradicts the common assumption that these catalysts are functioning as electrocatalysts, and shows that their role on our BiVO4 is mainly that of a surface passivation layer. Based on this we predict that the modification of BiVO4 surface with RuOx, a well-known (non-passivating) oxygen evolution catalyst, does not improve the performance; our experiments show that this is indeed the case. These results allow us to develop a modified carrier dynamics model, which will be discussed here. We further examined the nature of the surface states with a combination of cyclic voltammetry and spectroelectrochemistry. We also compared BiVO4 prepared by various deposition techniques—both chemical and physical routes—to demonstrate their influence on the nature of the surface of BiVO4. The semiconductor/co-catalyst/electrolyte interface was also investigated by depositing CoPi onto those differently prepared BiVO4 photoanodes. Finally, the diverse effects of porous (ion-permeable) and dense (non-ion-permeable) co-catalyst on the photoelectrochemical performance of such semiconductor/co-catalyst systems will be discussed. [1] Park, Y.; McDonald, K. J.; Choi, K.-S. Chem. Soc. Rev. 2013, 42 (6), 2321. [2] Kim, T. W.; Choi, K.-S. Science (80-. ). 2014, 343 (6174), 990. [3] Ma, Y.; Le Formal, F.; Kafizas, A.; Pendlebury, S. R.; Durrant, J. R. J. Mater. Chem. A 2015, 3 (41), 20649. [4] Zhong, D. K.; Choi, S.; Gamelin, D. R. J. Am. Chem. Soc. 2011, 133 (45), 18370. [5] Trotochaud, L.; Young, S. L.; Ranney, J. K.; Boettcher, S. W. J. Am. Chem. Soc. 2014, 136 (18), 6744.