Enabling Unassisted Solar Water Splitting By Hematite

As a photoelectrode material for solar water splitting, hematite has received tremendous attention. The appealing properties presented by hematite include earth abundance, stability against corrosion, suitable band gap, and nontoxicity. It, however, also presents a long list of challenges, such as notoriously short hole diffusion distance, poor catalytic activity, and low photovoltage. Among these challenge, the low photovoltage has received the least attention. We have shown that the low photovoltage is a result of combined effects of band edge position mismatch with water redox potentials and poor surface energetics. By interfacing with NiFeOx, an amorphous oxygen evolution catalyst, we have measured greatly improved photovoltages (from 0.24 V to 0.61 V). Importantly, the improved photovoltage was understood as a result of improvement in surface energetics, but not better water oxidation kinetics enabled by NiFeOx. To further test what is the true limit of the photovoltage generation capability by hematite, we recently carried out a detailed structural study on hematite and found that surface defects are responsbile for additional photovoltage lost. A facile regrowth strategy was developed to correct the deficiency. A record-low turn-on voltage of 0.45 V (vs. RHE) was achieved. The results permitted us to construct, for the first time, unassistant water splitting by hematite with a Si photocathode. An overall efficiency of 0.91% was measured.