Photoelectrolysis of water offers a convenient and scalable solution to store solar energy in the form of a chemical fuel, i.e., hydrogen. This can be achieved by hybrid PV-electrolyzers, which consist of highly efficient photovoltaic devices modified with suitable electro-catalysts for the oxygen- and hydrogen evolution reactions (OER, HER). However, an important disadvantage of all conventional hybrid PV-electrolyzers studied so far is the shadowing and scattering of light by the catalyst deposited on the transparent front-contact of the cell. We overcome this problem by using a triple-junction amorphous silicon solar cell (3-jn a-Si) in a superstrate geometry. This special geometry allows unhindered illumination via the front, since both the OER and HER catalysts are deposited onto the backside of the cell. Shadowing and scattering effects are thus completely avoided.

After some structural modifications of the back contact, the 3-jn a-Si solar cell shows excellent chemical stability in a 0.5M sulfuric acid solution, even after 20h of continuous illumination (1000W/m² at AM 1.5G). The next challenge is the deposition of the catalysts, which has to be done at low temperatures since the solar cell is destroyed at temperatures above 200°C. To achieve good mechanical adhesion at these low temperatures, we embed the catalytically active nanoparticles in a conducting polymer, PEDOT:PSS. To demonstrate this concept, we first used standard RuO₂ and platinum nanoparticles for the OER and HER. This resulted in a solar-to-fuel efficiency (SFE) of 3.7%, and and stable performance for more than 18 h.

In order to replace the expensive platinum, we proceeded to use MoS₂ as an alternative HER-catalyst. To achieve an optimal dispersion, the MoS₂ was deposited on multi walled carbon nanotubes (MWCNT) using a wet chemical process [1]. The carbon-supported molybdenum disulphide particles were characterized by X-ray diffractometry, transmission electron microscopy and cyclic voltametry. As transition metal sulphides are known to degrade and evolve dihydrogen sulphide (H₂S), differential electrochemical mass spectroscopy was also applied in order to distinguish between HER and corrosion processes. Surprisingly, no dihydrogen sulphide evolution was detected for our MWCNT/MoS₂ composite.

Finally, the MoS₂ particles were embedded in PEDOT:PSS and deposited on the back contact of the superstrate 3-jn a-Si solar cell to make a complete water splitting device. The SFE for the combined system, in the absence of an externally applied bias, was 2.7% under AM1.5G solar radiation.

[1] A. Laurson et al., Energy & Environmental Science, 2012, 5, 5577