One of the most promising ways of storing solar energy is converting it to electrochemical fuels [1]. The solar redox flow cell (SRFC), which uses a photoelectrochemical (PEC) cell to photocharge fast-kinetics redox couples, follows this concept. In future applications, SRFCs will be able to cogenerate dispatchable electric and thermal energies, fitting the needs of nearly zero energy buildings (NZEB) [2] with overall energy conversion efficiencies above 70 %. Research in the field of SRFCs is currently focused on the development of high performance photoelectrodes (PE) and redox couples whose energy-level matching should allow for unbiased and efficient photocharge (efficiencies of ca. 20 % were reported very recently [3]). Despite these efforts, little attention has been given to device architecture and to the upscaling of PEs, which is expected to strongly contribute in improving the performance of SRFCs and accelerate its implementation [4]. At present, most reported SRFC devices have photoabsorber active areas of ≤ 1 cm² and were developed for lab-scale proofs-of-concept.

This work addresses the challenges related with the design, manufacture and upscale of SRFCs and goes beyond the state-of-the art, proposing an innovative and user-friendly 25 cm² device, the SolarFlow25 cell. To aid the device’s design, a CFD model was developed to implement the following key features: i) the front window of the cell is the 25 cm² PE for maximized radiation absorption by the semiconductor; ii) an optimized and continuous electrolyte recirculation is promoted to ensure effective diffusion and convection of redox species; iii) the counter electrode and membrane are positioned in a zero-gap back-to-back configuration, guaranteeing minimal ionic transport resistances. To demonstrate the photocharging performance of the device, a hematite (α-Fe₂O₃) photoelectrode connected in series with a dye sensitized solar cell was chosen as photoactive material. The electrolytes consisted of alkaline aqueous solutions of sodium ferrocyanide (Na₄Fe(CN)₆) and anthraquinone-2,7-disulphonate (2,7-AQDS). Connecting the proposed system with a conventional redox flow battery (RFB) allowed the electrochemical fuels to be photocharged and discharged in continuous operation, converting the stored energy into dispatchable electricity. An average unbiased photocurrent of ca. 11 mA was recorded (during photocharge) under 1000 W∙m^-2, with Coulombic efficiencies >80 % for 10 cycles. The solar-to-output-energy efficiency remained stable at ca. 0.15 % and no leakage or electrical problems were detected. These results validate the proposed device as a powerful tool for testing and developing different materials using close-to-commercialization working conditions.