Thermoelectric–Photoelectrochemical Water Splitting under Concentrated Solar Irradiation

Photoelectrochemical devices could play a crucial role toward fuel production in a circular economy. Yet, light absorption suffers losses from thermalization and the inability to use low-energy photons. Here, we demonstrate that photoelectrochemical reactors can utilize this waste heat by integrating thermoelectric modules, which provide additional voltage under concentrated light irradiation. While most single semiconductors require external bias, we already accomplish unassisted water splitting under 2 sun irradiation by wiring a BiVO4 photoanode to a thermoelectric element, whereas the photocurrent of a perovskite-BiVO4 tandem system is enhanced 1.7-fold at 5 sun. This strategy is particularly suitable for photoanodes with more positive onset potentials like hematite, with thermoelectric-perovskite-Fe2O3 systems achieving a 29.7× overall photocurrent increase at 5 sun over conventional perovskite-Fe2O3 devices without light concentration. This thermal management approach provides a universal strategy to facilitate widespread solar fuel production, as light concentration increases output, reduces the reactor size and cost, and may enhance catalysis.

After that, the pH was adjusted to 1.20 using concentrated nitric acid. A second solution consisting of benzoquinone (0.292 g, 2.7 mmol, 0.3 M) in absolute ethanol (9 mL) was also sonicated for 3 min. The two solutions were mixed and stirred for 30 min at room temperature to obtain a dark brown BiOI precursor solution. The orange BiOI layer was then S3 electrodeposited onto the active area of the FTO slides, by maintaining a potential of -0.3 V against a Ag/AgCl reference electrode for 5 s, followed by -0.1 V for 180 s. A vanadyl acetyl acetonate (VO(acac)2) solution was prepared by sonicating VO(acac)2 (0.530 g, 2.0 mmol) in 5 mL DMSO for 5 min. 40 μL cm -2 of the VO(acac)2 solution was drop-casted onto the BiOI active areas, before heating the FTO slides at 723 K for 60 min, with a ramp rate of 1 K min -1 .
After the glass slides were left to cool down to room temperature, a NaOH (0.4 M) aqueous solution was used to wash the brownish V2O5 crust from the surface resulting in a bright yellow BiVO4 photoanode. A [Ti4O(OEt)15(CoCl)] precursor was synthesized and spin-coated as previously reported, to form a TiCoOx (TiCo) O2 evolution catalyst. 3 Fe2O3 was prepared by a hydrothermal procedure. 4  A smooth perovskite layer was then deposited by spin coating 0.3 mL of the precursor solution in a two-step procedure, first 10 s at 1000 rpm and then 35 s at 6000 rpm, using 0.6 mL chloroform as the antisolvent ~7 s before the end. The perovskite layer was then annealed at 373 K for 30 min. A thin PCBM layer was deposited as ETL by spin coating ~0.2 mL of a 35 mg mL -1 PCBM solution in chlorobenzene at 3000 rpm for 45 s. Next, a 3.87 μL mL -1 PEIE solution in isopropanol (0.4 mL) was spin coated under ambient conditions at 3000 rpm for S4 30 s, before storing the samples under inert atmosphere. Lastly, a 100 nm silver layer was evaporated through a custom-made mask to form the top electrical contact.
The inverse-structure perovskite cells were then encapsulated with graphite epoxy (GE) paste and a copper foil to form PVK. Graphite powder was mixed with Araldite Standard two component epoxy in 3:4 mass ratio to create a GE paste. 6 The paste was evenly spread on top of the Ag contact layer of the PV device. The copper foil (1.0 mm thick, 44 cm 2 ) was then slightly pressed against the paste to form an electrical contact. The GE was allowed to settle for 24 hours. Finally, Araldite 5-Minute Rapid two component epoxy was used to seal the edges and left to settle overnight ( Figure S18).

Setup assembly.
A TE module is first attached to the aluminium wall of a 3D-printed water reservoir heat sink using thermal paste. The water reservoir is equipped with an additional cooling block, which connects to a chiller, to maintain a steady water bath temperature (25 ℃) on the cold side of the TE module under concentrated irradiation ( Figure S4). The same thermal paste is used to interface the bottom copper contact of the PVK to the hot side of the TE module. A BiVO4 photoanode is next placed on top of the perovskite PV cell. The PEEK reactor is next placed over the TE-PVK-BiVO4 stack and a lid tightens all components in place.
The BiVO4 photoanode is connected in series to a PVK, the TE module, and Pt cathode. In case of the Pt-TE-BiVO4 and Pt-TE-Fe2O3 arrangements, PVK was not used ( Figure S3). CompactStat.e potentiostat in a two-electrode configuration, with the photoanode and sputtered platinum cathode placed side-by-side. All the measurements were conducted in a custom made one-compartment PEC cell, under stirring (see Figures S3 and S4). Prior to the electrochemical S5 measurements, the cell was sealed with rubber septa and the solution was purged with N2 for 30 min. All purging needle holes were then sealed with Loctite superglue Universal adhesive.
Cyclic voltammetry scans were recorded between −0.5 and 1.3 V for the TE-BiVO4, −1.4 and 0.7 V for the TE-PVK-BiVO4, and -1.15 and 0.7 V for the TE-PVK-Fe2O3 (10 mV s −1 scan rate). The STH efficiency of this system was calculated using Equation (S1), where J0V is the bias-free photocurrent density, FY is Faradaic yield, and Ptotal is the total light intensity flux (100-500 mW cm −2 ). 1 (S1)