Selective CO production from aqueous CO2 using a Cu96In4 catalyst and its integration into a bias-free solar perovskite-BiVO4 tandem device

Porous dendritic copper–indium metal alloy foam catalysts are interfaced with a perovskite‖BiVO4 tandem device for solar CO2-to-CO conversion under bias-free conditions using water as an electron donor.


Synthesis of the catalyst.
The CuxIny alloy catalysts were synthesized via a template assisted electrodeposition method following previously reported protocols. 1, 2 Cu foil was used as substrate for electrodeposition.
Prior to electrodeposition, the substrates were electropolished in 50% ortho-phosphoric acid by applying +2.0 V for 90 s.
The electrodeposition of the bimetallic material was conducted in a clean glass beaker containing copper sulfate and indium sulfate precursor salts (0.025 M) in 1.5 M sulfuric acid solution. Different ratios of precursor salts were used to vary the metal composition in the alloy. A three-electrode set-up was used where the Cu foil substrate was used as working electrode, a Cu foil (4 x 4 cm) as the counter electrode and a leak-free Ag/AgCl (saturated NaCl, BASI) electrode as the reference electrode. For the galvanostatic deposition process, a current density of j = −3.0 A cm −2 was applied for 60 s. After electrodeposition, the catalysts were cleaned by dipping into Milli-Q water for 120 s and dried under N2 stream at room temperature.
Deposition of inverse structure perovskite. Inverse structure triple cation mixed halide perovskite photovoltaic cells were prepared by adapting a previously reported method. 3  annealed at 373 K for 30 min. The perovskite layer appeared as a transparent black film on top of the substrate. Then, a thin [6,6]-phenyl C61 butyric acid methyl ester (PCBM) electron transport layer was deposited on top of the perovskite filmby spin coating 35 mg mL -1 PCBM solution in chlorobenzene at 3000 rpm for 45 s. A PEIE film was also deposited on top of the PCBM coated perovskite layer by spin coating 3.87 µL mL -1 PEIE solution in IPA at 3000 rpm for 30 s to prevent interfacial degradation by reacting with the metal contact. A conductive silver layer was deposited by metal evaporation as electrical contact between the perovskite and the encapsulating graphiteepoxy. The 100 nm Ag layer was deposited in such a way that the active perovskite area becomes around ~0.5 x 0.5 cm. All the photovoltaic cells used in this study have an active area between 0.225 to 0.275 cm 2 .
Preparation of BiVO4‫|‬TiCo photoanode. BiVO4 photoanodes were prepared following a reported procedure. 3,4 At first BiOI was electrodeposited onto an FTO substrate (with a defined exposed Assembly of the tandem device. The BiVO4 photoanode was also encapsulated by epoxy after attaching a connecting wire with the help of conducting Ag paste. For an artificial leaf-type tandem assembly the perovskite‫|‬Cu96In4 cathode and the BiVO4 photoanode were attached to each other by epoxy glue. To prevent excess photoabsorption, the inactive area of the BiVO4-TiCo phonoanode was covered by black tape. The active surface areas of perovskite, BiVO4, and catalyst were measured before the assembly.

PEC and EC measurements and product quantification.
A certified Newport 1916-R optical power meter was used to calibrate the Newport Oriel 67005 solar light simulator with Air Mass 1.5 Global (AM 1.5G) solar filters to 100 mW cm -2 (1 Sun) prior to each PEC experiment. The lower light intensities were obtained by additionally employing neutral density filters with 50% and 20% transmission. All electrochemical and PEC experiments were conducted with a PalmSens Multi EmStat 3+ (multichannel potentiostat consisting of four separate channels) and Ivium CompactStat potentiostats. The reaction medium was an aqueous 0.5 M KHCO3 solution, which was purged for at least 30 min prior to the experiments with CO2 or N2 (with 2% methane internal standard). A three-electrode set up consisting of a Ag/AgCl (sat. NaCl) (BasiMW-2030) reference, a platinum mesh counter, and a CuxIny or a perovskite‫|‬CuxIny working electrode was used in a twocompartment gas tight cell for the electrochemical and PEC measurements. PEC measurements were performed under chopped light irradiation (50 min on, 10 min off). All the experiments were carried out at room temperature. A Selemion (AGC Engineering) anion exchange membrane was used to separate the cathode and anode compartments. An additional EC experiment was carried out with an ultrapure KHCO3 which was prepared by purging CO2 through 99.995% K2CO3 solution and then further purified by soaking for at least 24 h in regenerated Chelex 100 sodium form resin S6 (50-100 mesh (dry), Sigma Aldrich) to test for any effects from trace amounts of heavy metal ions (which might be present in the as-purchased electrolyte) on the catalytic activity of Cu96In4 material The lower total faradaic efficiency at the beginning of the experiment was presumably due to bubble trapping inside the porous catalyst architecture, and solubilized gaseous products inside the electrolyte solution. With time, the concentration of the gaseous products increases into the headspace, thus the total product efficiency is close to 100%.   column (b, d, f) shows the same for a Cu96In4 catalyst after 10 h biasfree PEC experiment. The initial ratio of (Cu + CuxO) and In components was 88 : 12 on the surface of the as-prepared catalyst. After the experiment, the ratio becomes 93 : 7. The decrease of about 40% in the In percentage supports the migration of the In phase from the surface to the bulk over the course of the experiment. Note that the surface elemental ratio from the surface-sensitive XPS analysis differs from the actual bulk composition determined by the EDX and ICP-OES analysis. 2). The carbonate / bicarbonate adsorption peaks can be observed along with the adsorbed C≡O peaks. Note that the CuxO peaks at 500 to 700 cm -1 range can be observed for both samples at OCP and they disappear after a cathodic potential has been applied. This suggests that the surface oxides are unstable under the cathodic environment and they are reduced to the metallic phase when the negative potential is applied. The spectro-electrochemistry was performed in a three-electrode configuration using a 633 nm laser. The experiments were carried out at room temperature.    View from the catalyst side when the buried perovskite-biased cathode is attached to a metal rod support. A BiVO4‫||‬perovskite‫|‬Cu96In4 tandem device before (d) and after (e,f) attaching to a metal rod support where (e) represents view from the catalyst side and (f) represents view from BiVO4 side. Dark adhesive tape is used to cover the area surrounding the BiVO4 to block excess light from reaching the perovskite active area.