Photocorrosion of WO₃ Photoanodes in Different Electrolytes

Abstract

Photocorrosion of an n-type semiconductor is anticipated to be unfavorable if its decomposition potential is situated below its valence band-edge position. Tungsten trioxide (WO 3 ) is generally considered as a stable photoanode for different photoelectrochemical (PEC) applications. Such oversimplified considerations ignore reactions with electrolytes added to the solvent. Moreover, kinetic effects are neglected. The fallacy of such approaches has been demonstrated in our previous study dealing with WO 3 instability in H 2 SO 4 . In this work, in order to understand parameters influencing WO 3 photocorrosion and to identify more suitable reaction environments, H 2 SO 4 , HClO 4 , HNO 3 , CH 3 O 3 SH, as electrolytes are considered. Model WO 3 thin films are fabricated with a spray-coating process. Photoactivity of the samples is determined with a photoelectrochemical scanning flow cell. Photostability is measured in real time by coupling an inductively coupled plasma mass spectrometer to the scanning flow cell to determine the photoanode dissolution products. It is found that the photoactivity of the WO 3 films increases as HNO 3 < HClO 4 ≈ H 2 SO 4 < CH 3 O 3 SH, whereas the photostability exhibits the opposite trend. The differences observed in photocorrosion are explained considering stability of the electrolytes toward decomposition. This work demonstrates that electrolytes and their reactive intermediates clearly influence the photostability of photoelectrodes. Thus, the careful selection of the photoelectrode/electrolyte combination is of crucial importance in the design of a stable photoelectrochemical water-splitting device.

Accessing In Situ Photocorrosion under Realistic Light Conditions: Photoelectrochemical Scanning Flow Cell Coupled to Online ICP-MS

“…Therefore, the S-number concept was adapted to account for any oxidized species using the total number of transferred electrons ( n (e – )) The S-number­(e – ) based on the second CA hold under illumination was calculated to be 8850 and is in the ballpark of previously reported values for WO 3 in HClO 4 conducted in our group . The calculated S-number­(e – ) for the dark CA hold equals to about 21 700.…”

“…However, it remains disputed whether OER is the sole reaction occurring on the WO 3 surface with HClO 4 as the electrolyte, as ClO 4 – can undergo decomposition reactions to form radical species . Therefore, the S-number concept was adapted to account for any oxidized species using the total number of transferred electrons ( n (e – )) The S-number­(e – ) based on the second CA hold under illumination was calculated to be 8850 and is in the ballpark of previously reported values for WO 3 in HClO 4 conducted in our group .…”

“…Previous PEC-ICP-MS setups utilizing a 385 nm LED detected an onset potential of dissolution around 1.0 V RHE for WO 3 . 7,36 With the emitted polychromatic spectrum from the solar simulator including wavelengths below 385 nm, it is suggested that the additional high energy irradiation leads to an altered dissolution characteristic compared to results using monochromatic light. Once the potential passes 1.2 V RHE (gray dashed line in Figure 4) the dissolution rate experiences a substantial incline indicated by the two red dashed lines in Figure 4c representing linear fits of the dissolution rate before and after 1.2 V RHE .…”

“…Such photocorrosion behavior in conjunction with the current response was not observed in previous PEC-ICP-MS studies with WO 3 . 7,36 Encountered discrepancies in the photocorrosion behavior between 385 nm and AM 1.5 G implies the importance of utilizing realistic irradiation conditions to understand photostability properties for photoelectrodes in real devices. Applying light to a CA hold at 1.5 V RHE has a more detrimental effect on the photoanode, causing a steep increase in dissolution until peaking around 0.2 ng cm −2 s −1 .…”

Photocorrosion of n‐ and p‐Type Semiconducting Oxide‐Covered Metals: Case Studies of Anodized Titanium and Copper

“…[ 22 ] In contrast, photocorrosion is an issue with this system. [ 23 ] Other relevant materials for which photocorrosion data has recently become available are tungsten oxide‐based photoanodes, [ 24 ] and zinc oxide single crystals. [ 25 ] Overall, special surface modification methods are used for inhibition.…”