Water Oxidation Catalysis Beginning with 2.5 μM [Co4(H2O)2(PW9O34)2]10-: Investigation of the True Electrochemically Driven Catalyst at ≥600 mV Overpotential at a Glassy Carbon Electrode

Evidence for the true water oxidation catalyst (WOC) when beginning with the cobalt polyoxometalate [Co-4(H2O)(2)(PW9O34)(2)](10)- (Co-4-POM) is investigated at deliberately chosen low polyoxometalate concentrations (2.5 mu M) and high electrochemical potentials (>= 1.3 V vs Ag/AgCl) in pH 5.8 and 8.0 sodium phosphate electrolyte at a glassy carbon working electrode-conditions which ostensibly favor Co-4-POM catalysis if present. Multiple experiments argue against the dominant catalyst being CoOx formed exclusively from Co2+ dissociated from the parent POM. Measurement of [Co2+] in the Co-4-POM solution and catalytic controls with the corresponding amount of Co(NO3)(2) cannot account for the O-2 generated from 2.5 mu M [Co-4(H2O)(2)(PW9O34)(2)](10-) solutions. This result contrasts with our prior investigation of Co-4-POM under higher concentration and lower potential conditions (i.e., 500 mu M [Co-4(H2O)(2)(PW9O34)(2)](10-), 1.1 V vs Ag/AgCl, as described in Stracke, J. J.; Finke, R. G. J. Am. Chem. Soc. 2011, 133, 14872) and highlights the importance of reaction conditions in governing the identity of the true, active WOC. Although electrochemical studies are consistent with Co-4-POM being oxidized at the glassy carbon electrode, it is not yet possible to distinguish a Co-4-POM catalyst from a CoOx catalyst formed via decomposition of Co-4-POM. Controls with authentic CoOx indicate conversion of only 3.4% or 8.3% (at pH 8.0 and 5.8) of Co-4-POM into a CoOx catalyst could account for the O-2-generating activity, and HPLC quantification of the Co-4-POM stability shows the postreaction Co-4-POM concentration decreases by 2.7 +/- 7.6% and 9.4 +/- 5.1% at pH 8.0 and 5.8. Additionally, the [Co2+] in a 2.5 mu M Co-4-POM solution increases by 0.55 mu M during 3 min of electrolysis-further evidence of the Co-4-POM instability under oxidizing conditions. Overall, this study demonstrates the challenges of identifying the true WOC when examining micromolar amounts of a partially stable material and when nanomolar heterogeneous metal-oxide will account for the observed O-2-generating activity.