No abstract

No abstract

results in improved kinetics for H2 evolution from the p-type semiconductor. Kinetics for H2 evolution was previously identified as a major problem at p-semiconductor aqueous electrolyte interfaces (1-3). Surface modification of electrodes has recently been an active area of research, and examples of electrocatalysts are emerging (4). To produce an efficient photosensitive interface from which to evolve H2 we take advantage of the light absorption and charge separating properties of a semiconductor (5-8). But additionally we employ surface modification to achieve good H2 evolution kinetics. A p-type semiconductor should serve as a photocathode, because the minority carrier is the electron, e-. As indicated in Fig. 1, the photoexcited electron becomes available at the interface as a reducing equivalent with reducing power no greater than the bottom of the conduction band at the interface ECB. Efficient solar-driven electrolysis requires: (i) use of a photocathode having a small Eg, 1-2 eV; (ii) optimization of the product of Ev and photocurrent; and (iii) a durable interface. Experiments show that Ev can be significant for small Eg photocathodes, but the kinetics for H2 evolution are so poor that little if any efficiency can be realized for p-type Si (Eg =

Smooth Pt and optically transparent SnO2 electrodes can be functionalized with redox active polymeric material derived from the hydrolytically unstable {N,N′‐bis[‐3‐(trimethoxysilyl)propyl]‐4,4′‐bipyridinium}dibromide, I. The redox polymer on the surface false[false(PQ2+)n]normalsurf. is reversibly reducible to false[false(PQ+)n]normalsurf. in either CH3CN/normalor H2O/normalelectrolyte solutions , and in CH3CN the reduction of false[false(PQ+)n]normalsurf. to false[false(PQ0)nfalse] is reversible. The E0′false[false(PQ2+/+)n]normalsurf.=−0.55±0.05V vs. SCE, and −0.45±0.05V vs. SCE in H2O and CH3CN , respectively. The E0′false[false(PQ2+/0)n]normalsurf.=−0.85±0.05V vs. SCE in CH3CN . The optical properties of false[false(PQ2+)n]normalsurf. (colorless), false[false(PQ+)n]normalsurf. . (purple, λmax≈545 normalnm , ∈545 normalnm≈107 cm2mol−1 ), and false[false(PQ0)n]normalsurf. (yellow, λmax≈375 normalnm , ∈375 normalnm ≈4×107 cm2mol−1 ) depend somewhat on solvent/electrolyte. The optical spectral features associated with the three redox states of the surface polymer are easily distinguishable with the naked eye for coverages corresponding to >10−8mol/cm2 of redox active centers. Electrodes bearing false[false(PQ2+/+)n]normalsurf. are very rugged and do not deteriorate upon repeated cycling between false[false(PQ2+)n]normalsurf. and false[false(PQ+)n]normalsurf. . Potential step/chronoamperometry experiments establish that the current for oxidation/reduction of the polymer is nearly proportional to t−1/2 , indicating that the rate of oxidation or reduction is controlled by a diffusion process with a diffusion constant, D , of 10−9 to 10−10 cm2/sec depending on electrolyte and its concentration. In practical terms a potential step from 0.0 to −0.80V vs. SCE results in ∼50% reduction of false[false(PQ2+)n]normalsurf. to false[false(PQ+)n]normalsurf. in a time as short as 5×10−3sec at a coverage of 1×10−8 normalmol/cm2 .

No abstract