Bulk photoexcitation

Upon illumination, photons having energy higher than the band gap (eg = ec — v) are absorbed in the semiconductor phase and the electron-hole-pairs (e //i+) are generated. This effect can be considered equivalent to the photoexcitation of a molecule (Fig. 5.57) if we formally identify the HOMO with the ec level and LUMO with the v level. The lifetime of excited e //i+ pairs (in the bulk semiconductor) is defined analogously as the lifetime of the excited molecule in terms of a pseudo-first-order relaxation (Eq. 5.10.2). 

The primary process following a photoexcitation of nltrosamldes XIV Is the dissociation of the N-N bond to form a radical pair XV and the ensuing chemical events are the reactions of amldyl and nitric oxide radicals In the paired state or Individually In the bulk of solutions. Naturally, secondary reactions, thermal or photolytic, have to be taken Into consideration under Irradiation conditions (21). First of all, the relatively straightforward chemistry of selective excitation In the n-ir transition band (>400 nm) will be discussed, followed by the chemistry of Irradiation with a Pyrex filter (>280 nm). As nitric oxide Is known to be rather unreactlve (23), primary chemical processes In the Irradiation with >400 nm light under... 

Photoexcitation of the RbAgi Is can be expected to result in the formation of electron-hole pairs. The photoelectrons will diffuse from the surface into the bulk because of their much greater drift mobility compared to the photoholes. A net positive space charge... 

Steady state photoelectrochemical behaviour of colloidal CdS For the purposes of the studies reported here, the photocurrent was taken to be the total current recorded at the ORDE from an illuminated colloidal dispersion of CdS minus the current recorded under identical condition from the same dispersion in the dark. In both studies, the photocurrents generated by CdS particles illuminated at the ORDE exhibited a wavelength dependence (action spectrum) identical to the absorption spectrum of colloidal and bulk CdS [166,168], unambiguously indicating that the observed photocurrent is due entirely to ultra-band gap photoexcited conduction band electrons. However, it should be noted that, unless stated otherwise (e.g. the action spectrum experiments), the particle suspensions of both studies were usually irradiated with white light from a 250 W quartz iodine projector lamp to maximise the photocurrents observed. 

As Powell emphasizes, differences in experimental details and, in many cases, the lack of precise information about these details and about the specimen surfaces make the correlation of much of the existing escape data impracticable. What is important here is that at electron energies of 1253 and 1486 eV, corresponding to the Mg and A1 Kx lines commonly employed in XPS, the escape depths are of the order of 20—40 A. Photoexcitation of deeply bound core levels of course corresponds to smaller E0 values and hence to smaller A s. As a rule an XPS experiment typically probes the outermost 15—20 A of the sample. On one hand, the bulk is probed on the other, XPS is a surface sensitive technique. 

The current theories of chemically induced magnetic polarization can therefore be summarized into the two basically different mechanisms the photoexcited triplet mechanism (PTM) responsible for the initial electron polarization and the observed Overhauser effect in nuclear polarization, and the radical-pair mechanism which, to date, accounts for almost the remaining bulk of the known polarization systems. We proceed to describe the simple physical models of these two mechanisms by beginning with the more sophisticated radical-pair theory. 

As already discussed in Sect. 2.2., the bandgap of semiconductor particles increases considerably when their size becomes smaller than about 100 A (Figs. 4 and 5). Accordingly, the position of energy bands is shifted, and it is expected that certain reactions should become possible with quantized particles which do not occur with bulk materials. This has been demonstrated for H2-evolution in 50 A PbSe- and HgSe-colloids, which has not been observed with large particles [181, 182]. An extreme negative shift of the conduction band by about 1.2 eV has been found with 50 A-CdTe-colloids due to their low effective mass. Since COa-reduction to formic add was observed with photoexcited CdTe-colloids, the conduction band must be at < - 1.9 eV, compared to the flatband potential of n-CdTe electrodes of — 0.6 V [181]. 

Layered compounds provide unique character for electron-transfer processes owing to their low dimensionality. Especially layered materials with ion-exchange and/or intercalation capabilities show behavior that is not seen in so-called bulk-type materials. Layered materials, which have been often used in studies of photoelectron transfer as well as photocatalysis, may be classified into two groups compounds in which the host layers work as an active component for the photoexcitation and electron-transfer reactions, and materials in which the layers are inert for electron-transfer processes. Examples of the former are layered titanates and niobates and of the latter are clays. In the latter case, photoactive materials are intercalated in the interlayer spaces. Recently, the exfoliation of various layered compounds has become possible and artificial assemblies consisting of these exfoliated sheets have been formed. Electron transfer in such assemblies is also an attractive subject in this field. 

Two celebrated early investigations of transmembrane oxidation-reduction were interpreted in terms of direct electron exchange between redox partners bound at the opposite vesicle interfaces. One involved apparent reduction of diheptylviologen [( 7)2 V +] in the inner aqueous phase of phosphatidylcholine liposomes by EDTA in the bulk phase that was mediated by membrane-bound amphiphilic Ru(bpy)3 + analogs the ruthenium complexes acted as photosensitizers and were presumed to function as electron relays by undergoing Ru(II)-Ru(III) electron exchange across the bilayer [105]. The other apparently involved direct electron transfer between photoexcited Ru(bpy)3 + and bound at the opposite interfaces of asym-... 

The most important point in this study is that the dipole-bound excited states of the 1 (CH3CN)2 cluster cannot be considered as a result of further solvation of the [I-CH3CN]- ion core. Instead, the nature of diffuse excited states is determined by the entire arrangement of solvent molecules around the solute anion, as mentioned above. This situation is more akin to CTTS in bulk solutions than solvated ion cores, which are common to most ionic clusters and electrolytic solutions without photoexcitation. 

The elemental Pt(0) Is dispersed throughout the surface polymer as determined by depth profile analysis,(7) and a representation of the interface is given in Scheme V. According to this view there is a certain amount of Pt(0) in contact with the thin SiOx overlayer on the bulk p-type Si. This is a relevant structural feature, since direct deposition of Pt(0) onto photocathode surfaces is known to improve the efficiency for the reduction of H2O to H2> Thus, we expect that, for an interface like that depicted in Scheme V, there will be a certain amount of the H2 evolution occurring by direct catalysis of the reaction of the photoexcited electrons with H2O at the Si0x/Pt(0) interfaces. In the extreme of a uniform, pinhole-free coverage of Pt(0) on p-type Si/SiOx one expects that the photocathode would operate as a buried photosensitive interface and in fact would be equivalent to an external solid state photovoltaic device driving a photoelectrolysis cell with a Pt(0) cathode. 

---