Abstract

Charge transfer (CT) between an adsorbed atom or molecule and its substrate is of direct importance for the understanding of photochemical surface processes and more generally of the adsorbate-substrate coupling. A direct measurement of its timescales is difficult as it is extremely fast (from less than a fs to some or some tens of fs, as deduced from indirect evidence). New methods have become possible with the availability of synchrotron light at very high resolution and intensity, which utilize core hole excitations to determine timescales for fast electron transfer processes. These techniques are based on the use of the core hole lifetime as an internal time standard. This approach can be applied in the time domain as well as in the frequency domain. This survey describes mainly the former approach which consists of the study of the decay spectra after resonant excitation of an adsorbate core hole, induced with excitation band widths below the core hole lifetime width (so called Auger resonant Raman conditions). Tuning through the resonance allows one to separate those parts in the decay spectra which correspond to decay before, vs. after, transfer of the locally excited electron from the adsorbate into the substrate. The ratio of intensities of these two types of spectra is connected to the ratio of the timescales of CT and of core hole decay via simple rate equations. Since the core hole lifetime is known, the CT time can be calculated from this spectral branching. For adsorbate resonances between the Fermi and the vacuum level of the substrate, values from below 1 fs (strong chemisorption), via the range of some fs to some 10 fs (physisorption with graded coupling), to above 50 fs (decoupled condensate) have been found; their dependence on the actual excitation energy can also be determined. The CT times of electrons screening a core hole on a chemisorbate—a process which happens at the Fermi level—on the other hand are even shorter, namely fast even on the timescale of sudden photoionization. This is demonstrated by the possibility to measure vibrationally resolved XPS peaks even for chemisorbed species (indicating the dominance of a final state of core ionization which is made neutral by CT). The observed total and partial suppression of the PCI (post collision interaction) effect for chemisorbates and physisorbates, respectively, is consistent with such timescales of screening.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call