Physical aspects of one- and two-photon photoemission from metals into electrolyte solutions

Abstract

The law of one-photon photoemission from metals into electrolyte solution which connects the photocurrent 4 with the light intensity ( L ), the energy of photons ( ħω ) and the electrode potential ( φ ) has been proved experimentally and can be expressed by the ratio: I ∼ L ω -4 ε 5/2 , ε . = ℏω - ℏω o The statistical treatment of the volt-ampere characteristics of the photocurrent from mercury into KCl, NaF, Na 2 SO 4 solutions with N 2 O as an electron acceptor has shown that they are in better agreement with ref. 4 than with the Fowler law. The frequency factor ω −4 is obtained by measuring the photocurrent at various wavelengths (300–600 nm for mercury) and at constant ɛ set up by the corresponding potential changes. The angular characteristics for the photocurrent from mercury with polarized light (the electric vector in the plane of incidence and normal to the plane, I ‖ and I ⊥ , respectively) are in line with the Fresnel formulas for the light energy absorbed in the volume of the metal which shows that the contribution of the volume photoeffect is essential. The fraction of the surface effect estimated from these measurements is about 20–30% of the total photocurrent. The diffusional dependence of the photocurrent on the concentration of the acceptor and the ψ ′-potential is proof of the electron emission from metal into solution beyond the dense part of the double layer. It is possible to observe a photocurrent proportional to L 2 in the 1–3 MW cm −2 region induced by the first harmonic ruby laser pulses and by the second harmonic Nd laser pulses. The transition from one- to two-photon photoeffect (from n =1 to n =2) induced by the ruby laser occurs on varying the electrode potential. The threshold difference of one- and two-photon photoeffects coincides with the photon energy. The two-photon photoemission probability obtained from these measurements is 3–4 orders in excess of that predicted from the multi-photon photoemission theory. The angular photocurrent dependence for I ‖ and I ⊥ is in agreement with the assumption of a volume character of the two-photon photoeffect. The long-lived metal-electrolyte interface states are supposed to take part in the two-photon photoemission. Unlike the low intensities of light, at L >0.5 MW cm −2 the measured current varies little at low acceptor concentrations for both one- and two-photon photoemission. The latter has been observed by Barker and interpreted as a change of the photocurrent mechanism in passing from the one-photon to the two-photon process. Apparently, such deviation of the diffusional dependence of photocurrent from c a is caused largely by the change in the parameters of the secondary processes of solvation, capture and the subsequent return of the emitted electrons to the electrode in the conditions of high temperature gradients which arise due to pulse heating of the cathode and the adjacent layer of the electrolyte.