Answers are grounded in this paper and cite their sources.
The experimental data we shall try to explain are shown in the Figs. 1 and 2. In Fig. 1 we see the current voltage for a p−n junction heavily doped with Sb donors. There is seen to be no current until a voltage is reached such that eV = ħω, where w is the frequency of a particular phonon needed to conserve momentum. For forward voltage four such onsets, corresponding to four different phonons, can be seen; for back voltage the current due to the first phonon becomes so large that the onset due to the other phonons is not seen. In Fig. 2 we see dI/dV versus V, which gives us peaks for those voltages at which a new phonon process becomes possible. Note that these peaks are completely smeared out for those junctions with donors other than Sb. This is because when Sb is substituted for a Ge atom in the crystal the effect is just to add a single positive charge on the Ge site, whereas the other impurities give a drastic change in the central cell potential as well. Why Sb does not have these central cell corrections is not known, but it is an experimental fact that of all the impurities only Sb has bound electronic energy levels obtained by assuming the impurity potential is equal to the potential Ge plus one proton. The central cell potentials, being localized in r space, contain high-k components which can scatter the tunneling electrons and supply the momentum which otherwise would have to be supplied by the phonons.
No takes yet. Share an insight, caveat, or question.
Kleinman et al. (1964) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: