Except where otherwise noted, all materials were used as received, including potassium hydroxide pellets (KOH, Macron Chemicals, ACS 88%), boric acid (H 3 BO 3 , Sigma-Aldrich, BioReagent ≥99.5%), potassium ferrocyanide trihydrate (K 4 Fe(CN) 6 • 3H 2 O, Acros, >99%), potassium ferricyanide (K 3 Fe(CN) 6 , Fisher Chemicals, certified ACS 99.4%), potassium chloride (KCl, Macron Chemicals, Granular ACS 99.6%), Br 2 (Sigma Aldrich, 99.999%), and CH 3 OH (EMD Millipore, >99.9%). Water with a resistivity of 18.2 MΩ•cm was obtained from a Millipore deionized water system. Single-side polished, Sn-doped (N d = 4.22-5.83×10 17 cm -3 ), (100)oriented, n-type InP wafers were obtained from AXT Inc. Nominally undoped (N d = 10 15 cm -3 ), (100)-oriented, n-type single-side-polished InP wafers were obtained from MTI Inc. Zn-doped (N a = 3×10 18 cm -3 ), (100)-oriented, p-type single-side-polished InP wafers were obtained from MTI Inc.
Growth of the p + n-InP junction:All samples were grown on InP (001) epi-ready substrates in a Varian Gen-II Molecular-Beam Epitaxy (MBE) system that had been modified to thermally crack gas-phase PH 3 to P 2 and H 2 .Elemental In and Be were used as group-III and p-type sources, respectively. At ~410 °C, the native oxide on the substrate surface was fully desorbed under the P 2 overpressure, as confirmed by streaky 2×4 surface reconstruction patterns observed using reflection high-energy electron diffraction (RHEED). The highly doped p + -InP layer growth was triggered when the In shutter was opened and the InP film growth rate was set at 1 μm h -1 (i.e., 1 monolayer s -1 ), as calibrated by the RHEED oscillations. The PH 3 flow rate was set at 6 sccm throughout the growth. The doping for these two samples was ~5×10 18 cm -3 , as calibrated by Hall measurements on InP test film samples grown on semi-insulating InP substrates. A p + -layer with a thickness of 100 nm was