Mapping electrostatic potentials across the p‐n junction in GaAs nanowires by off‐axis electron holography

The development of III−V materials on Si platforms, with the aim of reducing production costs while achieving high conversion efficiency, has been a continuing area of photovoltaic research in the last decades [1,2]. This process is challenging due to large lattice mismatches, the polar non‐polar interfaces and the differences in thermal expansion coefficients. The use of III–V nanowires (NWs) provides a novel method of integrating III‐V materials with Si, which avoids dislocations [3]. However the control of other parameters, such as vertical yield in a patterned array, crystal phase, dopant concentrations and electrostatic potential distribution, become challenging. The electrical performance of a semiconductor device relies strongly on how precisely the electrostatic potentials are distributed across the active region. An accurate measurement of this potential distribution is of vital interest to the semiconductor industry. The technique of off‐axis electron holography in the transmission electron microscope (TEM) is a powerful tool for fulfilling the required accuracy in mapping electrostatic potentials [4]. Here, we present electron holography measurements from single GaAs core‐shell nanowires with a p‐n junction, grown on a Si (1 1 1) substrate. The Ga‐assisted vapor–liquid–solid (VLS) growth mechanism on a silicon substrate was used for the formation of a patterned array of radial p‐i‐n GaAs NWs encapsulated in AlInP passivation. A cross‐sectional specimen for off‐axis electron holography was prepared perpendicular to the growth direction of the NW using focused ion beam milling (FIB) and the in‐situ lift‐out technique in an FEI Helios Dualbeam FIB/SEM, equipped with a micromanipulator. Holograms were acquired at 120 kV using an FEI Titan 80‐300ST TEM, equipped with a rotatable Möllenstedt biprism. The thickness of the specimen was measured to be around 280 nm by convergent beam electron diffraction (CBED). Fig. 1 shows the reconstructed phase and amplitude from the hologram of the cross‐sectional specimen. A core‐shell structure is observed, with the core being p‐type and the shell being n‐type. The phase shift across the p‐n junction is close to 1 radian, corresponding to a built‐in potential of 0.4 V, as shown in Fig.2. The potential variation measured by holography is used to quantify the actual doping densities in the n‐type layer and p‐type layer of the NW. This holography measurement indicates that the active dopant concentrations are lower than nominal values, causing a low built‐in potential. A greater control on the dopant concentration and distribution is required in order to achieve a higher efficiency of the NW solar cells.