Surface morphology of indium phosphide grown on silicon by nano‐epitaxial lateral overgrowth

Abstract

InP is grown on Si by nano‐epitaxial lateral overgrowth (NELOG or nano‐ELOG) on patterns consisting of net‐type openings under different growth conditions. Surface morphology is characterized with AFM and profilometer and optical quality assessed by Micro Photoluminescence measurements (μ‐PL). Results show that growth conditions affect both morphology and optical quality, with thicker layers generally corresponding to better surface morphology. Lower growth temperature seems to improve surface morphology irrespective of thickness, and ELOG layers exhibit significantly better morphology than the planar layer. (© 2009 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)

Morphological evolution during epitaxial lateral overgrowth of indium phosphide on silicon

“…We will elaborate on these phenomena further below by considering the case of sample 1-CD after treating the surface roughness. From Table 1, it is observed that the surface roughness of the coalesced layer decreases with increasing layer thickness in accordance with earlier observations of ELOG on mesh openings [16]. This study shows certain difference in surface roughness with respect to pattern type and opening angles in the same sample, also shown in Table 1.…”

Heteroepitaxial growth of Indium phosphide from nano-openings made by masking on a Si(001) wafer

“…The results show that the net type openings result in a better surface morphology of the grown layer. The surface morphology of an ELOG layer depends to some extent on how thick the sample is, confirming results from previous studies (6). The AFM characterization was done for all the field sizes on sample A and there is little difference on the surface roughness of the respective grown layer as the field size varies.…”

High-Temperature Annealing Effect of Si in Group-V Ambient Prior to Heteroepitaxy of InAs in Metal–Organic Vapor Phase Epitaxy

“…For this integration of III-V semiconductors on Si, heteroepitaxy through selective-area growth (SAG), especially epitaxial lateral overgrowth (ELO), [6][7][8][9][10][11][12][13][14] is a desirable approach because 1) it is a self-aligned process, therefore, the growth position can be decided in advance, 2) it is easy to obtain thin films by controlling the growth rate and/or growth time, and 3) dislocations and defects, generated by the difference in lattice constants and thermal expansion coefficients between a III-V layer and Si, can be avoided by limiting the growth area. Particularly in the case of III-V semiconductors on Si, domain boundaries in the III-V layer easily emerge owing to the coalescence of several threedimensional nuclei generated at the initial stage of growth because ideal layer-by-layer growth is rarely achieved.…”