We have used cross-sectional scanning tunneling microscopy to study the atomic-scale interface structure of InAs/Ga1−xInxSb superlattices grown by molecular beam epitaxy. Detailed, quantitative analysis of interface profiles obtained from constant-current images of both (110) and (11̄0) cross-sectional planes of the superlattice indicate that interfaces in the (11̄0) plane exhibit a higher degree of interface roughness than those in the (110) plane, and that the Ga1−xInxSb-on-InAs interfaces are rougher than the InAs-on-Ga1−xInxSb interfaces. The roughness data are consistent with anisotropy in interface structure arising from anisotropic island formation during growth, and in addition a growth-sequence-dependent interface asymmetry resulting from differences in interfacial bond structure between the superlattice layers.
We have used cross-sectional scanning tunneling microscopy ͑STM͒ to perform nanometer-scale characterization of compositional structure and interfacial properties within GaAs 1Ϫx Sb x /GaAs double-quantum well structures. An algorithm has been devised based on analysis of strain effects in STM data to obtain detailed, quantitative compositional profiles within alloy layers. Using this and other analysis techniques, we have assessed the influence of group V anion soaks at each heterojunction interface on interface roughness and abruptness. An As soak at the GaAs-on-GaAs 1Ϫx Sb x interfaces reduces interface roughness but leads to a slight loss of abruptness at the interface, while an AsϩSb soak at GaAs 1Ϫx Sb x-on-GaAs interfaces improves abruptness while leaving interface roughness largely unaffected. Significant compositional grading at the nanometer scale is observed within the GaAs 1Ϫx Sb x layers.
We have performed detailed characterization of atomic-scale interface structure in InAs/Ga 1Ϫx In x Sb superlattices using cross-sectional scanning tunneling microscopy ͑STM͒ and established a semiquantitative correlation between interface structure and transport properties in these structures. Quantitative analysis of STM images of both ͑110͒ and (110) cross-sectional planes of the superlattice indicates that interfaces in the (110) plane exhibit a higher degree of interface roughness than those in the ͑110͒ plane and that the Ga 1Ϫx In x Sb-on-InAs interfaces are rougher than the InAs-on-Ga 1Ϫx In x Sb interfaces. The roughness data are consistent with anisotropy in interface structure arising from anisotropic island formation during growth and, in addition, a growth-sequence-dependent interface structure arising from differences in interfacial bond structure between the two interfaces. Low-temperature Hall measurements performed on these samples demonstrate the existence of a substantial lateral anisotropy in mobility that is in semiquantitative agreement with modeling of interface roughness scattering that incorporates our quantitative measurements of interface roughness using STM.
Nanometer-scale compositional structure in InAs x P 1Ϫx /InN y As x P 1ϪxϪy //InP heterostructures grown by gas-source molecular beam epitaxy and in InAs 1Ϫx P x /InAs 1Ϫy Sb y /InAs heterostructures grown by metalorganic chemical vapor deposition has been characterized using cross-sectional scanning tunneling microscopy. InAs x P 1Ϫx alloy layers are found to contain As-rich and P-rich clusters with boundaries formed preferentially within ͑111͒ and ͑111͒ crystal planes. Similar compositional clustering is observed within InN y As x P 1ϪxϪy alloy layers. Imaging of InAs 1Ϫx P x /InAs 1Ϫy Sb y superlattices reveals nanometer-scale clustering within both the InAs 1Ϫx P x and InAs 1Ϫy Sb y alloy layers, with preferential alignment of compositional features in the ͓112͔ direction. Instances are observed of compositional features correlated across a heterojunction interface, with regions whose composition corresponds to a smaller unstrained lattice constant relative to the surrounding alloy material appearing to propagate across the interface.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.