Influence of growth conditions on tin incorporation in GaAs grown by molecular beam epitaxy

Abstract

Intentional perturbations applied to the growth parameters of Sn-doped GaAs layers grown by molecular beam epitaxy have been performed in order to investigate the tin incorporation mechanism. The start, the interruption, and the end of growth as well as a variation of fluxes or substrate temperature have been studied, using either the Auger electron spectroscopy (AES) measurement of tin accumulation on the surface, or C-V derived free-carrier concentration profile versus any of these growth parameters. The theoretical model proposed by Wood and Joyce, based on a time-delayed incorporation mechanism, has been found to fit the observed results, especially for As-rich surface, provided that an incorporation mechanism of second order is assumed. For Ga-rich conditions (Ts≳580 °C), a new result has been recognized, i.e., a significant reduction of carrier concentration as Ts is increased. The assumption of a partially acceptor nature of tin incident atoms under these growth conditions does not seem to fully explain this result. On the other hand, this may be better understood assuming a certain amount of tin atoms being re-evaporated in the high substrate temperature range. This behavior induces a temporarily weaker accumulation at the surface, and hence a relatively smaller incorporation rate.

Sn incorporation into InP grown by molecular beam epitaxy: A secondary-ion mass spectrometry study

“…However, the fact that such behavior is observed is not a guarantee that the incorporation is uniform in the grown layer, as is attested to by the fact that a linear dependence is also seen for Sn incorporation into GaAs, 5 where surface accunmlation during MBE is known to occur. 6,7 For the studies reported here, hcterostructures consisting oflayers of both InP and Gao.47Ino.53As lattice matched to InP were grown by HSMBE on (100) InP wafers. In the HSMBE version of gas source MBE (GSMBE) 8 the group V elemental molecular beams (As 2 and P 2) are derived from thermally decomposed AsH 3 and PH 3 while all other beams, induding the SIl beam, result from heating the elements in effusion cells.…”

The influence of aluminum composition of AlxGa1–xAs in distributed Bragg reflector on surface morphology

“…They obtained smooth surface layers for binary compounds GaAs and AlAs, and the roughness was found to be maximal for the ternary compound Ga 0.5 Al 0.5 As. It was believed that the consequence would be a reduced ability of Al atoms to find the right incorporation sites before subsequent Al atoms arrive, favoring local inhomogeneity of the Al content of the layer [12]. Also, it could be thought that roughening would intensify as the number of foreign atom in the host lattice increases.…”

Molecular beam epitaxy

“…It was originally suggested (Cho 1975, Ploog andFischer 1978) that the tin was segregating to the surface of the growing film, but subsequently Wood and Joyce (1978) proposed a surface rate limiting process in which the Sn is incorporated from a surface layer, not directly from the vapour phase. This model has been extended by Alexandre et al (1980) and Wood et a1 (1980) and m is the order of the incorporation process. Because the surface Sn has a tendency to form three-dimensional islands an apparent zeroth-order behaviour with respect to Jsn can also be observed (Harris et a1 1982).…”