Influence of crystal polarity on crystal defects in GaP grown on exact Si (001)

Abstract

For the implementation of optoelectronic devices on silicon, which could be realized by a combination of Si and direct-bandgap III/V semiconductors, a defect free nucleation layer of GaP on Si is essential. This paper summarizes the results of structural investigations carried out by transmission electron microscopy on defects, which can be observed in GaP films grown by metal organic vapor phase epitaxy on exactly oriented (001) Si substrates. Under optimized growth conditions the anti phase domains (APDs), which arise in the III/V semiconductor at the monoatomic steps on the silicon surface, show a specific typical shape. They self-annihilate on {112} planes in the GaP and can be observed in [110] cross-section, looking perpendicular to the steps on the Si surface. In contrast to that, the anti phase boundaries (APBs) lie on {110} GaP planes in the [−110] direction, parallel to the steps on the Si surface. From convergent beam electron diffraction one can show, that the GaP has Ga-polarity in the [110] direction, viewing perpendicular to the steps on the Si-surface. With the knowledge of the polarity and the shape of the APDs, we suggest a model for chemical composition of their boundaries. According to this model the APBs, which lie on {110} and {112} planes, consist of an equal amount of Ga-Ga and P-P bonds. Furthermore, when stacking faults and twins are observed, they only occur in the Ga-polar [110] GaP direction, and consequently lie on {111}A planes. With the knowledge of the structure of the defects that arise at the GaP/Si interface we suggest growth conditions and an optimum Si surface structure, which guarantee a defect-free GaP overgrowth layer after several 10 nms of III/V material, even on exact Si substrates.

Study of planar defect filtering in InP grown on Si by epitaxial lateral overgrowth

“…9 intersect the ELOG layer surface in the [110] direction, the fault must lie either on a (111) plane or a (1-11) plane; whichever it may be, it does not alter the model due to symmetry. investigation, and a similar SF polarity has been observed in other heteroepitaxial systems such as GaP on Si [27].…”

Effect of bulk growth temperature on antiphase domain boundary annihilation rate in MOCVD-grown GaAs on Si(001)

“…In addition to showing the preferential annihilation of APBs along {112} planes [30], Beyer et al, observed that in similarly grown GaP-on-Si samples APBs oriented along {110} planes displayed a tendency to jump one or more atomic planes, thereby creating a finite thickness to the APB [31]. Thus, increasing growth temperature can increase the jump probability for APB planes and enhance the annihilation rate.…”

GaP-nucleation on exact Si (001) substrates for III/V device integration

“…It will be reported elsewhere [38] that the APBs which kink on {1 1 1} or {1 1 2} planes can consist of P-P false bonds. Therefore, the V/III ratio used during overgrowth might play a role in the kinking of the APBs.…”

“…Therefore, specific care should be taken when judging defects from cross-sectional TEM investigations, as one need to choose the proper /1 1 0S direction for preparation to image these defects. A detailed investigation of the polarity of the GaP layers deposited on Si and its dependence on growth conditions, as well as of the defects which are formed in these layers and their dependence on the crystal polarity, will be published elsewhere [38].…”