R. Mucciato et N. Lovergine, Detailed thermal boundary conditions in the 3D fluid-dynamic modelling of horizontal MOVPE reactors, J CRYST GR, 221, 2000, pp. 758-764
A three-dimensional (3D) detailed heat transfer model describing the effect
ive thermal behaviour and fluid dynamics of a horizontal metalorganic vapou
r-phase epitaxy (MOVPE) reactor chamber is reported. Both H-2 conduction an
d convection were considered, along with the presence of a cooling H-2 flow
outside the main MOVPE chamber, the quartz heat conduction through the cha
mber walls and the convective heat transfer to the ambient air. Computation
al fluid-dynamic (CFD) simulations indicated that relatively large gas temp
erature gradients occur normal to the wafer surface both in the process and
cooling gas. In the former a fully developed thermal profile is observed.
Reduced temperature gradients occur instead in the lateral directions withi
n the process gas flow, giving rise to a good lateral uniformity of the tem
perature field close to the susceptor. However, large gradients build up in
the cooling gas as a result of heat transfer to the surrounding ambient by
the free convection. Correspondingly. large buoyancy appears in the coolin
g flow. A laminar behaviour is instead obtained for the process flow, altho
ugh the substrate rotation introduces an asymmetry of the gas pathlines abo
ve the wafer. Comparing simulated susceptor-ceiling temperature differences
(AT) with experimental ones indicates that a free convection parameter alp
ha approximate to 7.5 W/m(2) K occurs for our system. This is consistent wi
th Ra similar to 10(5) for the ambient outside the MOVPE chamber, suggestin
g subcritical-free convection conditions. CFD simulations performed as a fu
nction of susceptor temperature show close agreement between simulated and
experimental DeltaT, indicating the usefulness of a 3D detailed heat transf
er approach for the correct simulation of temperature fields in a horizonta
l MOVPE reactor. (C) 2000 Elsevier Science B.V. All rights reserved.