Chris R. Kleijn

Return flows in horizontal MOCVD reactors studied with the use of TiO2 particle injection and numerical calculations

Flow visualization experiments with the use of TiO2 particles were performed for a horizontal MOCVD reactor which was resistance-heated from below and water-cooled on top. It is shown that, due to the heating of the incoming gas, very pronounced return flows can occur at the leading edge of the heated susceptor. The formation of these return flows was systematically studied as a function of pressure, flow velocity and temperature. We show that for typical flow rates, carrier gases and reactor design used in MOCVD growth of GaAs and AlGaAs, it is possible to avoid return flows by decreasing the pressure to about 0.2×105 Pa. Besides these experimental studies also numerical calculations on gas flows based on the differential equations for the conservation of mass, energy and momentum were performed, using a two-dimensional finite difference procedure. Parameters in this study were again pressure, flow velocity and temperature, and in addition also the height of the reactor and the type of carrier gas were varied. In comparing the results obtained by the flow simulation experiments and the numerical calculations, it is found that the occurrence of return flows is predicted accurately by the numerical flow calculations. Small differences can be ascribed to three dimensional effects which are not included in the 2D numerical calculations. Both flow visualization experiments and numerical calculations show that the occurrence of return flows is mainly determined by only two dimensionless hydrodynamic numbers: the Grashoff number Gr and the Reynolds number Re. It is possible to define a number αcrit such that no return flow occurs for GrReκ<αcrit. Here κ equals 1 for small Reynolds numbers (Re ≤ 4) and goes to 2 for larger Reynolds numbers (Re ≥ 8), being independent of all other hydrodynamic parameters. The critical value αcrit was found to depend slightly on the temperature difference and is independent of all other parameters. The criterion is confirmed by both the TiO2 experiments and the numerical calculations for 0.1 < Re < 10 and 10 < Gr < 10 000.

Computational modeling of transport phenomena and detailed chemistry in chemical vapor deposition - a benchmark solution

Two-dimensional simulations are presented of the full multi-component transport phenomena and the multi-species, multi-reaction chemistry in thermal chemical vapor deposition (CVD) of silicon in a rotating disk/stagnation flow reactor. The detailed description of the problem definition and model equations allows for easy reproduction of the presented simulations, which were validated against the well-known one-dimensional SPIN code from SANDIA by Coltrin and co-workers. Thus, the simulation results can be used as a benchmark against which (commercial) multi-dimensional CVD equipment simulation codes can be validated.

A STUDY OF 2- AND 3-D TRANSPORT PHENOMENA IN HORIZONTAL CHEMICAL VAPOR DEPOSITION REACTORS

2- and 3-D mathematical models for epitaxial growth in horizontal chemical vapor deposition reactors have been developed and modeling results have been compared with experimental results. The models include the partial differential equations describing the balance of mass, momentum, heat and species concentration, including multicomponent diffusion and thermodiffusion. The deposition process is assumed to be in the transport-limited regime. A detailed model for calculating the side wall temperature including thermal radiation, is included in the 3-D model. The equations are solved numerically, using a control volume based finite-difference method. The model is applied to MOCVD of GaAs from TMGa and AsH3. By varying the reactor height, inflow velocity and carrier gas, a large range of Rayleigh (Ra) and Reynolds numbers is covered. The calculations for the deposition rates in the midplane of the reactor are compared with experimental results from the literature. The 2-D model is found to give accurate results for large and medium width-to-height aspect ratio reactors operated at subcritical Ra only. For supercritical Ra conditions, 3-D calculations with detailed modeling of side wall temperatures give accurate predictions of experimental results. Modeling results are very sensitive to the use of correct thermal boundary conditions on the side walls. Deposition on the side walls is found to have little influence on the axial growth rate distribution. The relative importance of convection, diffusion and thermodiffusion for the species transport is studied. The exact Stefan—Maxwell equations for multicomponent diffusion are compared with an approximate approach, which was found to be very accurate.

On multiple stability of mixed-convection flows in a chemical vapor deposition reactor

Abstract The laminar mixed convection flow and heat transfer in a Chemical Vapor Deposition reactor are studied through numerical simulation. It is found that the non-linear interaction between forced and free convection may lead to the existence of multiple stable flows. Arclength continuation techniques, implemented within the framework of the finite volume discretisation, have been applied to determine the causes for this multiplicity. It is shown that the relevant dimensionless groups are Gr/Re and Re×Pr. When both these groups are sufficiently large, multiple stable flows may exist.

A surface and a gas-phase mechanism for the description of growth on the diamond(100) surface in an oxy-acetylene torch reactor

A gas-phase and a surface mechanism were developed, suitable for multidimensional simulations of diamond oxy-acetylene torch reactors. The gas-phase mechanism was obtained by reducing a 48 species combustion chemistry mechanism to a 27 species mechanism with the aid of sensitivity analysis. The surface mechanism for growth on monocrystalline (100) surfaces developed, was based on literature quantum-mechanical calculations by Skokov et al. It consists of 67 elementary reaction steps and 41 species, and contains CH3 and C2H2 as gas-phase growth precursors and atomic hydrogen and oxygen to etch carbon from the surface. The gas-phase and surface chemistry models were tested in one-dimensional simulations, yielding dependencies of the growth rate on feed composition and surface temperature that are in qualitative agreement with the experiments. A more detailed study of the surface chemistry showed that, compared to CH3, acetylene contributes very little to diamond growth. Furthermore, molecular and atomic oxyg...

Subcritical flow past a circular cylinder surrounded by a porous layer

A study of the flow at subcritical Re = 3900 around a circular cylinder, surrounded at some fixed small distance by a porous layer with a hydraulic resistance typical for that of textile materials, has been performed by means of direct numerical simulations. The flow in the space between the porous layer and the solid cylinder was found to be laminar and periodic, with a frequency locked to that of the vortex shedding in the wake behind the cylinder. Time averaged flow velocities underneath the porous material were in good agreement with experimental data from laser Doppler anemometry.

On turbulent flows in cold-wall CVD reactors

With the increase in diameter of the wafers and the tendency to increase operating pressure of chemical vapor deposition reactors, the flow in these reactors may turn turbulent as a result of buoyancy. The effect of turbulence on the CVD process in a single wafer reactor has been studied numerically with large eddy simulations. It is found that the free-convection induced turbulence increases the heat flux, whereas the conditions are uniform in a large part of the reactor as a result of turbulent mixing, in principle offering the possibilities for a high, uniform deposition rate.

Mixed convection in radial flow between horizontal plates — I. Numerical simulations

Abstract A forced radially outward flow with secondary, buoyancy induced convection has been studied numerically in an axisymmetric geometry, consisting of two differentially heated, horizontal, coaxial, circular plates with a diameter of 25 times their mutual spacing. A forced laminar flow is supplied through the centre of the upper plate. The onset of thermal instability, leading to axisymmetric and three-dimensional rolls, has been determined as a function of the Reynolds, Prandtl and Rayleigh numbers.

Effects of wavy surface roughness on the performance of micronozzles

Recent trends in small-scale (~1 dm 3 ) satellites motivate the further development of microscale propulsion subsystems. In the present paper, we focus on flow dynamics simulations of conical convergent-divergent micronozzles and on the increased importance of wall effects due to the decrease in the characteristic length of such small systems. The inefficiency associated with viscous losses due to the developing boundary layer and the effect of sinusoidal surface roughness due to the employed microelectromechanical-system fabrication techniques are studied through computational fluid dynamics simulations for nonturbulent, nonrarefied flow conditions. Depending on the specific nature of the surface roughness, the formation and reflection of several weak shocks and, as a consequence, a decreased performance are observed.

A stereo vision method for tracking particle flow on the weld pool surface

The oscillation of a weld pool surface makes the fluid flow motion quite complex. Two-dimensional results cannot reflect enough information to quantitatively describe the fluid flow in the weld pool; however, there are few direct three-dimensional results available. In this paper, we describe a three-dimensional reconstruction method to measure weld pool surface features based on a single high-speed camera. A stereo adapter was added in front of the high-speed camera lens to obtain two images in the same frame from different view points at the same time. According to machine vision theory, three-dimensional parameters can be reconstructed based on two such images. In this work, three-dimensional velocity fields have been obtained using this method. Based on the calibration technique employed, the associated error is estimated to be less than 11.4%. Quantitative experimental results are useful for understanding the flow pattern, and possibly for controlling the flow of liquid in the weld pool.