This paper presents an integrated physiconeural network approach for the modeling and optimization of a vertical MOCVD reactor. The basic concept is to utilize the solutions obtained from a physical model to build an accurate neural network (NN) model The resulting model has the attractive features of self-adaptiveness and speed of prediction and is an ideal starting tool for process optimization and control. Following this approach, a first-principles physical model for the reactor was solved numerically using the Fluid Dynamics Analysis Package (FIDAP). This transient model included property variation and thermodiffusion effects. Using software developed in house, neural networks were then trained using FIDAP simulations for combinations of process parameters determined by the statistical Design of Experiments (DOE) methodology. The outputs were the average and local deposition rates. It is shown that the trained NN model predicts the behavior of the reactor accurately. Optimum process conditions to obtain a uniform thickness of the deposited film were determined and tested using the physical model. The results demonstrate the power and robustness of NNs for obtaining fast responses to changing input conditions. A procedure for developing equipment models based on physiconeural network models is also described.

1.
Ando
S.
, and
Fukui
T.
,
1989
, “
Facet Growth of AIGaAs on GaAs With SiO2 Gratings by MOCVD and Applications to Quantum Well Wires
,”
J. Crystal Growth
, Vol.
98
, p.
646
646
.
2.
Berkman, S., Ban, V. S., and Goldsmith, N., 1978, “An Analysis of the Gas Flow Dynamics in a Horizontal CVD Reactor,” Heteroepitaxial Semiconductors for Electronic Devices, G. W. Cullen and C. C. Wang, eds., Springer, New York, Ch. 7, p. 264.
3.
Bird, R. B., Stewart, W. E., and Lightfoot, E. N., 1960, Transport Phenomena, Wiley, New York.
4.
Calmidi, V., and Mahajan, R. L., 1995, “Mixed Convection on a Horizontal Surface Within a Partial Enclosure: presented at ASME Int. Mech. Eng. Cong. Exp., Nov. 13–18, San Francisco.
5.
Cockerill
T. M.
,
Honig
J.
,
Forbes
D. V.
,
Beernink
K. J.
, and
Coleman
J. J.
,
1992
, “
Characterization of Electrical and Optical Loss of MOCVD Regrowth in Strained Layer InGaAs-GaAs Quantum Well Heterostructure Lasers
,”
J. Crystal Growth
, Vol.
124
, p,
553
553
.
6.
Dilawari
A. H.
, and
Szekely
J.
,
1989
, “
Computed Results for Deposition Rates and Transport Phenomena for an MOCVD System With a Conical Rotating Substrate
,”
J. Crystal Growth
, Vol.
97
, p.
777
777
.
7.
Dilawari
A. H.
,
Szekely
J.
, and
Daly
J.
,
1990
, “
Experimental Measurements and Theoretical Predictions for the MOCVD of Gallium Arsenide Using a Barrel-Type Reactor
,”
J. Crystal Growth
, Vol.
102
, p.
635
635
.
8.
Durst
F.
,
Kadinskii
L.
,
Peric
M.
, and
Schafer
M.
,
1992
, “
Numerical Study of Transport Phenomena in MOCVD Reactors Using a Finite Volume Multigrid Solver
,”
J. Crystal Growth
, Vol.
125
, p.
612
612
.
9.
Evans
G.
, and
Grief
R.
,
1987
a, “
Effects of Boundary Conditions on the Flow and Heat Transfer in a Rotating Disk Chemical Vapor Deposition Reactor
,”
Numer. Heat Transfer
, Vol.
12
, p.
243
243
.
10.
Evans
G.
, and
Grief
R.
,
1987
b, “
A Numerical Model of the Flow and Heat Transfer in a Rotating Disk Chemical Vapor Deposition Reactor
,”
ASME JOURNAL OF HEAT TRANSFER
, Vol.
109
, p.
928
928
.
11.
Eversteyn
F. C.
,
Severin
P. I. W.
,
vida Brekel
C. H. J.
, and
Peck
H. L.
,
1970
, “
A Stagnation Layer Model for the Epitaxial Growth of Silicon From Silane in a Horizontal Reactor
,”
J. Electrochem. Soc.
, Vol.
117
, No.
7
, p.
925
925
.
12.
Fotiadis
D. I.
,
Kremer
A. M.
,
McKenna
D. R.
, and
Jensen
K. F.
,
1987
, “
Complex Flow Phenomena in Vertical MOCVD Reactors: Effects on Deposition Uniformity and Interface Abruptness
,”
J. Crystal Growth
, Vol.
85
, p.
154
154
.
13.
Fotiadis
D. I.
,
Kieda
S.
, and
Jensen
K. F.
,
1990
, “
Complex Flow Phenomena in Vertical MOCVD Reactors: Effects on Deposition Uniformity and Interlace Abruptness
,”
J. Crystal Growth
, Vol.
102
, p.
441
441
.
14.
Freeman, J. A., and Skapura, D. M., 1991, Neural Networks, Addison-Wesley.
15.
Grew, K. E., and Ibbs, T. L., 1952, Thermal Diffusion in Gases, Cambridge Univ. Press, Cambridge, United Kingdom.
16.
Hirschfelder, J. O., Curtiss, C. F., and Bird, R. B., 1967, Molecular Theory of Gases and Liquids, Wiley, New York.
17.
Houtman
C.
,
Graves
D. B.
, and
Jensen
K. F.
,
1986
, “
CVD in Stagnation Point Flow. An Evaluation of the Classical 1D Treatment
,”
J. Electrochem. Soc.
, Vol.
133
, No.
5
, p.
961
961
.
18.
Jensen, K. F., 1994, “Transport Phenomena in Epitaxy Systems,” Handbook of Crystal Growth, Vol. 3b, D. Hurle, ed., Elsevier, Amsterdam.
19.
Jones
C. R.
, and
Furry
W. H.
,
1946
,
Rev. of Modern Phys.
, Vol.
18
, No.
2
, p.
151
151
.
20.
Kelkar, A.S., 1993, “Neural Networks for the Modeling of Diabetes Mellitus,” Master’s Thesis, Univ. of Colo., Boulder.
21.
Kleijn, C., 1991, “Transport Phenomena in Chemical Vapor Deposition Reactors,” Ph.D. Thesis, Netherlands.
22.
Kosko, B., 1992, Neural Networks and Fuzzy Systems, Prentice-Hall.
23.
Kusumoto
Y.
,
Hayashi
T.
, and
Komiya
S.
,
1985
, “
Numerical Analysis of the Transport Phenomena in MOCVD Process
,”
Jap. J. App. Physics
, Vol.
24
, No.
5
, p.
620
620
.
24.
Mahajan
R. L.
, and
Wei
C.
,
1991
, “
Buoyancy, Sorer, Dufour and Variable Property Effects in Silicon Epitaxy
,”
ASME JOURNAL OF HEAT TRANSFER
, Vol.
113
, p.
688
688
.
25.
Mahajan, R. L., 1993, “Process Optimization and Control in Semiconductor Manufacturing,” invited paper, presented at the ASME Winter Annual Meeting, New Orleans, LA, Nov. 30.
26.
Mahajan, R. L., 1996, “Transport Phenomena in Chemical Vapor Deposition Systems,” Advances in Heat Transfer, Academic Press, Vol. 28, p. 339.
27.
Marwah, M., Li, Y., and Mahajan, R. L., 1996, “Integrated Neural Network Modeling for Electronic Manufacturing,” J. Electronic Manufacturing, Vol. 6, No. 2, June.
28.
Moffat
H.
, and
Jensen
K. F.
,
1988
, “
Three-Dimensional Flow Effects in Silicon CVD in Horizontal Reactors
,”
J. Electrochem. Soc.
, Vol.
135
, p.
459
459
.
29.
Montgomery, D. C., 1991, Design and Analysis of Experiments, Wiley, New York.
30.
Naganuma
M.
,
Notomi
M.
,
lwamura
H.
,
Okamoto
M.
,
Nishida
T.
, and
Tamamura
T.
,
1990
, “
In GaAs/InP Quantum Well Wires Fabricated by GSMBE, MOCVD, and Selective Chemical Etching Techniques
,”
J. Crystal Growth
, Vol.
105
, p.
254
254
.
31.
Patnaik
S.
,
Brown
R. A.
, and
Wang
C. A.
,
1989
, “
Hydrodynamic Dispersion in Rotating-Disk OMVPE Reactors: Numerical Simulation and Experimental Results
,”
J. Crystal Growth
, Vol.
96
, p.
153
153
.
32.
Reid, R. C, Prausnitz, J. M., and Poling, B. E., 1987, The Properties of Gases and Liquids, 4th ed., McGraw-Hill, New York, p. 587.
33.
Ristorcelli, J. R., Jr., and Mahajan, R. L., 1987, “Silicon Deposition and Dopant Incorporation in Epitaxial Processes,” Proc. 2nd ASME-JSME Thermal Engineering Joint Conference, HI, Mar. 22–27.
34.
Seshadri
S.
,
Guido
L. J.
,
Moise
T. S.
,
Beggy
J. C.
,
Cunningham
T. J.
, and
Barker
R. C.
,
1992
, “
Non-equilibrium Al-Ga Interdiffusion in MOCVD Reactor Annealed AlGaAs Quantum Well Heterostructures
,”
J. Electronic Materials
, Vol.
21
, No.
1
, p.
33
33
.
35.
Simpson, P. K., 1992, “Foundations of Neural Networks,” Artificial Neural Networks, IEEE Press.
36.
Stock
L.
, and
Richter
W.
,
1986
, “
Vertical Versus Horizontal Reactor: An Optical Study of the Gas Phase in a MOCVD Reactor
,”
J. Crystal Growth
, Vol.
77
, p.
144
144
.
37.
Wang
X. A.
, and
Mahajan
R. L.
,
1995
, “
CVD Epitaxial Deposition in a Vertical Barrel Reactor: Process Modeling and Optimization Using Neural Network Models
,”
J. Electrochem. Soc.
, Vol.
142
, No.
9
, p.
3123
3123
.
38.
Wang
X. A.
, and
Mahajan
R. L.
,
1996
, “
Artificial Neural Network Model-Based Run-to-Run Process Controller
,”
IEEE Trans. CPMT, Part C
, Vol.
19
, No.
1
, p.
9
9
.
39.
Weber
C.
,
Opdorp
C. van
, and
Keijser
M. de
,
1990
, “
Modeling of Gas-Flow Patterns in a Symmetrical Vertical Vapor-Phase-Epitaxy Reactor Allowing Asymmetric Solutions
,”
J. Appl. Phys.
, Vol.
67
, No.
4
, p.
2109
2109
.
This content is only available via PDF.
You do not currently have access to this content.