Upload
others
View
2
Download
0
Embed Size (px)
Citation preview
-A185 £99 EQUILIBRIUM ANALYSIS OF THE PREPARATION OFIN(X)GA((i>-X)AS BY THE VPE-HY (U) ROME AIRDEVELOPMENT CENTER GRIFFISS AFS NY K P QUINLAN NOV 86
UNCLASSIFIED RADC-TR-6-i79 F/G 72 N
.Ehhhh10 0 010 iEE11E10EEEEE
L11! 1! 3=2
11111 1.0 L6 128 . 25
11 1 11 _ _-6
LL~IlU IA' ul
MICROCOPY RESOLUTION TEST CHART
NATIONAL BUREAU OF STANDARDS-1963-A
rm,, FtLE
'-- RADC-TR.6-179In-House Report
Ln November 196
EQUILIBRIUM ANALYSIS OF THEPREPARATION OF IlnxGai1xAs BY THEVPE-HYDRIDE TECHNIQUE; METHODTO PREDICT THE COMPOSITION OFTHE TERNARY
k ;\~. SEP 29 1987u
Kenneth P. Quinlan 0lAPPROVED FOR PUBLIC RELEASe" DISTRIBUTION UNLIMITED
ROME AIR DEVELOPMENT CENTERAir Force Systems Command
Griffiss Air Force Base, NY 13441-5700
Unclassified
SECURITY CLASSIFICATION OF THI PA Form ApprovedREPORT DOCUMENTATION PAGE OMB No. 0704-0188
la. REPORT SECURITY CLASSIFICATION lb. RESTRICTIVE MARKINGSUnclassified
2a. SECURITY CLASSIFICATION AUTHORITY 3. DISTRIBUTION/AVAILABILITY OF REPORTApproved for public
2b. DECLASSIFICATION /DOWNGRADING SCHEDULE release; distribution unlimited.
4. PERFORMING ORGANIZATION REPORT NUMBER(S) S. MONITORING ORGANIZATION REPORT NUMBER(S)
RADC-TR-86- 179
6&. NAME OF PERFORMING ORGANIZATION 6b. OFFICE SYMBOL 7a. NAME OF MONITORING ORGANIZATIONRome Air Development (if applicable)
Center ESM
6c. ADDRESS (City, State, and ZIP Code) 7b. ADDRESS (City, State, and ZIP Code)
Hanscom AFBMassachusetts 01731
8a. NAME OF FUNDING/SPONSORING 8b. OFFICE SYMBOL 9 PROCUREMENT INSTRUMENT IDENTIFICATION NUMBERORGANIZATION (if applicable)
8c. ADDRESS (City, State, and ZIP Code) 10. SOURCE OF FUNDING NUMBERSPROGRAM PROJECT TASK WORK UNITELEMENT NO. NO. NO ACCESSION NO.
I 62702F 4600 17 0711. TITLE (Include Security Classification)
Equilibrium Analysis of the Preparation of InxGal_XAs by the VPE-Hydride Technique:Method to Predict the Composition of the Ternary
12. PERSONAL AUTHOR(S) Kenneth P. Quinlan
13a. TYPE OF REPORT 13b TIME COVERED 14. DATE OF REPORT (Year, Month, Day) 15. PAGE COUNTIn-House FROM TO 1986 November 22
16. SUPPLEMENTARY NOTATION
17. COSATI CODES 18 SUBJECT TERMS (Continue on reverse if necessary and identify by block number)
,EL GROUP SUB-GROuP Equilibrium analysisI I III-V Semiconductor
=q. I I VPE -hydride19. ABSTRACT (Continue on reverse if necessary and identify by block number)
The equilibrium of the deposition of InxGal_xAs in the VPE-hydride technique wasanalyzed in order to derive a method to predict the composition of the ternary from initialpartial pressures. An equilibrium expression is derived where the constant is the productof the equilibrium constants for the deposition of the binaries and the activity of theternary, InxGalxAs, is taken as one. The relationship requires the determination of theequilibrium amounts of InxGal-'xAs deposited in order to determine the composition.Values for In"Gal'_xAs deposited in terms of partial pressure were dete-rmined at threearsenic pressures. A linear relationship was observed between the partial pressure ofthe In AslxAs deposited and the initial concentration of arsenic iP/). Good agreement
is achieved between the compositions calculated by the present method and theexperimental values reported in the literature.
20. DISTRIBUTION / AVAILABILiTY OF ABSTRACT 21 ABSTRAC SECUITYCLASSIFICATIONC uNCLASSIFED.UNLIMITED 6 SAME AS RPT D OTIC USERS Unclassifie
22a NAME OF RESPONSIBLE INDIVIDUAL 22b TELEPHONE (Include Area Code) 22c OFFICE SYMBOLKenneth P. Quinlan (617) 377-4015 1 ADC/ESM
DD Form 1473, JUN 86 Previous editions are obsolete SECURITY CLASSIFICATION OF THIS PAGE
111 M 1111Pil
iAccesio ForNTIS CRA&M {
DTIC TAB 1)Urannouniced [-Justification
By .......................Distribution I
Aviibaity Codes ContentsDitAvai 'ridIor
Dist
1. INTRODUCTION
2. EQUILIBRIUM ANALYSIS 2
3. RESULTS AND DISCUSSION ] i 5
4. CONCLUSIONS co" 13
REFERENCES 15
Illustrations
1. Mole Fraction of Gallium Arsenide, 1-x, in In Ga As as a Furi:tionx ]-X
of the Partial Pressure of Gallium Monochloride (P_;aC __.
Calculated With Eq. (13) Using PIn Ga As - 1.7 Y 10- 4 atm
atP 4.3 X l0-4 atm x -xPAs 4
2. Mole Fraction of Gallium Arsenide, 1-x, in In Ga l xA as a Fun, !ion
of the Partial Pressure of Gallium Monochloride (P(;
Calculated With Eq. (13) Using PIn Ga A - 2.6 - j)-4 amn at0 -3 -PAs4 1.3 X [3 atm -x
iii
Illustrations
3. Mole Fraction of Gallium Arsenide, 1-x, in InxGa 1 xAs as a Function of
the Partial Pressure of Gallium Monochloride (Pa
Calculated With Eq. (13) Using Pln Ga = 3 .7 X 10 atmat P 0 = 3.9 X 10- 3 atmx 1-X
As4
4. Calculated Equilibrium Values of In Ga I-xAs Deposited
(PIn xGa 1 -xAs atm) Versus Initial Arsenic Partial Pressures (PAs 4 )
5. Calculated Mole Fraction of Indium Arsenide, x, in In Ga 1 lAs as a
Function of the Ratio of the Initial Partial Pressure of IndiumMonochloride to the Total Partial Pressure of the III-monochlorides(P0 /P 0 + P ) 10
Ic I GaCI6. Calculated Variation of Indium Arsenide Mole Fraction, x, in In Galx As
as a Function of the Initial Partial Pressures of Arsenic (Ps) at TwoP04,,. PInC l
,_+ Values 11InCl GaCI
Tables
1. Comparison of Experimental x Values (In xGalx As) With ThoseCalculated by the Present Method 13
iv
Equilibrium Analysis of the Preparation
of InxGai _ x As by theVPE-Hydride Technique;
Method to Predict the Composition of theTernary
1. INTRODUCTION
Epitaxial layers of In0. 5 3 Ga0. 4 7As have electrical properties that make them
important materials for microwave and optoelectronic devices. 1-4 Epitaxial
growth of In xGalx As has been accomplished by various techniques: VPE-chloride,
VPE-hydride, 6 MOCVD, 7 MBE 8 and LPE. 9 The versatility and advantages of the
VPE-hydride method have been aptly described by Olsen and Zamerowski. 10
Mac rander and Strege 11 have demonstrated that InxGa I-xAs epitaxial layers of
excellent quality can be prepared by the VPE-hydride technique. The VPE-hydride
technique will undoubtedly be applied more frequently to prepare III-V semiconductor
devices.
Equilibrium models for the deposition of Inx Ga As have been reported by
various investigators. The primary objective of these equilibrium studies is to give
the investigators a better estimate of the initial partial pressures of the reactants
to produce a desired composition of the epitaxial layer. Many of these approaches
are complicated and require a computer code for the calculation of the equilibrium-12 .13
of the multiphase system. The papers of Nagai et al and Minagawa et al are
examples of these techniques applied to the growth of In xGaIx As. Less
(Received for publication 12 December 19S6)
(Due to the large number of references cited above, they will not be listed here.See References, page 15.)
1
complicated methods for the deposition of In xGaIx As have been reported by14 15 16 x XKajiyama, Nagai and Jacobs et al. However, these methods required
assignment of arbitrary values to the initial partial pressures or iterative proce-
dures to determine the composition of the epitaxial layer.
A simplified method has been developed in the present study where the
equilibrium expression can be used directly to analyze the reactor system perform-
ance, for example, composition of epitaxial layer, effect of arsine pressure on
composition, and so on. Evaluation of the equilibrium expressions requires values
of the amounts of In xGalx As deposited to be known. Values of In Ga l_xAs
deposited at three arsine pressures were determined from previously reported0
data. A linear plot of the initial concentrations of arsenic (PAS) versus amounts
of deposit expressed in atmospheres allows the determination of PIn Ga As at
various initial arsenic concentrations. The fairly consistent agreementX 1-x
achieved between the experimental values of other investigators and the present
approach indicates the method is useful in estimating the composition and in under-
standing the In xGa ixAs deposition system.
2. EQUILIBRIUM ANALYSIS
In the present study of the equilibrium analysis only the following species are
considered: InCl, GaCI, As , HCI, H2 , and In Ga lxAs. Other probable species
exist only in minor amounts. 17,18
The overall equation for the deposition of InxGa l xAs in the VPE-hydride
method is depicted in Eq. (1):
xlnCl + (l-x)GaCt + I/4As4 + /2H 2 In xGa -xAs + HCI (1)
14. Kajiyama, K. (1976) Vapor pressure dependence of the relative compositionof III-V mixed crystals in vapor-phase epitaxy, J. Electrochem. Soc.
123:423.
15. Nagai, H. (1979) A simple analysis of vapor-phase growth; citing an instanceof Ga xInIx As, J. Electrochem. Soc. 126:1401.
16. Jacobs, K., Simon, I., Bugge, F., and Butter, E. (1984) A simple method forcalculation of the composition of the VPE grown Ga xInl1xAs layers as a
function of growth parameters, J. Crystal Growth 69:155.
17. Ban, V. S. (1972) Mass spectrometric and thermodynamic studies of the CVDof some III-Vcompounds, J. CrystalGrowth 17:19.
18. Ban, V. S., and Ettenberg, M. (1973) Mass spectrometric and thermodynamicstudies of vapor-phase growth of In lxGa xP, J. Phys. Chem. Solids,43:1119.
2
where x varies from 0 to 1. The equilibrium expression for Eq. (1) is
K a I Ga 1- A '*PHCI 21 p X ,;x 0 P"/ 4 * (2)2
nCI ~GaCl As 2
where a is the activity and the P's are the equilibrium partial pressures. Since
ln XGa 1 -XAs can be considered a pure chemical solid, its activity may be taken as
one. 19 Equation (2) reduces to Eq. (3):
K1 P I~ HCl P' (3)
1 ;In~ GaCI As 4 H2
The equilibrium constant, K1, can be shown to be the product of the equilibriumconstants for the formation of the binaries. The equations for the binary formation(InAs and GaAs) and their equilibrium expressions are depicted in Eqs. (4) and
(5) respectively:
xInCl + x/4 As 4 + x/2 H 2 ----- 0 xmnAs + xHGI (4)
(1-x)GaCl + (1-x)/4 As 4 + (1-x)/2 H 2 - (l-x)GaAs + (l-x)HCI) (5)
a X PK 2 InAs HC1 (6)
Ic As 4 H
al1X * l-XK- GaAs HCI (7)
3 lp-x ,-I~-x/4 . lP--XFGaCI A 4 H
InL-s and GaAs in Eqs. (4) and (5) are pure chemical substances and therefore their
activities in Eqs. (6) Lind ()maY also be taken as one. The product of 1-qs. (6)
and (7) is therefore
K2~ 3 1 ,x 0 1) 1 -X. I/ 4 0 -j-,1 2
lnutl ('a( I As Ii.4 2
19). Klotz, 1.-%1 (1950) ( hemPni-.i The rmrnk nam ics P rentice-H[a 11. Inc.,* p. 268.
The right hand sides of Eqs. (3) and (8) are identical and therefore K 1 is equal to
K 2 K 3 .
The equilibrium values of the reactants and products in Eq. (8) [or (3)] may
be defined in terms of the initial partial pressures and the amount of Inx Ga As
deposited expressed in atmospheres. The following relations give the material
balances for the equilibrium partial pressures in Eq. (8).
1
PHCI = HCI + PIn xGal_xAs (9
I
where PHCI is the partial pressure of the residual hydrogen chloride that remains
after the reaction of HCI with indium and gallium in the source zones.
PinGa As is equivalent to the HCI produced in Eq. (1). Values for PX P a 1-xInx 1-x A nl al
--- - p1/4and PAs are
4
pX 0 xP XInC1 (P Pncl XPIn xGa 1 As) (10)
Sl-X =pO _X)Pnx 1 x s-x
GaC ( GaCI- (I Ga As (11)
1/4 = - 1/4 )1/4 (12)PAs 4 (As 4 PIn xGa 1 As
where P 's are the initial partial pressures.
The only variable in the above equations that needs to be known to calculate
the equilibrium partial pressures is the amount of In Ga -x s In xGa _As)
deposited. Values of the variable (Pin Ga xAs) at three different As 4 pressuresx l-x 2
were determined from the data of Weyburne and Quinlan. The study showed thatonly one value of the In xGaIx As deposit was necessary to describe the experi-
mental data at each arsenic pressure. A plot of PIn Ga As versus initial arsenicx 1-x
pressures (P 0 ) gave a linear relation that enables values of PInxGa-AS to beAs4 Inx 1 -x A
determined at any arsine pressure. The procedure and results are described in
20. Weyburne, D.W., and Quinlan, K. P. (1985) The Effects of Arsine Pressureon the Compositions, Carrier Concentrations, Mobilities and Growth Rateso--hpitaxial Layers of Gan_-' xe--e-- - -MTTn-PE-Hydride Technique,
RADC-TR-85-238, AD Al-5771, Rome Air Development Center, Air ForceSystems Command, Griffiss Air Force Base, NY. 13441-5700
4
the following section. The initial partial pressure of hydrogen, PH 2 is used in
Eq. (8) since the final pressure does not differ significantly from the initial
pressure.
3. RESULTS AND DISCUSSION
The determination of the composition of In xGa _xAs prepared by the VPE-
hydride technique can be accomplished with the expanded form of Eq. (8) as pre-
sented in Eq. (13)
PHCI + PIn XGa IxAs (13)
, 2K X= (P O o l-x - I/4 P 114 p 1/Z (S I nl - XPinXGa IXAS1
GaCI InxGa I -xA s As 4 In x Ga -x As pli2
where x and 1-x are the mole fractions of indium arsenide and gallium arsenide,
respectively. The terms in Eq. (13) have been defined in the preceding section.
P0 and PGaC the initial partial pressures of the Group III-monochlorides, were
calculated from the starting hydrogen chloride pressure in the Group III-element
source zones and the equilibrium constants for the following reactions:
In + HCI mInCl + 1/2 H 2 (14)
Ga + HCI - GaCI + 1/2 H 2 (15)
Values for the equilibrium constants of reactions (14) and (15) were calculated
from the thermodynamic data of Seki et al2 1 and Kirwan, 22 respectively. PCI'
the residual HCI from Reactions (14) and (15), was determined from the difference
between the starting HCI pressures and the III-monochlorides. The POAs was 1/4
of the initial arsine partial pressure. K2 and K 3 were determined from the thermo-
dynamic relationships derived by Jacobs et al. Values of In Ga As were
found by determining those values of PIn Ga I-xAs in Eq. (13) that gave the best fit
to describe the data of Weyburne and Quinlan at the three arsine pressures.
21. Seki, H., and Minagawa, S. (1972) Equilibrium computation for the vaporgrowth of InxGa lxP crystals, Japan J. Appl. Phys. 11:850.
22. Kirwan, D. J. (1970) Reaction equilibria in the growth of GaAs and GaP by thec:hloride transport process, J. Electrochem. Soc. 117:1572.
5
Figures 1, 2 and 3 show the calculated curves v th the experimental data of
Weyburne and Quinlan 2 0 at the three arsenic pressures (Ps4 The solid lines4-4
were calculated with Eq. (13) using values of P InxGal-xAs 1.7 X 10S4 4 14
2. 6>< 10 -4 and 3. 7 X 10 atm at arsenic pressures (Ps of 4. 3 X 10
1. 3 X 10 and 3. 9 X 10 atm, respectively. The figures demonstrate that
Eq. (13) describes the data accurately when one value of PIn xGa l-As is used at
one arsenic pressure. The straight line indicates that changing the gallium mono-
chloride pressures over the range indicated has little effect on the equilibrium
amount of In xGa l-xAs deposited. These results clearly illustrate the fact that the
equilibrium expression [Eq. (13)] can be used with confidence to interpret the
deposition of the ternary in the 0.4-0.5 value range for 1-x.
0.7
0.6
- 0.500
00
) 0.4- es"4.3 x I5 4 atm.
0.3 -PO 4.3 xI063atm.ICI
C' 2 1 1 -3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 x 104
POaC (atm.)
Figure 1. Mole Fraction of Gallium Arsenide, l-x, in InxGa I-xAs as a Functionof the Partial Pressure of Gallium Monochloride (P . Calculated
(1)UigP . - aClWith Eq. (13) Using PinGa A 1.7 X 10 atm at P 4.3 X 10- atm.
In Ga 1 As -AsExperimental points are from Reference 20 and are indicated by 0
6
0.6
0o.5oJ- 0K
0
0 0
, p: -1.3x lO 3 atm.- As4 -
0.3 PO =4.3x10"3 atm.mnCI
0.2 1 I I I I I I -
3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 x I04
PO I(atm.)
Figure 2. Mole Fraction of Gallium Arsenide, 1-x, in In xGa lxAs as a Functionx
of the Partial Pressure of Gallium Monochloride (P ac ). : Calculated
With Eq. (13) Using PInxGal_xAs = 2.6X 10- 4 atm at As 4 = 1.3 X 10 - 3 atm.
Experimental points are from Reference 20 and are indicated by 0
Curve fitting of Eq. (13) with the experimental data shows that increasing the
arsenic pressure increases the equilibrium amounts of In xGa -XAs that are
deposited. This is not surprising since equilibrium considerations mandate this
chemical behavior. The calculated equilibrium amounts of In xGa lxAs deposited
as a function of the initial arsenic pressures are exhibited in Figure 4. A straight
line is drawn through the data points that enables values of PIn xGa -- As to be
determined at any partial pressure of P . The data points in Figure 4 indicate
a slight parabolic trend as implied by Eq. (13),(PsP1 / 4 = fin Ga0 s4 inx Ga -x
calculations with the linear relationship at low PAs 4 values gave more acceptable
values.
The calculated lines in Figures 1, 2 and3 extendover a limited range of gallium
monochloride pressures. Figure 5 illustrates the effect of extending the partial
pressures of gallium monochloride on x, the mole fraction of indium arsenide at
7
0.6
,) 0.5 0
0 0:- 0.4
X P,- 3.9x I0 3 atm.
0.3 PO =4.3xlO 3 atm.
0.2 I I I I5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5x 10- 4
PO (atm.)
Figure 3. Mole Fraction of Gallium Arsenide, 1-x, in In xGa As as a Functionx -x
of the Partial Pressure of Gallium Monochloride (P0 ) : CalculatedG PaCI -3
With Eq. (13) Using P Ga xAS 3.7 X 10 atm at PO 3.9 X 10 atm.In G AsAs 4
Experimental points are from Reference 20 and are indicated by 0
two P . Values of x are plotted as a function of the ratio of the partial pressure
P 0
of InCl to the total III-monochloride pressures InCI The partial_ PGacI+PInC1
pressure of In(*l wa held constant at 4. 3 X 10 atm and that of GaC1 was varied
from 1.9 X 10- 3 to 4.4 X 10- 5 atm. The figure shows that decreasing the partial
pressure of GaCl by approximately 2. 5X decreases the mole fraction of Ga in
In xGa lxAs by -13 percent. These curves are similar to those reported by
Kajiyama. 14 The data of Conrad ct al23 and Enstrom et al24 are plotted to show
23. Conrad, R. W., 1-oyt, P. L., and Martin, D. D. (1967) Preparation ofepitaxial Gax In xAs, J. Electrochem. Soc. 114:164.
24. Enstrom, R.E., Richman, D., Abrahams, M.S., Appert, J.R., Fisher, D.G.,Sommers, A. 1., and Williams, B. F. (1970) Vapour growth of Ga 1 xln xAs
for infrared photocathode applications, Proc. 3rd Int. Smp. on GalliumArsenide, Conf. Series Number 9, nst_ ?'5TPE7Y- " -'chen, Germany,
8
that the present method of analysis accurately describes the trend of the experi-mental values of other investigators. The deposition temperatures used by theseinvestigators are higher (725 and 745 0 C) than the temperature of 7000C used in thepresent study. Figure 5 shows that at high GaC1 partial pressures
o -3(P GaC > 1. 9 X 10 atm), the mole fraction of indium arsenide (x) calculated by thepresent method approaches zero rather abruptly. This is different from the experiment-
al results where x slowly approaches zero. A possible explanation for this differenceis that the equilibrium amounts of InxGaI-xAs deposit do not remain constant butincreases at high GaCI partial pressures.
- 10-
1.0 2.0 3.0 4.0 X 10-4
PlnxGoajxAs (atm.)
Figure 4. Calculated Equilibrium Values of In Ga 1 xAs Deposited (PIn Ga Asx - x al-x
atm) Versus Initial Arsenic Partial Pressures (PI 0 P Values areA 4 InxGa I-x A s
Calculated Using Eq. (13) (see text)
92=2
I .0PAs 4 (atm.)
0.9 ... 4.0 x 10"3
- 2.5 x 104
0.8
~/q
/
0.6 /4€ /
K, /
0.5 /Xc I )
C /:< 0.4 /I
II
0.3 I
I
0.2 i
0.1 /
0.0.0.5 0.6 0.7 0.8 0.9 1.0
PlnC IPO + PO
InCI GaCIFigure 5. Calculated Mole Fraction of IndiumArsenide, x, in In xGa lXAs as a Function of the
Ratio of the Initial Partial Pressure of IndiumMonochloride to the Total Partial Pressure of
III-Monochlorides (P0C/ no + p0in InC I GaCl~
Experimental values are from References 23and 24. Those from Reference 23 areindicated by 0; those from Reference 24 areindicated by r0
10
I I I I I I I
0.6
0.5
0.4-
0=-8o 0.3xC
0.2PO
Y/ y PnCl
InCI GaCI0.1
0 .0 I , I I ,2xlO 4 I0" 5xlO3
PO (atm.)
Figure 6. Calculated Variation of Indium Arsenide Mole Fraction, x, inIn xGa lxAs as a Function of the Initial Partial Pressures of Arsenic
x P0
(P 0 aCI Values(ps)at Two pOnC pO
InCA + GaCI
The effect of arsenic pressure on the composition of the ternary is observed in
Figure 5. The mole fraction(x) of indium arsenide is increased in the ternary with
increasing arsenic partial pressures. This effect is shown more dramatically in Fig-
ure 6 where POs 4 is plotted against the mole fraction(x) of indium arsenide in the
ternary at constant GaCI and InCl partial pressure. With increasing P s 4 the
11
mole fraction of indium arsenide increases until a value is obtained that remains
constant with further increases in arsenic pressure. Experimental studies have not
been reported at these high P 0 to see whether constant indium arsenide mole frac-As416 14
tions are observed. The method proposed by Jacobs et al and Kajiyama show
the indium arsenide mole fraction remains constant at high P 4 values while that
15 Aof Nagai indicates an increase in indium arsenide mole fraction with increasing
arsenic pressures. The present method derived from experimental data clearly
shows that increasing arsenic pressures increases the indium arsenide in
In xGax As. A possible explanation for this observation is that as the pressure of
As 4 increases, the number of collisions between indium atoms and As 4 molecules
increases resulting in an increase in the number of effective collisions. The effec-
tive collision number between Ga and As 4 probably remains maximum even at low
As 4 pressures. This difference in chemical reactivity gives rise to In xGaIx As
containing greater amounts of indium at high As 4 pressures.
The present method is valuable only if the composition of the ternary can be
predicted from the reactants' partial pressures. Table I shows the compositions
of In xGaI-XAs calculated with the present method along with those determined
experimentally by Hyder et al25 and Jacobs et al. 16 Exceptionally good agreement
is observed if one considers the variability that each reactor demonstrates. Al-
though the deposition temperatures of Hyder et al (688*C) and Jacobs et al (727 0C)
are different from the temperature (700 0 C) used in the calculations, the results
illustrate that this temperature difference has little effect in the present compari-
son. Not all the experimental data studied by the present method gave such good11
agreement as reported in Table 1. For example, Macrander and Strege reported
initial partial pressures of GaCI, InCl and AsH 3 to prepare In 0.53Ga 0.47As. Using
these same values, the present method gave a ternary composition corresponding
to In0. 3 4 Ga 0.66As. This difference in composition can be attributed to a number
of factors, such as reactor characteristics, errors associated with mass flow con-
trollers, and errors in analyses of gas mixtures. The value of the present method
lies in its ability to determine with simplicity the partial pressures of GaCl, InCt,
and AsH 3 to be used in preparing a desired composition of a ternary, for example
In 0.53Ga 0.47As.
25. Hyder, S. B. , Saxena. R. R.. (Chiao, S. H., and Yeats, R. (1979) Vapor-phaseepitaxial growth of InGaAs lattice-matched to (100) InP for photodiodeapplication, Appl. Phys. Lett. 35:787.
12
Table 1. Comparison of Experimental x Values (In xGax As) \WithThose Calculated by the Present Method
CalculatedExperimental x
Sample x (Present Method) Reference
22-1 0.53 0.52 Hyder et al 2 5
18-22 0.53 0.46 Hyder et al 2 5
24-2 0.53 0.59 Hyder et al 2 5
Fig. 1 0.54 0.59 Jacobs et al 1 6
Fig. 1 0.55 0.60 Jacobs et al116
Fig. 1 0.52 0.59 Jacobs et al 1 6
4. CONCLUSIONS
A method has been developed to estimate the composition of the ternary,
In x Ga l x As , prepared by the VPE-hydride technique. The method requires know-
ledge of the equilibrium amounts of Inx Ga deposited. A plot is given where
equilibrium amounts of In Ga ixAs deposited can be determined at any initial arsenicpressure. The composition of the ternary is estimated from the initial partial
pressures of the reactants and the amount of InxGa 1 xAs deposited. Comparison
of the results derived from the present method with experimental compositions
exhibited good agreement. The method may be applied to other ternary systems,
for example, In xGa 1-x P, to understand their deposition process.
13
References
1. Chen, C.Y., Cho, A.Y., and Gaebinski, P.A. (1985) A new In 0 . 5 3 Ga 0.47Asfield effect transistor with a lattice-mismatched GaAs gatefor high speed circuits, IEEE Electron Devices Lett. EDL-6:20.
2. Gardner, P.D., Narayan, S.Y., Yun, Y.H., Colvin, S., Paczkewski, J.,Dornan, B. , and Askew, R.E. (1982) Ga 0 4 7 1n0 5 3 As deep depletion andinversion mode MISFETs, Conf. on Galliumn Arsenide and RelatedComeounds, Ser. No. 65, Inst. oTTPEys., Albuquerque, N. Mex.,pp. 339-4U5.
3. Chand, N., Houston, P.A., and Robson, P.N. (1985) Novel high-speed
In0. 5 3 Ga0. 4 7 As lateral phototransistor, Electron. Letters 21:308.
4. Tell, B., Liao, A.S.H., Brown-Goebeler, K.F., Bridges, T.J.,Burkhardt, E.G., Chan, T.Y., and Bergano, N. (1985) Monolithic integra-tion of a planar embedded InGaAs p-i-n detector with InP depletion-modeFET's, IEEE Trans. Electron Devices ED-32:2319.
5. Cox, H.M., Koza, M.A., and Kermamidas, V.G. (1985) Vapor phaseepitaxial growth of high purity InGaAs, InP, and InGaAs/InP multilayerstructures, J. Crystal Growth 73:523.
6. Erstfeld, T.E., and Quinlan, K. P. (1985) The growth of epitaxial layers ofGa0. 4 7 In0. 5 3As by the vapor-phase epitaxy-hydride method using agallium-indium alloy, J. Electrochem. Soc. 131:2722.
7. Kuo, C.P., Cohen, R.M., and Stringfellow, G.B. (1983) OMVPE growth ofGaInAs, J. Crystal Growth 64:461.
8. Cheng, K.Y., and Cho, A.Y. (1982) Silicon doping and impurity profiles inGa0. 4 7 In0. 5 3As and Al0. 4 8 In 0 . 5 2 As grown by molecular beam epitaxy,
J. Appl. Phys. 53:4411.
15
References
9. Oliver, J. D., and Eastman, L. F. (1980) Liquid phase epitaxial growth andcharacterization of high purity lattice-matched Ga In xAs on <111 > B InP,J. Electron. Mater. 9:63. x 1-
10. Olsen, G. H., and Zamerowski, T. J. (1980) Crystal growth and properties ofbinary, ternary and quarternary (In, Ga) (As. P) alloys grown by thehydride vapor-phase epitaxy technique, in Progress in Crystal Growth andCharacterization, B. R. Pamplin, Ed., Vol. II, Pergamon Press Ltd..London, pp. 309-375.
11. Macrander, A. T., and Strege, K.E. (1986) X-ray double-crystal characteriza-tion of highly perfect InGaAs/InP grown by vapor-phase epitaxy,J. Appl. Phys. 89:442.
12. Nagai, H., Shibata, T., and Okamoto, H. (1971) Thermodynamical analysis forthe vapor growth of GaxInl1xAs crystals, Japan J. Appl. Phys. 10: 1337.
13. Minagawa, S., Seki, H., and Eguchi, H. (1972) Thermodynamic calculation forthe vapor growth of In xGa l _ x As: the In-Ga-As-Cl-H system,
Japan. J. Appl. Phs. 11:855.
14. Kajiyama, K. (1976) Vapor pressure dependence of the relative composition ofIII-V mixed crystals in vapor-phase epitaxy, J. Electrochem. Soc. 123:423.
15. Nagai, H. (1979) A simple analysis of vapor-phase growth; citing an instanceof GaxIn 1 -xAs, J. Electrochem. Soc. 126:1401.
16. Jacobs, K., Simon, I., Bugge, F., and Butter, E. (1984) A simple method forcalculation of the composition of the VPE grown Ga XIn -XAs layers as a
function of growth parameters, J. Crystal Growth 69:155.
17. Ban, V.S. (1972) Mass spectrometric and thermodynamic studies of the CVD ofsome III-V compounds. J. Crystal Growth 17:19.
18. Ban, V.S., and Ettenberg, M. (1973) Mass spectrometric and thermodynamicstudies of vapor-phase growth of In Ga P. J. Phys. Chem. Solids,43:1119. l-x x
19. Klotz. I.M. (1950) Chemical Thermodynamics, Prentice-Hall, Inc., p. 268.
20. Weyburne, D. W., and Quinlan, K. P. (1985) The Effects of Arsine Pressureon the Compositions, Carrier Concentrations, Mobilities and Growth Ratesof Epitaxial Layers of Ga xnl_x As Prepared by the VPE-Hydride Technique,
RADC-TR-85-238, AD Ai-52179, Rome Air Development Center, Air ForceSysteins Command, Griffiss Air Force Base, NY, 13441-5700.
21. Seki, H. , and Minagawa, S. (1972) Equilibrium computation for the vaporgrowth of InxGa 1-xP crystals, Japan J. Appl. Phys. 11:850.
22. Kirwan, D.J. (1970) Reaction equilibria in the growth of GaAs and GaP by thechloride transport process, J. Electrochem. Soc. 117:1572.
23. Conrad, R.W., Hoyt, P.L., and Martin, D.D. (1967) Preparation of epitaxialGa xInlx As, J. Electrochem. Soc. 114:164.
16
References
24. Enstrom, R. E. , Richmian, D., Abrahams, M. S., Appert. J. R., Fisher, D. G.,Sommers, A. H., and Williams, B. G. (1970) Vapour growth of Ga -In xAs
for iafrared photocathode applications, Proc. 3rdlInt. ymp * on GalliumArsenide, Conf. Series Number 9, Inst.oTT7 ,acen Grmay,
25. Hyder. S. B.,* Saxena, R. R. , Chiao, S. H., and Yeats, R. (1979) Vapor-phaseepitaxial growth of InGaAs lattice-matched to (100) InP for photodiodeapplication, AppI. Phys. Lett. 35:767.
17
MISSION* Of
Rame Air Development CenterRAVC ptans& and execute.6 4e,6eatch, de'eeopment, te.tand zetec-ted acqwu-'..Non pLogtam.6 in 6uppo'ut o6
*Command, Cont,%t, Communicationz and Intetgence(C31) acivitez. TechniLcat and engiLneevLing4&uppotrt within ata o6 compe.tence iA~ pkovided toESV Pto4o9a 066ice. (P06) and o-the4 ESV etementz~to pet~o4m e66ective acquiation o6 C31 4yqen.Tke a'reaz o6 technicat compe-tence inctudecommunica.tZan., command and cont~~o, batttemanagement, indotmation p'Lace.6.6ing, 4A4vettanceAenoL, intetZ-Lgence data cottection and handting,.6oZid s.tate 4cienceAi, etettomagne-tics, andp4opagatcon, and etecttonic, maintainabitity,and com'pati.bitity.
Printed byUasIed Stites Air ForceHeec... APS, Mass. 01731
~'%%