-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

~'%%

MOB-look