N A S A T E C H N I C A L

MEMORANDUM

oovOOSCO

NASA HLXJ268. c.tN COPY! RETURN TO

VL TECHN1CA' ' ""-\RY

K1RTLAND AF1

THERMAL EXPANSION IN THE

NICKEL-CHROMIUM-ALUMINUM AND

COBALT-CHROMIUM-ALUMINUM SYSTEMS TO 1200° C

Carl E. Lowell, Ralph G. Garlick, and Bert Henry

Lewis Research Center

Cleveland, Ohio 44135

N A T I O N A L A E R O N A U T I C S AND SPACE ADMINISTRATION • WASHINGTON, D. C. • AUGUST 1975

https://ntrs.nasa.gov/search.jsp?R=19750020136 2018-05-08T03:45:06+00:00Z

TECH LIBRARY KAFB, NM

1. Report No.

NASA TM X-32682. Government Accession No.

01520853. Recipient's Catalog No.

4. Title and Subtitle THERMAL EXPANSION IN THE NICKEL-

CHROMIUM-ALUMINUM AND COBALT-CHROMIUM-ALUMINUM SYSTEMS TO 1200° C

5. Report DateAUGUST 1975

6. Performing Organization Code

7. Author(s)

Carl E. Lowell, Ralph G. Garlick, and Bert Henry8. Performing Organization Report No.

E-8323

9. Performing Organization Name and Address

Lewis Research CenterNational Aeronautics and Space AdministrationCleveland, Ohio 44135

10. Work Unit No.

505-0111. Contract or Grant No.

12. Sponsoring Agency Name and Address

National Aeronautics and Space AdministrationWashington, B.C. 20546

13. Type of Report and Period Covered

Technical Memorandum14. Sponsoring Agency Code

15. Supplementary Notes

16. Abstract

Thermal expansion data were obtained on 12 Ni-Cr-Al and 9 Co-Cr-Al alloys by high-temperatureX-ray diffraction. The data were computer fit to an empirical thermal expansion equation de-veloped in this study:

LP -. i , F -

where LP™ is the lattice constant at any temperature T in °C, LP2_O_ is the lattice constantat 25° C, and R is an expansion constant. The fit was excellent to good. The expansion con-stants depended on phase but not on composition. For phases in the Ni-Cr-Al system, the ex-

4 4 4pansion constants were as follows: 19. 2x10 * for y/y', 19. 9x10 for /3, and 13. 4x10 fora-Cr. For phases in the Co-Cr-Al system, the expansion constants were as follows:20. 9x10 for a -Co and 17. 8x10" for /3. Only a-Cr had an expansion constant lowminimize oxide spalling or coating cracking induced by thermal expansion mismatch.

17. Key Words (Suggested by Author(s))

19. Security dassif. (of this report)

Unclassified

18. Distribution Statement

Unclassified - unlimitedSTAR Category 26 (rev. )

20. Security Classif. (of this page)

Unclassified21. No. of Pages

19

22. Price*

$3.25

For sale by the National Technical Information Service, Springfield, Virginia 22151

THERMAL EXPANSION IN THE NICKEL-CHROMIUM-ALUMINUM

AND COBALT-CHROMIUM-ALUMINUM SYSTEMS TO 1200° C

by Carl E. Lowell, Ralph G. Garlick, and Bert Henry

Lewis Research Center

SUMMARYV

The effect of temperature on the lattice parameters of phases in 12 nickel-chromium-aluminum (Ni-Cr-Al) alloys and 9 cobalt- chromium -aluminum (Co-Cr-Al) al-loys was determined by high-temperature X-ray diffraction (HTXRD). The temperaturerange was 25° to 1200° C. The data for each phase of each alloy were computer fit to anempirical thermal expansion equation developed in this study:

where LP_, is the lattice constant at any temperature T in °C, LP 0 is the latticeconstant at 25° C, and R is an expansion constant. The fit was excellent in nearly allcases. An expansion constant was derived for each phase. Comparing expansion con-stants revealed that, for a given phase, R was independent of alloy composition. Forphases in the Ni-Cr-Al system, the expansion constants were as follows: 19. 2x10"for y/V, 19. 9x10 "4 for 0, and 13. 4xlO"4 for o--Cr. For phases in the Co-Cr-Al sys-tem, the expansion constants were as follows: 20. 9x10 for a -Co and 17. 8x10 for 0.Of all the phases, only or-Cr in the Ni-Cr-Al system had an R sufficiently low to re-duce to an unimportant level the stress induced by thermal expansion mismatch betweenoxide and substrate or coating and substrate.

INTRODUCTION

The nickel-chromium-aluminum (Ni-Cr-Al) and cobalt-chromium-aluminum(Co-Cr-Al) systems, commonly called M-Cr-Al systems, are becoming increasinglyimportant in high-temperature applications where combined oxidation and hot corrosionresistance are required. Such applications include coatings for blades and vanes in ad-

vanced gas turbine engines (ref. 1) or matrices for dispersion-strengthened alloys insimilar turbine engine applications (ref. 2). M-Cr-Al systems are highly oxidation re-sistant because of their initial rapid formation of a thin protective oxide layer (mostlyaluminum sesquioxide (AUOo)) upon high-temperature exposure. In intermittent ser-vice, however, there is a tendency for this protective alumina-rich scale to spall offduring each cooling cycle. Such spalling requires a renewed formation of the oxide uponheating. After many such heating and cooling cycles, alloy surfaces become sufficientlydepleted in aluminum (unless resupplied) so that other less protective oxides form(ref. 3). The oxidation process is then accelerated until ultimately the material is soseverely attacked that it is subject to surf ace-induced failure in service.

The most commonly accepted cause for the spalling of oxides during thermal cy-cling can be described as follows:

(1) The oxide scale forms at temperature and is coherent with the substrate.(2) As the material cools at the end of a cycle, the oxide and the metal contract dif-

ferently.(3) This results in a stress in the substrate and a stress in the oxide.(4) When the stress in the oxide exceeds its strength, it fails and the oxide cracks

and/or spalls off.If this mechanism is the primary cause of spalling, reducing the difference in the coef-ficients of thermal expansion (CTE) between the substrate and the oxide would lessen thestress in the oxide and hence decrease the tendency for the oxide to spall. A similarargument has been used to explain M-Cr-Al coating failure by cracking. Here the im-portant difference in CTE is between the M-Cr-Al-based coating and the substrate.

The work'described in this report is part of a larger program designed to identifythe optimum M-Cr-Al composition ranges based on a number of evaluation criteria.The program includes work in the following areas:

(1) Cyclic oxidation(2) Cyclic hot corrosion(3) Coefficient of thermal expansion (CTE)(4) Ductility(5) Diffusion

In this report the CTE's in the Ni-Cr-Al and Co-Cr-Al systems are surveyed for thosecompositions which show the greatest potentials for combined oxidation and hot corro-sion resistance. To this end, the thermal expansion of individual phases for a numberof single and multiphased materials (12 Ni-base and 9 Co-base) were determined byhigh-temperature X-ray diffraction (HTXRD). The data were then fit by computer to athermal expansion equation which allowed the CTE for each phase of each alloy to becharacterized by a single constant. These constants were then assessed to determinewhether or not optimum alloys could be chosen to minimize the CTE effect in spallingand/or coating cracking.

MATERIALS AND SAMPLE PREPARATION

The distributions of compositions selected for casting are shown in figures 1 and 2.The original intent was to cast nine different compositions in each system ranging from6- to 30-at. % Al and 10- to 22-at. % Cr. (Several Ni-base alloys had to be recast toget closer to these nominal compositions.) The distribution was chosen to cover thecompositions expected to have good oxidation and corrosion resistance. A "star" ar-ray was selected to give the most information on the effects of trends in compositionwith the least number of data points (ref. 4). All alloys were melted in zirconia cru-cibles and cast under argon into zirconia molds. Each casting consisted of a "tree" of10 coupons whose dimensions were 2. 5 cm by 5. 1 cm by 0. 25 cm. Each coupon had asmall triangular riser attached which was removed after casting and used for chemicalanalysis by atomic absorption spectroscopy. The results of these analyses for Cr and Alare shown in table I.

A coupon from each casting was cut into several 1. 3-cm by 0. 95-cm by 0. 25-cmpieces and glass-bead blasted. These coupons were used to make the expansion meas-urements.

EXPERIMENTAL PROCEDURES

The high-temperature X-ray diffractometer (HTXRD) has been described fully inseveral publications (e. g. , ref. 5). In summary, each sample was mounted in the dif-fractometer, briefly heated in helium; to 1200° C, and slowly cooled to room tempera-ture before measuring in order to relieve stress from the glass -bead blasting. A com-plete diffractometer scan was made at room temperature in order to identify the phasespresent and to obtain room-temperature lattice constants. The samples were reheatedto 50° C and then heated in 50° C increments to 1200° C. At each temperature a latticeconstant was determined for each phase by using the highest 29 diffraction lines pos-sible. The samples were then cooled in 50 C increments. Lattice constants were ob-tained at each temperature to a precision of ±0. 0001 nm (±0. 001 A). Upon cooling toroom temperature, another complete scan was made to check for possible changes in thephases or for excessive oxidation.

When all lattice constant data were collected, the lattice constant values for eachphase of each alloy were computer fit to an equation developed for this study:

(1)

where LP™ is the lattice constant at any temperature T in C, LP o is the lattice°

3

constant at 25 C, and R is an expansion constant. Also calculated was the mean coef-ficient of thermal expansion (CTE) over the temperature range 25° to 1200° C. Thisprocedure is detailed in appendix A.

RESULTS

Exposing the test alloys to 1200 C in static helium resulted in only minor amountsof oxides being formed from residual oxygen in the gas and possible leaks in the system.The oxidation that did take place had little effect upon the lattice parameter data, whichshowed hysteresis for only one alloy, Ni - 19-at. % Cr - 24-at. % Al. In all cases thealloys gave the same lattice constants before and after the CTE run. The phases thatwere found and for which CTE's were determined in the Ni-Cr-Al system were y (nickelsolid solution), y' (NioAl type), 0 (NiAl type), and «-Cr (chromium solid solution).This was expected from the phase diagram (ref. 6). When y' is present in an alloy,the presence of y cannot be determined by X-ray diffraction. The reason is that thediffraction pattern of y' contains all the lines of y plus a few extra lines. Both phaseshave the same lattice constant and apparently the same CTE because no splitting of thediffraction lines was seen even at 1200° C. Only the cv-Co (Co solid solution) and /3(CoAl type) phases were found in the Co-Cr-Al system. This may, in part, account forthe greater precision found in the Co-Cr-Al data.

Figures 3 to 6 are representative of the data obtained for the solid solution (y, y\and of-Cr) and NiAl type (/3) phases of the Ni- and Co-Cr-Al alloys. Even when a phasewas present over a limited temperature range (fig. 7), excellent fits, which would beconsistent with more complete data sets, were obtained and the scatter was small. Inthe very few cases where the scatter was larger (as in fig. 8), a reasonable fit and ex-pansion constant were found.

Tables H and III contain values for LP 0 , R, and mean CTE from 25° to 1200° Cfor all phases in each alloy. For the Ni-Cr-Al system, multiple linear regression com-bined with analysis of-variance (ref. 7) showed that, with a rejection level of 0. 05, ex-pansion constants for y/y' (average 19.2x10 ) and /3 (average 19.9x10" ) were notcomposition dependent. The same analysis showed the expansion constants for y/y'and j3 to be the same, while those for a-Cr (average 13. 4xlO~ ) differed from thosefor either y/y' or 0. The analysis is detailed in appendix B. In the Co-Cr-Al systemthere is a significant difference between the expansion constants of a-Co and /3,20. 9xlO"4 and 17. 8x10 , respectively.

DISCUSSION

Thermal expansion data and their implications are discussed both generally and asthey apply to oxidation spallation and coating cracking. In general, the various methodsof expressing thermal expansion data are compared. Also the expansion coefficients ob-tained in this investigation are compared with previously published data.

General

There is no generally accepted method of presenting thermal expansion data. Theyare often expressed as the polynomial

LPT = LPQ(1 + AT + BT2 + CT3 . . .) (2)

although the number of terms varies from two to more than five (ref. 8). This approachoften leads to excellent approximations of the data, but it has the disadvantage of makingit difficult to compare one material to another since several constants are involved. Inaddition, extrapolation from limited data sets is very unreliable. Another widely usedmethod is to reduce the data to a mean CTE. This has the advantage of yielding a singleconstant for ease of comparison, but the value is valid only for the quoted temperaturerange and gives no clue as to the shape of the thermal expansion curve.

The method used in this investigation has the advantages of both these techniquesbut none of their drawbacks. It allows a material's expansivity to be expressed by oneconstant, R. Also it describes the shape of the original lattice constant - temperaturecurve even with limited data sets.

That this technique gives data similar to those obtained by conventional dilatometrictechniques can be shown by comparing these data with current work by Felten, Friedrich,and Strangman of Pratt & Whitney Aircraft under contract NAS3-18920. Their meanCTE at 1200° C for Ni-18Cr-5Al and Ni-18Cr-9Al are (extrapolated from 1100° C) 20. 7and 18. 2, respectively. These values are in reasonable agreement with the data ob-tained by HTXRD.

One source of concern in using HTXRD is that the values obtained for the individualphases might be influenced by interphase constraint. That is, phases with differing ex-pansion coefficients might prevent each other from expanding in the same amount asthey would as single phases. However, this effect was not found. Where the HTXRDdata could be compared (at 20° C) with literature values, as for a-Cr, a-Co, and Ni(ref. 9), little difference was noted. In addition, this effect should be a function of therelative amount of the phases. A glance at tables II and El shows no evidence of suchinteractions.

Assessment of Specific Results

Spalling and CTE mismatch. - There are several equations that can be used for cal-culating the stress induced in an oxide scale from thermal expansion mismatch (refs.10 to 12). However, all give similar results. The equation quoted by Douglass (ref. 11)is

E AT(CTE - CTE )_. _ ox ux m /o\r\~v ^~~~~~~*~~~^~~~~ \ /OX TT /t \i+2M^

Em\W

where

2o\ stress in oxide, N/mox

n

Env elastic modulus of oxide, N/mOX

AT difference between oxidizing temperature and temperature to which sample iscooled, °C

CTE CTE of oxide over AT, °C"1ox '

CTEm CTE of metal over AT, °C"1

2E elastic modulus of metal, N/m

t thickness of oxide, cmox

t thickness of metal, cm

The primary assumptions are that only CTE controls spalling and that neither thermalshock nor growth stresses are important. For oxidation of Ni-Cr-Al, t „ « t andox m •EQX/Em < 2. Therefore,

If the oxidation temperature is 1200° C and the material is cooled to room temperature(25° C), AT equals 1175° C. Since the most protective oxide is Al9O.j, which is known to

10 2form on M-Cr-Al systems, its elastic modulus in bulk form will be used (37x10 N/m ,ref. 13). Equation (3) then becomes

aox = -43xl° (CTEox

As CTE^, is greater than CTE^, the oxide is in compression. Then putting in them ox Q ncompressive strength of a solid body of Al^Oo (31x10 N/m , ref. 13) allows the calcu-lation of the maximum allowable change in thermal expansion coefficient ACTE tomaxavoid failure of the oxide:

ACTE = -7. IxlO"6 °C"1

The value for CTE_,Y is 8. IxlO"6 °C"1 (ref. 14). Therefore, the maximum CTEm isR n 1

15x10" C . In tables H and III only the a-Cr phase in the M-Cr-Al systems is be-low this value. As far as CTE is concerned, the way to reduce spallation in the Ni-Cr-Al system is to produce alloys with a high volume fraction (at least 67 percent) ofa-Cr. There seems to be little possibility of doing this in the Co-Cr-Al system.

Coating, cracking, and CTE mismatch. - When the effect of CTE mismatch on acoating-substrate system is being considered, the calculations are a little more nebu-lous. The ratio of thicknesses is not insignificant and the elastic moduli and strengths ofM-Cr-Al systems are not known. For a coating-substrate system, the subscripts oxand m become coat (coating) and subs (substrate), respectively. Assuming a modulus

1 0 9of 21x10 u N/m'2 and a tcoat/tsubs of 0. 1 yields a stress in the coating of

CTcoat = 18xl°10 (AT)(ACTE) N/m2 (5)

Coating deposition temperatures range from 900° to 1100° C; an average AT may betaken to be 1000° C. In this case, equation (5) reduces to

CTcoat = 18xlo? ACTE N/m2

where ACTE is in units of 10"6 °C"1.For example, at 1000° C, the CTE of an experimental blade material (a direc-

tionally solidified eutectic), y/y' - 6, is 14xlO~6 °C"1 (Pratt & Whitney Aircraft, NAS3-18920). This CTE is below those of all phases of the M-Cr-Al systems except o--Crof Ni-Cr-Al system and is due to the large volume fraction of NioNb in this alloy.Therefore, M-Cr-Al coatings would be in tension, except for those with very highamounts of a-Cr. Because its CTE is slightly lower than that of y/y' - 6, coatingsrich in a -Cr would be in slight compression. Without better knowledge of the moduliand yield strengths of M-Cr-Al systems, nothing further can be decided about the CTEeffect on coatings. In summary, the a-Cr phase apparently could minimize thermalstresses in a coating on a low-CTE alloy and could reduce spalling of aluminum oxide.

SUMMARY OF RESULTS

The results of using high -temperature X-ray diffraction (HTXRD) to determine thecoefficients of thermal expansion (CTE) of Ni-Cr-Al and Co-Cr-Al alloys from roomtemperature to 1200° C may be summarized as follows:

1. Expansion data for these systems can be well described by an equation developedin this study with only one constant, R, for a lattice parameter at any temperature:

2. The expansion constants R for y/y* and p in the Ni-Cr-Al system were nearlythe same, 19. 2xlO~4 and 19. 9xlO~4, respectively.

3. The expansion constant R for a -Cr was 13. 4x10 .4. There was a slight but significant difference between the expansion constants for

a -Co and /3 in the Co-Cr-Al system, 20. 9xlO"4 and 17. 8xlO"4, respectively.

CONCLUSIONS

From the thermal expansion data obtained on alloys in the Ni- and Co-Cr-Al sys-tems, the following conclusions may be drawn:

1. The or-Cr phase in the Ni-Cr-Al alloys is most desirable for minimizing thermalstresses induced by thermal expansion coefficient (CTE) mismatch either between anoxide and an alloy or between y/y' - 6 and an M-Cr-Al coating.

2. In the Co-Cr-Al system, little can be done to minimize CTE mismatch becausethe reduction in going from a-Co to /3 would be "slight.

3. The equation L?T = LP 0 (1 + R)L1+(T/273)-J ' , where LP is the lattice pa-rameter at any temperature T, LP 0 is the lattice parameter at 25° C, and R is anexpansion constant, is most useful in describing thermal expansion and comparingthermal expansion data. It fully describes the shape of the curve and allows thermal ex-pansion data to be compared by using a single coefficient for each material.

Lewis Research Center,National. Aeronautics and Space Administration,

Cleveland, Ohio, May 9, 1975,505-01.

APPENDIX A

DEVELOPMENT OF MATHEMATICAL MODEL

In scanning the data, one of many reasonable candidates for an appropriate mathe-matical model appeared to be

• y = A(B)X (Al)

where y = LPT and x = T. In this case the parameters A and B could be estimated bylinearizing the model in the form

lny = lnA + x l n B

and performing a linear regression.In observing the results, two problems became apparent. The first difficulty arose

from the fact that the dependent-variable differences were several magnitudes smallerthan those of the independent variable. As a result, the parameter In B hovered nearzero but remained positive. This led to the observation that a compound growth equa-tion

y = A(l + r)x (A2)

was, in fact, a model that was more realistic and that could better cope with the problemof scaling since it could be treated as

y = A(l + qr)x/q

where r is the rate of growth and q could be chosen a priori. Since x was given in°C, a first choice for q was 273. This choice proved to be successful. It could beused both as a scale factor and as a constant to convert from Celsius to Kelvin. Thus,the independent variable became

x _ T + 273q 273

The second problem hinged on the need to provide an improved fit. The fit was im-proved by raising the independent variable to some power. Since many physical rela-tionships depend on T ' , an exponent of 1. 5 seemed to be the most likely candidate.

These modifications resulted in the model

y =

which could be linearized to

or in terms of this study

/ T\1.5In LPT = in LP + 1 + -L-] ln(l + R)

1 ^ <- \ 273/

or

1.5

The least-squares method could then be used to determine the parameters In.A andIn B. In addition, error estimates of the parameters as well as an overall standarddeviation were calculated for each set of data. v—

10

APPENDIX B

MULTIPLE LINEAR REGRESSION WITH ANALYSIS OF VARIANCE

A multiple linear regression analysis was performed on the expansion constants forNi-Cr-Al listed in table n according to the model equation

R = bQ + ± S. E. E. (Bl)

by using the technique described in reference 7 with dummy variables. In the presentinvestigation,

Because there were two duplicate runs available, these were used for the mean-squareerror terms. The final equation for an a rejection level of 0. 05 was

R = 19.46 - 6.06 x2 ±1.43

The hypothesis that all the y/y' expansion constants are the same was tested by thelack-of-fit term in the following analysis of variance (ANOVA) table:

Source

Lack of fitreplication

Total re-sidual

Sum ofsquares

44. 4045.5650

44.9695

Degrees offreedom

192

--

Mean square

2. 2202.2825

2.0441

1/2The standard estimate of error is (2. 0441) ' and the variance ratio test is

f-, Mean-square lack of fit „ 0[-nr = -* = /. BoyMean-square replication

At the 0. 05 significance level, the value of F for 19 and 2 degrees of freedomshould exceed 19. 4. Therefore, the lack-of-fit term, which is a measure of differencesin R, is not significant (i. e., all R's are equal).

11

REFERENCES

1. Felten, E. J.; Strongman, T. E.; and Ulion, N. E.: Coatings for DirectionalEutectics. (PWA-5091, Pratt & Whitney Aircraft; NAS3-16792.), NASA CR-134735, 1974.

2. Timbres, Donald H.; Nonis, L. F.; and Clegg, Maurice A.: Improvement of theOxidation Resistance of Dispersion Strengthened Nickel-Chromium Alloys.Sherritt Gordon Mines, Ltd. (AFML-TR-72-50; AD-748266), 1972.

3. Barrett, Charles A.; and Lowell, Carl E.: Comparison of Isothermal and CyclicOxidation Behavior of Twenty-Five Commercial Sheet Alloys at 1150° C. NASATN D-7615, 1974.

4. Johnson, Norman L.; and Leone, Fred C.: Statisitics and Experimental Design inEngineering and the Physical Sciences. Vol. II, John Wiley & Sons, Inc., 1964.

5. Lowell, Carl E.; and Drell, Isadore L.: High-Temperature X-Ray DiffractometerStudy of Oxidation of Two Superalloys, WI-52 and IN-100. NASA TM X-2002,1970.

6. Taylor, A.; and Floyd, R. W.: The Constitution of Nickel-Rich Alloys of theNickel-Chromium-Aluminum System. J. Inst. Metals, vol. 81, 1952-53, pp. 451-464.

7. Draper, Norman; and Smith, Harry: Applied Regression Analysis. John Wiley &Sons, 1966.

8. Graham, M. G.; and Hagy, H. E., eds.: Thermal Expansion - 1971. AmericanInst. Physics, 1972, pp. 87-95.

9. Metals Handbook. American Society for Metals, 1961, pp. 48-49.

10. Kingery, W. D.: Introduction to Ceramics. John Wiley & Sons, Inc., 1960, p. 483.

11. Douglass, D. L.: Exfoliation and the Mechanical Behavior of Scales. Oxidation ofMetals and Alloys. American Soc. Metals, 1971, pp. 137-156.

12. Tylecote, R. F.: The Adherence of Oxide Films on Metals - A Review of Informa-tion. J. Iron Steel Inst., vol. 195, pt. 4, Aug. 1960, pp. 380-385.

13. Latva, John D.: Selection and Fabrication of Nonmetallics - Oxides, Beryllides andSilicides. Materials Progr., vol. 82, no. 5, 1962, pp. 97-102.

14. Baldock, P. J.; Spindler, W. E.; and Baker, T. W.: An X-Ray Study of the Varia-tion of the Lattice Parameters of Alumina, Magnesia and Thoria up to 2000° C.AERE-R 5674, United Kingdom Atomic Energy Authority Research Group, 1968.

12

TABLE I. - CHEMICAL ANALYSIS OF AS-CAST MATERIALS

Alloy

1

2A2B345A5B67

8A8B

Base

Nickel

'

Chromiumcontent,

at. %

15.9811.5012.4413. 1918.4114.3516.8919. 1515.8118.8720.84

wt%

15.9612.0012.7712.7017.7414.8518. 1820.0114.7320.9820.81

Aluminumcontent,

at. %

17.5425.5822.7212.0711.0623.6529. 1924. 16

5.7726.9916.52

wt%

9.0913.8512.106.035.53

12.7016. 30.13. 102.79

14.908.56

Alloy

9

1011

1213

141516

1718

Base

NickelCobalt

Chromiumcontent,

at. %

9.7315.8611.5512.8518.7818.9310.4613.0715.9022.11

wt%

9.6215.8611.6812.3318. 1119.6110.3813.5414.7422. 12

Aluminumcontent,

at. %

17.1818.3020.9712.0511.5923.2518.2524.55

5.4116.95

w t %

8.819.50

11.006.005.80

12.509.04

13.202.608.80

13

I<

o

tuo§

m

O)01nt

PH

>> £0 £

<

O

6

CO.

-\

58

5

•g § s |rt ^_. ^ o"Q rt •% a

JillQ.

° = 1 §u .2 S "SC O O Crt r- js rtQJ *1] •*"" Or

5 O ', i X"=• O OJ

co .t_r

1 1 Krt co "g" §U y

g 1 U c

s 1 ° "i"rt S CM _ c

a. d ^

T-J C O J?u o o Srt T' *3 cu•a rt -jj g

i -^ "a Go c c o0 .S g '3S •- u So> 2d -" a.S » -3 £

§ ^r'w c

c3 ^Q- GX O

1H

"eti ra CM _ c

&

•g § s «"rt 3 S cu

-§ 1 f h

a

o "c a oO QJ C ---

•~< I-. CO 1G O QJ Crt ^ j3 SCU *S •*-• Qjj

2 'o CD

q

a qX OW u

S •£ ^ ,§-

tS « CM -^

a d J

o

0°°

J

1C

CXJ

D

O °

J

3l

D

0°°

s

1c>•fl

c

CD1C

>C

cXCC

cCs

c-irc^

c

<CN

C.tr

iZ

•>4

i* C4 1-

)

}4

> C

ir

cc\

sr

irc^

<,—

cCC

i2

) C"

* C

> CC

CC

CC

—

IT

ITV

•—

CC1

CCC

12

ioXO CD LCM O C

O

CD

oX

1oXOl CM *•

CO CO Ci-t r* r

O O CCO CO CCO CO CCM CM C

CM10

XO C - C - C O C D C D t - Ci - » i - H O O O - ^ T ' - ' C

o

CD1oT-H

XC O I > C O C M O d O Ct-H i-t »— 1 CM C M * - I C M C

O^H

Xt~ O ^ lO G5 rH CD ̂

co' c- co' CM as ai o cT - H i - H r - t C M r H ^ H C M C

i— 1 ^ CM if5 C O C M O ULO CD LO "̂ lO CD CO U

C M C M C M C M C M C S I C M C

O

- a s c o c M a j c o c D c o T r r - i ^

j co to co co ^ CD co to ai as

2 S 2 2 2 5 S g ^ £

^ ^ n ^ ^ ^ o o N - * ^

^ t— CO CD CO ^ ^ t~ ^ tr~ CCO ^H ^H CM CM CM CM "̂* CM CM C*

1 1 1 1 1 1 1 1 1 1

U U O U U O O O U U C .

i i i i r i i i r i

OD

OH

O CO

noM

PM

MM

r 01M oiM i-H

o00M

03

3sJ

» S14 '** OJ

5 <

14

TABLE III. - THERMAL EXPANSION OF Co-Cr-Al ALLOYS

Alloy,at. %

Co-16Cr-18AlCo-12Cr-26AlCo-19Cr-23AlCo-13Cr-25AlCo-22Cr-17AlAverage

Phase

a -Co

Latticeparameterat 25° C,LP25°C'

nm

O occc

.3564

.3563

.3574

.3566

.3571

Expansionconstant,

R

A

21.621.821.621.821.720.9

Mean co-efficient

of thermalexpansion,

CTE,oc-l

onvon1 fton

21

n-6

Standarddeviationof lattice

parameter,

°LP

-2

.20

.16

.20

.20

.15

/»

Latticeparameterat 25° C,LP25°C'

nm

0.2861.2861.2864.2861.2865

Expansionconstant,

R

17. 7xlO"4

17.717.818.017.917.8

Mean co-efficient

of thermalexpansion,

CTE,oc-l

17xlO"6

Standarddeviationof lattice

parameter,aLP

0. 05X10"2

.06

.06

.08

.06

A Nominal compositionsO Actual Ni-base compositions

* 20ro

c:-SJ

§ 150 IJ

E

£5 10

5

L

_ 0I

i

<

Q

— Qi

i

Q_O

o

O

° 0i

\ \ \

25

-- 20s

~ 15

10

!— A Nominal compositionsO Actual Co-base compositions

OA

-CA

,o

' 10 15 20 25 30Aluminum content, at %

Figure 1. - Distribution of projected and actual castingcompositions in Ni-Cr-AI system.

i 10 15 20 25 30Aluminum content, at. %

Figure 2. - Distribution of projected and actual castingcompositions in Co-Cr-Al system.

15

.365

.364

.363

e .362C

I .361Ero

a -360O

3 .359

.358

.357

.356i

O Measured values, healing

Calculated values

J I100 200 300 400 500 600 700 800 900 1000 1100 1200

Temperature, °C

Figure 3. - Typical thermal expansion curve for y/y1 in Ni-Cr-AI system(alloy Ni-18Cr-llAI). Standard deviation of lattice parameter, a. P,0.0010.

.3651—

.364 —

.363 —

.362 —

f .361Enj

1 -560^U

2 .359

.358

.357

.356

Measured values, heatingMeasured values, cooling

• Calculated values

0 100 200 300 400 500 600 700 800 900 1000 1100 1200Temperature, °C

Figure 4. - Typical thermal expansion curve for o-Co in Co-Cr-Al sys-tem (alloy Co-13Cr-12AI). Standard deviation of lattice parameter,OLP, 0.0007.

16

.293

.292

.291

.290

P .289

E .288

.287

.286

.285

Measured values, heatingMeasured values, cooling

Calculated values

i 100 200 300 400 500 600 700 800 900 1000 1100 1200Temperature, °C

Figure 5. - Typical thermal expansion curve for p in Ni-Cr-AI system(alloy Ni-19Cr-27AI). Standard deviation of lattice parameter, OIP,0.0011.

.292r-

.291

.290

£ .289

to

a .288<DO

ra-1 .287

.286

.285

Measured values, heatingMeasured values, cooling

• Calculated values

0 100 200 300 400 500 600 700 800 900 1000 1100 1200Temperature, °C

Figure 6. - Typical thermal expansion curve for p in Co-Cr-AI system(alloy Co-16Cr-18AI). Standard deviation of lattice parameter, OLP,0.0005.

17

.292

.291

.290

= .289i—O>

O>

| .288a<D

I •287

.286

.285

.284

O Measured values, heatingQ Measured values, cooling

Calculated values

I I I I I I I I I0 100 200 300 400 500 600 700 800 900 1000 1100 1200

Temperature, °C

Figure 7. - Curve fitting with limited data - p in Ni-Cr-AI system (alloyNi-17Cr-29AI). Standard deviation of lattice parameter, OLP, 0.0020.

.293j

.292

E -291

c

•I -290

IS. .2890>O

^ .288

.287

.286

O Measured values, heatingD Measured values, cooling

Calculated values

AO? I I I I I I0 100 200 300 400 500 600 700 800 900 1000 1100 1200

Temperature, °C

Figure & - Curve fitting to worst data set - p in Ni-Cr-AI system (alloyNi-19Cr-24AI). Standard deviation of lattice parameter, OLP, 0.0046.

18NASA-Langley, 1975 E-8323

NATIONAL AERONAUTICS AND SPACE ADMINISTRATION

WASHINGTON. D.C. 2O546

OFFICIAL BUSINESS

SPECIAL FOURTH-CLASS RATEBOOK

POSTAGE AND FEES PAID

NATIONAL AERONAUTICS AND

SPACE ADMINISTRATION

773 001 C1 U C 750725 S00903DSDEPT OF THE AIE FORCEAF WEAPONS LABORATORYATTN: TECHNICAL LIBPAPY (SOL)KIRTLAND AFB NM 87117

POSTMASTER If mil.Return

''The aeronautical and >/ :tics of the United States shall beconducted so as to contribute . . . t<j >:sion of human knowl-

mena in the atmosphere and space. The Admi.shall provide for ti .'c and appropriate disseminationof in fon- of."

—NATIONAL AERONAITICS AND SPACE ACT OF 1958

NASA SCIENTIFIC AND TECHNICAL PUBLICATIONSTECHNICAL REPORTS: Scientific andtechnical information considered important,complete, and a lasting contribution to existingknowledge.

TECHNICAL NOTES: Information less broadin scope but nevertheless of importance as acontribution to existing knowledge.

TECHNICAL MEMORANDUMS:Information receiving limited distributionbecause of preliminary data, security classifica-tion, or other reasons. Also includes conferenceproceedings with either l imited or unlimiteddistribution.

CONTRACTOR REPORTS: Scientific andtechnical information generated under a NASAcontract or grant and considered an importantcontribution to existing knowledge.

TECHNICAL TRANSLATIONS: Informationpublished in a foreign language consideredto merit NASA distribution in English.

SPECIAL PUBLICATIONS: Informationderived from or of value to NASA activities.Publications include final reports of majorprojects, monographs, data compilations,handbooks, sourcebooks, and specialbibliographies.

TECHNOLOGY UTILIZATIONPUBLICATIONS: Information on technologyused by NASA that may be of particularinterest in commercial and other non-aerospaceapplications. Publications include Tech Briefs,Technology Utilization Reports and

inology Sur\<.

Details on the availability of these publications may be obtained from:

SCIENTIFIC AND TECHNICAL INFORMATION OFFICE

N A T I O N A L A E R O N A U T I C S A N D S P A C E A D M I N I S T R A T I O N

Washington, D.C. 20546