FIh1L~E COPA .

"MTL MS 87-3 AD Ii-

SALUMINUM OXYNITRIDE SPINEL (ALON):00

, A REVIEW

4 DTICE-LECfTESEP 1 61987

CU

NORMAND D. CORBINMATERIALS CHARACTERIZATION DIVISION

p

July 1987

Approved for public release; distribution unlimited.

US ARMY"LABORATORY COMMANDMATERIALS TECHNOLOGY U.S. ARMY MATERIALS TECHNOLOGY LABORATORY-L\BORATORY Watertown, Massachusetts 02172.0001O 1 .

87

* o

The findings in this report aro not to be construed as an official

Department of the Army position, unless so designated oV other

authorized documents.

Mention of any, trade names or rnanufactuiers in this report

%hall not be construed as advertising nor as an official

or indlorsement or approval of such products or companies by

the United States Governme~nt.

Pr.

CiIS1POSITION I NSTR•UCTIONS

Testroy this report when not io be onscr needed

Do not return it to the a oginfator.

UNCLASSIFIED - ---"SECURtTY CLASSIFICATION or THIS PAGE (Ahon Date Fn.,ed.r

REPORT DOCUMENTATION PAGE READ INSTRUCTIONS__ BEFORE COMPLETING FORM

SREPORT NUM]IFP 2. GOVT ACCESSION NO. 3 RECIPIENT'S CATALOG NUMBER

MTL MS 87-3

4 TITLE (-nd S-btitle) S TYPE OF REPORT & PERIOD COVERED

ALtMINLYM OXYNITRIDE SPINEL (ALON): Fina G Report

A REVIEW ___.6 PERFORMING ORG. REPORT NUMBER

7 AuTHOR(S) 6 CONTRACT OR GRANT NUMBER(T)

Normand D. Corbin*

9 PERFORMING ORGANIZA! ION NAMF AND ADDRESS 10. PROGRAM ELEMENT. PROJECT. TASK

AREA & WORK UNIT NUMBERSU.S. Army Materials Technology LaboratoryWatertown, Massachusetts 02172-0001 D/A Project: 1L161102AH42SLCMT-O1fl' ---

TI. CONTROLLING OFFICE NAME AND ADDRESS 12. REPORT DATE

U.S. Army Laboratory Command July 19872800 Powder Mill Road 13. NUMBER OF PAGES

Adelphi, Maryland 20783-1145 1714 MONITORING AGENCY NAME & ADDRESS(of differeof trom Controlllng Office) 15. SECURITY CLASS. (of this report)

Unclassified

15a. DECL ASSI FIC ATION!DOWN GRADING

SCHEDULE

16 DISTRIBUTION STATEMENT (of thisg Report)

Approved for I)ublic release; distribution unlimited.

17 DI6STRIBUTION STATEMENT (of the ebstract entered In Block 20. it different from'n Report)

IS SUPPLEMENTARY NOTES

*Presently at Norton Company, High Performance Ceramics,

Northboro, Massachusetts 01532-1545

19 KEY WORDS (Co-Iinue on -- 01er3C . de if nec .,ery -nd identify by block nu..ber)

Spinel Transparent materialsNitrides Aluminum oxynitridesAluminum oxides SinteringCeramic materials Radomes

20 ABSTRACT (Continue on reverse side If nerese.,ry end ianId ify by block number)

(SEE REVERSE SIDE)

FORMDD I JAN 73 1473 EDITION OF I NOV 6 5 IS OBSOLETE UNCILASSIFIED

SECURITY CLASSIFICATION OF THIS PAGE (".en Date Entered)

UNCLASSIFIEDS(CuRiTY CLASSIFICAr[ON OF THIS PAGE "oen Oar.o tniv Oe

Block No. 20

ABSTRACT

Aluminum oxynitride spinel (ALON) is a relatively new ceramic that can

be processed into fully dense transparent material. There are a variety of

applications where this material could replace alumina, especially where

optical transparency and strength are important. This report reviews the

current status of this riaterial. including phase equilibria, processing, and

properties.

UNCLASSI FIED ___

SECURiTY CLASSiFICAYIGHN OW Tw,1 PAGE (^o~n [)at& Entrrtdj

<,

CONTENTS

Page

INTRO UCTION . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . 1

ALON HISTORY. . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . .. 1

A1N-AI 2 03 -PHASE RELATLONS....................... 2

ALON COMPOSITION. ............................... 3

ALON CRYSTAL STRUCTURE ....................... ........................... 4

ALON PROCESSING

Using Alumina ...... ............. ................ . 7

UsIng Alum num .............. . . . .... ... . . . 7

Using Gases ...... ...................................... 8

ALON PtzOPEPTiES

Optical aNd Dielectric . . . . . ........................ 8

Mechanical . . . . . . . . . .................... . . . iO

Thermal .......................................................... 1.

Oxidation and Chemical Resistance ..... ........ . . . . . . . . . . 10

ALON-Free Energy of Formation .. .. ......... 11

FUTURE ALON RESEARCH .. .. .. . . ... . .. ... . . . .. . ... 12

CONCLUSIONS ...................... ............................... ... 12

LITERATUM CITED. ............. . ................. .. 13

Accesion For- NTIS CRA&j

Q,,IALj DIC TABtSP CTED. ,Eri o,;nld

By ...........Di..t. Jb;t,tcov I

P .:i~tab:I~ty (,:0~C5

ViDIA V ., - -•I,,

INTROUUCTION

Polycrysalline aluminum oxynttride spinel (ALON, also referred to as Y-ALON)is a jelatively new ceramic that can be processed into fully dense transparent mate-rial, L as shown i, Figure 1. Since it has an isotropic structure, it may haveadvantages over polycrystailine alumina for applications where true optical trans-parency and high strength are required. Alumina cannot be processed into a fullytransparent material because of its anisotropic structure. This causes its refrac--tive index to vary as a function of direction through a crystal. In a polycrystal-line alumina body, the mismatch of the refractive indices at grain boundariesresults in scattering. This produces a material that is translucent and not trans-parent. Aluminum oxynitride spinel can also be thought of as a nitrogen stabilizedalumina sigce only 9.1 atomic percent nitrogen is required to stabilize the spinelstructure. Current research on this material is involved vith p~operty evaluationsto determine the potential applications for this material. '/,•

Figure 1. Transparent ALON plate (courtesy of Raytheon Company).

ALON HISTORY

It has been observed9 since 1946 that a spinel form of A120 3 can be stabilizedat high temperatures (>1000C)& Initially, it was believed this phase was stabi-lized by reduciqg the a uminum -12 (AIt to A1+

2 ) as when producing Fe3 04 spinelfrom Fe 2 O3 (Fetj to F ). In 1959 Yamaguchi reported that the ;pinel phase was

stabilized not by Al but by nitrogen. During the 19&Os, French investigat rsstuaied dif¶erjnt aspects of ?4ynitride materials i•1uding phase relations1 and

processing. Adams et al and Long and Foster also studied various aspectsof oxynitrides in the United States during this time. Not until interest wassparved in England on materials in the silicon-aluminum--oxygen-nitrogen chemical

V N1

v'v r~wy'~ >$\v~.~QN~..$>~-: ~ ~ .".v~ ~AN V-t0- *J ;>'

system "SIALONS"' 2 1 did aluminum oxynitride materials become of major interest. The

suzge of interest that followed focussed on obtaining a better uider.tanding of the

AiN-A1 2 0 3 phase diagram since it would give a better understanding of the St-Al-0-N

system. The first work reported on producing ALON spinel as a dense transparent ma-

terial was published by mcSuley and Cof 1979. Since then, repoits dealing

with phase equilibria, processing, , and properties4' have been pub-

lished.

AlN-AM2 03 -PHA$L RELAtIONS

In order to produce single phase ALON, it is important to understand the phase

rLlationships in the Al-0-N system. To date, all studies have been along the AIN

and Al 2 03 join. Experimental results have shown that 13 different phases may

occur. Table I lists these phaesaloggith their nomenclature, approximate corn-

positions, and investigators. 19'20 . .' 2

Table 1. ALUMINUMIN, OXYNITRIDE PHASES

'AIN : Polytypes of AIN i Spinel i

'21 21 20-H 27-R:160-H'. 21-R i 12n 8 A1203

o l-- sVe AIN 19 5 35.7: -21.0 16.7 7. 4-10 0

LA - Adam (1962) X i X

Long (1961) X I -114 IX X

I Henry (09691 -I--- -- --- ------ 2 xX----- .•

Nlichel (1972) X I Xi acke (176) x i / 4xI X I XX,

... . _ ._ . . . __ _

,McCauley (1981) I, XI / 'X XI :X

- z1llj----*-_x--i-- -_-i-

"_. I__... .dentified that a phase or phases existed over this composition range

The c monly termed "Aluminum Nitride Polytypes" are phases 2 ajd on the AINstructure. Since these phases are wmpositionally controlled,' they are not

true polytypes, as are SiC polytypes. The different AIN polytypes are listed in

Table 2 along with their compos iis, assuming that they are stable only at spe-

cific meta-to-nonmeta ratios, i allowing for charge neutrality and assuming nosilicon is in the structure. Most of the studies on these "polytype" phases are a

result of work in the Si-AI-0-N system. As a result, all reported information 3 n

these phases show silicon to be pzesent. Unreported data of Zangvil and Doser

have shown that several of these phases can be stabilized without silicon.

The comp9sition of the spinel phases Y-A ON and 9 '-ALON in Table 1 were taken

from the spinel model evaluation I SCauley. Phase Y'-ALON spinel was foWd upon

the oxidation of ALON by Goursat. ' Phase ý-ALON was observed by Michel anddetermined to be monoclinic. The O-ALON phase was observed by many investigators;

its reported composition ranges from 10-4 mole % AIN, depending on the researcher.

2

1 _t-

Table 2 AIN "POLYTYPE PHASES"

Ramsdell Mole % 1Phase Notation M X Fg-mula AIN

2-H I: 1 1"" 00

20-H 10: 11 A11003 N8 88.9

2-H8 >9: 10 - >87.527-R 9:10 A19 03 N7 87.5

16Ht 8 9 A18 03 N6 85.7

21-R 7: 8 A~1O3N5 83.3

12-H 6 7 A1603N4 80.0

• Bartrom only M CationtSakai only X Anion

The first phase diagram for this system was reported by Lejus. 14 Six phaseswere incorporated into the diagram, With a single phase designated "X" encompassingall the AIN polytype phases. The ALON solid solubility field was centered aromd25 mole %,2IN. Since this early work, 2 he diagram has been modified by Sakai,Gauckler, I and McCauley and Corbin. 1 For the most part all diagrams agree, witheach newer version becoming more complex as it improves the preceding work.

Figure 2 is the most recent diagram as reported by McCauley and Corbin. 2 4 Theyincorporated the vapor phase because of its importance to the stability of AIN. Thenitrogen-rich portion of the di ram agrees well with the temperature stabilityrP1e!tInsh1ps renprted by Sakai and the composition relationships rccordcd inTable 2. The composition of the ALON solid solubility field was determined to bemore nitrogen-rich than reported by Lejus (35.7 vs 25.0 mole % AIN). ALON was foundto melt congruently at 2165 0 C in one atmosphere of nitrogen. Since their work didnot focus on 6-ALON, it has not been entered on the diagram.

Dre3 5 ,36 37This phase diagram was thermodynamically evaluated by Dorner and Kaufman

using thermodynamic data and assuming the stable ALON composition was equivalent to

25 mole % AIN. Both determined that ALON spinel phase was the only stable interme-diate phase in this system. However, aue to their different assumptions, which willbe highlighted in the section on ALON-Free Energy of Formation, their results do not

agree. Dorner has ALON incongruently melting at 2000*C and only stable above 1600GC,while Kaufman has it melting congruently at 1940°C and stable at room temperature.Kaufman's results better agree with experimental data on the stability of ALON.

ALON COMPOSITION

T-e true limits of ALON solid solubility have yet to be accurately determinedby a detailed evaluation of the final reacted products. This is probably the majorreason for the discrepancies between investigators as shown in Table 3. A majorfinding upon which all recent investigators agree is that the solid solubility limit

does not include the ideal stoichiometric spinel composition Al 3 03 N (eg• 1:1, asin MgAl 2 0 4 , Fe 3 04 , FeAl 2 0 4 ). All models for this oxynitride spinel, ' asshown in Table 3, assume a constant anion lattice of nitrogen and oxygen with a de-crease in spinel cation vacancies with increasing nitrogen content.

3

30O0. 3000

2900 2900

2800 28&00

2700 2100

2600VAO 2600AIN + 27R +V

250 VAO 2502400 2400

'4_

L /H

2300 -+ 2300 z3

2200 - + 2165o+15UC - 200= I \ •• L + ALON I "

E 2100 20850C 2100

12HR + ALON

A + NAI20OL + a--_2low0 ALON (y) - 1900

1800 AIN + ALON ALON + a-AI20 3 _ 1800

1700 1700

27R 21R 12H AIN + a-A1203 ALON (y-)

100 90 80 70 60 50 40 30 20 10 0AIN mole % AIN A1203

Figure 2. Phase diagram for the A12 0 3 -AIN pseudo-binary (McCauley, 1981.

ALON CRM'JAL STRUCTURE

From the very early results of Yamaguchi,13 it was known, hat nitrogen would

stabilize a spinel form of A12 0 3 . Recently, Goursat, et a1323 have determined the

crystal structure of ALON by neutron diffraction to be (Fd3m) spinel. They con-

firmed that oxygen and nitrogen occupy the anion sites of a Epinel lattice. The

aluminum cations were found to occupy both octahedral and tetrahedral sites of thespinel structure. This spinel phase was found to be stable only when a d i sorderedvacancy is in the spinel octahedral position. Since a trivalent ion (Al+ ) is in

the tetrahedral positions of the spinel structure, ALON can be considered as an

4h_--

'Vr

Table 3. Y ALON COMPOSITIONS

Experimental RangeOxynltlde Spinel Formulas Mole %= AIN References

Al (64+x) IS] (8-Xil32.x Nx 40-27 McCauley (1978)

(0 !<x !ý-8)

Al (8.._+x) (l-x) 04-X NX 33-16 Lejus (1964)3 3

(WS x < 1)

Al (2+x) [ (3-4x. 0 (3-x) Nx 50-33 Adams (1962)12W0 !S x _< 3/4)

A170 9 N1 25 Kaufman (1979)Dormer (1981)

A13 03N 50 Yamaguchi (1959)'Henry (19691

A13 04 ,Misinterpreted Yamaguchi (1950iFifonenko (1957)Vert (1957)Is -Spinel cation vacancies

x • Limits from AI30 3N to A1203

inverse spinel. T6,ý14 2 • pares stotchiometric spinels to nonstoichiometricoxynitrLde sptniels."1' 32

The variation of lt•tice parameter with composition of this material has pre-viodcily baŽen L.4,.Jeu d aiu LdLgCeZ ±tLU 7.95.1 A (rnii.rogern rich) ro 7.938 A (oxygenrich). Figure 3 shows the variation of lattice parameter as a function of oxyge 9

content (determined by neutron activation analysis) for fully reacted specimens.

Table 4. SPINiL STRUCTURES

Stochiometric Nonstochlometric

Type Normal Inverse IOxynitrile° Y-Oxynitride Y-uxynitride Y-Alunilna

Formulat AB20 4 B(AB)0 4 BB20 3 N 33.3 mro AIN 35.7 r/ol AIN B8r304

Example MgAI2O4 ,,gFe20 4 A1303 N AI5C •N A123 027 N5 A1203

Anion Sites (32) 0-2 (32) 0-2 (24) 0-2 (27.44) 0-2 (27) 0-2 (32) 0-2

32 Total (8) N-3 (4.56) N-3 15i N-3

Cation Sites (16) AI+3 (8) Fe+ 3 (16) Al+ 3 (14.88) AI+ 3 (15) AI"+3 (13.3) AI" 3

16 (oct.) (8) , %+2 (1.12) M (1) [ 2 (2.7) E

8 (ten.) (8) Mg+2 (8) Fe+ 3 (8) AI+3 (8) A1t3 (8) AI+ 3 (8) AI+3

Reference Kingery (1q761 , Goursat1981) McCauley (1918) Lejus (1964)

*Not stabletA -"A+ : B 8 +3; 0 0 -2; N - N-3

5

7.950 Oxygen by neutron act!vdtioI 3nalysis

of fully rvduted spCdc[iC.

7.940

7 930

38 39 40 41 42 43 44Weight Percent Oxygen

Figure 3. ALON lattice parameter as a function of oxygen content.

ALON PROCESSING

Three basic methods have been used to produce aluminum oxynitride phases. The

most common of these involves the simultaneous reduction and nitridation of Al 2 0 3 .

A second method uses aluminum metal, and oxynitrides it by a combustion-type

reaction, Finally, these phases have also been produced by the reaction of gde5.

Table 5 lists the reactions used to produce these phases, along with references.

Table 5. ALUMINU, OXYNITRIDE SPINEL REACTIONS

- General Equatiow - Required' References

McCauley (19791 Harnett (198111 IA1203 (s! + AIN(s) -- T-ALON1(s) >16500 C Goursat (19701 Sakai (19/8. 1982)

I I Lejus (1964) Adanms (1962)

2 I +?Rafan'ello (1981)A 0+N(g -ALONs O . Isti-ShalOm (1980)r-- + S +_ _! Filonenko (1951)

3 A120 3( s) + C(s) + Air - '-ALON(s) + CO(g) a10u0C Vert (1957)__________ ___________________Yarnaguchi (19591

4 AI203(s) + Al(p I N2(9) - X-ALON(s] >1500(0)C Michel (1966)

5 At-3 + Al, Michel (19721A12031st + AIlS + Air - Y-ALON(s) Collongues (19671

6 Al203(Os) + NM 3(g) + H2(g) ----. Y-ALON(s) H2 0 ? '1650°C Collongues (1962)

A )+ Air Y-LN()O Michel (1972)SAI Air Y-ALON(s1 (Bourlannes (1974)

8 AICI3(g) + C02(g) + NH3(g) + N2(g) -9 -- 00C- Silvestri (1975)

Y-ALON(1 j + CO(1) 4 N2(g) + HCL~g9

6

aLU!i ny M uilina

It ha.; been reported by Colloagues17 that alutmini [ s;table in nitrogen unlessa reducing agent is present. In the preseiu-c of both a reducing agent and nitrogen,aluminum oxynitrides and aluminum nitride will form. ReducinglT agents that have beensuccesslul include aluminum, carbon, ammonia, and AIN (see Table 5, Eqs. 1--6).

The developmen. of vapor species may be an Important part of this reduction/nitridation process. It has be 9 0ported that the \,latility of alumina is muchgreater in reducing conditions. ' " The formation of vapor 4pecies also appears 4 So

have reported that, during the formation of alumlntum oxycarhides by alumina/carbon

reactions, vapor transport is important. It is antLicipated that the same followsfor oxynitride formation from alumina.

Reacting A2S 1 d A120 powder mixtures aboye 1650'C has been used to produce

ALON 21K reaction sintered samples, and reaction hot pressed spec-imensU. 'C , 4 t4, The major difficulty in this technique is to obtain A1M powderhaving high purity and a fine particle size. It has been shown that the character-istics of the AIN powder (especially un ý,acted Al, metal impurities, and particlesLze) are important for sintering AWN.

Rafaniello and Cutler 2 5 and Ish-Shalomn have produced ALON by heating mix-tures of A1 2 0 3 and carbon in nitrogen above 1700°C. This process has also beenused for producing AMN by adding enough carbon to remove all the oxygen in A1 2 03 .26

Early Jiu•les.. in which these mixtures were heated in air, also produced the spi nelpiiase. Lnis author has observed Lite Formation of the spinel phase on thesurface of u-A1 2 0 3 powder pellets when heated in an atmosphere of nitrogen whileexposed to the carbon heating elements above 1800'C. Thu%, carbon need not be indirect contact wtth the alumina in order to produce ALON.:c otb

Aluminum oxynitridy phases have also been produced by reactlng1 lylina and alu-mtnum mixtures in air and in nitrogen Collongues, et al ' producedALON and 6--ALON by reacting alumina with an ammonia plus hydrogen gas mixture above1650 0 C.

Vardelle and Besson4 7 havn produced a spinel phase by arc-plasma sprayinga-AI 2 0 3 in both hydrogen/nitrogen and hydrogen/argon plasma-gas mixtures. Thisauthor believes there Ls a good possibility that the material produced in the

plasma could be ALON.

The production of ALON from an alumina precursor is desirable because of thelow-cost, high-purity, fine-partlcle alumina powder readily available. The majorproblem with using alumina as a starting component is due to the high temperatures(>1650%C) required for synthesis.

Using A1 uwinuin--

Aluminum oxynitrides have been produced from aluminum metal by Michel andHluber 4 8 and Bouriannes.49,50 Michel noted that, at high temperatures, aluminumliquid can react with air to form various aluminum oxynitrides. Bouriannes formedALON by rapidly induction-heating aluminum spheres in air until combustion occurred(initiated'1300 0 C). The resulting phases were found to depend on the air pressureused (pressure > 70 Bar ALON, < 70 Bar = cx-A1093). This method of synthesis is il

7

many ways similar Lo a techniqule, termedk Self -Propagat tng hligh--Temperat tre Synthesis(SIIS), used to pvodlicý 1aI a rrety of refractory materials including borides, car-bides, and nitrides. $5

Uising Gases

Various 3lumiTIum oxytiltridc c~v-sitions have been stsccess ulty depositedby chemical depositioln techniques. Irenf- and Silvestri. were able to useReaction 8 in Tabb:_ 5 Lu prodilce 11thin polycrystalltine spinel layers on siliconsubStLates. The ALON phase was observed at 900 0C, while, at 770%C, a zeta-A 2 0 3-type structure was observed (possibly zeta-ALON). This work produces ALON spinelat the lowest temperatures.

ALON PROPLRTIES

Recently, several reports dealing, exclusiP y 7 w~th the properties of aluminumoxynitride spinel (ALON) have- been published.4 I I Data from these and earlierworks were used to produce Table 6. Currently, all property data available on ALONis on polycrystalline material or powders. Because of this, none of the data areintrinsic to this material. Essentially, all property data are extrinsic, limitedby such variables as grain size, residual porosity, and impurities. Future studieswith single crystals of ALON would be valuable in determining the intrinsic limita-tions of this new material.

Table 6. ALUMINUM.1 OXYNITRIDE SPINIL PROPER'IES

.OPTICAL IDIELICTRICII

19I Cut-Off 5.? jsm Hiifltntt (1m8) IRefractive Index 1.77 -. 1.80 Corbin (1981)

j Loss Tangent O_.(), (25o. 1l7 [jZ) Corbin (1991)Dielectric Constant 8.5 (250. 102' Hz) jCorbin (1981)Magnetic Susceptibility -0.34' x 10-6)9 (1300 Gauss.) .Yamaguchi (1959),

l-ir~nss 650 -1800 Kg/mm? P faniello (1981)Eta-tic Morlulus 47 0 psi I C.,bin (19811Poisson Ratio 0.'49 -. 0.263 Cortin 11981)

I Fract_-re Streng'.n 35 -50 -K19 psi ICorbin (1931)Fracturt: -vghness 1 5 2.9 MN/riii.) . Quinni (1984)

*THERMAL

I Thermal Shock (6Ti) 175 i 50C Quinn (1984)I Thermal Conductivity (RT) 1 0-030 cal (crn.S.K0)-l Hartnett (1984)I Thermal Expansion Coif 7.9x160 5- aai (19801

I5.23 x 1O660C (20020 Corb in (1981i

adD e Tr h Secimic Diffusivity I 0 00 to0 C2A037 cmI windo mateoui ;994

Melting PitnN? 2165 1 150C h~c.Caulr'. Ml.OxdtonRsstne Protective La3yer 12000C Cc~rtin j;1

Starts at 65n0C Coursai[Chemfical Attack _Stbl~eVersus Acids & Bases IYamnaju,, (V-y

Since AOisbeing cniedasa potential electromagneticwidwmtrlisoptica' and dielectric properties are of special interest. Hlartnett, et al

b,

8

have reported the inline transmission of a 2.4 mm thick specimen to have an infraredcutoff at 5.12 pm and an ultraviolet cutoff near 0.27 wm. They have also reportedthat samples >99% theoretical density show visible light transparency (Figure 1).The refractiye indices for ALON vary with composition from 1.770 (30% AIN) to 1.875(37.5% AIN). Figure 4 shows the power absorption coefficient of a 3.43 mm thickspecimen as a function of wavelength.59 In the mm-wavelength region, ALON has verylow absorption. The variation in the dielectric constant and loss tangent as afunction of frequency and temperature have previously been reported.7 Figures 5 and6 are reproduced from this work. The magnetic susceptibility was determine 1 3 by theGouy method and found to be -0.34 x 10- 6 /g with an applied field of 1300 G.

7.0

3.43k-f-n-thuck Specimen 7-3- 5-0F

Figure ,.ALON power absorption coefficient U.S. wave number.

1-.0 A-

35000 ExtrapoIated value at 1011 (--1 MM X)

9 0.1K0.000 Loss Tangent

3.o

. -

0 .01 -

0.001K

- -I I i iI_

102 103 104 1o. ]06 107 108 109 1010 11 1012

Frequency (1-12

Figure 5. Loss tangent vs frequency (Corbin, 1981).

9

14*O ALON (30 mlo AIN,

13-50CP

S12

S10 400P

io• 103 ]04 105 106 107

Frequency (Hzl

Figure 6. Dielectric constant vs frequency (Corbin, 1981).

Mechani calMhair n vale• of 1650-1800 Kg~mm2 were obtained when using a 100-g load

with a knoop indenter. , Quinn, et al have reported a variety of mechanicalproperties on polycrystalline material, including an e~astic modulus of 47.3 x 106psi, a Poisson ratio of 0.263, and a K ot Z.5 MN/m -°. Their four-point £ractutestrengths range from 44.4 x 103 psi at room temperature to 38.7 x 10 psi at1000*C. In most cases, a critical flaw (large grain or pore) limited the strengthof this material. All the data was obtained on relatively coarse-grained material(25-100 im).

Thermal

The the mal expansion coefficient of ALON varies from 5.23 x 10- 6 /.C (25-200-C)to 7.0 x 12- (20-980 0C) for reaction sintered materials, 8 while, for hot pressedspecimens,60 it is 7.59 X 10-6/ 0 C (25-10000C). The values for its room temperaturethgrpl conductivity, thermal diffusivity, and specific heat are listed in Table6. ' Results from water quench experiments have determined the critical thermalshock temperature (AT ) to be 175*C ; 50C. 7 ' 8 Tlis is somewhat lower than the valuedetermined for a-Al 2 03 in the same study. Quinn has proposed this is a result ofthe differences in the termal conductivities between these two materials (ALON/n-Al2 03 = 1/3).

Oxidation and Chemical Resistance

There is some disagree ent on tne stability of ALON in an oxidizing environ-. ment. Corbin and McCauley, testing bulk samples in air, 6 ound the oxidation to

produce a protective oxide layer up to '1200*C. Billy, however, found theoxidation of ALON to start as low as b650C in oxygen (p - 32 torr) and not producea protective layer. It is reasonable to expect that the partial pressure of oxygenand nitrogen plays an important role on the stability of ALON at elevated temper-atures. For example, in air, PN2 0.7, while in the oxygen environment it is anorder of magnitude less. N1

10

Work by Goursat 3 2 ' 3 3 investigated the oxydation kinetics of ALON and the for-mation of a new phase, y'-ALON, during the oxidation process. Figure 7 is a thermo-gravimetric analysis (TGA) trace taken from their work, showing the oxidationbehavior of a 20-50 Pm powder. Oxidation starts near 650'C and reaches a maximumweight gain at 1150*C. Heating beyond this temperature reduces the total sampleweight gain. They attribute this to the oxidation of ALON below 1150C with no lossof nitrogen. This results in the formation of the oxygei.-rich phase Y'-ALON. Above1150%C, oxidation occurs with the loss of nitrogen resulting in CC-Al 2 0 3. They alsofound that when a nitrogen-rich ALON composition is used as starting material, theY'-ALON phase is not observed.

8-

7-

K Figure 7. Thermogravametric analysis_ (TGA) of -ALON in oxygen (32 torr)

3 3 (Goursat, 1976).

Ift 2 -

600 800 1000 120 1400TOC

Resistane oftis material to other forms of chemical attack have also beeninvestigated. - 3 The results show that ALON is resistant to various acids,bases, and HO.

ALON-Free Energy of Formation

The ree energy of formatioa for ALON can be evaluated from the data of Dorner,et al 3 5 ,3 and Kaufman. Dorner tabulated enthalpies and entropies of formationfor ALON as a functlon of temperature. He determined these values by using theeutectJ~d decomposition of ALON into a-A12 0 3 and AIN at 1600*C, as determined byLejus, and assuming a small positive value (I Joule/!K-gram atom) for the entropyof formation from c-Al 2 0 3 and AIN. Kaufman listed the free energy of reaction asa function of temperature for producing ALON by reacting a-Al 2 0 3 and AIN. He calcu-lated his values by using lattice stability information along with solution andcompound phase parameters which he selected from observed thermochemical propertiesand phase diagrams. Both researchers assumed the composition for ALON to be at25 mole % AIN (A1 7 0 9 N). This composition 2ctually lies outside the experimentallydetermined solid solubility field of ALON. 2 4

Table 7 was determinq by using their data along with thermochemical informa-tion in the JANAF tables. Under column (L), the differences between the freeenergy of formations when using Dorner vs Kaufman data are listed. Even thoughthese differences appear small, they are extremely influential in evaluating thestability of ALON below 1600'C. When using these values to determine the stabilityof ALON below 16000 C, Dorner's data projects ALON as unstable, whereas Kaufman'sshow it to be stable at room temperature. As previously discussed, the predictionsof Kaufman are in better agreement with the experimental data.

I]i

Table 7- y-ALON" FRFE ENERGY OFFORMATION (A•C0

Kcal moleI-

T0K TIC Dorner Kautman A

170U 1427 -841.905 -843.242 -1.337

190 1627 -791.497 -791-667 -0.170

2100 1827 -741.090 -740.455 +0.635

2300 2027 -690.83 -89.568 + 1.115

2500 2227 -640.216 -638.998 + 1278

'7 AI(j, + 912 02 + 112 N2 • AIO9N (25 mlo AIN)

FUTURE ALON RESEARCH

Since aluminum oxynitride spinel is a relatively new material, very little in-

formation on its intrinsic properties is available. In order to fully utilize andfind new application for this material, a series of basic research studies would be

in order for the future. The following lisc highlights some critical areas for

future ALON research:

- An experimental evaluation of ALON's thermodynamic properties,especially to determine its stability in a variety or differentchemical environments.

- A determination of the oxygen, nitrogen, and aluminum diffusion Icoefficients in ALON, as a function of temperature and pressure.

- An evaluation of the effects of impurities and dopants on thesintering behavior of ALON.

- Studies involved with producing high purity ALON powders.

CONCLUSIONS

Aluminum oxynitLride spinel (ALON) 18 a new material that has the potential to

replace O--Al 2 0 3 in a variety of applications where optical transparency, highstrength, and isotropic properties are important. This material has been sintered

into fully dense, transparent material by a variety of researchers using differenttechniques. Current property evaluations of this material show it to be optically

transparent from 5.12 to 0.27 pm, have high strength (4 point - 44.4 x 10 Ypsi), and

oxidation-resistant in air to 1200 0 C Future studies should evaluate the stability

of this material in a variety of environments to open up new areas where ALON can beutilized.

12

LITERATURL CITED)

I. 11Y.A I ý. an C 1( ORBIN. N.D I) hine Relation an(I Reac t'J~n Sjinfering of 7ranspaurcnt Cubici Al/umininu Oxi nirul, Sinnel(. I I 0A I I Ame r '. 'r Soc-v 62. no. 9-10, 1979. p. 4776.

2. lIAR I \ I11. 1 MN .t \ ;UI RI 1- 11 A- G TI I. MAN, R. L.. C'ORBIN. ND..) and McCAU1IA LY. W A!/uniniiin Vý ni inri(Ie Spinlet(I11.0\) A \c 0 Vpm at andl Mtinoncude l~ inotd ii'alaernal ('et. Eng. atid Sci. Pro,,:., v- 3. 1n0, 1-2. I982. 1P (,

3. S\KAI- 1. liii sO vioitrrii/s inll Sth stseil -1-/A .150 Sittten.ng - Tlteorv aol Practice, Niate:nal Scicnce Monogr-aphs.Kolir. Pcjo~i k. and Rdi~t- ed. I sevier Sc:ientific PUblishi-ng'Co. v. 1 4. 1982, p- 591

4 lIARINIT. Ti r , mid ( ,I-\TISI. AN. R L Opncat and ,¶ec/ieuanwa Prnpcr-ties of Highly Tranqwvrnt Si-jonel amd AEONA DormesTo he publish1 j in Pron. SI'lII Arnnual MeeLting, 1984.

5 R1101)1S. A K.. SISLLLRS. DTJ. anid VASILOS, T. ttoi-IWnrking of 4tiontnion Oxidle 11. Optical Properties. J. Arner. (er. Soc..v. 58. 11975. P. 31.

6- MCAULEY. J.W- A Sinqplu .1/ode ! or Alumoinium Ovitnitridle S5pine/s J. Anwr. Cc~r Soc., v- 6 1, ito. 7-8, I1978, p. 372.7. CORBIN. NI) . and \IcCAUI IXY. I .W. irrogen-.Niabiizcd ,l/uninwon Oxide Srinet (ALOA). Pro%:. SPII: Annual Meeting (1981),

%, -")l, Eiicr-ging Optical Maiteutals. 1992. 1P. 21.8. QHlNN (.I. (-QRB; N. N.DI.. mnd McCAUlIY. I- J.W. Thciernoniec/ianiaa Properties (if A/oinniuon Oxvmntride Sinnet Amine Cer.

SOC Bull. v. 6(3. 110. 5. I984. P). 723.-9 X AM ACI.'CI GI . Re -,ra(n:it P / ower (fteLt .r t'i/int AJlume~itnu Ion (Al" or A /+') in the (.r-tsal. Bull. (hem. .'oc. Japian.

\. 23. 1950. 1) 89.90.10. Y5AM A GU LCI . G. J. cc. Avo.c. .aaltmi uYogyo-Kyolzi-Shi., v. 55. i947. P. 42.

II -ILONIKO N.E. LAVROV., IN. ANL)RFLVA. O.V., and I tLVINER, RUL. Ahiitiinont Spinel, A10 A1203 Acid ofn 01 Sincsof USSR. Proc. Cnemi. Tech. Sec.. . 115. 1957. p. 41.

12. V!iRT. Z.L.. KAMLI.N SIX. MV.. K (I:RY.XVT ISEV. V-1., ard SOKIIOR. MI. Redttlio'[I; of Ill ,V3 hj Carbon Academy of' Sciencesof USSR. Proc. Cimein. Tech. Sect , % 1 10, 1957. p. 75.

13- Y'AM\AGUCIII. G' andl NYA.NA( 'ID)A. It. Study onl t/iC Reductive Sp.inet - A Vew Spitiel Porniota A/IN-A/9-05 Inistead of the hereeolisOne 11?04 Chemit. Soc. of Japan Boll.. v. 32. no. ! , 1959. p. 1264.

14. E-LIUS. AM.%I Sur /a 1-urination a flitau Temnperattire de Spinet/es tion Swoec/nomnetriqnes et de Phase, Deriiees. Rev. liautes Temnpur.cit Retract. v 1. 1004. p- 53.

17 MICHI:' L. ID.. Ph RLZ y JORBA. MI.. and (OLI.ONGUES. R Ox)nuitrce IFormtiiion During Alumina Fusion in Air in Pie Presence of,1/iioliliiooi C. R. Acad. Sci.. SCFr. v. 263. 1966, p. 1366.

lo. (O0ILLON( U'I I- . -(': L LES. J C.. [-1-1 US, AM..1 PE REZ )JO0RBA, M.i, arid s'ICH EL.. D. Rec/terchie.s sur le.; Oxynitenures illetal/iques.Mat. Res Boll.. v. 2. 1907. p. 837.

17. (COLLONGULS, R.. COLIN. F'.. THFIRY. J-, MICHIEL, D., andi PEREZ y IORBA. NI. Redoction and ,Vitridairon Reactionis inAulol;u~i-Baieci (ei-aoiics Bull. Soc. Fr. Cer..v 77. 1 96- ). p 1. I

1i k.- jI- t.Urd, LAt '). K . J,11.:. I( .. and 1-iU.AM. Action (le / Amnnoltta stir Differents 0,j'e/es Super-Refractaires. Soc. Cluiniquetic France Bull., 1962. p. 2113.

91. ADAMS. I.. AuCOIN, IR. Tand WOLFF. G;.A. lhitonii'sceitie :ni the Sistleit 41pO 3 -AIA' J. Electfo ChIem. Soc.. v. 109. 1962, p. 1050-20 LONG;. G;.. anti FOSTER. L.M. Cetsrtat Phases ill the Si steint AH -t. I. Amr03r-o..v 2,16 .25

'I. JACK. K Ii- Re'tieo, Sia/Ons and Relaiead-.Vii'ogen Ceramiics J. Mat. Set.. v'. 11, 19)76. p. 1135.2 2. SAKA.I, T. ltor.Preesing iof [lie -1/N-AiO1? S -vi'ein J. Ccr. Assoc. Japiatn (Yoggo-Kyokau-Slii). v. 86. 1978. p. 125.23. (CAUCK l.!.L.J ,attd PIATZOW. G. Repriesenatonit of A/tu/ticoliptpitent Silitont Nitride-Based SiVsenins Nitrogen Ceramics,

1: L. Riley. ed.. Nniirdhloff Publishing Co.. The Nethltelands, 1977, p 4124. M~cC.AULI-Y. J.W.. arid CORBIN, N-D. ttigit 7enipraoiwe Reactiric' attd .1/i-rosuititciroes itt tile .'/23-l.\ ytem. Progress itt

Nit rogett Cet attits. F.IU Riley. ed.. Mlartittos NijhtofI Publisher-, 1983. p. Ill-25. RA E-ANII'LLO, W., and CUTLE R. 1.B. Prep~aration of.Sintrcahte Cubi--,litoni~oin Ovetnijride bi the Car/jnt/ertnal Nitridririon of

A/tiotinutn Oxide CottuttL~ictmtatinn of thle Anicr. Cer. Soc., v. 54, 1981, p. C-I 28.26. ISH1-SlIALOM. MI. I-ortitation of ,11tomiion oxt snitride by Carbot/terinta Redutinttottc A/utntttutn Oxide- illt ireogeut. J . Mlat. Sci.

Letters, v. 1. 1982. P) 147.27. HIENRY. J.H., RUSSEL, J.J., arid KELLY. l1.J. The Systeot At4 C'3-Al/A-A12O3. U.S. Department of the Inte.i~or, Bureau of Mines

Repomt No. 7320. Novemiber 1969.28. BARTPAMI. S.F., and SLACK, G.A. Allj 6 N8 0 3 andi 1119,'703. Tweo ewRep~eated-La yir Stmntrctiie lin r/ic AfiN-A!S)0 Sy~stem.

%- B-35, 1979, p. 2281-29. JACK. K.11 The Relauions/tip of Phase Diagramn.s to R-.search and tDrre/opioent of Sta/on~s Phase Diagrarms Matcnals Science and

Technology. A.M. Alper, ed., Academic Press. Ness York, v. V, 1978, P). 241.3G. KINGERY. W D.) BOWEIN, IlK., and UFILMANN, l).R. lnit)I tewiont to (eramti.s 211L Ld.. J0o1n Wiley & Sons. Itic., 19 76, 1 81.31. ZANGVIL. A.. and DOSI:R, R.W. Tranmttission Electron iuicroscoptv of A/-)O/.-A/V Stuoiplte Final Report. ANMMRC Contract

No. DAAG46-81I-M- 1377.32. GOL'RSAT. P.. BILLY. M.-, COFURIOT. P.. I-ABBE. I.C.. VILLECHIINOUX, JM.ROULI. , and BARI)OLLF., J. (onttrihuitiemn

aI Iftude du S'.s:en'e AI/V,/N /Ii Retenttiont d..lzrte dans les l'rwoehtip ii Vs tiitrtire d .1littioiiui -Y Mat. Cltctn.. v. 6. 1981. p. S1.33. GOURSAT. P., COLURIOT. P., and BILLY, NM. Con~trjbiittin a IT'tdid? (lii ,S.1 vnie At/UI/% L. Reactirite di 'V"Oxtnitrure d 'Atutnittutn

'y Mat. Chtein.. ,. 1. 1976. p. 131.34. MICHf:L. D. Cownuiri/iiu a / Etude de PI'eloonieiies d/" 0dtosancenotntt fIe L)ejaiut~s itani. (tee Aonoeri-iaux die Afatenaii.t Refra:taires a

B'ase (! tutttinin cl cle Zircone Rcv, lnt. lajuttes I etnp~r. et Rd hac.. v. 9, I1972. 1) 225.35. DORNER. P., GAUCK LER. [ .,. KRIEC'(, 11l., LUKAS, HLL., PF.TZOW, G.. arid WEISS, J. Ott tie Calculationt aiid Rep~resentutattn of

,flluticoot/onetlt S ivste'ms CALIPIIAD. v~. 3. 1979 7)p. 241.36. 1)ORNER. P.. G.XU(KLER. I 1. KRIEC '. 11., LUKAS, HI.L.. PIITZOW, G., arid WEISS. J. Calculatioun of Iletet-ogetieot6 Phase lEqui-

/ihria int the SIAIONV Svuetnz J. Mlat. Sci.. v. 16. 1981. p. 935.37. KA UFMAN., L. (.alcidawion or., Quasi/idiocy atid Quesiternaty Vvni~titidc Sysium's - 1ll. PreseNtd at CALPiIAD ViII, Stockhtolmti

Sw; dctt, 1979.38. KINGFIRY, W). D- OWEN, uI.K ., in'l UILMANN. D)R. Infeodlieno to) Cerantics, 2tnd d.. John Wiley & Sons, Inc., 1976, p,. 62.39. BURNS, F-. and CORBIN. NBD. D~ata obtainted at AMNIRC40. BREWER. L., aatd SEARCY, A.W. 7/ti (;aseotu Spiecie~s of the A/-Al-03 Si'stelnt J. Anter. Cliem.- Soc.,.v. 73, 1951, p). 5308.

41. YANAGIDA. If., and KROGE:R, FLA. The Svse'em Al-C' I- Aawr. (Cr. Soc.. v. 51. 1,,, 12. 1968. 1) 700-42. CORBIIN, N.D. The' In i luencc3 oj (-arh)n, Nitrop-n. ,and Argon oil A4luminumn Os i-etriele .$peeil I'oro inem %tasters 'Iliesis. Massa-

cht set Is I osti tutc of lTvcliicolo,ýy I)epar meuit ofI Matc oak Scic 1ce andlL nC ieg Jime IQA98243- PASCO, WIN)., and [JORI NIUS. B.H. Cur,Norherenic Ree ciiap o)] Atluminea. Pali U. 7Jee-rtooeheeeistre (;ci1rai Flecticc C'o.. lcchlin-

cal Report No. 82CRD1)1,I May 1982.44. KIEFFER. R . WRUSS. W. , od WILLER, B. Praprietus Pi'esiquees ci -Iecaniclucs detCc (er'ooqu- 4AI-A 120. Obtetoins par LaoepreN

* ,slon a C/laud Rev, lot. If Lutes. Temper. ct Retract_. v. 13, 1 970. p)945. TURPI N-LAUNAY. D)., Ti EVLN0l. F., Ml)EVOYI.. 1-.. and 130Cl1. P. keaire Jleet-pressoeg of y-YAheeooiieloe Oxý oinadk (Crainni

Pcwders. Vincenzini. :d.. Flsevier PUbliShing Co., A~nisteredaiec 11)83 p). 891.46- CORBIN. N.D., ande McCAULE-1Y, J W. Further Studie's oi ALO.\ and Ollicr Maiteriais in the Al\AI ',03 Sesiene. Anice. Cc:i So,

Bull.. v 59. 1980. p. 373.47. VARDELLLE, MI., and BF 'SSON. J.LI. tliomnina Obtained by Art Plasma Sprujiog /I Srwhd of thle Optimiezation of Sprauectg (enieetiops.

Cer Int., v. 7 . 981, p. 48.48. MICHIIEL. D., and HUIBfER. Nt. Etude %ur lonocoistaux de I Ordopineamemeoeee dt-5 Vejezes dines l'Ox iatrurL d*Ahoiaeeoeo -f Rev. 'hit.

Hastes. Temper. -.:t RetraLct - v 7, l910, P). 145.49. BOý.URI ANNE -S, R. Etadv3 IExperimeii'ieik de la ('onilestjoni de l'Aleawnonee dones hs Mlelainges OxygeneArgone. dawe "I Azote et duaps

Tenieraeer Rev cethses Aatt Soriet:r etela Refact Prelen 10,en Mat7rial p.r 1g andSci

I0 B~RENE. l.A. R.I.IHARDY, A-). adi~ MANONIO E N .CmiSaine PropertAluie um Sphereeav Vapor Idepsitedri oHines Total~ eN,Ceja. R Jc .Ele ctroni Ca., %. 2 4, 1975. p. 409.

541. CRIDVESIR, V.J.,Sl I-Ropagt, l.A., Temperatur S., and hss A P S oSJ.D C'ierialoVaor rduc ng jtio~ fain: 4tera.N~ Films J. aElecticMatc.. v. 4, no-75,1p. 4929 . 19

52. McAREK ,F.. .1-W. O, COBI, N-. and ZYKOV, TAM..PrpaandiONGe cef AtonuinaeOpideu Piteeprattnete and Seleýasei ofe Patrkial. i MljetU Lercniregrd, v 57. l~ng. p.d 1450. , -3 o 91.SptibiO:tbr182 )553. DREYER. L.,AR. E 1. KOLASKA.RI V.J, and REITERU , N.. Somena Proetiest No. ,8 Ceua1,5Va4. Juel);ie 4,1980.110 N.

57. co J.NKA I'eiro ndc YMaMt.,vO, 1975,apans Pa4t0o91,8.Api.,9958. SILVSIIIZAWA, ., aRnd , SIOA., ZI. apneSe . actnt No. 3S,9. Januar 3heia Vao1980.o fAlX0--1-isJ Eetoi

59MIRHa JRt Dt.o, vCsupie-atralcore, t aioa PyiclLaatr, cpi-on idl42,9ntd igdm560. SAkKAI F. DEIIV., and WATANABE A. Theraeparasion fn ofer Aluminimerides Nird fin the' Gas~e AhasA 123 bli er. Soc.KBlln.. .

Cc. 2, 198, p. 5? 1(-85. p4056. SAKAI. '1., K.' GRIYAMA If., KLAS~KAl, T., and KEITERA N. Gfeema Paente No.eu 2,851,584, June 4he 1980.n ae heTeeoalCn

57. tvtyc I A A/N Pa- ersand YAA .TO Y. Aso.Japan.se(PateotKNo.a41,282, Apri 86. 197o. 4 98.462. BISIlYAW, Z., aindtco as-cudA I.Jaatinese Nitroen Ceoamics.25. Jan iley, 31 ,NodofPbihngC. ~cNtiia, 1977,

63. SAKUI, TM. Praantd pAArAE R.Teamlctpnion af S~a ut tPinctaere Prairietes inte SyOvenitA'r\-Ae03 dAneeiicner Socit Builel., v.5eFranc 8, 1ll. .1962,.21.

64. citt JANAF T Per oelvemieal Tabes 2nd Asdo. Dr.apland, i Progopheta-i, Project Dir 4,tors 8,Wapei.gtu74 .. ,1762 ILM ieiso I-oi ecos.Ntoe eais .. Rly dNodhf ulsigC. h elVlns 97p.23

Frnc ul..162 p 1.364 AVFTemceia abe,2dE. r tl n H rpePoetDrc.SW.IiitrDC,17

I&I-

FM

L J~

a' a,

I., >A ~n

t' a' EA o Ia

E. CLE >,C E"

a-. :3

0- a,.' a'- a.'

3> L 0 >CA0c LLw a' LL'-r a

432f a'- I -

Q Cu L -C u

(x F- a> I-- Ec - a>

.'C-~ ~~~~~ o) a'- Lm.'-- -0 a O D'005 C'- - C aQ 0'-

Lý 0c-'a --- L I Aa

OVA. -o 0-AO-.A' -'kL

fir--C~~c 41 dC,'-~'- C La'--- 3" c.'o I ' lo--Aa~a

c. o L

a'a

COL ~ ~~a OLcW-. C S

-LA o OC -- f

rc -L a' >,. caS -' aOau w a- E--CCIE.f

-J o a'- :3a'~~~~~~.,-a. 4 ' S C 42"

-~ CC' C

-o r- o - o

'IL' 1.am -ý

L a' L. oa CDLA

I~~~~c w >, 0 fl- 0 f-

rE -L 0 rj Z

mQy v. o A1o

lC mV c) a,

:m~C -c c .'o QAoW 0u EW3- aI