Creep and Microstructural Evolution at High Temperature of Liquid-Phase-Sintered Silicon Carbide

Juan J. Melendez-Martınez,z Miguel Castillo-Rodrıguez, and Arturo Domınguez-Rodrıguez**

Departamento de Fısica de la Materia Condensada, Universidad de Sevilla, 41081 Sevilla, Spain

Angel L. Ortiz,*,w and Fernando Guiberteau

Departamento de Electronica e Ingenierıa Electromecanica, Universidad de Extremadura, 06071 Badajoz, Spain

The compressive creep characteristics at 16251C of liquid-phase-sintered silicon carbide ceramics containing 5 and 15wt% of crystalline Y3Al5O12 (YAG) as the secondary phasewere studied. In the two cases, strains between 10% and 15%were reached without failure. The creep behavior was charac-terized by a stress exponent n�2, and the proportion of sec-ondary phase was related to the creep resistance of thematerials. The microstructural evolution during creep consistedfirstly in the re-distribution of the secondary phase, probably asa consequence of its viscous flow at the creep conditions, andsecondly an extensive nucleation and growth of cavities, whichwas more important for the highest YAG content. The latterreflects the carbothermal reduction that the secondary phaseundergoes during creep.

I. Introduction

DURING the last decade, non-oxide ceramics have taken onan unquestionably important position. Silicon nitride is a

good example of this.1 But silicon carbide (SiC) has also recentlybeen shown to exhibit exceptional features for structural appli-cations. The set of properties of SiC includes great hardness andstrength with moderate toughness,2 high resistance againstaggressive chemical attack,3 high resistance to oxidation andwear,4,5 high thermal conductivity and thermal-shock resist-ance,6 and excellent strength retention at high temperatures.7

These make SiC-based materials suitable for such industrialapplications as muffle furnace linings, kiln parts, and especiallyas parts of high-temperature heat exchangers.8

The densification of SiC ceramics is generally achieved by li-quid-phase sintering with the help of oxide additives.2 Theadditives promote the formation of a liquid phase at relativelylow temperatures (17501–20001C), and densification occurs byliquid rearrangement and solution-precipitation. After cooling,the liquid remains as a secondary phase, located at multigrainjunctions and as intergranular films, which affects the mechan-ical properties of the resulting material. For instance, liquid-phase-sintered (LPS) SiC ceramics processed with mixtures ofAlN, Y2O3, and RE2O3 (RE5 rare earth) as sintering aids havebeen shown to possess excellent fracture toughness, creep resist-ance, and strength at high temperatures when compared withsolid-state-sintered (SSS) SiC ceramics.7,9 One of the most widely

studied additive systems is Y2O3–Al2O3, especially when thesecomponents are chosen in the ratio 3:5 to form yttrium–alumi-num garnet (YAG, Y3Al5O12)

10,11: the presence of crystallineYAG has been demonstrated to improve the fracture toughnessat room temperature of LPS SiC.2

The high-temperature compressive creep of SiC has been thefocus of several works before,12–24 although it has yet to be sys-tematically studied. In particular, to the best of the authors’knowledge, the correlation between creep behavior and micro-structural evolution in LPS SiC containing crystalline YAG asthe secondary phase has only been studied once before, showingthe effect of sintering time.24 As a complement to that study, thepresent work was aimed at shedding light on that correlation bystudying two LPS SiC ceramics containing different proportionsof crystalline YAG, both processed under the same heat-treat-ment conditions. In particular, it will be shown that, eventhough their microstructure is greatly degraded during creepdue to the evolution of the secondary phase, the materialsexhibit interesting characteristics for high-temperature applica-tions.

II. Experimental Procedure

The starting powders were commercial, ultrafine powders of a-SiC (UF-15, H. C. Starck, Berlin, Goslar, Germany), Al2O3

(AKP-30, Sumitomo Chemical Company, New York, NY), andY2O3 (Fine Grade, H. C. Starck). These powders were combinedin the proportions given in Table I, resulting in a-SiC compos-ites with 5 and 15 wt% Y3Al5O12 (YAG; i.e., 3.6 and 11.1 vol%,respectively). The powder mixtures were then ball milled inethanol for 24 h using zirconia balls media and polyethylenebottles. The resulting slurries were subsequently dried on hotplates while being stirred. The dried powders were next crushedand sieved to remove the hard agglomerates. Compacts weremade from each of the powder batches by uniaxial compactionin graphite dies at 50 MPa, followed by isostatic compaction at350 MPa in water within latex globes. The pellets were thenplaced in graphite crucibles containing coarse Al2O3 and SiCpowders (used as powder beds to prevent mass loss during sin-tering), and subsequently pressureless sintered in a graphite fur-nace. The sintering conditions were: a peak temperature of19501C, heating and cooling rates of 6001C/h, a hold time atpeak temperature of 2 h, and a flowing argon atmosphere. Thesintered materials were ground down 1 mm to remove the ad-hered powder bed layer, and were then rinsed clean with acetoneand ethanol. The indentation hardness (HV) and toughness(KIC) of the two LPS SiC ceramics at room temperature weremeasured by Vickers tests at 98 N load (10 separate indentationsper each material), and the obtained values are given in Table I;elastic modulus (E) values determined using Hertzian indenta-tions were used in the KIC calculations.

The mechanical behavior of the processed materials was in-vestigated by uniaxial compression creep tests on samples cut

S. Wiederhorn—contributing editor

This work was supported by the Ministerio de Ciencia y Tecnologıa (Government ofSpain) and the Fondo Europeo de Desarrollo Regional (FEDER) under grant Nos. CICYTMAT 2004-05971 and UNEX00-23-013.

*Member, American Ceramic Society.**Fellow, American Ceramic Society.wAuthor to whom correspondence should be addressed. e-mail: alortiz@unex.eszPermanent address: Departamento de Fısica, Universidad de Extremadura, Avda. de

Elvas s/n, 06071 Badajoz, Spain.

Manuscript No. 21409. Received January 19, 2006; approved July 24, 2006.

Journal

J. Am. Ceram. Soc., 90 [1] 163–169 (2007)

DOI: 10.1111/j.1551-2916.2006.01350.x

r 2006 The American Ceramic Society

163

and ground as parallelepipeds of approximate dimensions 5 mm� 3 mm � 3 mm. The tests were performed on a prototypedead-load creep machine fitted with SiC push rods and a hemi-spherical furnace. The temperature was continuously monitoredthroughout the test (within711C) with a W-10% Re/W-26%Re thermocouple placed near the sample. The instantaneouslength of the sample was recorded (within72 mm) by a linear-variable differential transformer (LVDT) located outside thefurnace. Additional details of the experimental device have beengiven elsewhere.25 The tests were carried out at 16251C, atstresses ranging between 40 and 90 MPa, approximately, andwere stopped without failure of the samples. To avoid the oxi-dation of the samples, all the tests were performed in a con-trolled argon atmosphere.

The microstructures of the undeformed and crept sampleswere examined by scanning electron microscopy (SEM; ModelXL30, Philips Research Laboratories, Eindhoven, the Nether-lands) using both secondary (SE) and back-scattered electrons(BSE). For the undeformed samples, the observation surfaceswere chosen arbitrarily; for crept samples, they were chosenparallel to the compression axis. In all cases, the observationsurfaces were first ground and then polished with diamond pasteto a 3 mm finish. Subsequently, they were plasma etched using amixture of CF4 and O2 in the ratio 1:6 for 135 min to reveal thegrain boundaries, and then gold coated. The grain diameter [d;defined as d5 (4�Area/p)1/2], and form factor [F; defined asF5 4p�Area/(Perimeter)2] were measured from SEM micro-graphs using a semi-automatic image analyzer (Kontron MOP30, Zeiss Kontron Elektronik Gruppe, Kontron Bildanalyse,Oberkochen, Germany) on no less than 300 grains.

III. Results

(1) Microstructural Characterization of the UndeformedSamples

Figure 1(A) shows a typical SEM micrograph of undeformedSiC–5YAG. After sintering, this material was fully dense(r5 3.26 g/cm3). The microstructure consists of equiaxed SiCgrains (dark phase), among which the secondary phase (crystal-line YAG),26 in bright contrast, is homogeneously distributed aspockets located at triple points of the grain structure. Anamorphous film at the grain boundaries is likely to be present,27

although it was not observed by SEM. The characteristic core-rim structure of the SiC grains, resulting from grain growthduring sintering,28 is clearly observed. The grain size (taken asthe planar equivalent diameter) and form factor of this materialwere 0.770.3 and 0.9570.05 mm, respectively.

The microstructure of undeformed SiC–15YAG is similar tothat described above, as shown clearly in Fig. 1(B); in particular,about the same grain sizes and form factors were measured (cf.Table I). The main differences result from the absolute density(r5 3.36 g/cm3 for SiC–15YAG, fully dense) and the distribu-tion of the secondary phase, which is somewhat less homoge-neous than in the material containing 5 wt% YAG.

(2) Mechanical Behavior

The creep curves (log _e vs. e) at 16251C for SiC–5YAG and SiC–15YAG are shown in Figs. 2(A) and (B). For comparison pur-poses, the applied stress and the strain attained at each stage ofthe creep tests were similar for the two materials. The creepcurves reveal some interesting properties. In all cases, the steady-

state regime is reached almost immediately after loading; sig-nificant transient states appear mostly at the first loading stage.At each stress stage, once the steady state has been reached, thestrain that the materials are able to undergo can be high (see, forinstance, the first stage in Fig. 2(B), with e5 6.9%). The totalstrain (without failure) is higher than 15% for SiC–15YAG, andabout 10% for SiC–5YAG. These values compare well withprevious results in LPS SiC containing Y2O3 and Al2O3 as sin-tering aids12,29 and also for some SSS SiC.14,17 As will be clearbelow, it is also important to note the duration of the tests: thiswas about 120 h for SiC–5YAG and more than 150 h for SiC–15YAG.

The observed range of strain rates lies between about 10�7 s�1

(SiC–5YAG under 49.4 MPa) and about 6 � 10�7 s�1 (SiC–15YAG under 90.6 MPa). These strain rates are also in goodagreement with previously reported results in similar sys-tems.12,24 In addition, the materials studied here are significant-ly more creep resistant than some SSS SiC 14,17 and superplasticLPS SiC.30 Figure 3 compares the strain rate versus stress loga-rithmic plot at the creep temperature for the two materials. At agiven stress, the strain rate increases with the YAG content: thehigher the proportion of secondary phase, the lower the creepresistance of the material. No evidence for accelerated creep,generally associated with damage development, was observedin any case. One can hence state that SiC–5YAG and SiC–15YAG exhibit reasonably good ductility creep resistance at ahigh temperature. The experimental results, therefore, demon-strate that the applicability of YAG-sintered SiC as a structural

Fig. 1. Scanning electron micrographs of as-processed (A) SiC–5YAGand (B) SiC–15YAG showing the grain morphology and phases present.

Table I. Characteristics of the Two Materials Prepared in this Study

Material�

Powder batch composition (wt%)

Grain size d (mm) Form factor (F) Elastic modulus E (GPa) Hardness HV (GPa) Toughness KIC (MPa �m1/2)a-SiC Al2O3 Y2O3

SiC–5YAG 95 2.147 2.853 0.770.3 0.9570.1 41073 24.070.6 2.470.1SiC–15YAG 85 6.441 8.559 0.870.3 1.070.3 37575 19.470.3 3.170.1�SiC–xYAG, where x refers to YAG proportion in wt%.

164 Journal of the American Ceramic Society—Melendez-Martınez et al. Vol. 90, No. 1

material is not restricted to temperatures below the 13451C eu-tectic temperature, as has been pointed out elsewhere.4

The creep results are analyzed in terms of the classical power-law creep equation:

_e1sn (1)

where _e is the steady-state strain rate, s is the applied stress, andn is the stress exponent for creep, which can be calculated at eachstress change.31 The measured stress exponents for SiC–5YAGand SiC–15YAG are included in Fig. 2. An average value ofn�2 was obtained for both materials under the present exper-imental conditions (see Fig. 3), in good agreement with previ-ously reported values for SiC.12,17,20,24

(3) Microstructure of the Deformed Materials

Figures 4(A) and (B) show typical SEM micrographs of SiC–5YAG and SiC–15YAG after the creep tests, respectively. Acomparison of these micrographs with the corresponding onesin Figs. 1(A) and (B) reveals that during creep, no dynamic graingrowth took place, and the grains maintained their mean sizeand their equiaxed shape. However, creep affected the micro-structure of the materials in two principal ways. Firstly, it wasevident that the secondary phase is expelled from some grainboundaries (presumably those subjected to compressive stres-ses), and rearranged in others. Figures 4(a) and (B) show this forSiC–5YAG and SiC–15YAG, respectively.

The second main microstructural evolution consisted of theextensive growth and coalescence of large, isolated cavities. Fig-ures 5 and 6 ((A) close to the center of the samples, to (C) nearthe surface) show this feature for SiC–5YAG and SiC–15YAG,respectively. The surface density (i.e., number of cavities per unitarea) and the size of the cavities increase rapidly from the centerto the surface of each sample. The cavities show a harsh elonga-tion in a plane perpendicular to the compression axis: this effectbecomes even more pronounced as one approaches the surfaces.Instead, the cavities that form in the absence of applied load areequiaxed and smaller, as seen in Fig. 7 for specimens placed in-

20 40 60 80 100 120 140

1625 °C

n = 2.0

n = 2.1

SiC-5YAG

SiC-15YAG

STRESS (MPa)

10−8

10−7

10−6

STR

AIN

RA

TE

(s−1

)

Fig. 3. Strain rate versus stress logarithmic plot for SiC–5YAG andSiC–15YAG under the experimental conditions used in this study. Theaverage stress exponent obtained by lineal fit is shown for each case.

0 84 12 16 20STRAIN (%)

0 42 6 8 10STRAIN (%)

STR

AIN

RA

TE

(s−1

)

10−8

10−7

10−6

10−5

STR

AIN

RA

TE

(s−1

)

10−8

10−7

10−6

10−5

2.2 1.9 1.7

σ (MPa)

43.9 65.8 65.881.0

n (B)

SiC-15YAG1625 °C

1.8 2.1 1.9

SiC-5YAG1625 ºCσ (MPa)

49.4 1.471.47 90.6

n

SiC-5YAG1625 °C

(A)

Fig. 2. Creep curve at 16251C for (A) SiC–5YAG (stresses between 49and 91 MPa up to about 10% of strain) and (B) SiC–15YAG (stressesbetween 43 and 81 MPa up to about 16% of strain). The values of thestress exponent obtained from each stress change are shown.

Fig. 4. Grain morphology of crept (A) SiC–5YAG and (B) SiC–15YAG, showing evidence of re-distribution of the secondary phase.The arrows indicate the compression axis.

January 2007 Creep and Microstructural Evolution at High Temperature of LPS SiC 165

side the creep machine during the tests but not loaded. It wasalso observed that cavitation was significantly more importantin SiC–15YAG than in SiC–5YAG. Similar results have beendescribed elsewhere24 for an LPS SiC containing 10 wt% YAG,although the cavitation was much less severe than that observedhere. Finally, cavitation was not observed in the undeformedsamples, even though sintering was performed at the highertemperature of 19501C in an argon atmosphere.

IV. Discussion

To ascribe the measured stress exponents to microscopic de-formation mechanisms requires a prior detailed analysis of themicrostructural evolution. This can be done in terms of a seriesof concurrent kinetic processes accompanying creep. Let us firstconsider the redistribution of the secondary phase mentioned

above. One can argue in terms of the well-known viscous flowcreep models,32,33 as the viscosity of the secondary phase at thetesting temperature is low enough to allow its squeezing fromthe grain boundaries under compression. According to thismechanism, the secondary phase is squeezed out from grainboundaries under compression and accumulates in multigrainpockets that are under only slight compressive stresses. Conse-quently, the flow of the secondary phase can explain the ap-pearance of large pockets of secondary phase and relativelyextensive regions of interlocked grains depleted of secondaryphase. In addition, if viscous flow of the secondary phase pro-ceeds for a relatively long time (i.e., if high final strains areattained), then one would expect the aforementioned agglo-merations to locate preferentially near the outer surfaces ofthe samples, where they would be larger in size. Therefore,viscous flow is compatible with some of the observed micro-structural features, although it cannot alone explain the

Fig. 5. Low-magnification scanning electron micrographs of SiC–5YAG after creep, showing the severe cavitation (A) close to the cen-ter of the sample, (B) between the center and surface of the sample,and (C) close to the surface of the sample. Cavitation was observed toincrease in importance as one approaches the surface of the sample.Also observed is the elongation of the cavities in a plane perpendicularto the compression axis, represented by the arrows.

Fig. 6. Low-magnification scanning electron micrographs of SiC–15YAG after creep, showing the severe cavitation (A) close to the cen-ter of the sample, (B) between the center and surface of the sample,and (C) close to the surface of the sample. Cavitation was observed toincrease in importance as one approaches the surface of the sample. Alsoobserved is the elongation of the cavities in a plane perpendicular to thecompression axis, represented by the arrows.

166 Journal of the American Ceramic Society—Melendez-Martınez et al. Vol. 90, No. 1

observed severe cavitation. In addition, it is also incompatiblewith the creep characteristics cited above: due to its inherentlytransient nature, high strains in the steady state cannot beattained by this mechanism.

Second, YAG-containing LPS SiC may undergo chemicaltransformations at the high temperatures required for creep.For instance, passive oxidation of SiC may occur.34,35 Oxida-tion of SiC is expected to give rise to a narrow layer of oxide atthe surface of the samples: this surface was not clearly observedhere because of the extensive damage. For the present work, themost important chemical change consists of the reactionsbetween YAG and SiC at the creep temperatures, as was alsoobserved in the previous work.24 It has been reported that sub-stantial weight loss may occur during sintering of Y2O3–Al2O3–SiC containing free Al2O3. This has been attributed to thereaction between the free Al2O3 and the SiC, resulting in theformation of volatile species. According to the literature, themost probable reaction is11

SiCðsÞ þAl2O3ðsÞ2Al2OðgÞ þ SiOðgÞ þ COðgÞ (2)

Other reactions between SiC (s) and Al2O3 (s) to form Al2OC(g) and/or Al4O4C (g) have also been reported elsewhere.36 Areaction similar to (2) involving Y2O3 is also likely to occur.

In addition, yttrium aluminates (Y3Al5O12 (YAG), YAlO3

(YAP), Y4Al2O9 (YAM)) have been shown to decompose in thepresence of SiC at high temperatures, even in the absence of freeAl2O3. Indeed, Mah et al.10 showed that YAG was highly stableat up to 16501C in air and under vacuum, but underwent severedecomposition at temperatures as low as 15001C under vacuumin the presence of SiC. The decomposition of YAG was accom-panied by the formation of aluminum-containing gaseous spe-cies, which are most likely to escape from the specimen. Theauthors explained this effect as being associated with the car-

bothermal reduction of YAG, which proceeds via the followingdecomposition reaction

Y3Al5O12ðsÞ þ 4COðgÞ23YAlO3 þAl2OCðgÞþ 3CO2ðgÞ (3)

being the source of gaseous CO the active oxidation of SiCunder low oxygen partial pressures:

SiCðsÞ þO2ðgÞ2SiOðgÞ þ COðgÞ (4)

Grande et al.11 have also identified this process at highertemperatures.

Now consider the aforementioned agglomerations of crystal-line YAG generated by viscous flow. The formation of volatilespecies accompanying their carbothermal reduction may thenwell be at the origin of the observed extensive nucleation andspectacular growth of cavities. The same carbothermal reduc-tion of YAG may explain why cavitation becomes even moreimportant as one approaches the outer surfaces of the samples(since, as mentioned above, the secondary phase tends to flowoutward as creep proceeds). This scenario is also consistent withthe fact that cavitation is much more severe in SiC–15YAG: thismaterial contains more secondary phase, and in addition wassubjected to high temperatures over a much longer time. Fur-thermore, this scenario is also consistent with the absence ofcavitation in the undeformed samples: for sintering, the pelletswere embedded into powder beds (coarse SiC plus Al2O3) inside

Fig. 7. Low-magnification scanning electron micrographs close to thesurface of (A) SiC–5YAG and (B) SiC–15YAG after heat treatment at16251C under an argon atmosphere in the absence of applied load.

O Al

Si

YC

after creep

before creep

SiC-5YAG

Rel

ativ

e in

tens

ityR

elat

ive

inte

nsity

(A)

0.0 0.5 1.0 1.5 2.0 2.5 3.0

OAl

Si

YC

after creep

before creep

SiC-15YAG

Energy (keV)

0.0 0.5 1.0 1.5 2.0 2.5 3.0Energy (keV)

(B)

Fig. 8. Energy-dispersive X-ray spectra (0–3 keV) from scanning elec-tron analysis for (A) SiC–5YAG and (B) SiC–5YAG before and afterthe creep tests. The peak at 0.68 keV is attributed to F, which is due tothe plasma etching.

January 2007 Creep and Microstructural Evolution at High Temperature of LPS SiC 167

graphite crucibles with screwable lids, which is known to preventthe formation of volatile species within the sample.37

The validity of the above argumentation was evaluated byenergy-dispersive X-ray spectroscopy (EDX) analysis. Com-pared in Figs. 8(A) and (B) are the EDX spectra of SiC–5YAG and SiC–15YAG before and after the creep tests, re-spectively. As can be seen in Fig. 8, there is no discernible Alpeak in the EDX spectra of the two crept specimens, which isdirect evidence of loss of Al during the creep tests. Thus, theEDX analysis is consistent with the existence of carbothermalreduction of YAG under the creep conditions. Consequently,the secondary phase in the crept samples is no longer YAG.

Most of the microstructural characteristics described abovemay be explained in terms of a viscous flow plus carbothermalreduction of a YAG-coupled process. At the creep temperature,viscous flow makes SiC grain boundaries under compressioncome into contact as well as agglomerating the YAG squeezedout from these grain boundaries into large pockets. The size anddensity of these pockets are expected to grow as one approachesthe surfaces of the samples. Simultaneously, the YAG in thepockets undergoes carbothermal reduction, which gives rise toseveral volatile species and nucleates cavities in the large pocketsformed by viscous flow. Finally, these cavities, which are initiallymost likely to have a nearly spherical form, are deformed, re-sulting in lenticular cavities because of the action of the appliedstress. Recall that pure compression load also causes deform-ation normal to the direction of the applied load, and that theSEM observations were made in sections containing the loadaxis.

From the analysis of the complex kinetic processes that areinvolved in the creep of the materials, the identification of mi-croscopic deformation mechanisms is formidably difficult. Vis-cous flow cannot be responsible for the creep behavior, due to itstransient nature. Instead, grain boundary sliding, consistentwith the measured stress exponents and with the lack of eitherdynamic growth or change of shape of the grains, is likely to bethe active mechanism. In addition, the growth and change ofshape of the cavities may also contribute to both the strain andthe strain rate. Finally, dislocation glide and climb have alsobeen identified by transmission electron microscopy (TEM) dur-ing creep of LPS SiC,12,19,22,24 although such TEM observationswere not performed in the present study. Thus, it is reasonableto think that the creep behavior of these materials cannot beexplained by the classical scheme of deformation mechanismplus accommodation (rate-controlling) mechanism. Instead, theobserved steady-state creep seems to result from the contribu-tions of a set of simultaneous and concurrent processes (grainboundary sliding1viscous flow of the secondary phase1chem-ical transformation of YAG1formation and growth of cav-ities1(likely) dislocation activity).

In spite of the obvious difficulties associated with the identi-fication of creep mechanisms, and from the viewpoint of thepossible structural applications of these materials, it should benoted that the materials exhibit relatively good creep character-istics (especially creep resistance and ductility) without evidenceof failure. The suitability of YAG-containing SiC for high-tem-perature applications is thus demonstrated.

V. Conclusions

The compressive creep characteristics at 16251C and the micro-structural evolution during creep of two LPS SiC polycrystalscontaining 5 and 15 wt% crystalline YAG as the secondaryphase were studied. Steady-state creep extended to strains up toabout 15% without evidence of failure: stress exponents close to2 were measured. The microstructural evolution during creepconsisted, first, of the redistribution of the secondary phase and,second, of the extensive nucleation and growth of large cavities.This evolution was explained in terms of a coupled viscousflow1carbothermal reduction of the YAG process. Finally,steady-state creep results from the contribution of a set of sim-ultaneous processes instead of from the classical scheme of

deformation mechanism1accommodation mechanism. It wasnoteworthy that, in spite of the degradation of the microstruc-ture, the materials exhibited good creep properties (ductility andcreep resistance), which compare well with the results reportedfor other SiC ceramics.

Acknowledgments

JJMM thanks the Consejerıa de Infraestructuras y Desarrollo Tecnologico(Junta de Extremadura) for supporting his stays at the University of Seville. Fruit-ful discussions with Prof. A. Munoz-Bernabe of the University of Seville aregratefully acknowledged.

References

1J. J. Melendez-Martınez and A. Domınguez-Rodrıguez, ‘‘Creep of SiliconNitride,’’ Prog. Mat. Sci., 49 [1] 19–107 (2004).

2N. P. Padture, ‘‘In-Situ Toughened Silicon Carbide,’’ J. Am. Ceram. Soc., 77 [2]519–23 (1994).

3K. A. Schwetz, ‘‘Silicon Carbide-Based HardMaterials’’; pp. 683–740 inHand-book of Ceramic Hard Materials, Edited by R. Riedel. Wiley-VCH, Weinheim,Germany, 2000.

4R. P. Jensen, W. E. Luecke, N. P. Padture, and S. M. Wiederhorn, ‘‘High-Temperature Properties of Liquid-Phase-Sintered a-SiC,’’ Mater. Sci. Eng., A282[1–2] 109–14 (2000).

5V. S. R. Murthy, H. Kobayashi, N. Tamari, S. Tsurekawa, T. Watanabe, andK. Kato, ‘‘Effect of Doping Elements on the Friction and Wear Properties of SiCin Unlubricated Sliding Condition,’’ Wear, 257 [1–2] 89–96 (2004).

6L. S. Sigl, ‘‘Thermal Conductivity of Liquid-Phase-Sintered Silicon Carbide,’’J. Eur. Ceram. Soc., 23 [7] 1115–22 (2003).

7Y.-W. Kim, M. Mitomo, and T. Nishimura, ‘‘High-Temperature Strength ofLiquid-Phase-Sintered SiC with AlN and RE2O3 (RE5Y, Yb),’’ J. Am. Ceram.Soc., 85 [4] 1007–9 (2002).

8S. J. Dapkunas, ‘‘Ceramic Heat-Exchangers,’’ Am. Ceram. Soc. Bull., 67 [2]388–91 (1988).

9Y.-W. Kim, M. Mitomo, and T. Nishimura, ‘‘Heat-Resistant Silicon Carbidewith Aluminum Nitride and Erbium Oxide,’’ J. Am. Ceram. Soc., 84 [9] 2060–4(2001).

10T.-I. Mah, K. A. Keller, S. Sambasivan, and R. J. Kerans, ‘‘High-Tempera-ture Environmental Stability of the Compounds in the Al2O3–Y2O3 System,’’J. Am. Ceram. Soc., 80 [4] 874–8 (1997).

11T. Grande, H. Sommerset, E. Hagen, K. Wiik, and M. A. Einarsrud, ‘‘Effectof Weight Loss on Liquid-Phase-Sintered Silicon Carbide,’’ J. Am. Ceram. Soc., 80[4] 1047–52 (2004).

12A. Gallardo-Lopez, A. Munoz, J. Martınez-Fernandez, and A. Domınguez-Rodrıguez, ‘‘High-Temperature Compressive Creep of Liquid-Phase-SinteredSilicon Carbide,’’ Acta Mater., 47 [7] 2185–95 (1999).

13S. Honda, T. Nagano, K. Kaneko, and H. Kodama, ‘‘Compressive Deform-ation Behavior of Al-Doped b-SiC at Elevated Temperature,’’ J. Eur. Ceram. Soc.,22 [6] 979–85 (2002).

14Y. Shinoda, M. Yoshida, T. Akatsu, and F. Wakai, ‘‘Effect of Amount ofBoron Doping on Compression Deformation of Fine-Grained Silicon Carbide atElevated Temperature,’’ J. Am. Ceram. Soc., 87 [8] 1525–9 (2004).

15S. M. Wiederhorn, B. J. Hockey, and J. D. French, ‘‘Mechanisms of Deform-ation of Silicon Nitride and Silicon Carbide at High Temperatures,’’ J. Eur.Ceram. Soc., 19 [13–14] 2273–84 (1999).

16M. L. Duval-Riviere, C. Carry, and J. Vicens, ‘‘Microstructural Study of De-formation Mechanisms in Polycrystalline a-SiC Deformed at High Temperature,’’Phys. Status Solidi, 155, 6–82 (1996).

17Y. Shinoda, M. Yoshida, T. Akatsu, and F. Wakai, ‘‘Compression Deform-ation Mechanism of Silicon Carbide: I, Fine-Grained Boron- and Carbon-Dopedb-Silicon Carbide Fabricated by Hot Isostatic Pressing,’’ J. Am. Ceram. Soc., 87[10] 1919–26 (2004).

18A. Djemel, J. Cadoz, and J. Philibert, ‘‘Deformation of Polycrystallinea-SiC’’; pp. 381–94 in Creep and Fracture of Engineering Materials and Struc-tures, Edited by B. Wilshire and D. R. J. Owen. Pineridge Press, Swansea, U.K.1981.

19J. E. Lane, C. H. Carter Jr., and R. F. Davis, ‘‘Kinetics and Mechanisms ofHigh-Temperature Creep in Silicon Carbide: III, Sintered a-Silicon Carbide,’’J. Am. Ceram. Soc., 71 [4] 281–95 (1981).

20H. Tanaka and Y. Inomata, ‘‘Diffusional Creep in Sintered Silicon Carbide,’’J. Jpn. Ceram. Soc., 93 [1] 45–50 (1985).

21G. Grathwohl, T. H. Reets, and F. Thummler, ‘‘Creep of Hot-Pressed andSintered SiC with Different Sintering Additives,’’ Sci. Ceram., 11, 425–31 (1981).

22R. D. Nixon and R. F. Davis, ‘‘Diffusion Accommodated Grain BoundarySliding and Dislocation Glide in the Creep of Sintered Alpha Silicon Carbide,’’J. Am. Ceram. Soc., 75 [7] 1786–95 (1992).

23Z. C. Jou A.V.Virkar and R. A. Cutler, ‘‘High Temperature Creep of SiCDensified Using a Transient Liquid Phase,’’ J. Mater. Res., 6 [9] 1945–49 (1991).

24M. Castillo-Rodrıguez, A. Munoz, and A. Domınguez-Rodrıguez, ‘‘Correla-tion between Microstructure and Creep Behavior in Liquid-Phase-Sintered a-Sil-icon Carbide,’’ J. Am. Ceram. Soc., 89 [3] 960–7 (2006).

25J. J. Melendez-Martınez, D. Gomez-Garcıa, M. Jimenez-Melendo, andA. Domınguez-Rodrıguez, ‘‘Creep Mechanism of Gas-Pressure-Sintered SiliconNitride Polycrystals I. Macroscopic andMicroscopic Experimental Study,’’ Philos.Mag., A84 [31] 3375–86 (2004).

168 Journal of the American Ceramic Society—Melendez-Martınez et al. Vol. 90, No. 1

26A. L. Ortiz, F. L. Cumbrera, F. Sanchez-Bajo, F. Guiberteau, H. Xu, and N.P. Padture, ‘‘Quantitative Phase-Composition Analysis of Liquid-Phase-SinteredSilicon Carbide Using the Rietveld Method,’’ J. Am. Ceram. Soc., 83 [9] 2282–6(2000).

27T. Nagano, H. Gu, G. D. Zhan, and M. Mitomo, ‘‘Effect of Atmosphereon Superplastic Deformation Behavior in Nanocrystalline Liquid-Phase-Sintered Silicon Carbide with Al2O3-Y2O3 Additions,’’ J. Mater. Sci., 37, 4419–24 (2002).

28L. S. Sigl and H.-J. Kleebe, ‘‘Core/Rim Structure of Liquid-Phase-SinteredSilicon Carbide,’’ J. Am. Ceram. Soc., 76, 773–6 (1993).

29M. Castillo-Rodrıguez, A. Munoz, and A. Domınguez-Rodrıguez, ‘‘Effect ofAtmosphere and Sintering Time on the Microstructure and Mechanical Propertiesat High Temperatures of a-SiC Sintered with Liquid Phase Y2O3–Al2O3,’’ J. Eur.Ceram. Soc., 26 [12] 2397–405 (2006).

30T. Nagano, K. Kaneko, G.-D. Zhan, M. Mitomo, and Y.-W. Kim, ‘‘Super-plastic Behavior of Liquid-Phase Sintered b-SiC Prepared with Oxynitride Glassesin an N2 Atmosphere,’’ J. Eur. Ceram. Soc., 22 [2] 263–70 (2002).

31D. Gomez-Garcıa, J. Martınez-Fernandez, A. Domınguez-Rodrıguez, A.Eveno, and J. Castaing, ‘‘Deformation Mechanisms for High-Temperature Creepof High Yttria Content Stabilized Zirconia Single Crystals,’’ Acta Mater., 44 [3]991–9 (1996).

32J. R. Dryden, D. Kucerovsky, and D. S. Wilkinson, ‘‘Creep Deformation dueto a Viscous Grain Boundary Phase,’’ Acta Metall., 37 [7] 2007–15 (1989).

33M. M. Chadwick, D. S. Wilkinson, and J. R. Dryden, ‘‘Creep due to a Non-Newtonian Grain Boundary Phase,’’ J. Am. Ceram. Soc., 79 [9] 2327–34 (1992).

34J. W. Hinze and H. C. Graham, ‘‘The Active Oxidation of Si and SiC in theViscous Glass-Flow Regime,’’ J. Electrochem. Soc., 12 [7] 1066–73 (1976).

35M. Balat, G. Flamant, G.Male, and G. Pichelin, ‘‘Active to Passive Transitionin the Oxidation of Silicon Carbide at High Temperature and Low Pressure inMolecular and Atomic Oxygen,’’ J. Mater. Sci., 27, 697–703 (1992).

36A. Gadalla, M. Elmasry, and P. Kongkachuichay, ‘‘High Temperature Reac-tions within SiC–Al2O3 Composites,’’ J. Mater. Res., 7 [9] 2585–92 (1992).

37V. V. Pujar, R. P. Jensen, and N. P. Padture, ‘‘Densification of Liquid-Phase-Sintered Silicon Carbide,’’ J. Mater. Sci. Lett., 19, 1011–4 (2000). &

January 2007 Creep and Microstructural Evolution at High Temperature of LPS SiC 169