24TH INTERNATIONAL CONGRESS OF THE AERONAUTICAL SCIENCES

1

AbstractA new concept of Ceramic Matrix Composite(CMC) material, mainly based on the use of aself–sealing technology for matrix and the useof a multilayer woven reinforcement, has beendeveloped by Snecma for achieving highperformance levels targeted by future jetengines. The driving force for this developmenthas been to increase both lifetime andtemperature capability of previous C/SiC andSiC/SiC composites materials using amonolithic SiC CVI matrix and finishingtreatment against oxidation.

1 INTRODUCTION

In order to meet high performance levelstargeted by future jet engines components [1],Snecma Propulsion Solid (SPS), has developeda new concept of self-sealing Ceramic MatrixComposites (CMC) materials namedCERASEP®A410 and SEPCARBINOX A500combining improvements of fiberreinforcement, interphase and matrix [2]. Thesematerials, manufactured by SPS are designedfor thermo-mechanical applications such as longterm parts for gas turbine engine. Thedevelopment of these CMCs has beenmilestoned by steps, which as specified indifferent papers [3].

The aim of this paper is to presentthermomechanical characteristics and analysisvalidated by sub-element testing of these CMCs.After reviewing the material development story,the CMC's development phases, we focus on themeasured thermomechanical properties and life

time evaluation by fatigue testing. Thischaracterization program provides a basicdatabase used for the stress analysis of CMCcomposites through mechanical behaviormodels using Continuum Damage Mechanics. Athermomechanical analysis is presented with thecomparative behavior of two materials. A subelement testing validate the orthotropic non-linear law.

2 Materials for gas turbine engine

The application of advanced materials hasplayed a vital role in the development of gasturbine engines over the last fifty years. Anestimated 50% of the increase in efficiency andperformance has been directly attributed tomaterial development. Material developmentscome in two forms:• the easiest being the study development of a

known material system through iterations ofmaterial and processing : In this way, thetemperature capability of nickel basesuperalloys has increased by over 350°Csince the 1940’s,

• the other way is the introduction of newtypes of advanced material, tailored to thespecific requirement of gas turbine engines.

This approach is more difficult and requiresconsiderable continuous investment over manyyears. The introduction a new materials takestypically a minimum of fifteen years. Thenecessary levels of safety and reliability dictatethat all risk associated with the introduction of anew technology are adequately assessed andaddressed. In gas turbine engines thecomponents are subjected to many potentially

AN IMPROVED LONG LIFE DURATION CERAMICMATRIX COMPOSITE MATERIAL FOR JET AIRCRAFT

ENGINE APPLICATIONSL. Baroumes, E. Bouillon, F. Christin

Snecma Propulsion Solide

L. BAROUMES, E. BOUILLON, F. CHRISTIN, Snecma Propulsion Solide

2

detrimental factors including high temperatureand loads, vibrations and load cycling thermalshock, oxidation, corrosion, abrasion, erosion,handling and occasional impacts. In this contextvery new materials, however promising, cannotinstantly be put into enginesThe silicon bases ceramics, primarily SiliconNitride and silicon Carbide have been the focusof much interest and development during thelast twenty years. These materials retainstrength at over 1200°C where there has been noother material available (Fig. 1).

Fig. 1 Usable strength for high temperature materialsfor gas turbine engines.

However monolithic ceramic technology wasconsidered unsuitable for structural components: in an environment when catastrophic failure isto be avoided at all cost. The very low defecttolerance and low failure strain of this materialwas considered to impact too high of a risk andconsequently has not been specified inproduction engines. Composite materialcombining the advantages of ceramics with amaterial capable of offering an improvedtolerance to strain have captured all the recentefforts to use ceramics in larger aero-engines.The incorporation of continuous ceramic fibersoffer the ultimate in the ability to control theproperties of the material.

3 BACKGROUND

SiC/SiC (Silicon Carbide fiber/ siliconcarbide matrix) composite material series A300developed by SPS (originally SEP) before 1992were made of a 2D reinforcement (2D : 2directions of reinforcement), using NicalonTTMM

C.G. from Nippon Carbon, with pyrocarboninterphase, and a SiC matrix deposited bychemical vapor infiltration (CVI).

CERASEP A373 (A300 serie) materialwas characterized by a high specific strength, atroom temperature, of about 300MPa and a non-brittle behavior, with an enhanced failure strain,of about 0.5%. This material, protected by anappropriate finishing treatment againstoxidation was also characterized by a relativelygood life time capability, in flexural creep,under air, at temperature between 873 to 1473K.Nevertheless, testing in more severe conditions,especially for tensile/tensile fatigue testsperformed at a stress level of 120MPa, failure ofthe material occurred rapidly, due to a partialoxidation of the interphase. This was illustratedby a life time duration of about 10 to 20 hours at873K and less than 1 hour at 1123K. Inconclusion, for these series of SiC/SiCcomposites life time duration in air, at moderateand high temperature was considerably reducedbeyond the elastic yield point (about 80MPa). Afirst evaluation of the SiC/SiC material seriesA300 behavior was established in realconditions by the design, manufacturing andbench testing of Rafale engine M88-2 innernozzles flaps in 1992. This highlighted earlycracking of some components which resultedfrom local stresses. These stresses weregenerated by the combination of non-uniformthermal gradient and higher temperatures thananticipated by thermal analysis. It was foundthat these stress levels were twice than that ofthe SiC/SiC series A300 design allowable (80MPa) and induced micro-cracking of thematerial leading to an oxidation rateincompatible with long life span.

In the same period, Snecma developed aC/SiC (SEPCARBINOX , Carbon fiber/Siliconcarbide matrix) material for engine nozzlesoperating at intermediate temperature and lifespan objective over one thousand hours. Atypical example of application is the M88-2engine outer flaps.

SEPCARBINOXA262 is made of amultilayer reinforcement with carbon fiber, apyrocarbon interphase, a SiC CVI matrix and an

Specific Ultimate Strength - Mean value

0

5

10

15

20

25

30

35

40

100 350 600 850 1100 1350 1600 1850 2100temperature (F)

Spec

ific

ultim

ate

stre

ngth

(Ksi

)

Nickel base alloysMax temperature for steady-state use between 600°C (Inco718) and 900°C (Rene77)

Titanium alloysMaximum temperature for steady-state use up to 400°C (Ti6.2.4.2.)

Cobalt base alloysMaximum temperature for steady-state use up to 950°C (Mar-M 509)

Ceramic Matric Composite A 500

Ceramic Matric Composite A 410

3

AN IMPROVED LONG LIFE DURATION Ceramic Matrix CompositeMATERIAL FOR JET AIRCRAFT ENGINE APPLICATIONS

enhanced finishing treatment to improve itsbehavior in oxidative atmosphere.SEPCARBINOXA262 is characterized by anon-brittle behavior with failure strength ofabout 250 MPa at room temperature, and lifetime duration, for temperatures below 973 K,matching the application requirements.Qualification of the SEPCARBINOXA262technology was pronounced in 1996 for theM88-2 engine outer flaps, after completion ofendurance testing based on accelerated missioncycles. The lifetime objective of the componenthas now been demonstrated on the M88-2engine and serial production is underway.

Moreover, destructive evaluation, afterrepresentative engine life duration (1000 hoursengine test, resulting from 375 hours ground testand 625 hours flight test), performed by tensileproperties measurements, at room temperature,has confirmed the excellent capabilities of suchmaterial :

− σ/σinitial > 0.90,− ε/εinitial > 0.90,− E/ E initial > 0.90

4 NOVEL SELF-SEALING MATERIALS

4.1 General approachSPS decided in 1992 to engage the

development of a new CMC material concept,matching long life span at high temperature withhigher stress allowance.

The initial objective was to develop aSiC/SiC material useable beyond its mechanicalyield point for temperature range up to 1373K,and capable of operating life exceeding 1000hours in oxidative atmosphere integrating 100hours in severe conditions (stress level up to120MPa). The material concept was mainlybased on the use of a novel self-sealingtechnology for the matrix and the use of amultilayer woven reinforcement to reducedelamination sensitivity during themanufacturing process. The developmentapproach for the composite has already beendescribed [2].

More recently, considering the highpotential of the self-sealing matrix, and the

know-how achieved on C/SiC A262, thecombination of a carbon reinforcement with aself-sealing matrix has been studied. Thisapproach provides better economical outlookconsidering the gap between the cost of carbonand carbide fibers.

4.2 A410 and A500 description

Plane multi-layer reinforcement namedGUIPEX has been developed, in order toprevent the natural delamination sensitivity of2D materials. GUIPEX preforms are made oflayers, linked together and the technology isapplicable to carbon and ceramic fibers.Number of layers is adjustable, in order toobtain composite thickness range between 2 and7 mm, or variable thickness composite. Thesereinforcements have been optimized to obtainorthotropics composites, with in-planecharacteristics close to a 2D material.

A hardening step is necessary to obtainthe desired shape, with the desired fiber volumefraction, before the matrix infiltration. SpecificCVI route, combining both interphase andhardening process has been selected.

A self-sealing route has been selected forthe matrix in order to eliminate finishingtreatment. The principles of the self-sealingapproach are to consume part of the incomingoxygen and to prevent access of the residualoxygen to carbon interphase through themicrocracks. A novel matrix technology,combining carbides deposited by CVI processwith specific phases, has been developed.The functioning of the self-sealing matrix isillustrated in Figure 2, where the consumptionof a carbide (black area) sequence by oxidationand formation of a sealing glass in a microcrackis well shown after tensile fatigue test, in air, athigh temperature.

L. BAROUMES, E. BOUILLON, F. CHRISTIN, Snecma Propulsion Solide

4

Fig. 2 Micrograph of self-sealing matrix aftertensile fatigue at 600°C

5 MATERIAL CHARACTERIZATION

5.1 Experimental procedure

CERASEP®A410 and SEPCARBINOX A500test coupons are machined from standard200x200 mm2 plates. The first manufacturingstep of these plates is the interphase depositionand the CVI hardening. The second stepconsists in performing, after demolding, twoCVI cycles for carbides deposition. Testspecimens for high temperature mechanicaltesting in oxidative atmosphere, are machinedafter the first CVI cycle, in order to efficientlyprotect all the specimens edges. Most tests areperformed according to current EuropeanStandard, as already mentioned [3]. In-planetension, creep and tensile fatigue coupons aredogbone specimens with a width of 16 mm inthe gage section in order to limit scattering. Lowcycle fatigue tests (LCF) are carried out usingefficient strength and not apparent strength. Infact, it is assumed that the seal-coat does notplay a structural role. Hence, the thickness ofthe coupons after the first CVI cycle is takeninto account to calculate the load. Databasegeneration is focused, first on basic mechanicaland thermal properties versus temperature. Inparallel, assessment of the durability, in air,based mainly on tensile fatigue at 0.25Hz(trapezoidal cycle), typical engine cycle fatigueand tensile creep are carried out, taking intoaccount environmental effects, in order todetermine design allowable.

Table I. CMCs physical properties

Materials Fiber Vol.Fraction (%)

Density(g/cm3)

Porosity(%)

- A410- A500

3540

2.20 - 2.301.90 - 2.10

12 - 1412 - 14

Physical properties of these two CMCs aredescribed in Table I.

5.2 Mechanical and thermal database

A410 material is characterized by mechanicalproperties similar to those obtained on previousSiC/SiC materials (Table II). At roomtemperature, 40 coupons corresponding to 8manufacturing batches have been tested.Standard deviations obtained are 20 MPa for theultimate strength and 25 GPa for the elasticmodulus. These monotonic tensile properties donot change significantly up to 1473 K. Typicalstress-strain curve with unload/reload loop isshown in Fig.3, for the A410.

As already mentioned by otherresearchers [4], three domains can be observed:a linear elastic portion, a non-linear elasticdomain due to microcracking starting at amacroporosity level followed by a quasi lineardomain due to microcracking at the tow level.Residual strain after loading and hysteresisloops is negligible which is typical of adamageable linear elastic behavior.

A500 material shows tensile propertiessimilar to those obtained on

0

50

100

150

200

250

300

350

0 0,2 0,4 0,6strain (%)

stre

ss (M

Pa)

Fig. 3 Tensile stress-strain curve, at RT and dir.1 forA410

5

AN IMPROVED LONG LIFE DURATION Ceramic Matrix CompositeMATERIAL FOR JET AIRCRAFT ENGINE APPLICATIONS

SEPCARBINOXA262 (Table II). It ischaracterized by a low elastic modulus incomparison with the A410. In addition, the as-produced A500 is microcracked due to thethermal expansion mismatch between carbonfiber and carbide matrix. This explains theabsence of a linear elastic domain on the stress-strain curve (Fig.4). At high temperature, moreparticularly at 1473 K, the A500 ischaracterized by a slight increasing of the elasticmodulus in agreement with microcracks partialre-closing

Table II. Results from tensile test in dir. 1Temp.

(K)Numbercoupons

σ (MPa)

ε (%)

E(GPa)

CERASEP A410RT8731473

4059

315320325

0.500.600.60

220210205

SEPCARBINOX A500RT8731473

1533

230230230

0.800.800.70

657090

The two CMCs exhibit a significant differenceof in-plane thermal expansion coefficient . Infact, the respective values are about 5 10-6K-1

for A410 and about 2.5 10-6K-1 for the A500, at1273 K (Fig.5).On the other hand, the thermal conductivity ofthe A500 material is slightly higher than the

thermal conductivity of the A410 material athigh temperature for directions 1 and 2 (Fig.6).

5.3 Determination of the life time duration inoxidative atmosphere

Characterization of the tensile fatiguebehavior versus temperature, using a trapezoidalcycle at 0.25Hz has been performed on twoCMCs.

Illustration of LCF tensile behavior ofCERASEP A410, using trapezoidal cycle at0.25 Hz is presented in figure 7. At 873 K,duration of 1000 hrs without failure is reached

Fig. 4 Tensile stress-strain curve, at RT and dir.1 forA500

0

50

100

150

200

250

0,0 0,2 0,4 0,6 0,8 1,0strain (%)

stre

ss(M

Pa)

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0 500 1000 1500Temperature [°C]

Ther

mal

exp

ansi

on [%

]

A410

A500

Fig. 5 Thermal expansion of CMCs

Fig. 6 Thermal conductivity of CMCs

0

2

4

6

8

1 0

1 2

1 4

0 200 400 600 800 1000 1200 1400

Temperature [°C]

Ther

mal

con

duct

ivity

[W/m

.K]

A500, dir.1

A500, dir.3

A410, dir.1

A410, dir.3

L. BAROUMES, E. BOUILLON, F. CHRISTIN, Snecma Propulsion Solide

6

for a stress level of 170 MPa, and a durationhigher than 100 hrs at 200 MPa. At 1473 K, aduration of about 100 hrs is still obtained at 160MPa. The potential of the CERASEP A410,decreases at 1673 K, in accordance with thefiber thermal capacity. Furthermore, duration ofabout 100 hrs are still observed for stress levelslower than 100 MPa.

A comparison of the SEPCARBINOX

A500 behavior tested in tensile fatigue at 0.25Hz and in tensile creep is provided in figure 8.The life time duration, in tensile fatigue, isaround 100 hrs for a stress level closed to 160

MPa and in the temperature range of 873 – 1473K. In tensile creep, life time durations aredebited, but still remain significant.

6 THERMOMECHANICAL ANALYSIS

The life design of CMC components isbased upon both thermal and thermomechanicalanalyses using MARC© Finite Elementmodeling. Thermal analyses are transient inorder to determine the change with time of thetemperatures during representative jet enginecycles. Thermal analysis also yields importantresults for thermomechanical calculations. Hightemperature gradients are responsible of internalloads due to differential thermal expansion(Fig.9). Thermomechanical analysis also takesinto account pressure loads and interfaceconditions between parts.

In order to size CMC components, anaccurate knowledge of the non-linearorthotropic behavior throughout the entire rangeof operating temperature is needed. That ismean experimental characterization withintension, compression and shear for allreinforcement directions of the composite.While stress-strain relationship of in-planetesting is relatively easy to measure, it is oftenmore difficult in the case of shear and Poissoneffects. Because of the damaging character ofCMC behavior, tensile testing must beperformed using incremental load-unloadcycles. In compressive testing, stress-straincurves are generally linear. In addition to the setof curves defining the stiffness matrix, it isnecessary to add the 3 thermal expansioncurves.

Fig. 7 A410 T/T fatigue, in air, 0.25 Hz

80

100

120

140

160

180

200

220

240

10 100 1000Lifetime (hrs)

Stre

ss (M

Pa)

873 K1473 K1673 K1573 K

Fig. 8 A500 T/T fatigue and tensile creep, in air

80

100

120

140

160

180

200

220

240

10 100 1000Time (hrs)

Stress

(MPa

)

873 K - T/T fatigue 1473 K - T/T fatigue873 K - Creep 1473 K - Creep

Fig. 9 Thermal field on a nozzle flap

Gaspathside

Coolingflow side

7

AN IMPROVED LONG LIFE DURATION Ceramic Matrix CompositeMATERIAL FOR JET AIRCRAFT ENGINE APPLICATIONS

6.1 Comparative thermomechanical behavior

For components submitted to heterogeneousthermal flow, like nozzle flaps, the combinationof thermal properties and elastic modulus ofA500 material, is very favorable to decreasingthermal gradients and as a result to loweringthermal stresses [5]. An example ofthermomechanical analysis, performed on acomponent representative of a nozzle flap, andtaking into account the specific characteristicsof each material is given in Fig.10 and Fig.11.

Fig. 10 thermomechanical stress field on a nozzle flap

σ11 σ22

0

25

50

75

100

125

150

0 100 200 300 400 500 600location from forward end (mm)

Ther

mom

echa

nica

l stre

ss,

sxx

(MPa

)

A410

A500

60 MPa / ε=0,07%

145 MPa / ε=0,12%

The lower sensitivity to thermalgradients of SEPCARBINOX A500 comparedwith A410 is a way to increase its designmargins as outlined by the following example :referring to the same design case, Fig.11 showsmax. longitudinal stresses respectively of 60MPa and 145 MPa for A500 and A410materials. The relevant stress allowable for a

life time objective of 100 hours, issued from thefatigue and creep curves plotted in Fig. 7 & 8,are respectively 160 MPa and 190 MPa forA500 and A410 materials. These data end up ina sizing coefficient of 2.7 for A500 materialwhich is twice than that of the A410 sizingcoefficient.

6.2 Coupling damage and oxidationSnecma Propulsion Solide has developed

an uncoupled orthotropic elastic non-linearbehavior law, which take into account a networkof strain stress curves [6]. More recently,coupled mechanisms of damage and inelasticstrains have been taken into account using adamage model, named ODM, and developed byONERA [7]. This approach is essential toimprove modeling of CMC behavior. Fig. 12shows the tensile-compressive stress-straincurve with unload-reload loop, obtained withdamageable model, for A500.

Fig. 12 Tensile-compressive stress-strain ODM curve, atRT and dir.1 for A500

-200-150-100

-500

50100150200250

-0,2 0 0,2 0,4 0,6 0,8Strain (%)

Stre

ss (M

Pa)

The quality of CMC component sizing,based on thermomechanical analysis depends onthe understanding of both fatigue and creepyields and failure limits. Once again, fatigue andfailure limits are easily obtained for in-plane,uniaxial and isothermal conditions, whileanisothermal, multiaxial and non-proportionalloading need non standard characterization.

In contrast with metallic materials, there isno general validated failure criterion. Theanalysis could be based on known criteria takenfrom the literature, as Hill, Hoffman or Tsai-Wu. The current criterion is based on maximalstress or strain, which is directly compared to

Fig. 11 Longitudinal stress - A410 and A500 cases

Gaspathside

Coolingflow side

L. BAROUMES, E. BOUILLON, F. CHRISTIN, Snecma Propulsion Solide

8

stress-duration curves issued from fatigue andcreep testing. In most cases, stress distribution ispreferentially uniaxial (Fig.10). But, criterioncurrently used is sometimes insufficient,because knowledge of the failure mechanismsand associated characterization are still inprogress.

To improve lifetime justification with self-sealing matrix, introduction of couplingbetween macroscopic damage mechanism andoxidation is needed. This work had beeninitiated with a macroscopic model, which takesinto account the relationship between oxidationkinetics and fissuring mechanisms atmicroscopic scale [8]. This new method givesfirst good results concerning fatigue-creepbehavior for A410 database. Improvements ofthe method will be extended and validated forA500 material and complex thermodynamicconditions.

7. SUB-ELEMENT TESTING

Required changes in the method, i.e.multiaxial fatigue-failure criteria and life timepredictive laws, must be validated by sub-element testing. Sub-element testing meansinstrumented CMC component with a geometryas near as possible of the engine component oneand taking into account both boundaryconditions and representative loading of jetengine application.

A first example of sub-element testing atroom temperature is shown on figures 13 & 14.

Fig. 13 flexure-torsion sub-element testing and itsinstrumentation

Fig. 14 deformed shape for flexure-torsion sub-elementtesting(a) Beginning of testing (b) Max loading

The A500 plate was instrumented with 10strain gauges. This testing was dedicated tovalidate both orthotropic non-linear elastic andODM laws using Finite Element analysis. Theplate was loaded by an uniform pressure on thetop and was simply supported on the other side.One of the supports had a longitudinal deviationequal to 10% or 20% of the transversal size.These boundary conditions consisted of acombination of flexure and torsion loading.Several cycles with incremental pressure havebeen carried out

Mechanical analysis has also been carriedout, as shown on fig. 15 for displacement andfig. 16 for transversal stress and translaminarshear.

Fig. 15 deformed shape (x10) and normal deviationfield for flexure-torsion analysis

Fig. 16 transversal stress and translaminar shear fieldfor flexure-torsion analysis

The comparison of calculated strains andmeasured strains by a 10 gauge set shows goodagreement. Figure 17 gives the strain response

9

AN IMPROVED LONG LIFE DURATION Ceramic Matrix CompositeMATERIAL FOR JET AIRCRAFT ENGINE APPLICATIONS

of a [0°,45°,90°] strain gauge, implemented nearthe center of the plate (Fig. 13). Similarly theOrthotropic Non-Linear Elastic Law (ONLEL)load responses are correctly simulated.However, because of elastic assumption,unloading and residual strains are not reached.

Fig. 17 Calculated and measured strain, at RT and dir.0°, 90° &45° for flexure-torsion testing (ONLEL)

0

50

100

150

200

250

300

350

400

450

500

0 0,05 0,1 0,15 0,2 0,25 0,3 0,35 0,4strain (%)

forc

e (d

aN)

test - dir. 0°test - dir. 90°test - dir. -45°FEM - dir. 0°FEM - dir. 90°FEM - dir. -45°

0° 90° -45°

When introducing damage effects, it ispossible to obtain a more realistic CMCbehavior, taking into account damage historyand permanent strains, as shown on fig. 18, foran unidirectional gauge at the center of theplate. In this case and concerning A500, sub-element analysis validates the identification ofboth coupling parameters and damage kineticsof ODM behavior law.

Fig. 18 Calculated and measured strain, at RT and dir.90° for flexure-torsion testing (ODM)

050

100150200250300350400450500

0 0,05 0,1 0,15 0,2 0,25 0,3strain dir. 2 (%)

Forc

e (d

aN)

measured

calculated(ODM)

In a second time, the same sub-element hadbeen tested under torsion loading at 873 K inair, as shown on Fig. 19. The plate was simplysupported on two adjacent sides and a deviationequal to 10% of the transversal size was appliedon the opposite corner. Testing was performed

during 550 hours, with unloading every 18seconds (110 000 cycles). No apparent failurewas visible after this high duration. Before andafter torsion fatigue, the plate has also beentested at room temperature with a 10 gauge set(Fig. 19) in order to validate behavior law ofboth virgin and older CMC. A final flexure-torsion testing at maximum pressure anddeviation was applied without any failure of thesub-element.

Fig. 19 Torsion sub-element testing at ambienttemperature (static) and 873 K (fatigue).

Translaminar shear field is shown on fig.20. Throughout the entire plate surface, shear isin the range from 70% to 90% of meanIosipescu ultimate shear measured on A500specimen. Once again, orthotropic non-linearbehavior law led to model only the first loading.As shown on fig. 21, strain responses fromODM analysis are in a good agreement withexperimental results. While damage andresidual strain after loading is reached,hysteresis loops are not negligible anymore.

Fig. 20 translaminar shear field for torsion testing

L. BAROUMES, E. BOUILLON, F. CHRISTIN, Snecma Propulsion Solide

10

Fig. 21 Calculated and measured strain, at RT and dir.45° for torsion testing

-400

-200

0

200

400

600

800

-0,4 -0,3 -0,2 -0,1 0,0 0,1 0,2 0,3 0,4 0,5strain dir. 45° (%)

forc

e (N

)

measured

calculated

The complete results of sub-element testingdemonstrates substantial improvement:Ø A500 component shows strong potential for

providing high life duration consideringmultiaxial loading at 873 K in air;

Ø The validation of orthotropic non-linearelastic law , which is the current method usedin the most CMC jet-engine development, isconfirmed;

Ø The ODM damageable behavior law is anexcellent candidate in order to perform futuredetailed sizing analysis and to improve bothfatigue-failure criterion and life timepredictive law with oxidation coupling.

8. Conclusion and Perspectives

A new concept of Ceramic Matrix Composite(CMC) material based on Self Sealing Matrix isexpected to become an emerging technologywhich opens "new frontiers and horizons" in the21st century. Field service evaluations of Sealsare planned in the near future. It will be a newchallenge for these materials to be tested in"flight". Consistent efforts are underway toimprove the technology readiness level ( TRL )as well as towards an improvement of thematerial life capability above the performancesto date. To this end, modeling tools addressingmaterial aging and life prediction are good helpfor long term projections.

References

[1] Spriet P and Habarou G. Applications of continuousfiber reinforced ceramic composites in militaryturbojet engines. ASME Paper No. 96-GT-288.

[2] Bouillon E, Abbé F, Goujard S, Pestourie E, HabarouG. Mechanical and thermal properties of a sel-sealingmatrix composite and determination of the life timeduration. 24th Annual Conference on Composites,Advanced Ceramics, Materials and structures,Vol.21, Issue 3, pp 459, 2000.

[3] Bouillon E, Lamouroux F, Baroumes L, cavalier J.C,Spriet P, Habarou G. An improved long life durationCMC for jet aircraft engine aplications. ASME PaperNo. GT-2002-30625.

[4] Guillaumat L. PhD Thesis, Univ. Bordeaux I, France,1994.

[5] John R, Zawada L P, Kroupa J L . Stresses due totemperature gradients in Ceramic-Matrix-CompositeAerospace Components. Jour. of Am. Cer. Soc., [1]pp 161-168, 1998.

[6] Lapp P and Quesada B. Analysis of solid RocketMotor Nozzle. AIAA/SAE/ASME/ASEE 28th JointPropulsion Conference and Exhibit, Nashville, AIAA92-3616

[7] Maire JF and Lesne PM, A damage model forceramic matrix composite, Aerospace Science andTechnology, no4, pp 247-257, 1997

[8] Ladevèze P, Letombe S and Cluzel C. A CMCdamage model based on micro and macromechanicsfor high temperature and complex loading. HTCMC4– 4th International Conference on High TemperatureCeramic Matrix Composites, Munich