Issues for the life prediction of Ceramic Matrix Composite components

Transcription

1 C a c h a n Issues for the life prediction of Ceramic Matrix Composite components Jacques Lamon LMT, CNRS/ENS/Paris-Saclay University, France lamon@lmt.ens-cachan.fr

2 Summary CMCs are of interest to thermostructural applications. They were developed initially for military and aerospace applications. Effort on CMCs in France started in the seventies They have now a high level of technological development. The issues have moved from processing methods and basic characteristics to issues relative to resistance to high temperatures and aggressive environments, life of material and components, predictive models and simulation. Predictions and control of life are fundamental issues for safe introduction and reliable use of Ceramic Matrix Composites (CMCs) in industrial systems running at high temperature or in aggressive environments.

3 Summary - Applications - Some exceptional properties of CMCs: Features of fast fracture Features of delayed fracture at high temperature - Significance of microstructure/properties relationships - Multiscale Modeling for lifetime prediction and composite design by tailoring properties to service conditions

4 APPLICATIONS Ceramic Matrix Composites (CMC) : FIBRES + Interphase + MATRIX (SiC, C) (PyC, BN) (SiC, C) High temperature structural applications Space and Defence Aeronautical applications Nuclear reactors Issue : lifetime control and prediction for long term applications 4

5 NUCLEAR POWER PLANTS Fuel cladding Control rods (SiC/SiC) Core Lay-out Gas Fast Reactor Core GFR Core Vessel

6 SPACE CRAFT THERMAL PROTECTION PROPULSION ARIANE LAUNCHER

7 HERMES EUROPEAN SPACE SHUTTLE PROJECT Thermal protection components

8 AERONAUTICAL APPLICATIONS

9 C/C FORMULA 1 BREAKS

10 CERAMIC MATRIX COMPOSITES CMC continuous fiber reinforced ceramics display remarkable properties - resistance to high temperature - resistance to high temperature fatigue - versatile stiffness - damage tolerance - crack arrest capability - decreased flaw sensitivity - quite infinite toughness/notch insensitivity - reliability - versatility

11 STRESS (MPa) VERSATILITY of CMCs Tensile behavior: influence of Young s modulus contrast 400 2D-SiC/SiC cvi 2D-C/C 300 2D-SiC/C 200 2D-SiC/CAS STRAIN (%) (after A. G. Evans)

12 Microstructure damage behavior relationship macropore longitudinal tow transverse tow 0.5mm layer 2D woven SiC/SiC microstructure Matrix damage and tensile behavior (Guillaumat, Lamon, 1993)

13 DAMAGE: Load transfer from matrix to fibers during matrix damage in 2D woven SiC/SiC RELATIVE YOUNG S MODULUS A: Ni/(PyC20/SiC50)10/SiC D: Ni/PyC100/SiC F: Hi-Ni/(PyC20/SiC50)10/SiC G: Hi-Ni/(PyC100/SiC bonded zone load fibre/matrix interfacial shear E / E debonded zone matrix crack F A E f.v f 2.E D G bonded zone load matrix fibre Strain (%) TOP DOWN PROCESS (Bertrand, Pailler, Lamon, 2001)

14 Tensile strength: tow controlled fracture of 2D composite Tensile strength of composite s TS = V L s tow Tensile strength of tow s = F N 1 R tow 2 t ac f Predictions for 2D Nicalon / SiC composites N t = 500 a C = 0.17 V L 0.2 R f = 7.5 mm Minimum Tow strength F(N) Composite strength s TS (MPa) ELS LLS RLLS (b r = 0.35) RLLS (b r = 1) 25 68

15 Density (1/MPa) Reliability: Ultimate failure of CMCs: from single filaments to woven composites Composites (2D) Composites (1D) Tows Fibers Stress (MPa) Strength density functions for SiC fibers (NLM 202), SiC fiber tows, SiC/SiC (1D) minicomposites and 2D SiC/SiC composites (Calard, Lamon, 2002)

16 Reliability: Ultimate failure of CMCs: statistical distributions of failure strengths limited effects of stress-state s 3p / s R 1.15 (Calard, Lamon, 2002)

17 Multiscale modeling of tensile behavior

18 Ultimate failure of filaments and multifilament tows Statistical distribution Damage mode Tensile behavior [Calard, Lamon, 1996]

19 Matrix fragmentation in CMCs Flaw-induced stochastic process [Calard, Lamon, 1998] [Lebrun, Lamon, 1996]

20 Matrix damage: Fragment dichotomy based model Strength data have Weibulll distribution P Vi rup V 1 S exp i l i 1 V 0 V i s s 0 m dv l di l di l di l di s m Fragment volume (equivalent to length l i ) is a statistical variable such as 0 2l i fragment Fibre i l di 2l i -l di 2l i x Stress-state in fragment i P ( x) x ld l 2l i d (Lissart, Lamon, 1997, 2009, 2010)

21 Force(N) Prediction of composite tensile behavior Matrix damage mode Longueur du minicomposite : tau = 84 MPa V f = 0.4 E m = 400 GPa E f = 300 GPa m m =5.03 m f =4.2 S 0m =5.7 MPa S 0f =6 MPa S rf =0 MPa nombre de fibres : 500 rayon des fibres : 7 microns expérience simulation Tow damage mode Déformation(%) [Pailler, Lamon, 2004]

22 Damage tolerance and notch sensitivity

23 Resistance to crack propagation 2D woven SiC/SiC fabricated By polymer conversion process (Kagawa, Goto (1997)) C/C and C/SiC are notch insensitive SiC/SiC may be notch insensitive

24 Notch sensitivity A. G. Evans (1997)

25 Damage sensitivity Post impact tensile strength s R /s R (a=0) 1 s = F/(w-a)b w a this study 0.2 Notch insensitivity: s R = s R (a=0).(1-a/w) a= diameter of equivalent hole Residual strength after static fatigue s R /s R (a=0) a/w without Fatigue after Fatigue at 450 C a/w

26 Resistance to crack propagation 2D SiC/SiC composites Fracture toughness: Material property which characterizes the initiation of fracture from a sharp crack (obtained by fatigue cracking under plane Strain conditions) (Droillard, Lamon, 1996)

27 High Temperature behavior: - monotonous loading rate - static fatigue

28 Tensile behavior at high temperature (Forio, Lamon, 2001)

29 Delayed failure and lifetime at high temperature Mechanical loads Matrix cracking Degradation of interphases Fiber overloading Temperature Environment High temperature oxidation creep of the matrix T>1200 C degradation Fiber weakening Delayed failure of fibers oxidation Creep T>1200 C slow crack growth T<1000 C Ultimate failure Durability

30 Crack healing in 2D woven SiC/SiBC composite Cyclic fatigue (20 Hz) at 1200 C (Carrere, Lamon, 1999) Static fatigue at 1200 C (load150 MPa)

31 Lifetime (s) Lifetime (s) Lifetimes in fatigue at high temperature Stress on fiber (MPa) 700 C Minicomposite BN interphase [Morscher 1998] 600 C NLM 207 treated 700 C NLM 202 as-received [Lavaire 1999] 600 C 1D SiC/Si-B-C Composite 2D 120 MPa 600 C NLM 207 treated 600 C 1D SiC/Si-B-C 600 C 2D SiC/Si-B-C 1000 Composite 2D 220 MPa Stress on fiber (MPa)

32 Static fatigue: slow crack growth in SiC fibers SiC tows SiC/SiC composite Slow crack growth Slow crack growth Surface oxidation Growth of oxide layer at fiber surface Protection of fiber by oxide layer and SiC matrix: slowing down SCG phenomenon 500 C 900 C 1000 C 1200 C Creep

33 Static fatigue: slow crack growth in SiC Nicalon tows Silica thin layer Fracture surface of a fiber after static fatigue on Nicalon tows at 700 C flaw Fiber (Forio, Lamon. JACS, 2004) Stress intensity factors estimated from crack sizes Penny shaped cracks : a KI 2s 2MPa m KIC crack length K I K I /K IC

34 lifetime ( s ) ddv (h) lifetime (s) lifetime (h) Static fatigue of SiC filaments and tows tows 500 C tows 800 C single fibres 800 C applied stress (MPa) n 8.4 A 0 = 5, Ea = 182 kj.mol mois sem y = 3, x -7, j y = 3, x -7, h ,1 100 Fils Hi - Nicalon S 600 C 1 min Fils Hi - Nicalon S 800 C 10 0,01 Fils Hi - Nicalon S 800 C 0, (Gauthier, Lamon, J. Am. Ceram. Soc. 2009) 34 stress ( MPa ) Hi-Nicalon S tows at 600 C and 800 C n 7,2 A 0 = 7, Ea = 178 kj.mol ,1 0,01 0,001 t s n = A t s n = A 0 exp (Ea/RT) Hi-Nicalon tows at 500 C and 800 C

35 Theory: fiber lifetime distribution - Subcritical crack growth V = da dt = V * - Stress-strength-rupture time relation - Lifetime distribution t = æ ç è K I K IC ö ø n 2 2K é IC æ s f ö ê ç V * Y 2 s 2 (n - 2) ë ê è s ø n-2 ù -1ú û ú s fj = s 0 (- v 0 v Ln(1 - P m j )) é æ P(t,s,v) =1 - exp -ç v ö æ ç s m ö ê æ 1 + t n - 2ö ç ê è v o ø è s 0 ø è t * 2 ø ë 2 t* = K IC V * s 2 Y 2 (R Mili, Lamon, 2011, 2012) 1 m n-2 ù ú ú û

36 Probability Distribution of lifetimes under constant stresses MPa 700 MPa MPa Critical fiber 0 1.E-04 1.E-01 1.E+02 1.E+05 1.E+08 Rupture time (hours) Hi Nicalon S C in air (R Mili, Lamon, 2015)

37 Size dependence of rupture time é æ P(t,s,v) =1 - exp - v ö æ ç s ö ê ç ê è v o ø è s 0 ø ë m t æ 2 = L ö 1 t ç 1 è ø L 2 æ ç 1 + t è t * n-2 m n - 2ö 2 ø m n-2 ù ú ú û PyC coating t 1 t 2 Fibre Fibre L 1 L 2 Influence of oxidation of PyC interphase in SiC/SiC: size effects On lifetime of SiC filament (P=0.1) at 500 C under 700MPa (R Mili, Lamon, 2011, 2012)

38 Rupture time (h) Stress-Probability-Time diagrams for 2D SiC/SiC p 90% Experimental data Predicted data % t s 2 = cste 25% 0% 75% 0, Stress on composite (MPa) Static fatigue at 500 C (Loseille, Lamon, 2010)

39 CMCs are versatile and smart materials Significance of microstructure/properties relationships Theoretically, composites can be designed with respect to end use applications Empirism still prevails But, composite design can be based on models Multiscale bottom up models are required: damage processes, failure mechanisms at pertinent scales, constituents properties, interface mechanics, and scale to scale changes Interface engineering, processing, treatment and new fibres and matrices (?)

40 ACKNOWLEDGMENTS Financial support: CNRS, Snecma, CEA, Conseil Régional d Aquitaine, European Commission. Ph. D. students: S. Bertrand, K. Rugg, L. Guillaumat, N. Godin, P. Forio, S. Pasquier, S. Pompidou, F. Pailler, P. Carrère, C. Droillard, F. Rebillat, V. Calard, C. Sauder, O. Loseille, V. Calard, J. El Yagoubi, M. R Mili, A. Laforêt

41