Department of mechanical engineering.

Imperial College
Dr. N. P. O'Dowd
Exhibition Road, LONDON SW7 2BX
United Kingdom

Micro-scale and macro-scale Finite Element Analysis

The project involves the development of new lightweight materials for applications at high temperatures such as automotive engines. The project builds on existing expertise in the area of ceramic and intermetallic materials to develop through simulation and testing a novel ceramic intermetallic composite (CMIC) with improved mechanical and corrosion resistance characteristics

Physically based crystallographic slip system-based models have been used successfully to predict montonic, creep and cyclic response of intermetallic materials such as TiAl and NiAl, [1]. For ceramics, constitutive modeling of high temperature deformation has focused on grain boundary and bulk diffusion as the primary deformation mechanism, e.g. [2]. However, there has been limited work on the study of deformation and failure processes in intermetallic-ceramic composites as is proposed here. Strategies for the study of two phase materials are relatively well established, e.g. unit cell modeling [3] or Voronoi cell type models [4]—it is expected that the former approach will be adopted in this work though the applicability of the latter will also be examined. Similarly, while models for sintering of ceramics and metal matrix ceramic composites are quite well- developed [5], [6] a literature survey has found no papers in the area of model development for sintering of a ceramic-intermetallic composite. Finally the failure mechanisms of the CMIC under high temperature wear and impact conditions will be examined. The approach used will be a multiscale modelling procedure spanning from the atomisitic level through the microscale up to the macroscopic level. A similar approach is currently being applied in Brite EuRam program BE97-4283, to examine the wear of coated metal components at ambient temperatures.

[1] “A Dislocation Mechanics-Based Crystallographic Model of a B2-Type Intermetallic Alloy”, Busso, E. P. and McClintock, F.A., International Journal of Plasticity, 12, pp. 1-28, 1996. [2] “Effect of grain-size on the stress and velocity fields ahead of a crack in a material which deforms by Coble creep”, Pan, J., Cocks, A.C.F., International Journal of Fracture, 60, 2, pp.121-134, 1993 [3] “Computational Modelling of the Cyclic Deformation of Aluminium and Aluminium Matrix Composites,” N.P. O’Dowd, Lim, L.H. and P.E. McHugh, Computational Materials Science 5, pp. 187-194, 1996. [4] “Multiple scale computational model for damage in composite materials”, Lee, K., Morthy, S. and Ghosh, S. in Modeling and Simulation based Engineering, Eds. S.N. Atluri and P.E. O'Donoghue, Tech Science Press, USA, (International Conference on Computational and Engineering Sciences, Atlanta, USA), pp. 1426-1431, 1998. [5] “Constrained Sintering of Layered Ceramic Structures, Part II: Constitutive Framework and Residual Stress Predictions”, E.P. Busso, L. Chandra and R.P. Travis, submitted for publication, 1999. [6] “Hardness modelling for Al-6061/SiC MMCs”, Huda, D., ElBaradie, M.A., Hashmi, M.S.J, Key Engineering Materials, 104-107, pp.825- 836, 1995.

The main goals of Imperial College are to be counted among the leading institutions of the world for research, teaching and education in science, engineering and medicine. to educate our students in a way that fosters technical excellence, originality and breadth of vision and to provide an environment within which both original research of all kinds and its application to useful purposes flourish. The role of the department of mechanical engineering of Imperial College in this project is strongly linked with these goals. Our work will involve the development of computational tools for studying the processing and mechanical behavior of the new CIMC. The work at Imperial College will be carried out by a Post Doctoral researcher, jointly supervised by two experienced members of the permanent staff of the department of mechanical engineering. Both members of staff have extensive expertise in the development and use of advanced computational methods to analyse a wide range of engineering problems—of particular relevance is their experience in the modeling of high temperature mechanical behavior of ceramics, intermetallics and composites and in the development of sintering models for ceramics.

WORKPACKAGE TITLE: 2.1 MACROSCALE STUDIES: loading analysis STARTING DATE (Month #) 18 DURATION (Months) 14 Effort (MM) 10 OBJECTIVES To model the behavior of the CMIC automotive valve

DESCRIPTION OF WORK / TASKS The overall goal of this task is to develop macroscale models of failure which can be used to predict the lifetime of the CMIC valve. A computational analysis of the new CMIC automotive valve will be carried out. Finite element models of the valve will be generated using standard commercial finite element codes. This work will be carried out in close collaboration with the industrial partners. The contact and tribological properties will be determined from work package 2.4 and 5.4 (initial studies will be based on representative properties). The high temperature creep properties will be obtained from work package x. The critical local loading conditions will be determined from this macroscale analysis and used as input into the microscale deformation studies of work package 2.2.

DELIVERABLES Local stress and strain state in the CMIC valves. Lifetime predictions for CMIC valves.

MILESTONES AND CRITERIA Month 6: Development of FE model to represent loading and contact conditions in CMIC valve Month 8: Comparison of results with experimental observations, retune models if necessary Month 12: Final validated macro-scale model of CMIC valve.

INTERRELATION WITH OTHER WORKPACKGES Tribological properties obtained from work package 2.4 and 5.4. Link with experimental analyses of work package x.x.

WORKPACKAGE TITLE: 1. MICROSCALE STUDIES STARTING DATE (Month #) 13 DURATION (Months) 23 Effort (MM) 16 OBJECTIVES To model the constitutive behavior of the CMIC at the microscale using unit cell models

DESCRIPTION OF WORK / TASKS It is expected that failure in these materials will be by (i) surface wear leading to failure of the coating (ii) brittle microcracking close to the region of contact or (iii) void growth at triple grain junctions in the ceramic phase leading to propagation of existing microcracks. Having determined the macroscopic loading state from Task 2.1 modeling will be carried out at the microstructural level using representative unit cell type models. The unit cell will model details of the microstructure including the grain morphology of the ceramic and the reinforcing intermetallic. The effect of the surface coating will also be included in the analysis. It is expected that at the temperatures of interest the ceramic phase will deform by grain boundary diffusion and the intermetallic phase by slip along crystallographic planes. Where possible in the development of the constitutive models for the ceramic and intermetallic phases contact will be made with the atomistic simulations of workpackage x.x. The local conditions required for fracture initiation will be determined from the finite element studies based on the mechanisms discussed above. Having determined a microscale model for failure this model can be incorporated into the macroscale studies of Task 2.1 to provide life predictions for the CMIC valve.

DELIVERABLES A micro-scale model which includes a criterion for crack growth

MILESTONES AND CRITERIA Month 17: Constitutive models for ceramic and intermetallic phase developed Month 21: Microscale failure criteria under typical operating conditions Month 24: Incorporation of microscale failure criteria into macroscale model

INTERRELATION WITH OTHER WORKPACKGES Link with macroscale modeling, workpackage 2.1 Link with atomistic scale modeling of workpackage X.X. Link with microstructural analyses of workpackage X.X to determine most likely failure mechanisms in the CMIC.

WORKPACKAGE TITLE : Powder Design Optimisation, Sintering analysis STARTING DATE (Month #) 1 DURATION (Months) 13 Effort (MM) 13 OBJECTIVES To develop a model of the sintering process which can provide guidance on optimum sintering parameters to give a dense CMIC

DESCRIPTION OF WORK / TASKS The CMIC will be formed by sintering, a process in which the two powders are consolidated by their prolonged exposure to elevated temperatures and, external pressure. The use of two powder phases may result in inhomogeneous shrinkage causing residual stress development in the sintered product and relatively high porosities. A constitutive model based formulation has been developed that can describe the microstructural changes (e.g. density, grain size) which a sintering material undergoes and can allow the processing conditions to be evaluated and altered if necessary. Such an approach can expedite optimization of processing conditions and hence enhance product development. The formulation relies on internal state variables related to the evolution of the microstructure during sintering, namely the average grain size and mean relative density. This work package will involve the simulation of the evolution of relative density and average grain size and predict the stress state of the material under constrained sintering conditions. Fully three dimensional analysis of the sintering process will be carried out. The conditions, e.g. temperature, pressure which result in an optimum sintered CMIC will be identified using the above sintering models.

DELIVERABLES Sintering model of a ceramic intermetallic composite. Predicted optimum sintering conditions.

MILESTONES AND CRITERIA Month 4: Initial simulations of two phase sintering carried out Month 10: Recommendations for sintering conditions of CMIC Month 12: Final calibrated model for sintering of CMIC

INTERRELATION WITH OTHER WORKPACKGES Link with all partners involved with workpackage 3.

N.P. O’Dowd, W.C. Tang, E.P. Busso and P.E. McHugh, “Modelling of Deformation and Failure of Metal/Ceramic Composites”, in Constitutive and Damage Modelling of Inelastic Deformation and Phase Transformations, Ed. A.S. Khan, Neat Press, Fulton, USA, (Proc. of Int. Conf. on Plasticity), pp. 181-184, 1999. E.P. Busso, “Oxidation Induced Stresses in Ceramic-Metal Interfaces, Journal de Physique IV, 1999. (In press). E. P. Busso, F. Meissonnier, N.P. O’Dowd, “Length Scale Effects on the Geometric Softening of Precipitated Single Crystals”, Journal de Physique IV, 8 pp. 55-61, 1998. E.P. Busso, Y. Lei, N.P. O’Dowd, and G.A. Webster, “Mechanistic Prediction of Fracture Processes in Ferritic Steel Welds within the Transition Temperature Regime”, J. Eng. Materials and Technology, 120, pp. 328-337, 1998. E.P. Busso, and F. A. McClintock, “A Dislocation Mechanics-Based Crystallographic Model of a B2-Type Intermetallic Alloy”, International Journal of Plasticity, 12, pp. 1- 28, 1996. N.P. O’Dowd, L.H. Lim, and P.E. McHugh, “Computational Modelling of the Cyclic Deformation of Aluminium and Aluminium Matrix Composites,” Computational Materials Science 5, pp. 187-194, 1996. E. P. Busso, and F.A McClintock, “Mechanisms of Cyclic Deformation of NiAl Single Crystals”, Acta Metallurgica et Mat., 42, pp. 3263-3275, 1994.