What are Nanocomposite Ceramics?

Over the past half century ceramics have received significant attention as candidate materials for use as structural materials under conditions of high loading rates, high temperature, wear, and chemical attack that are too severe for metals. However, inherent brittleness of the ceramics has prevented their wide use in different applications.

Significant scientific effort has been directed towards making ceramics more flaw-tolerant through design of their microstructures by incorporation of fibers or whiskers which bridge the crack faces just behind the crack tip; by designing microstructures with elongated grains which act as bridges between crack faces just behind the crack tip; by incorporating second phase particles which deflect the crack making it travel a more tortuous path; and by incorporating secondary phases which undergo stress induced volume expansion that forces the crack faces together. However, one of the most recent development has been the distribution of multiple phases in a ceramic composite at the nanoscopic length scale. Owing to prevalence of nanoscopic features, such composites are referred to as ceramic nanocomposites.

The definition of nanocomposite material has broadened significantly to encompass a large variety of systems such as one-dimensional, two-dimensional, three-dimensional and amorphous materials, made of distinctly dissimilar components and mixed at the nanometer scale. The general class of nanocomposite organic/inorganic materials is a fast growing area of research. Reducing the sizes of structural features in materials leads to a significant increase in the portion of surface/interface atoms.

The surface/interface energies essentially control the properties of a solid. Interfaces provide a means to introduce non-homogeneity in the material. This non-homogeneity acts as a significant modification of both thermal and mechanical properties of the composites. Selective mixing of materials in a highly tailored morphology with high percentage of interface area, leads to materials with enhanced properties.

The properties of nano-composite materials depend not only on the properties of their individual parents but also on their morphology and interfacial characteristics. The nanocomposites find their use in various applications because of the improvements in the properties over the simpler structures. Few of such advantages can be summarized as:

• Improved Mechanical properties e.g. strength, modulus and dimensional stability
• Decreased permeability to gases, water and hydrocarbons
• Higher Thermal stability and heat distortion temperature
• Higher Flame retardancy and reduced smoke emissions
• Higher Chemical resistance
• Smoother Surface appearance
• Higher Electrical conductivity

For components used in a gas turbine engine, a lifetime upto 10000 h and a retained strength of ~300 MPa at a temperature of 1400 °C have been postulated, together with negligible creep rate. Furthermore, at elevated temperatures, the material must exhibit high resistance to thermal shock, oxidation, and subcritical crack growth. Ceramic nanocomposites have been shown to be extremely important for such future applications.

Advanced bulk ceramic composite materials that can withstand high temperatures (>1500 °C) without degradation or oxidation can also be used for applications such as structural parts of motor engines, catalytic heat exchangers, nuclear power plants, and combustion systems, besides their use in fossil energy conversion power plants. These hard, high-temperature stable, oxidation-resistant ceramic composites and coatings are also in demand for aircraft and spacecraft applications.

One such material system in this class of composites, Silicon Carbide/Silicon Nitride (SiC/Si₃N₄) composites, have been shown to perform very well under high temperature oxidizing conditions. Interest in such nanocomposites started with experiments of Niihara² who reported large improvements in both the fracture toughness and the strength of materials by embedding nanometer range (20-300 nm) particles within a matrix of larger grains and at the grain boundaries. A 200% improvement in both strength and fracture toughness, better retention of strength at high temperatures, and better creep properties were observed.

An advanced nanocomposite microstructure such as that of polycrystalline Silicon Carbide (SiC)-Silicon Nitride (Si3N4) nanocomposites, Figure 1, contains multiple length scales with grain boundary (GB) thickness of the order of 50 nm, SiC particle sizes of the order of 200-300 nm and Si₃N₄ grain sizes of the order of 0.8 to 1.5 µm¹. Designing the microstructure of such a composite (and similar others such as TiN-Si₃N₄, SiC-Al₂O₃, SiC-SiC, Graphene/CNT+SiC, and Carbon Fiber+SiC nanocomposites) for a targeted set of material properties is, therefore, a daunting task. Since the microstructure involves multiple length scales, multiscale analyses based material design is an appropriate approach for such a task.

Figure 1. Actual microstructure of a SiC-Si3N4 nanocomposite1

The ceramic nanocomposite work in Multiphysics Lab at Purdue focuses on (1) Understanding Performance of Carbide and Nitride Based High Temperature Ceramic Nanocomposites for Extreme Environments found in power generation cycles Including Nuclear Applications, (2) Multiscale Modeling and Characterization in Oxide Ceramic Materials and (3) Understanding thermal conduction and thermal issues in materials for thermoelectric power generation. A description of major areas of interest and contributions is as follows:

• Understanding thermal conduction and thermal issues to develop materials with low thermal conductivity^3-5: This work focuses on understanding atomistic mechanisms of operation of nanocomposites for thermoelectric power generation such that materials with low thermal conductivity could be developed. Explicit molecular simulations using molecular dynamics (MD) are performed to understand how morphology alterations can be used to reduce thermal conductivity in nanocomposites. We have found certain biomimetic arrangements that could achieve significant reduction in thermal conduction. We are in the process of making and testing such materials.
• Understanding Performance of Carbide and Nitride Based High Temperature Ceramic Nanocomposites for Extreme Environments Including Nuclear Applications^6-12: This research work focuses on understanding mechanisms of room temperature and high temperature operations of advanced nanocomposite ceramic materials that can enable power plant operation at temperatures in excess of 1750 K leading to efficiencies of almost 70% and significant reduction in the plant emissions. As an off-shoot, this project also focuses on thermal properties of these materials for possible use as high temperature multifunctional materials, high temperature structural materials in nuclear applications or heat sensors in nuclear applications.
• Multiscale Modeling and Characterization in Oxide Ceramic Materials^13-18: Focus during this work has been on understanding multiscale thermomechanical behavior of advanced composite materials such as multifunctional Al+Fe₂O₃ nanocrystalline composites and high-strength Al₂O₃/TiB₂ ceramic armor composites. This research on atomistic deformation analyses of Al+Fe₂O₃ multifunctional nanocomposites using MD is one of the first in the area of atomistic deformation analyses of advanced ceramic composite nanomaterials. In this work large scale MD simulations of nanocrystalline Al+Fe₂O₃ multifunctional composites, of single crystalline Al, of single crystalline Fe₂O₃, and of various interfacial configurations of single crystalline Al and Fe₂O₃ are performed. In the case of Al₂O₃/TiB₂ ceramic armor composites, we have developed and used a new cohesive finite element method (CFEM) for quantitative characterization of dynamic fracture.

The above contribution is strongly based on a collaborative multiscale modeling-material design-experimental processing approach. A snapshot of the overall collaborative research approach on modeling, design, and fabrication highlights is provided below.

Multiscale Modeling of Ceramic Nanocomposites: An Example of Work in SiC-Si₃N₄ Ceramic Nanocomposites

Our multiscale analyses (at nanometer and micrometer length and time scales) based on a combination of CFEM and MD based techniques have revealed that high strength and relatively small sized SiC particles act as stress concentration sites in Si₃N₄ matrix leading to inter-granular Si₃N₄ matrix cracking as a dominant failure mode. CFEM analyses have also revealed that due to a significant number of nano-sized SiC particles being present in micro-sized Si₃N₄ matrix, the SiC particles invariantly fall in wake regions of micro-cracks leading to significant mechanical strength. This finding was confirmed in the MD analyses that revealed that particle clustering along the GBs significantly increases the strength of these nanocomposites. While some nanocomposite morphologies have sharply defined SiC-Si₃N₄ interfaces¹⁹, other nanocomposite morphologies have diffusion of C, N, or Si atoms at the interfaces²⁰.

In the case of SiC-Si₃N₄ nanocomposites, MD analyses have also revealed that the second phase particles act as significant stress raisers in the case of single crystalline Si₃N₄ phase matrix affecting the strength significantly. However, the particle's presence does not have a significant effect on the mechanical strength of bicrystalline or nanocrystalline Si₃N₄ phase matrices. The strength of the SiC-Si₃N₄ nanocomposite structures showed an uncharacteristic correlation between the grain boundary (GB) thickness and temperature.

The strength showed decrease with increase in temperature for structures having thick GBs having diffusion of C, N, or Si atoms. However, for structures with no appreciable GB thickness (no diffusion of C, N, or Si atoms), due to the particle clustering and increase in SiC-Si₃N₄ interfacial strength with temperature, the strength improved with increase in temperature. Figure 2 shows snapshots of fracture propagation analyses in such nanocomposites obtained using the CFEM.

Figure 2. Snapshots of mesoscale crack propagation and damage propagation in the Si3N4 nanocomposites

Figure 3 displays snapshots obtained using MD. Current research work focuses on obtaining experimental images of the ceramic nanocomposites developed by collaborators, developing nanoscale CFEM meshes on such images, and performing failure analyses using the combination of MD and CFEM techniques.

Figure 3. Snapshots of atomistic damage and failure propagation in two different SiC (particle) and Si3N4 (matrix) nanocomposites at two different temperatures.

Petascale Computing Based Material Design

Atomistic analyses at the nanoscale can impart important information about the effect of critical features such as a GB, an interface, or a triple junction, etc. on mechanical deformation behavior of a small nanoscale (~ few nm) sample. In multiscale modeling such information is used to formulate macroscale (>few µm) material models for understanding microstructure dependent deformation behavior of a material sample such as the one shown in Fig. 1.

Appropriate mathematical models of microstructure property relations allow to relate performances like fracture toughness, ultimate strength, fatigue lifetime etc., to key material microstructure parameters like volume fraction, particle size, and phase composition. Since a typical nanoscale test sample is much smaller and is subjected to varied surroundings in a typical microstructure (e.g. Fig. 1), the incorporation of nanoscale information in macroscale models is subjected to statistical uncertainty.

If a complex microstructure is to be designed for a targeted set of properties, it is important that such uncertainties be correctly quantified and incorporated within a robust material design framework. We have pioneered the development of a variable fidelity model management framework that can incorporate material behavior analyses at multiple length scales in a design optimization framework^21-24, (Collaboration with Prof. John Renaud's group at the University of Notre Dame).

Figure 4 details the process flow of a petascale multi-physics model management tool for multiscale material design. Deployed on a petascale machine, the design tool developed in this research, that integrates atomistic and mesoscale analyses using a variable fidelity model management framework, will facilitate a significant reduction in nanomaterials' development cost and time with a simultaneous increase in the possible different combinations of individual composite material phases to achieve desired material performance.

The model management framework^21,22, besides managing the models and scales, is also well suited to control hierarchical parallelism. The natural hierarchy is MD within CFEM within design under uncertainty, using a mixed programming model SHMEM^TM by SGI for CFEM and MPI for MD and the uncertainty modeling. Both MD and the uncertainty quantification (via quasi-Monte Carlo integration) can use 1000 processors, and CFEM 10, so 1000 uncertainty quantification groups of 10 CFEM groups of 1000 HMC processors is 10⁷ processors, nearing exascale.

Figure 4. Schematic Petascale Material Design Framework

Preliminary material design analyses of the model system have been performed to understand the morphology related parameters that must be controlled for optimal targeted set of properties. The application of design tool is focusing on the continuous fiber ceramic composite (CFCCs) models of SiC-Si₃N₄ nanocomposites, Fig. 5. The second phase (circles and cylinders) are the SiC fibers that have higher elastic modulus and higher creep resistance (E) but lower yield stress and fracture toughness, than that of the primary Si₃N₄ phase. The problem is to design the most suitable CFCC, with maximum strength and creep resistance for a set of external temperatures T, where the number of design variables will depend on whether the simulation tests are run on the 2-dimensional (2-D) or 3-dimensional (3-D) model. The design variables to be considered in the nanocomposite design optimization problem, for the 2-D model, are the fibers diameter (d) and the external temperature (T). And for the 3-D model the design variables to be considered are the fibers diameter (d), the length of fibers (l) and the external temperature (T). The problem definition in standard form is given below:

Figure 5. High and low fidelity models for the CFCC nanocomposites

Figure 6 illustrates normalized (0-100) function values for the strength and creep strain rate as a function of design variables for the high fidelity model (3-D) and low fidelity model (2-D). Figure 6 (left) shows an increase in the CFCC strength and a corresponding decrease in the creep strain rate as the design variable d increases. Similarly for the high fidelity model, Fig. 6 (right) shows an increase in the CFCC strength and a corresponding decrease in the creep strain rate as the design variables d and l increase.

Figure 6. (left) Strength and creep strain rate at 1500oC as a function of the design variable width-height (d) for the 2-D low fidelity model. (right) Strength and creep strain rate at 1500°C as a function of the design variables width-height (d) and length of fibers (l) for the 3-D high fidelity model.

Fabrication

Focus during this activity is on forming a collaborative modeling-deign-processing framework where complex ceramic nanocomposites for targeted set of mechanical and non-mechanical properties could be produced without wasting significant trial-and-error time and money. We are collaborating with Prof. Rajendra K Bordia's group at the University of Washington-Seattle. Polymer derived ceramics (PDCs) are an attractive approach to make material design predicted morphology of ceramic nanocomposites. First Niihara and his coworkers and then others used this approach to make high performance nanoscale reinforced composites^25-27.

Continued research in this area has led to the development of a range of nanostructures. One particularly interesting class of materials have predominantly amorphous Si-O-C nanodomains containing nanoscale SiC and C reinforcements. These materials have the desired characteristics for a broad range of high-temperature applications while offering greater control over processing, compositions and nanostructure. PDCs are produced by pyrolyzing preceramic polymers and are typically amorphous up to very high temperature but provide very intriguing ceramic-like properties, such as good creep and oxidation resistance^28,29.

Some of their unique properties are associated with in-situ formation of nanodomains and lack of grain boundaries in their microstructures. Due to the polymeric nature (thermoset) of the precursors, this family of materials is easily processable as fibers, matrices for composites, porous structures and coatings^30,31. Most studied PDCs can be categorized into three main groups: (i) silicon carbide (SiC) (ii) silicon oxycarbides (SiOC) and (iii) silicon carbonitrides (SiCN). SiOCs and SiCNs are distinctive due to their hybrid molecular composition between SiO₂ and SiC and between SiC and Si₃N₄, respectively with additional level of "free" carbon as schematically illustrated in Fig 7 for Si-O-C system.

Figure 7. Schematic of Phase Relations in the Si-O-C System

A unique nanostructural feature of these materials is that the controlled excess carbon is dispersed as graphene layers with domain size of a few nms. Control of, and understanding of development of such nanostructural features, using an integrated experimental and atomistic simulation approach, is the focus of our collaborative research.

Acknowledgement

The related research work in our lab has been made possible by support from the US-Air Force Office of Scientific Research (Program manager: Dr. Joan Fuller), the US-Department of Energy, and the US-National Science Foundation

References

1. Weimer, A.W. and Bordia, R.K., Processing and properties of nanophase SiC/Si3N4 composites. Composites Part B: Engg, 1999. 30: p. 647-655.
2. Niihara, K., New design concept for structural ceramics-Ceamic nanocomposites. J. Ceram. Soc. Jpn: The centennial memorial issue, 1991. 99(10): p. 974-982.
3. Samvedi, V. and Tomar, V., Analyses of interface thermal boundary resistance of Si-Ge superlattice system as a function of film thickness and periodicity. Nanotechnology 20 (2009) 365701 (11pp) (Special Mention by Editors and Reviewers), 2008.
4. Samvedi, V. and Tomar, V., Role of heat flow direction, monolayer film thickness, and periodicity in controlling thermal conductivity of a Si-Ge superlattice system. J. Appl. Phys. (also featured in Virtual Journal of Nanoscale Science and Technology), 2008. 105: p. 013541.
5. Samvedi, V. and Tomar, V., Role of straining and morphology in thermal conductivity of a set of Si-Ge superlattices and biomimetic Si-Ge nanocomposites. Submitted to Journal of Physics-D, Applied Physics, 2009.
6. Tomar, V. Multiscale simulation of dynamic fracture in polycrystalline SiC-Si3N4 using a molecularly motivated cohesive finite element method. in 48th AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics, and Materials Conference (April 23-26, 2007) Honolulu, Hawaii. 2007: Paper No. AIAA-2007-2345.
7. Tomar, V., Analyses of the role of the second phase SiC particles in microstructure dependent fracture resistance variation of SiC-Si3N4 nanocomposites. Modelling Simul. Mater. Sci. Eng. , 2008. 16: p. 035001.
8. Tomar, V., Analyses of the role of grain boundaries in mesoscale dynamic fracture resistance of SiC-Si3N4 intergranular nanocomposites. Eng. Fract. Mech., 2008. 75: p. 4501-4512.
9. Tomar, V. and Gan, M., Temperature dependent nanomechanics of si-c-n nanocomposites with an account of particle clustering and grain boundaries. submitted to Int. J. Hydrogen Energy, 2009.
10. Tomar, V., Gan, M., and Kim, H., Effect of temperature and morphology on mechanical strength of Si-C-O and Si-C-N nanocomposites. Submitted to the Journal of European Ceramic Society, 2009.
11. Tomar, V. and Samvedi, V., Atomistic simulations based understanding of the mechanism behind the role of second phase SiC particles in fracture resistance of SiC-Si3N4 nanocomposites. International Journal of Multiscale Computational Engineering, 2009(DOI: 10.1615/IntJMultCompEng.v7.i4.40, 277-294 pages).
12. Tomar, V., Samvedi, V., and Kim, H., Atomistic understanding of the particle clustering and particle size effect on the room temperature strength of SiC-Si3N4 nanocomposites. to appear in Int. J. Multiscale Comp. Engg. special issue on Advances In Computational Materials Science, 2009.
13. Tomar, V., Molecular Modeling of Al-Fe2O3 Nanomaterial System. 2009: VDM Verlag Dr. Müller Aktiengesellschaft & Co. KG, ISBN 978-3-639-15858-8.
14. Tomar, V. and Zhou, M. An empirical molecular dynamics potential for an Al+Fe2O3 reactive metal powder mixture. in 45th AIAA/ASME/ASCE/AHS/ASC structures, structural dynamics and materials conference. 2004. Palm Springs, CA USA: AIAA.
15. Tomar, V. and Zhou, M., Deterministic and stochastic analyses of dynamic fracture in two-phase ceramic microstructures with random material properties. Eng. Fract. Mech., 2005. 72: p. 1920-1941.
16. Tomar, V. and Zhou, M., Classical molecular-dynamics potential for the mechanical strength of nanocrystalline composite fcc-Al+a-Fe2O3. Phys Rev B, 2006. 73: p. 174116 (1-16).
17. Tomar, V. and Zhou, M., Tension-compression strength asymmetry of nanocrystalline a-Fe2O3+fcc-Al ceramic-metal composites. Appl. Phys. Lett., 2006. 88: p. 233107 (1-3).
18. Tomar, V. and Zhou, M., Analyses of tensile deformation of nanocrystalline a-Fe2O3+fcc-Al composites using classical molecular dynamics. Journal of the Mechanics and Physics of Solids, 2007. 55: p. 1053-1085.
19. Bill, J., Kamphowe, T.W., Mueller, A., Wichmann, T., Zern, A., Jalowieki, A., Mayer, J., Weinmann, M., Schuhmacher, J., Mueller, K., Peng, J., Seifert, H.J., and Aldinger, F., Precursor-derived Si-(B-)C-N ceramics: thermolysis, amorphus state, and crystallization. Appl. Organometallic Chemistry, 2001. 2001(15): p. 777-793.
20. Jalowiecki, A., Bill, J., Aldinger, F., and Mayer, J., Interface characterization of nanosized B-doped Si3N4/SiC ceramics. Composites Part A, 1996. 27A: p. 721.
21. Gano, S.E., Agarwal, H., Renaud, J.E., and Tovar, A., Reliability based design using variable fidelity optimization. Structure and Infrastructure Engineering, 2006. 2(3-4): p. 247-260.
22. Gano, S.E., Renaud, J.E., and Sanders, B. Variable fidelity optimization using a krigin based scaling function. in 10th AIAA/ISSMO Multidisciplinary analysis and optimization conference. 2004. Albany, New York.
23. Mejia-Rodriguez, G., Renaud, J.E., and Tomar V., A variable fidelity model management framework for designing multiphase materials. ASME Journal of Mechanical Design, 2007. 130: p. 091702-1 to 13.
24. Mejia-Rodriguez, G., Renaud, J.E., and Tomar V. A methodology for multiscale computational design of continuous fiber SiC-Si3N4 ceramic composites based on the variable fidelity model management framework. in 3rd AIAA Multidisciplinary Design Optimization Specialist Conference (April 23-26, 2007) Honolulu, Hawaii. 2007: Paper No. AIAA-2007-1908.
25. Kroke, E., Li, Y.-L., Konetschny, C., Lecomte, E., Fasel, C., and Riedel, R., Silazane derived ceramics and related materials. Mat. Sci. and Engr.: R: Reports, 2000. 26(4-6.): p. 197-199.
26. Niihara, K., Izaki, K., and Kawakami, T., Hot-Pressed Si3N4-32%SiC nanocomposites from amorphous Si-C-N powder with improved strength above 1200O C. J of Materials Science Letters, 1990. 10: p. 112-114.
27. Wan, J., Duan, R.-G., Gasch, M.J., and Mukherjee, A.K., Highly creep-resistant silicon nitride/silicon carbide nano-nano composites. J. Am. Ceram. Soc., 2006. 89(1): p. pp. 274-280.
28. Raj, R., An, L., Shah, S.R., Riedel, R., Fasel, C., and Kleebe, H.-J., Oxidation kinetics of an amorphous silicon carbonitride ceramic. J. Am. Ceram. Soc., 2001. 84(7): p. 1803-10.
29. Rouxel, T., Soraru, G.D., and Vicens, J., Creep viscosity and stress relaxation of gel-derived silicon oxycarbide glasses. J. Am. Ceram. Soc., 2001. 84(5): p. 1052-1058.
30. Riedel, R., Mera, G., Hausner, R., and Klonczynski, A., Silicon-based polymer-derived ceramics: Synthesis properties and applications-a review. J. Ceram. Soc of. Japan, 2006. 114(6): p. 425-444.
31. Torrey, J.D., Bordia, R.K., Henager Jr., C.H., Blum, Y., Shin, Y., and Samuels, W.D., Composite polymer derived ceramic system for oxidizing environments. Journal of Materials Science, 2006. 41: p. 4617-4622.