Influence of processing route on electrical and thermal conductivity of Al/SiC composites with bimodal particle distribution
- 291 Downloads
Al/SiC composites with volume fractions of SiC between 0.55 and 0.71 were made from identical tapped and vibrated powder preforms by squeeze casting (SC) and by two different setups for gas pressure infiltration (GPI), one that allows short (1–2 min) liquid metal/ceramic contact time (fast GPI) and the other that operates with rather long contact time, i.e., 10–15 min, (slow GPI). Increased liquid metal–ceramic contact time is shown to be the key parameter for the resulting thermal and electrical conductivity in the Al/SiC composites for a given preform. While for the squeeze cast samples neither dissolution of the SiC nor formation of Al4C3 was observed, the gas pressure assisted infiltration led inevitably to a reduced electrical and thermal conductivity of the matrix due to partial decomposition of SiC leading to Si in the matrix. Concomitantly, formation of Al4C3 at the interface was observed in both sets of gas pressure infiltrated samples. Longer contact times lead to much higher levels of Si in the matrix and to more Al4C3 formation at the interface. The difference in thermal conductivity between the SC samples and the fast GPI samples could be rationalized by the reduced matrix thermal conductivity only. On the other hand, in order to rationalize the thermal conductivity of the slow GPI a reduction in the metal/ceramic interface thermal conductance due to excessive Al4C3-formation had to be invoked. The CTE of the composites generally tended to decrease with increasing volume fraction of SiC except for the samples in which a large expansive drift was observed during the CTE measurement by thermal cycles. Such drift was essentially observed in the SC samples with high volume fraction of SiC while it was much smaller for the GPI samples.
KeywordsHigh Volume Fraction Particle Volume Fraction Squeeze Casting Shrinkage Porosity Effective Volume Fraction
One of the authors (G. S.) acknowledges funding by the EU FP6 integrated project “ExtreMat” under the contract no. NMPCT-2004-500253. Disclaimer: The work reflects only the views of the authors and the EU in not liable for any use of the information contained therein. J. M. Molina also acknowledges funds from the Generalitat Valenciana (project GVPRE/2008/244) and Universidad de Alicante (project GRE08-P13). J. M. Molina is also grateful to the Spanish Ministerio de Educación y Cultura for his “Juan de la Cierva” contract.
- 1.Zweben C (1992) J Met 44:15Google Scholar
- 2.Zweben C (1998) J Met 50:47Google Scholar
- 3.Zweben C (2005) Adv Mater Process 163:33Google Scholar
- 4.Zweben C (2006) Power Electron Technol 32:40Google Scholar
- 5.Occhionero MA, Fennessy KP, Adams RW, Sundberg GJ (2009) http://www.alsic.com/papers/cps_igbt_2001.pdf, viewed May 14
- 6.Occhionero MA, Hay RA, Adams RW, Fennessy KP (1998) Proceedings of SPIE—The international society for optical engineering 3582, p 687Google Scholar
- 7.Occhionero MA, Hay RA, Adams RW, Fennessy KP (1999) In: Wong CP (ed) IMAPS Advanced packaging symposium. Chateau Elan, Braselton, GeorgiaGoogle Scholar
- 10.Lloyd DJ (1994) Int Mater Rev 39:1Google Scholar
- 12.Han G, Feng D (2000) J Mater Sci Technol 16:460Google Scholar
- 13.Zulfia A, Hand R (2000) J Mater Sci Technol 16:867Google Scholar
- 19.Bass J (1982) In: Bass J, Fischer KH (eds) Landolt-Börnstein: numerical data and functional relationships in science and technology new series III/15: metals: electronic transport phenomena. Springer-Verlag, Berlin, p 5Google Scholar