A new technique of introducing dispersions of oxide particles into cast metals
- PDF / 1,391,009 Bytes
- 5 Pages / 603 x 783 pts Page_size
- 6 Downloads / 141 Views
M E T A L S and alloys with oxide dispersions are generally made by powder metallurgy r o u t e s - i . e . mechanical mixing, sintering and infiltration techniques. A more direct route is to add the particulate material to the molten m e t a l and cast the resulting dispersion. However, under these conditions most oxides are not wetted which makes their retention in the liquid difficult. One approach developed to overcome this problem was to coat the particles p r i o r to casting, 1 others t o promote particle wetting by saturating the melt with anions of refractory particles 2 or with wetting agents3 or to add the nonwetted particles t o a vigorously agitated, partially solidified slurry of the alloy.4 A more r e c e n t method of introducing the particles into m e l t s is that developed by Davies and Dennison5 in which a fine dispersion of oxide particles pref o r m e d in an internally oxidized c a r r i e r m e t a l is transferred into melts by dissolving the c a r r i e r immediately p r i o r t o casting. This method has been highly successful in controlling the as cast structure of certain alloys. The present paper describes attempts to identify the parameters controlling the d e g r e e of dispersion attainable by such a technique and t o a s s e s s the distribution of particles in the resulting ingots. In o r d e r to avoid complexities the work was confined t o simple dispersions, namely single phase particles dispersed in a single phase m a t r i x . Therefore, copper was selected for the latter, as its internally oxidized binary alloys provide a l a r g e selection of oxide particles, such as Cr203, A1203, SiO2, TiO2 which are all recognized as highly stable at the standard holding and transfer temperature, 1543 K, employed in the process currently described. Although the amounts of dispersed phases introduced in these experiments are of the o r d e r of 0.1 pct (wt) the process is not necessarily limited t o such levels.
J. P. DENNISON is Senior Lecturer, Department of Metallurgy and Materials Technology, University College, Swansea, U. K.,and S. E. KISAKUREK is Post-Doctoral Fellow,Oxford University, Department of Metallurgy and Science of Materials, Parks Road, Oxford, U. K. Manuscript submitted August 17, 1976. METALLURGICAL TRANSACTIONS B
EXPERIMENTAL
PROCEDURE
M e t h o d of S t u d y of D e g r e e of Dispersion and Particle Distribution in Castings The b a s i c experimental approach was t o m e a s u r e the effects of cooling r a t e , holding time, transferred volume fraction of particles and mean particle size, v a r i e d between two levels by employing half replicates of a 24 factorial design. In o r d e r to a s s e s s the importance of oxide density, the technique was a p plied to dispersions of Cu-SiO2 (density 2.32 x 103 Kg m -s) and Cu-TiO2 (density 4.20 × l0s Kg m-3). The details of each design and the levels of factors are given in Table I. Dispersions of Cu-A1203 and CuCr203 were excluded from such a systematic study mainly because A1203 particles were in acicular form wh
Data Loading...
