The nucleation and solidification of Al-Ti alloys
- PDF / 3,826,870 Bytes
- 9 Pages / 613 x 788.28 pts Page_size
- 88 Downloads / 189 Views
tectics, attributing grain refinement to the action of TiC in our system, x there seems to be no m y s t e r y for these alloys. A brief s u m m a r y is given at the end of the paper. We argue that nearly all the observations made to date in the A1-Ti system can be understood in the light of the results presented here.
J. CISSI~,formerlyof the Ford Scientific Staff, is now Senior Engineer, Soci6t6National d'Etude et de Construction de Moteurs d'Aviation, Gennevilliers, France. H. W. KERR, is Associate Professor, Department of Mechanical Engineering, University of Waterloo, Waterloo, Ontario, Canada. G. F. BOLLINGis associated with Product Planning and Research, Ford Motor Co., 20,000 Rotunda Dr., Dearborn, Mich. 48121. Manuscript submitted May 21, 1973.
Rapid freezing leads to dendritic growth. These v e r y rapid freezing experiments occurred too quickly to give meaningful inflection points on the cooling curves, but from the time for complete liquid to solid t r a n s f o r m a tion (approximately 20 st we estimate the velocities of the dendrite tips to be between 0.1 and I c m / s . Titanium profiles obtained with the electron microprobe are
METALLURGICALTRANSACTIONS
I. E X P E R I M E N T A L
TECHNIQUES
Specimens of various titanium concentrations, made up from aluminum of 99.99 pct purity and a commercial titanium " h a r d e n e r " were melted and cast from 100~C above the liquidus temperature, usually into a cylindrical graphite mold of 2 cm radius. The temperatures close to the bottom of each specimen were monitored using a fine chromel-alumel thermocouple, which projected 2 mm up from the bottom of the mold with its tip covered by a thin layer of alumina cement. A high speed (up to 2 c m / s chart speed) r e c o r d e r , with one or two milltvolts full scale, was employed. The overall cooling rates were varied by preheating the mold, allowing still air cooling starting with a r o o m - t e m p e r ature mold, or cooling the mold during and after c a s t ing with a water or air spray. The rate of cooling was not constant, and we have chosen to measure it from the tangent as close as possible to, and above, an inflection point. Microsections of regions very close to the t h e r m o couple tip were made by various techniques for optical microscopy. Counts of 60 to 100 s were made with a three spect r o m e t e r ARL-EMX electron microprobe on points of interest, and the raw data analyzed by the Multi 8, Magic 4, or Ford p r o g r a m s ~ for absorption, fluorescence and atomic number corrections. These p r o g r a m s differ in their handling of the atomic number corrections, but there remains considerable c o n t r o v e r s y about which p r o g r a m is most c o r r e c t f We have chosen to average the results of three p r o g r a m s in giving our results. II. RAPID FREEZING: Ti SUPERSATURATION AND DISTRIBUTION
VOLUME 5, MARCH 1974-633
shown in Fig. 1. The effective resolution was 4 microns. This can be supported by noting that the finest profile at 0.5 wt pct Ti was obtained a c r o s s a de
Data Loading...
