Solubility of nitrogen and carbon in CaO-Al 2 O 3 melts in the presence of graphite
- PDF / 471,305 Bytes
- 6 Pages / 603 x 783 pts Page_size
- 71 Downloads / 190 Views
TECHNIQUE
The experiments were performed in a similar manner as described in previous publications from this laboratory.1,3,4 A T a m m a n n furnace was used. The reaction tube of recrystallized alumina was inserted into the carbon tube of the furnace and closed gas tight at both ends. The gas phase was prepared by mixing nitrogen, carbon monoxide and argon with capillary flow meters. The m a j o r part (250 cc per min) of the gas was introduced into the reaction tube through a packing of graphite granules located under the crucible, a s m a l l part (15 cc per min) was bubbled through the melt (via a graphite tube) to ensure stirring. The starting s l a g s were premelted from reagent g r a d e CaCO3 and A1203. The graphite crucibles were machined from high p u r i t y (99.98 pct) graphite. An amount of 100 g of slag was used in the individual experiments. The slag was heated to 1600°C under
KLAUS SCHWERDTFEGER and HANS GEORG SCHUBERT are with Max-Hanck-Institut far Eisenforschung GmbH, Metallurgical Department, 4000 Dtisseldorf, Germany. Manuscript submitted October 26, 1976. METALLURGICAL TRANSACTIONS B
pure CO and held under the same gas for 1 h. Then the N2-CO-Ar gas was admitted. Total reaction time was between 30 and 100 h, during which samples of about 3 g were taken from the melt by suction into 4 mm ID graphite tubes, in r e g u l a r intervals. The samples wer,e sand blasted (to r e m o v e adhering graphite) and analyzed for total carbon, total nitrogen, and cyanide contents. In some of the samples also the acid soluble carbon was determined which was practically identical to total carbon, in agreement with previous measurements? Hence, the amounts of graphite in the slag are negligibly s m a l l . The applied analytical procedures have been described in a preceding p a p e r ? Further, the Ca and A1 contents were analyzed in o r d e r t o check whether changes of Ca/A1 r a t i o occ u r r e d during the long reaction times. EXPERIMENTAL RESULTS AND DISCUSSION The typical concentration-time c u r v e s are shown in F i g s . 1 and 2. The zero point of the time s c a l e is at the time when the carbon monoxide atmosphere was changed to the N2-CO-Ar gas mixture. In the upper d i a g r a m s the contents of total nitrogen (N)total, of cyanide (CN) and of total carbon (C) total are shown. S i m i l a r curves were obtained for all the other experiments. In the lower d i a g r a m s the analyzed Ca/A1 ratios are plotted against time. Because of the r e l a tive unprecision of the individual Ca/A1 values a v e r age values are given for all the experiments conducted with the same m a s t e r s l a g . E r r o r bars indicate the scatter of the individual points. With s l a g s of nominal composition 40 wt pct CaO/60 wt pct A1203 constant (N) total, (CN) and (C) total contents are attained after some time (e.g. Fig. 1), and the Ca/A1 r a t i o does not change within the experimental e r r o r . With the CaOrich slag of 50 wt pct CaO/50 wt pct A1203, however, the curves for carbon and cyanide have m a x i m a
Data Loading...
