• PDF / 2,010,065 Bytes
• 8 Pages / 603.28 x 783.28 pts Page_size
• 62 Downloads / 222 Views

DOWNLOAD

REPORT

---

THE pyrometallurgical refining of copper generally involves preferential oxidation of impurities in the metal phase which are removed to a slag as oxides. To predict the feasibility of any such process from chemical thermodynamic considerations, distribution of the constituents between the metal and oxide phases should be known. Thus the refining of copper under liquid conditions depends on the elements removed having a greater affinity for oxygen than does copper. Among the elements that are slagged in this way cobalt lies between nickel and iron. The plant experiments of Wanjukoff ] and Waehlert 2 (quoted in Ref. 3) showed that higher distribution coefficients of cobalt can be obtained during fire refining in reverberatory furnaces with acid linings than in furnaces with basic linings. Later, Johansen, 3 Wang, 4,5 Kuxmann, 6 Komorava, 7 Burylev, + and Zakharov 9 studied the equilibrium between metal containing cobalt and slag of various compositions. The findings for slagging of cobalt from alloy to slag were in poor agreement among the investigators. This may be because of experimental techniques which they used a n d / o r because of low concentrations of cobalt in equilibrium with liquid copper which caused inconsistencies in the analysis. This paper examines the removal of cobalt from liquid copper in the presence of Cu20-SiO 2 slags held in silica crucibles. EXPERIMENTAL METHOD The experimental technique adopted for this study has been described in detail. ]~ The starting materials

consisted of copper and cobalt of better than 99.97 pct purity, and of analytical grade cuprous oxide and silicic acid. The oxygen for the experiments was supplied by the cuprous oxide. Silicic acid was calcined at 1473 K for one hour. Experiments were done in a vertical tube furnace heated by silicon carbide elements with a 75 mm constant temperature zone at 1523 K. Equilibrium studies were done in silica crucibles with a neutral atmosphere of purified argon. The temperatures of liquid alloy and slag were measured by dipping a shielded Pt - (Pt + 10 pct Rh) thermocouple in the 20 ml crucible shown in Figure 1. Several experiments were carried out to determine time needed to reach equilibrium. The charge consisted of 30g (5 pct Co) of metallic charge and 10g (11 pct SIO2) of oxide mix. The results are presented in Fig. 2, from which it was decided that a two hour reaction time sufficed. Further experiments were done with a 30 g metallic charge consisting of Cu + (1 - 10 pct) Co, but with the same 10 g oxide charge, at 1523 K. Following

--13 1

A,eon

4

Flow

5 6

Fu+n~s Suppor~me Rod

~

;~12

Meter

7

Ineu[lting

8

Alloy 9 $1=~

9

Silir

Br+ck

C r ur

10 Alumm= Sheath 11

Mulllte

12

Coppar W=te,-9 End Jlcket B

5

Tube

13 Therm~ouple 9r

RAMANA G. REDDY, formerlyat The Universityof Utah, is now Research Associate, Dept. of Chem. and Met. Eng., Universityof Nevada,.Reno, NV 89507. GEORGE W. HEALY, Research Professor, Department of Metallurgy and Metallurgical Engineering, University of Utah, Salt Lake City, UT 84112. Man