Page images


DISCUSSION OF THIS PAPER IS INVITED. It should preferably be presented in person at the New York meeting. February, 1916, when an abstract of the paper will be read. If this is impossible, then discussion in writing may be sent to the Editor, American Institute of Mining Engineers, 29 West 39th Street, New York, N. Y., for presentation by the Secretary or other representative of its author. Unless special arrangement is made, the discussion of this paper will close Apr. 1, 1916. Any discussion offered thereafter should preferably be in the form of a new paper.

The Behavior of Stibnite in an Oxidizing Roast


(New York Meeting, February, 1916)

The leading antimony mineral is stibnite. In smelting stibnite ore two processes are available, precipitation and roasting-reduction. The former is suited only for high-grade ores. As low-grade ores are more common than high-grade, roasting-reduction is of greater importance than precipitation. In the roasting process the aim may be to leave the oxidized antimony in the ore, or it may be to volatilize as much of the antimony as possible, collect the volatilized oxide as a rich intermediary product and smelt it for antimony, leaving the gangue poor enough to be considered a waste product.

Whichever way the roast is conducted certain difficulties inherent in stibnite are encountered. These are:

1. The low melting point of stibnite which, according to Pélabon,' is 550°C., according to Wagemann? 540°C., and according to Borgström3 546°C.

2. The ignition temperature: According to Friedrich, stibnite, if heated in air, begins to oxidize at 290°C. if the size of a grain is 0.1 mm. in diameter; at 343°, if 0.1 to 0.2 mm.; and 430° if 0.2 mm.

3. The fusibility of a mixture of Sb2S3 and Sb203 which in the form of kermesite (Sb2S3)2.Sb203 melts at 517°C.5

4. The volatility of Sb2S3 and Sb203, for which no numerical data appear to exist, although practical experience has shown that they are volatile at low temperatures.

As regards the oxidation of metallic antimony, we have the experimental evidence of C. F. Plattnerm that when fused and brought to a red heat, it burns with a bluish-white flame to Sb203 which passes off as a

* Professor of Metallurgy, Massachusetts Institute of Technology.
† Analytical Chemist and Metallurgist.
1 Comptes rendus, vol. cxxxvii, p. 920 (1903).
2 Metallurgie, vol. ix, p. 518 (1912).
3 Chemical Abstracts, vol. ix, p. 2364 (1915).
* Metallurgie, vol. vi, p. 169 (1909).
5 Borgström. Loc. cit.
Die Metallurgischen Röst processe, Engelhardt, Freiberg, 1856, p. 162.

whitish fume, and that Sb20, can be further oxidized to Sb20, or Sb,05 If Sb2O4 is considered to be antimonious antimonate, the compound falls in line with metallic antimonates which are formed if a metallic antimonide or sulphantimonide such as pyragyrite, tetrahedrite, or jamesonite, is subjected to an oxidizing roast-or if Sb20, in contact with a finely divided metallic oxide is heated in a current of air.

The chemical change which takes place in roasting stibnite is usually expressed by the equation

Sb2S3 + 90 = Sb2O3 + 3802 but the Sb20, is further oxidized to Sb204. In carrying on a roast, the temperature of the ore is held at first at about 350°C., and the charge is rabbled more or less continuously to prevent or correct caking; later, when part of the Sb2S3 has been converted into Sb203 and Sb2O4, the temperature may be raised with advantage. Roasting changes the black sulphide into a yellowish-white oxide.

The aim of the present investigation was to study the changes stibnite undergoes when roasted at different temperatures; to determine the composition of the roasted product; and to ascertain the losses which occur.

The material used was crude stibnite of unknown origin from the metallurgical collection of the Massachusetts Institute of Technology. It was a compact grayish crystalline mass showing fine prismatic needles. The fresh fracture appeared silver-white and had a metallic luster. The general appearance was not unlike that of graphite. After grinding, it had a jet-black color. A chemical analysis gave 71.66 per cent. antimony and 0.96 per cent. insoluble. Calculating the antimony as the sulphide gives 100.3 per cent. Sb2S3; the excess over 100 is probably due to the presence of a small amount of metallic antimony.

The roasting furnace used was of the electric resistance type. It consisted of a porcelain tube, 10 in. long and 234 in. in diameter, wound with "nichrome" wire; this was surrounded by mineral wool, and the whole inclosed in a pipe of galvanized iron, 742 in. in diameter. Temperatures were measured with a thermo-electric pyrometer.

The ore was ground to pass a 20-mesh sieve; a 5-gram sample was used, as this covered the bottom of the porcelain boat and filled it to about one-third of its depth.

During a roast the ore was frequently rabbled with a wire bent to the form of a hook. A piece of moistened litmus paper held above the surface of the ore served as a test for sulphur dioxide. Any odor, or change in the appearance of the ore was noted. The temperature of the furnace was raised gradually to the point at which it was desired to finish a roast, and held there until no more changes in appearance and weight of the charge occurred.

Seven experiments were carried out, covering a range of finishing temperatures from 337.5 to 545.4°C. Experiment No. 3, in Table I, shows the manner in which the records were taken.

[blocks in formation]

A summary of the observations made during the six roasts is given in Table II.

TABLE II.--Summary of Observations in Roasting Experiments


°C. 196 219 281 336

Faint odor of SO2.
Litmus begins to turn pink.
Ore becomes dull; odor of SO, very distinct.
White ashes form on rabbling, and some Sb,03 is

Lowest temperature at which odor of SO, disap-

Ore becomes sticky if temperature is raised




The first observation made in each roast was the evolution of sulphur dioxide; this was followed by a change in the color of the litmus paper at a slightly higher temperature. At 280°C. the ore began to lose its luster and became dull. Then followed a strong odor of sulphur dioxide which directly preceded a sudden visible formation of Sb20, at 336°C. The oxide forms spontaneously at this temperature in the form of little yellow grains which become white upon cooling. As the temperature is raised above 336°C., the grains seem to break up into a grayish powder. If the temperature is raised too quickly before all the sulphide has been changed into whitish grains of oxide, the charge becomes sticky; this is not the case if the temperature is raised gradually. In roasts Nos. 6 and 7 a small amount of Sb203 was deposited on the rim of the heating tube, but the temperature and time at which the deposit was formed were not noted. Tables III, IV, and V contain data on the roasting experiments and the results obtained.

TABLE III.-Summary of Roasts



Raw Ore,



Loss of

Loss of
Per Cent.

Loss of

Sb, Per Cent.







None None 2.21 5.60* 2.93 4.21

* This excessive loss in weight is due to some experimental accident which cannot be definitely placed.

[blocks in formation]

TABLE V.Percentage of Antimony Compounds in Roasted Products

[blocks in formation]

A graphical representation of the results, given in Fig. 1, shows clearly the changes which stibnite undergoes when it is roasted. No change takes place until the gradual rise in temperature has reached 200°C. when SO, begins to be given off. The quick fall of the Sb2S3 curve shows that beginning with 200°C. a rapid oxidation takes place which is terminated at 400°C. It is fair to assume that the formation of Sb2O3 commences at the temperature when SO, is first detected. The Sb2O3 curve is therefore started at about 200°C; it has a rapid rise which corresponds to the quick fall of the Sb2S3 curve. Its highest point coincides on the temperature scale with the point at which the oxidation of Sb2S, is complete; the Sb2O4 curve starts at the same temperature. As the temperature rises, the Sb2O3 curve falls and is accompanied by a corresponding rise in the curve for Sb2O4. From this it is seen that in roasting Sb2S3 there is formed at first only Sb203, and that this is converted into Sb2O4 only when all the Sb S, has been oxidized. Thus the amount of Sb2O4 formed is dependent upon the temperature and the time of the roast. It is as-, sumed that there is no loss of weight below the temperature at which S02 begins to be evolved, i.e., about 200°C. From this point on, the loss in weight increases with the temperature and the formation of Sb2O3; it reaches a maximum at about 450°C. then falls and afterward becomes

« PreviousContinue »