Different parts of the world in which mineral separation by flotation is practiced experience vastly different natural temperatures, and in some districts there is a big difference between summer and winter temperatures. It will be shown in this paper that the temperature is of considerable importance where it is desired to separate pyrite or arsenopyrite from other minerals. It is probable that some of the differences obtained in the flotation of these minerals at plants working under otherwise similar conditions are due to temperature variations. The influence of temperature on the flotation of galena, chalcopyrite and sphalerite by xanthates is less marked than its influence on pyrite flotation.

Experimental Method

The methods used have been described previously. In each test the clean mineral specimen was immersed in a solution prepared by adding the reagents in the following order:

1. sodium carbonate and hydrochloric acid as necessary to buffer the solution and to give approximately the pH value desired,
2. sodium cyanide,
3. copper sulphate,
4. ethyl xanthate, followed by
5. a final adjustment of pH value.

After ½ hr. the mineral was tested for a xanthate film by bringing a bubble of air into apparent contact with the submerged polished surface. If true contact (spreading) was not obtained within ½ min. and after cleaning the surface by wiping with a small linen pad, it was concluded that the adsorption of a xanthate film had been prevented. Following the method already described (ref. 2), curves have been constructed such that for a solution composition corresponding to any point below the curve, contact between mineral and air is possible, whereas for a solution composition corresponding to any point above the curve, contact is impossible.

The temperature was controlled to within 1° C. of the specified value, not only during the period of immersion of the specimen in the solution, but also while the solution was being prepared and while the surface of the mineral was being tested.

Experimental Results

Ethyl Xanthate as Collector—No Added Copper Sulphate.— Fig. 1 shows for 10° C. the relationship between the pH value and the concentration of cyanide necessary to prevent contact between an air bubble and surfaces of chalcopyrite, pyrite and galena respectively. Fig. 2 shows the corresponding relationships for 35° C. As there is no activator present, sphalerite does not respond to xanthate at any point within the diagram.

Ethyl Xanthate as Collector—Copper Sulphate Present.—Figs. 3 and 4 show, likewise, the relationship between the pH value and the concentration of cyanide necessary to prevent contact at 10° and 35° C., respectively, for pyrite, chalcopyrite and sphalerite in the presence of 150 mg. CuSO4.5H2O per liter and with the standard concentration (25 mg. per

liter) of ethyl xanthate. Fig. 5 shows the effect on the curves for pyrite and sphalerite of lowering the xanthate concentration to 5 mg. per liter, the temperature being 35° C.

At pH = 9.0, 70 mg. per liter of cyanide is necessary under these conditions to prevent contact at 35° C. at a sphalerite surface. It has been found previously (ref. 3, p. 276, curve 2, Fig. 6) that at room temperature a lower concentration (30 mg. per liter) suffices. For higher xanthate concentrations the effect of raising the temperature to 35° has been found to be negligible.

Fig. 6 shows, for a single pH value, namely 9.0, the relationship between the temperature and the concentration of cyanide necessary to prevent contact at the surface of pyrite. From the shape of the curves

for pyrite in Figs. 3 and 4, it is apparent that the curve of Fig. 6 would need only slight modification for any pH value between 7 and 11.

The results of direct flotation tests with pyrite, carried out in stoppered cylinders, support the conclusions of the contact tests. The solutions used contained 25 mg. per liter potassium ethyl xanthate as collector, 20 mg. per liter terpineol as frother and 150 mg. per liter of CuSO4.5H2O. With 60 mg. per liter NaCN and at a pH value of 8.2, no flotation could be induced at 10° C., but at 35° C. flotation was excellent after a short conditioning period. With a more alkaline solution (pH = 10.0) and 5 mg. per liter of NaCN the results were similar. At 10° C. no flotation was possible even after a conditioning period of a day: the solution frothed

well, however. When the solution was warmed to 20° C., fair flotation was obtained within half an hour. When the solution was again cooled, flotation of the pyrite ceased and when heated to 35° C. flotation was restored. Subsidiary tests showed that low temperature cannot by itself prevent contact: alkali and cyanide are also necessary.

Amyl Xanthate as Collector—Copper Sulphate Present.—Fig. 7, which should be compared with Fig. 4, shows, for pyrite and sphalerite, the effect of substituting for 25 mg. per liter of ethyl xanthate an equivalent quantity of amyl xanthate, the temperature being 35° C. The corresponding curve for sphalerite at 10° C. was also determined: it proved to be practically identical with the curve for sphalerite (hot preactivated) previously determined at room temperature (ref. 3, p. 282) and thus lies somewhat lower than the curve of Fig. 7.

Contact Tests on Arsenopyrite

The decision to investigate the effect of temperature in relation to air-mineral contact arose from an attempt to determine, at room temperature, the usual cyanide pH value curves for arsenopyrite. It happened that the temperature fluctuated severely from day to day and results could not be duplicated. As soon as the temperature was controlled, however, results became reproducible. The results for 35° C. determined as described above (p. 3), indicated that the curve for arsenopyrite is almost identical with that given in Fig. 4 for pyrite. At pH values below 10 the curves for the two minerals are identical; at pH values above 10, arsenopyrite is a little more sensitive

to depression, but the difference is very small. The arsenopyrite curve for 10° C. was not determined accurately, but we have found that it differs little, if at all, from that given in Fig. 3 for pyrite.

In the absence of copper sulphate, contact between an air bubble and an arsenopyrite surface seems to be wholly unaffected by cyanide. In a 25 mg. per liter xanthate solution the critical pH value, 7.0, was unchanged by cyanide additions. Galena is the only other mineral with which we have experimented that behaves similarly in this respect. This critical pH value was determined at room temperature before the need for accurate temperature control was realized; in the absence of copper sulphate, however, there is less need for temperature control (see Figs. 1 and 2).

Discussion of Results

In the absence of copper sulphate, adsorption of xanthate is more easily prevented by alkali or by cyanide at 35° C. than at 10° C. There is, in fact, for each of the three minerals of Figs. 1 and 2, an increase in the critical pH value as the temperature is lowered, and there is likewise an increase in the critical cyanide-ion concentration.

The effectiveness of cyanide as a depressant for pyrite is amazing. For example, though contact is easily established in the absence of cyanide in a 25 mg. per liter ethyl xanthate solution at a pH value of 9, as little as 0.1 mg. per liter of sodium cyanide prevents contact. These tests were carried out in 20 c.c. of solution: consequently 0.001 mg. NaCN was sufficient to prevent contact with a pyrite specimen of which the total surface area was approximately 6 sq. cm. Direct flotation tests, carried out in stoppered cylinders, confirmed the extreme sensitiveness of pyrite to cyanide, but slightly more cyanide was necessary to prevent flotation. At a pH value of 9, flotation of pyrite was prevented by 0. 5 mg. NaCN per liter when using 25 mg. per liter ethyl xanthate as collector and 20 mg. per liter terpineol as frother. Flotation was hindered but not completely prevented by 0.1 mg. per liter. We had hoped to study the abstraction of cyanide by pyrite but have been unable to find a sufficiently sensitive method of analysis. Its depressant action on pyrite, in fact, is the most sensitive test known for cyanide. We had hoped, also, to show by analytical methods that the depressant action of cyanide is due to its preventing adsorption of xanthate by the pyrite. The small amount of cyanide necessary to depress the pyrite, though it would consume a little iodine, was not expected to prevent the estimation of xanthate by iodine titration. However, this method proved unsuitable because of the liberation of considerable amounts of reducing ions by the alkali necessary to raise the pH value to such a value that small concentrations of cyanide are effective. No satisfactory method could be found for the estimation of xanthate in the presence of these ions or for its separation from them.

Nevertheless by taking advantage of the fact that xanthate is an excellent collector, even when in very low concentrations, it was shown that alkali and/or cyanide greatly reduced, and probably completely prevented, the abstraction of xanthate by pyrite. Four grams of pyrite was ground for 5 min. in a porcelain mortar with 50 c.c. of a 25 mg. per liter ethyl xanthate solution containing 10 mg. NaCN per liter, and at a pH value of 10. After standing for ½ hr. in contact with the very finely divided mineral the solution was filtered, and both the mineral and the solution were tested for xanthate. The mineral was washed and then placed in a 100 mg. per liter solution of pure m-cresol which is a frother having no collecting properties for pyrite. Over a range of pH value from 8 to 4 it showed no tendency to float, except for a surface film of the finest material. The solution was brought to a pH value of 7, and after the addition of 100 mg. per liter of m-cresol, it proved to be an excellent flotation medium for some freshly ground pyrite. A control test without the alkali and cyanide showed that pyrite abstracts so much of the xanthate that the filtrate was unable to cause flotation of freshly ground pyrite whereas the original pyrite, when washed and placed in a cresol solution, floated excellently. It may therefore be concluded that the depression of pyrite by alkali and/or cyanide is due to their ability to prevent adsorption of the xanthate by the mineral.

From Figs. 1 and 2, it is apparent that the differentiation between pyrite and chalcopyrite is less at 35° C. than at 10° C. Figs. 3 and 4 show that this is also true in the presence of copper sulphate. It follows that from this aspect the separation of chalcopyrite from pyrite should be carried out at as low a temperature as is consistent with plant requirements, bearing in mind that at the lower temperatures the speed of the conditioning processes may be considerably reduced.

In the presence of copper sulphate and cyanide, temperature changes have a profound influence on the response of pyrite to xanthate. At low temperatures the copper sulphate-cyanide solution acts as a depressant and an amazingly low concentration of cyanide is effective. The copper itself is not a depressant, for, in the absence of cyanide, the critical pH value is raised by it from 10.9 to 11.3. Nor is the cyanide ion responsible for prevention of contact, since at the lower pH values copper sulphate decreases the amount of cyanide necessary to prevent contact. A complex copper cyanide ion, therefore, seems to be the active depressant for pyrite at 10° C. At 35° C., however, copper sulphate is an activator for pyrite and the curve of Fig. 4 for pyrite does not differ greatly from the corresponding curve for sphalerite. It is apparent, therefore, that under the conditions of these experiments the pyrite-chalcopyrite and pyrite-sphalerite separations are more easily effected at low temperatures. On the other hand, if pyrite or arsenopyrite has to be floated in a circuit containing cyanide, an increase in temperature, should be beneficial.

The main effect of raising the temperature to 35° having been to bring the curve for pyrite much closer to the curves for sphalerite and chalcopyrite, it was considered advisable to ascertain whether by raising the temperature still further, or by modifying the relative amounts of the reagents, the curves for the three minerals would become identical. However, we have found no conditions for which the curves are identical, nor have we obtained any evidence of the existence of such conditions.

It has been found previously that the amount of cyanide necessary to prevent contact with a sphalerite surface at any given pH value depends upon the ratio between the xanthate and copper sulphate additions (ref. 3, p. 275). The effect on the relative positions of the pyrite and sphalerite curves of varying the xanthate concentration, while keeping the copper sulphate addition constant, was therefore determined. Fig. 5 shows that 5 mg. per liter xanthate is insufficient, even at 35° C., to bring the curve for pyrite into the approximate region of the sphalerite curve; and, consequently, that even at this temperature the separation of sphalerite from pyrite would be possible by using minimal or “starvation ” quantities of the collector. Conditions favorable for this separation, when copper sulphate and ethyl xanthate are used, are: (1) low temperature, (2) low xanthate concentration, and (3) a pH value between 6 and 8 if low temperature is impracticable. Low temperature and low xanthate concentration are not without effect on sphalerite flotation, but their influence on pyrite flotation is greater and therefore should be of value in selective flotation.

Fig. 6 shows that beyond 35° C. a further increase of temperature does not greatly increase the amount of cyanide necessary to prevent contact at a pyrite surface. At pH = 9 the maximum concentration of cyanide that can be tolerated without preventing contact is just under 80 mg. per liter at about 40° C. Fig. 4 indicates that at constant temperature the amount of cyanide that can be tolerated is almost independent of the pH value within the limits 7 to 11. It seems certain, therefore, that there is no temperature or pH value at which an amount of cyanide substantially in excess of 80 mg. per liter could be tolerated. With chalcopyrite, however, Fig. 4 shows that over 100 mg. per liter can be tolerated and with preactivated sphalerite (ref. 3, p. 276), 90 mg. per liter. We have not found any conditions under which pyrite can be preactivated by copper sulphate. However, for sphalerite and for pyrite that have not been preactivated, the maximum cyanide concentration is about 80 mg. per liter, and does not seem to differ very greatly for the two minerals. It has been found also that under one set of conditions the curves for sphalerite and stibnite are almost identical.

It is of interest that whereas an increase of temperature from 10° to 35° raises the pyrite curve enormously, and the sphalerite curve slightly, it actually lowers the chalcopyrite curve slightly. However, as the curve for 35° C. is practically, identical with that previously obtained at room temperature, it seems that there would be little further decrease at higher temperatures.

Thus the effect of temperature change on the chalcopyrite curves is in the same direction in the presence as in the absence of copper sulphate: this suggests that the cyanide concentration itself is the controlling factor, through its ability to prevent adsorption of xanthate. On the other hand, where an increase in temperature raises the curve (or moves it to the right), for example, with pyrite and sphalerite, it is suggested that the function of the cyanide is to control the copper-ion concentration and thus to control activation. Activation must precede adsorption of xanthate and the copper-ion concentration necessary for activation probably differs from mineral to mineral.

It had been found (ref. 3, p. 282, Fig. 10) that at room temperature with solutions containing amyl xanthate and cyanide, copper sulphate is not such an effective activator for sphalerite within the pH range 8 to 12 as on either side of this range. When it was found that copper sulphate became more effective as an activator for pyrite as the temperature was raised, it was decided to ascertain whether raising the temperature to 35° would result in any change in the sphalerite-amyl xanthate system. Comparison of the sphalerite curve of Fig. 7 with that of the earlier paper (ref. 3) indicates that though there remains a small region of poor response to xanthate, this region is much smaller at 35° C. The corresponding curve for pyrite and amyl xanthate is included in Fig. 7. At this temperature cyanide is not an effective depressant at pH values above 8, whereas at room temperature (ref. 3, p. 282, Fig. 9) a few milligrams per liter of cyanide effectively prevented contacted at any pH value.

The results of the present investigation clear up some of the difficulties of previous work. Thus it was stated (ref. 2) that difficulty was experienced in fixing the lower left-hand portions of the pyrite and chalcopyrite curves, and that the former curve may have been 0.5 pH units in error. No such difficulty was experienced in determining the curves of Figs. 1 and 2, and it is probable that the earlier difficulty was due to lack of temperature control. It had been found (ref. 3, p. 287, Fig. 8) at room temperature that with 625 mg. per liter of ethyl xanthate present copper sulphate was within limits an activator for pyrite, as much as 40 mg. per liter of cyanide not preventing contact at pH = 9. This fact did not seem to fit in with the results for lower concentrations of xanthate and was inexplicable at the time. It is now apparent that the higher the xanthate concentration and the higher the temperature, the more likely is pyrite to float in the presence of copper sulphate and cyanide.

It had been shown (ref. 1) that the magnitude of the angle of contact is uninfluenced by temperature variations. The tenacity of air-mineral contact, therefore, is uninfluenced by temperature; provided, of course, that the temperature change does not prevent contact altogether.

The reasons for these effects of temperature on the formation of the xanthate film can only be considered in the broadest possible way. The factors that determine the direction and magnitude of any change of equilibrium with temperature are the heats of reactions of the various reactions involved. Since, however, there is not yet complete agreement concerning the nature of the reactions that occur, and moreover, we do not know the values of the heats of reaction (or adsorption) of the reactions in question, no progress can be made along these lines.

Summary

A study has been made of the influence of temperature on adsorption of xanthate at surfaces of galena, pyrite, arsenopyrite, chalcopyrite and sphalerite, and on the effects of alkali, cyanide and copper sulphate in hindering or promoting the adsorption. The development of a definite air-water-mineral contact angle has been used to indicate the adsorption of a xanthate film on the mineral surface.

1. In the absence of copper sulphate, a change in temperature from 10° to 35° merely alters slightly the amounts of depressants (cyanide and/or alkali) necessary to prevent contact, a greater concentration of depressant being necessary to prevent contact at 10° than at 35° C.
2. In the presence of copper sulphate, the influence of temperature on the amounts of cyanide and/or alkali necessary to prevent contact is slight for chalcopyrite and sphalerite, but is relatively great for pyrite and arsenopyrite. Under these conditions pyrite and arsenopyrite are much less influenced by the depressants at 35° C. than at 10° C.
3. The conditions most favorable for floating sphalerite away from pyrite are low temperature, low xanthate concentration, and if low temperature is impracticable, a small range of pH values which for the conditions used here is between 6 and 8.
4. The conditions most favorable for floating chalcopyrite or other copper-bearing sulphide minerals away from pyrite are likewise low tem-perature and low xanthate concentration.
5. There appears to be little advantage in raising the temperature of the system above 35° C. when pyrite or arsenopyrite is to be floated in the presence of copper sulphate and sodium cyanide.
6. The differentiation between sphalerite and pyrite when using amyl xanthate in a circuit containing copper sulphate and cyanide is much diminished as the temperature is raised.