Calcination – 1917, Mr H.W. Hutchin


This transcript of a lecture delivered to a meeting of the Cornish Institute of Engineers on September 17th 1917 by the President, Mr H.W. Hutchin, ARCSc, FGS MIMM. He was also Honorary Treasurer and had been since the formation of the Institute since its formation in 1913.
He said he had undertaken to introduce the subject of “Calcination” for discussion and he hoped much interesting and useful information would be forthcoming from those members who were engaged in applying Calcination to tin concentrates.

Calcination is a well-known operation in metallurgical operations; in particular it is an operation applied to intermediate tin concentrates in Cornish tin mines. The operation dates from great antiquity and may be best described as a distinct aid to tin dressing, as any mill foreman would demonstrate clearly in a few minutes on the vanning shovel with samples of roasted and unroasted concentrates. These remarks in the main are directed to explanations of the changes involved in calcination. It is unnecessary to discuss the various forms of calcining furnaces, except to point out that all are of one type, viz., the reverberatory type in which the fuel gases pass into, and over, the roasting hearth to a certain extent,

The effects of calcination may be best considered from two points of view: (1) chemical changes ; (2) physical changes The smelters’ requirements call for a high degree of purity in the final concentrate. Sulphur, arsenic, the base metals lead, zinc, copper, bismuth and antimony should be absent or of very low content. Moreover, a high silica content is not appreciated. These impurities may not be present in the intermediate concentrates of a given mine. Some rejoice in the possession of clean ore with little or no impurity. Others have perforce to treat less favourable material, rank or complex ores with compensations in the shape of valuable by-products.Between these two extremes there are other ores with impurities less pronounced, but calling for more special treatment than a clean ore.

The sulphide and sulpharsenide minerals undergo chemical changes when roasted in the presence of oxygen and for this reason it is essential to a successful roast that a full and abundant air supply should have access to the
roasting chamber and the escaping flue gases should contain from 8 to 9% of oxygen. Native copper, under these conditions would burn to oxide of copper. Chlorite and similar basic silicates undergo chemical changes not easily represented by chemical equation, but consisting in the splitting off of iron oxide. Of the sulphide and arsenide minerals there may be present :—iron pyrites, Fe2S (or marcasite); mispickel, FeS2.FeAs2; chalcopyrite, CuFeS2;  copper glance, Cu2S; zinc blende, ZnS ; and galena, PbS, are more or less

The  analysis of a sample of burnt complex concentrate reveals unburnt sulphur, and arsenic, sulphur in the form of sulphate of iron, copper, zinc or lead as the case may be, and some arsenic in the form of arsenate of the metals. Even when burnt sweet, i.e., to contain no unburnt sulphur or arsenic
there would still be a certain proportion of sulphur and arsenic present as sulphate and arsenate.

The tendency, however, is for the main reaction to take place in the direction of the elimination of the sulpher as the volatile gas, sulphur dioxide, e.g. with iron-pyrites:–
4FeS2 + 11O2  = 2Fe2 O3 + 8SO2

Similarly with mispickel:–
4FeS2 . FeAs2 + O  = Fe2O3 + SO2 = As2O3

with zinc blende;–
2ZnS + 3O2 = 2ZnO + 2SO2

with galena:–
2PbS + 3O2  = 2PbO + 2SO2

To a lesser extent the reaction may, and does, proceed in a different manner,
viz,  by a direct addition of oxygen to the sulphide molecule with the production of a sulphate :—
ZnS + 2O2  = ZnSO4
PbS + 2O2  = PbSO4
CuFe2 + 4O2 = CuSO4 + FeSO4
The reaction is more pronounced with zinc blende and galena by reason of zinc oxide and lead oxide being stronger bases. So too, if the concentrate contains strong bases like calcite or dolomite, the tendency towards fixation of sulphur a sulphate is increased. The production of sulphate begins in the region of low temperature and the sulphates produced undergo more or less decomposition towards the end of the operation in the hottest part of the calcination. Lead sulphate would not be affected but iron and copper sulphates would tend to change to oxides of the metals with elimination of sulphur trioxide (anhydrous sulphuric acid) which would pass away in the flue gases, e.g.,
Cu SO4  = CuO + SO4

Both sulphur dioxide and sulphur trioxide may be detected in the flue gases.
The presence of sulphates in the burnt concentrate is best illustrated by the pyritic copper minerals. The sulphate of copper produced is dissolved in the water used to wash the roasted concentrate and is readily detected by the deposition of copper on the shovel of the worker.

The arsenide minerals are similar to the sulphide, in that there is a tendency towards the formation of a higher oxide of arsenic, (As2O5) arsenic acid, with the result that the roasted concentrates contain arsenates of the metal but the
flue gases will contain no arsenic acid because at a temperature at which, say, ferric arsenate is decomposed, arsenic acid to decomposes into the lower oxide, viz, white arsenic (AS2O3.). The presence of strong bases in the concentrate will increase the tendency to form arsenates.

Important physical changes accompany the chemical reactions, the volume of the calcined pyritic minerals is increased and the density of the resultant oxides less than the original minerals. It is, as a consequence, less resistant to water, a fact which is of a distinct help in farther operations. Not only is there a change in density but the calcined impurities are less hard, a fact which is noted and appreciated by the assayer who has the preparation of roasted and unroasted concentrate for assay. The fact is also appreciated
in the supplementary operations of the tin yard.

There is one type of physical change induced by heat which tends to uneasiness of mind, vis., decrepitation. Decrepitation of the impurities in the calciner would be helpful, but the effect on the minerals cassiterite and wolfram would be disadvantageous. Information on this point is desirable.

There are some possible secondary chemical reactions which should have consideration. It is within my experience to receive samples of arsenic soot smelling strongly of hydrofluoric acid. There are two explanations of the circumstances but both depend upon the fact that some sulphuric trioxide (which, in the moist state, is sulphuric acid) is produced in calcination. On the one hand the flues of the arsenic chambers may contain stones of fluorspar in the walls, on the other hand the hydrofluoric acid may be generated in the calciner by the action of SO2 on the fluorspar contained in the concentrate. If the latter explanation is correct, it is a very undesirable constituent to come in contact with wolfram. I should expect wolfram subjected to hot hydrofluoric acid to suffer in density and to present circumstances tending to loss in further operations.

Calcination should not be conducted at too high temperatures. It is imperative that anything like fusion or fritting should not occur. Minerals of comparatively low fusibility are not infrequent in Cornish concentrates. Chlorite is most
common, and in some the presence of fluorspar would tend to make others more fusible. I have personally seen fritted fragments of rich concentrates, and the late Mr. J. J. Beringer called my attention to fused particles of cassiterite in a sample of black tin. The signs of fusion were clearly visible
under the microscope.  This aspect of calcination would not be complete without reference to the type of furnace used. Were the furnace of the muffle type, then fusion would be clearly due to excessive temperature alone. In the
reverberatory type the wall of flame may, and does, pass over the product, and is, in my opinion, an important factor. I have a suspicion, based on certain evidence, that it is an adverse factor. In one calciner coming under my observation the roof of the furnace was coated with oxide of zinc: the concentrate contained a proportion of zinc blende. For zinc oxide to appear on the roof of the furnace there must have been alternate reduction and oxidation, and is clearly evidence of the part played by the flame in the reverberatory type of furnace. In another instance, where the concentrate
contained carbonate of lime in the form of shell, the finished concentrate contained a small proportion of cassiterite soluble in hydrochloric acid. The effect of the flame on the mixture of shell and cassiterite had been to form calcium stannate, which is in accord with recent investigations in the assay of
tin ores. It would appear inadvisable to submit either cassiterite or wolfram to alternate reduction and oxidation for, even if there are no bases present to combine with the reduced product, the molecular condition of the. mineral will
have been destroyed and an alteration in density have resulted. I have endeavoured to trace changes of this kind in burnt leavings from different sources and obtained some evidence which support the view that cassiterite may be affected in calcination by the entrance of the flame, but unfortunately the matter was not carried to an incontestable conclusion. The present position is that a recommendation to use the muffle type of furnace could not be made with an assurance of benefits to follow. The strong point is that they would be best worked by producer gas and, where many calciners were in use, it might be more economical and the temperature more readily controlled.

The remaining point is a debatable one: with a clean ore would it be better to dress up to a very rich product, practically black tin, before calcination in preference to calcining a product containing about 25% metal as at present ?

I prefer to leave it as a question rather than to detail the
points for and against.
MR. H. R. BEEJNGER(sic) had concluded that each mine must work on a different basis because of the many differences existing in the ores mined in different properties. In some instances fuel might be in the ore and there might be no necessity for a huge fire. The Brunton calciner generally did very good work, though sometimes too high a temperature was reached and arsenic was lost. The more general use of the pyrometer was urgently needed. Heat should be moderated at will to ensure best results. It was not desirable to maintain uniform temperature throughout. A gradual increase in temperature should be aimed at.

MR. S, FURZE said that it was formerly thought that any man was good enough to take charge of a calciner.
They knew better. It was important work. The construction of a Brunton calciner was not always the same in detail. He had had occasion to draw plans for a new calciner. Crusts were formed under certain conditions. Where native copper occurred in the ore, it seemed impossible to have too high a temperature. Each mine had its difficulty. He had found it very difficult to regulate temperature in Brunton calciners.  He had experimented with various fuels and found that good bituminous coal generally gave the most satisfactory results.

MR. R. H. BERRYMAN experienced the greatest trouble in very rank arsenical ores. He would much like to find out the proper heat for complex ores containing arsenic, copper, tin and wolfram. The mispickel ignited. He
generally succeeded in burning completely off at first operation and got no crust.

MR. M. T. TAYLOR considered that calcination was not given that amount of attention it required. It was one of the most important operations in the treatment of Cornish tin ores. It was not possible to maintain uniform tem-
perature in a Brunton calciner. Ores changed in composition in different parts of the lode.   Fuel varied in quality. In experimenting he had sometimes observed thefusing of the finer particles; also that, with a low temperature, it was usual to get a good wolfram product, though more wolfram got into the tin concentrate. In increasing temperatures the quantity of tin in the wolfram increased to 3 or 4 per cent. It paid best to begin at a low temperature and
gradually raise it.

MR. JOSEPH BLIGHT said some calciners were operated under adverse circumstances because discharge was restricted. The products of combustion could not be got rid of fast enough. Difficulties of restricted discharge were sometimes further intensified to meet the requirements of Government
inspection; for example, where water towers were used.

MR. H. W. HUTCHIN had not always found low temperatures successful. He pleaded ignorance in the matter of correct temperatures for calcination. Temperature differed in different places in a calciner. A gradual increase might
save decrepitation.  Beginning at low temperatures and, later, raising them, also giving a full supply of oxygen, seemed to be quite correct practice.

A hearty vote of thanks was given to Mr. Hufchin for
his paper and it was decided to adjourn the discussion.


MR. H. W. HUTCHIN said he saw no necessity for repeating the substance of his remarks on September 17th.
He was glad to see such a general interest taken in calcination and was sure that the comparison of results obtained inthe operation by practical men would be very advantageous.

MR. HENRY BAILEY : Mr. Hutchin has performed a great service to the mining industry by initiating a discussion on a very important subject. There is no doubt that calcination has been dealt with in a very haphazard, hit-or-miss
fashion in some of our Cornish mines in the past. It is really an important metallurgical operation which requires much care and a knowledge of the chemical reactions involved.
Calcination has often been entrusted to an operator whose physical, or mental, incapacities rendered him altogether incompetent. Consequently the objects aimed at were either imperfectly attained or attained at an excessive expenditure of fuel and time.

The production of a perfectly sweet roast, free from sulphates and arsenates, with a minimum expenditure of fuel, is not a light undertaking when handling a mixture containing cassiterite, wolfram and complex sulphides. An ample supply of air, preferably heated, to every part of the charge, and a sharp rise in temperature when all the sulphides have been oxidised, are the important points to be aimed at. But it is often found that a high temperature alone is regarded as the the “all-in-all,” not from want of zeal on the part of the operators, but because the presence of an excess of air is not regarded as a necessity.
The heating of cassiterite, together with pyritic material, in an atmosphere in which an excess of oxygen is not present, or where the atmosphere is of a reducing nature, I have invariably found to produce stannous sulphide which,
being volatile at a red heat, is carried into the flues as vapour and is there oxidised to stannic oxide on coming into contact with any air which may find its way into the flues. I have recently protected a method for the separation of tin from its containing gangue, based on the experience outlined, large scale experiments having demonstrated that the chemical reactions mentioned will effect the removal of the tin, which can be collected as a bulky white powder, easy to smelt. The chief object in making these remarks is to warn those engaged in calcining tin ores, and to point out the necessity of carefully avoiding the presence, at any time, of a reducing atmosphere, or one in which there is not present an excess of oxygen.

CAPT. WM. THOMAS : The objects of calcination are not the same in all mines. Where tin is the only saleable product, concentrates are submitted to heat in order to facilitate the separation of tin from negligible quantities of
possible by-products and from what remains of the ordinary matrix. At such mines as South. Crofty, East Pool and Tincroft, wolfram, arsenic and occasionally copper are recoverable as well as tin. Indeed, if tin were the sole product, it is questionable whether those mines would pay to work. At Levant, wolfram is absent, but copper occurs in considerable quantities with tin and arsenic. Recently in the Alfred Mines, near Perranporth, and at Wheal Ellen,
Porthtowan, copper pyrites and zinc blende occurred in intimate association and, at least experimentally, the mixed ores were submitted to heat in order to increase the susceptibility of the pyrites to magnetic influences and thereby facilitate subsequent separation. Results on Wheal Ellen ore, using a small
reverberatory furnace and King’s magnetic separator were as follows :—

2 hours …… slow heat       ……     18 per cent. magnetic.

21/2 hours …increased heat  …..   30     „        „

3 hours  ……  red heat         …..     40     ,,        „

31/2  hours  ….red heat         ……  54     „        „

4 hours  ……  bright red heat ……     68     ,,        „

The Brunton calciner is not sufficiently supplied with oxygen when treating tin ores carrying a high percentage of sulphides. The men in charge of the calciners do not distinguish between temperature and combustion; and, finding ithing going wrong, are apt to stoke it up “thus innocently committing a folly”. Another defect in the Brunton when treating sulphide ores, is the inefficient stirring gear, the flukes. Portions of the charge are not brought into
contact with the heated air.

These defects are tacitly acknowledged in present practice; by roasting twice, incurring costs of extra fuel and handling, and by adding such substances as NaNO3 to the charge. But why buy solidified oxygen by the hundredweight
when it can be had in its free state from the air at little, or no cost? Some time ago it was suggested that air might be introduced into a Brunton calciner through inlet passages constructed through the brick-work of the fire-places. But a more effective plan would be to cast the three carriers supporting the flukes, hollow, with inlets from the air and outlets directed towards the revolving bed. A supply of heated air, direct upon the charge, would be automatically maintained, without any costs apart from those of construction. Experiments in this direction never got far. The air supply was introduced upon the coal fires instead of on the charge. The temperature was certainly
raised. This probably was not necessary. But oxidation of the sulphides was not hastened appreciably. The idea still awaits a real practical test. Under existing conditions, in mines producing sulphides with tin ores, too much fuel is consumed, and calcination is inefficiently done. To use an illustration from the laboratory, such ores require heating in an open tube, not in a closed tube!  A suitable calciner should perform the functions of an open tube. The Brunton calciners in our complex ore mines are not open tubes, they are just cracked closed tubes!   Somebody will, perhaps, suggest reverting to the Oxland calciner. That is certainly an excellent appliance for rapidly handling arsenical ores for arsenic; but, for well known reasons, the Oxiand is not
suitable for calcining tin ores.

Another point the President recently suggested that a good subject for informal discussion is magnetic separation. This subject is, locally, inseparable from calcination. Some years ago, when the Wetherill separator first came into prominence through its successful operation upon the zinc ores of New Jersey, a list of minerals was published with a scale of their magnetic susceptibilities. Heating in a calciner affects magnetic susceptibility. Has any systematic investigation been made locally to ascertain how the magnetic susceptibility of wolfram is affected by :—

(1) Duration of roast.

(2) Temperature.

(3) Size of particles roasted.

It is no secret that research has been stimulated of late.Important investigation is now being carried out.   The figures relating to losses of tin, published in Volume I, Transactions, have since been fully substantiated. No one will deny that losses in arsenic and wolfram are, at least, as serious as those in tin.   Every single pound of either commodity is well worth saving.

MR. M. T. TAYLOR: Being one of those who asked for an adjournment when calcination was previously the subject under discussion, I have endeavoured to collect a few facts conerning calcination effects upon tin and wolfram ores at East Pool Mine. Although I was able to get a register of the temperature at various points in the calciners and along the arsenic flues between the calciner and the stack, I am sorry to say that I was unable to get an apparatus to test the gases.

Starting with the calciner, the first test for a temperature registion was made immediately above the fire and gave 1340°F. The next test was in the centre of the revolving deck and gave a temperature of 1100°F. whilst, close to the flue on the outlet side, the temperature recorded was 1000°F. At 20 ft. along the arsenic flue away from the calciner, the pyrometer registered 600°F. and at 140 ft., the next point at in the test was made, 340°F. It is at this point that the deposition of the arsenic soot takes place. At a distance 240 ft. along the flues, a temperature of 297°F was registered and it is here that the best deposition takes place. At 700 ft. a considerable amount of moisture was observed on the outside of the test tube, the temperature registered being 118°F. Most of the arsenic soot was deposited before reaching this point. The next test was at the bottom of the outlet stack, which is 100 feet high, and the temperature at the base registered 80°F. so that the gases, from the time they leave the calciner, travel 1000 ft. through the flues before they are liberated in the atmosphere at the top of the stack. Up to 140 ft. away from the calciners, where the temperature registered 340°F., very little deposition took place but, from 140 to 240 ft., where the temperature ranged from 340°F down to 297°F., more than 90% of the arsenic soot wasdeposited.

Now, turning from the tests made under actual working conditions, the following tabulated statements on work carried out in the laboratory show the excessive amount of decrepitation that takes place in both wolfram and tin ores during calcination :—DECREPITATION OF WOLFRAM ORES. 1500°F.

Loss in
Burning.       —200     +200     +150     +100      +80       +60.

%            %        %         %         %         %       %

3.0            16.0   4.25       9.5              5             27.5      34.75

2.0            19.75   4.25       8.5          5        24.5           35.0

2.0            19.0     4.25       7.25           5.5         32.0        30.0

1.5            19.25    2.0         8.0        4.75        29.5        35.0

2.1            18.5        3.7      8.3            5.2        28.3     33.9 averages

2.5%        21%    4.5%   7.0%           5.5%       32.5%     27%


Loss in                                        *
Burning         —200.     +200.     +150.     +100    +80        +60.

%              %          %              %          %           %        %

1.5            2.5         .75        1.75     2.5        13         78
1.5            2.25         .75         2.0        2.5        12.5        80
1.5            2.25         .75        2.0         2.5        12.5        80
1.5            2.25      .75        2.0        2.5        12.5        80

15             2.25        .75        1.75        2.5        12.5        78.75 average

•5%            3.25%    .75%        2.0%    2.75%   12.75%    78%

Five tests were made on wolfram and four on tin ore. The testing temperature was 1500°F.  The original products for these tests were minus 40, plus 60, and it was found that the wolfram after calcination showed that 61.1% went through 60 mesh whilst 18.5 was minus 200. The temperature was then raised to 2000°F., the original products remaining the same, and the same ore samples were used, and the result was that 73% went through 60 mesh while 21% was minus 200. The latter product was extremely fine, resembling graphite, and on a vanning assay very little could be retained.

With the tin ore, the original products were the same as the wolfram, namely, minus 40, plus 60, and the samples were obtained direct from the mine, and had not been previously crushed or treated. The tin ore at 1500°F. showed that after calcination 21.25% passed through 60 mesh and 2.25 was minus 200 whilst, at a temperature of 2000°F, 22% passed through 60 mesh and 3.25 was minus 200.

In the wolfram treated at 2000°F very little sintering was noticeable to the naked eye, but with the tin product an excessive amount of fusing and sintering was in evidence.

The foregoing figures prove that calcination should not be carried out at a temperature above 1500°F. and that, the higher the temperature, the greater the decrepitation and, consequently greater losses and difficulties will be met with in the. later dressing operations. From the evidence so far obtained it looks as if a temperature between 1100 and 1300°F is a very suitable one for calcination.

Another point which should be borne in mind is that 1500°F is a suitable temperature for the volatilization of tin and, should sufficient chloride be present, considerable losses must take place through this medium.

MR. A. RICHARDS : If a temperature of 1400°F is exceeded there is a loss by volatilisation. Compared with the temperatures at which calcination is conducted in other parts of the world, the temperatures usual in Cornwall seem to be too high. Moreover, there is not a sufficient supply of
air going into the calciners.

MR. C. D. BARTLE : The observations on differences in temperature are very important. A great point in operating a calciner on Cornish tin ores is the adequate introduction of air, and, if a simple means of heating the air before its contact with the sulphides can be adopted, the results must certainly be beneficial.  The cost of experiments in this direction should not be excessive.

ME. E. H. DAVISON : Through microscopic examination of products I am able to confirm the statements made relating to to decrepitation. All my observations confirm the statements made by Mr. H. W. Hutchin.

CAPT. T. NEGUS : Effective calcination is a matter of business, and men in charge of the operation should be properly educated and trained to it.   I have observed in practice an appreciable escape of fine tin from the top of the calciner stack.   This may be partly checked by a water spray at the bottom of the stack. But the question is, need there be any such loss ? Doubtless better calcination results can be obtained if qualified men are selected to conduct the operation.

MR. S. FURZE : The President, in introducing the subject, said that it is very probable, when exceedingly high temperatures are reached, that reduction of cassiterite takes place, especially if crusts are formed, and I thought it worth
while to see if reduction takes place or not.

The calciner attendant was told to look out for crusts. A few were found and laid aside. One of these last crusts was’selected for examination. The outer portion was hard, forming a rough shell-like exterior, too hard to be crushed by the fingers, but powdered easily in a mortar. It appeared to be thoroughly burnt, no dark or unburn parts showing. This need not cause any surprise as the the crust was less than 3/4” diameter. After powdering in a mortar four
grams were taken and digested in 70 c.c. of strong HCl on the hot plate for about 20 minutes.  After diluting with about 100 c.c. water it was filtered and to the filtrate sufficient ammonia was added to precipitate the iron and tin, when it was again filtered and the residue carefully washed into a tin assay flask, HCl added, brought to boil, reduced with nickel coil, cooled and titrated in the usual way. In titrating 0.9 c.c. iodine were used, standard .510, equivalent to 2.295 Ibs. Sn per short ton.

An assay was made, without filtering or precipitating with ammonia ; that is, 2 grams were taken, digested in 60 c.c. HCl, reduced with nickel coil, and the usual practice followed. As might be expected, the liquor was somewhat turbid, but results were near the first, viz., 2.52 Ibs. Sn per short ton.

It may be asked,—why roast at such a high temperature? The answer is plain,—we have to. Those who are acquainted with Giew Mine know that a considerable amount of copper is found with the ore ; native copper, copper glance and some chalcopyrite. In the upper levels native copper is rather
common, the sulphides being lower down, but unfortunately the native copper is rather persistent and is often seen, even in the lower levels. The sulphides may be easily roasted to the sulphate, but the native copper presents difficulties. If the quantity of native copper were small so that, when
concentrated with the black tin, it would not exceed the second place of decimals in percentage, the trouble would be ended. Unfortunately this is not the case. Therefore roasting has to be regulated so as to deal with the native copper. Generally speaking all the sulphur is driven off in the calciner ; occasionally the shovels indicate the presence of copper sulphate ; this means that we roast our copper to an oxide.

Tin smelters have a strong objection to copper when present in black tin, therefore to get rid of the copper, and escape the smelters’ penalty, some form of acid treatment is necessary. As is well known, nitric acid will attack all forms of copper, but to use HNO3 for cleaning black tin is quite out of the question. Sulphuric acid, too, is ruled out these days but an effective substitute is found in nitre cake and this is used at the Giew Mill. The nitre cake is dissolved in hot water and is used as H2SO4 would be.

In one of the early years of the present working the black tin sold for the year averaged 1.07% Cn, whereas at the present time it is less than 0.03%.

More attention has been paid to the metallurgy of copper than that of tin ; consequently, something something may be learnt from what the metallurgist has accomplished in copper smelting. It is sometimes necessary to roast
ores before smelting them, and to this end large instalaltions of roasters have been laid down. The Washoe smelter of the Anaconda Copper Mining Company contains one of the largest roasting plants in the world. This plant consists of 64 McDougall furnaces contained in a single steel building 96 x 412 feet.  The furnaces are 18 feet 3 inches high and 16 feet external diameter. Each has 6 hearths,14 feet 6 inches diameter. The vertical shaft and rabble arms are water-cooled, each furnace requiring about 20 gallons forcooling water per min. to maintain the overflow temperatures at about 80°C.

These large roasting plants have been brought to a very high state of efficiency, both in operating costs and in reducing the sulphur contents of the ore. The following particulars have been taken from “Practice of Copper Smeltingby Peters. The first table is based upon temperature taken by inserting a pyrometer into the peep holes in the doors.

“Average temperatures of 21 Great Falls Meat Falls McDougall roasters taken by Mr. Pyne and of a considerable number of Washoe McDougal] roasters by Mr. Croasdale,

Hearth.         Great Falls Plant.       Washoe Pant.
No. 1               243°C.                      232°C.

2                     550                          538

3                     616                          621

4                     646                           738

5                 628                          678

6                 570                          649

It will be noted that the two batteries of roasters have approximately similar temperatures on the first three floors, while the Wathoe plant carries a considerable higher tem-perature during the remainder of the operation.

On May 10th, 1909, a series of observations was made at the Washoe smelter upon McDougall roaster No. 42. The furnace was running on ordinary feed—40 tons per 24 hours—with two doors open for admission of air on the lowest hearth, as customary. It appeared to be running a little hotter than usual, as the 12 hour barring out of the drop holes had just been completed.

A Chatelier pyrometer was used, connected for cold
junction temperature. Each result given is the average of
ten observations taken at half-minute intervals.

Ore.         Gases.
Uptake         ……              …..                 —             371°C.
Second hearth, outer edge       …..       610°C               —
Between second and third hearths            —             738
Third hearth, outer edge       ……           710         —
Fourth hearth, outer edge      ……            821          —
Between fourth and fifth hearths             —          771
Fifth hearth, outer edge        …..               788         —

Sixth hearth, outer edge        ……            705          —

On April 29th, similar tests were made on two other furnaces.
It is interesting to note the decided variations in tem-
perature occurring in furnaces which were running satisfactorily and uniformly, and which were all yielding a product containing about 71/2% sulphur.

Ore.            Gases
Uptake          …..                         —    —        338 377
Second hearth, outer edge              443   588        —    —
Between second and third hearths    —    —        638 749
Third hearth, outer edge                   538   593        —    —
Fourth hearth, outer edge              732   693        —    —
Between fourth and fifth hearths         —    —        649 705
Fifth hearth, outer edge                    732   710        —    —
Sixth hearth, outer edge                616   627        —    —

The length of time required for the ore to pass completely through one of these six hearth furnaces depends upon the speed of the stirrers, the angle of the plough blades, the size of the ore particles and various other less obvious
factors. It ranges from 21/2 hours to 4 hours, being about 31/4
hours under the conditions at Anaconda.

A somewhat elaborate experiment was made at the Washoe smelter to determine the relative rapidity of advance of the particles of different sizes. I have been furnished with the following results :

Two furnaces, each treating about 35 tons (31.74 met. tons), dry weight, of ore per 24 hours, and stirred every 35 seconds (one complete rev. every 70 seconds) gave results as follows :

Size of Particles.
Between         4.1 & 6.1. mm.   1 & 1.5 mm.
Average time in furnace               165 mins.        190 mins.
Minimum time in furnace               75 ,,           60 ,,
Maximum time in furnace              270 ,,          375 ,,

The operation of these furnaces was described in detail, particular attention being drawn to the fact that the heat of combustion was maintained by the sulphur contents of the ore; no carbonaceous fuel being used, only in the case of restarting a roaster from the cold or restoring the temperature to a bed that had by some means dropped below normal.

The following analyses are given :

Ore :—                Percentage

Moisture        ……       8.10.
Cu                  …            7.42
SiOs             …..          21.20
Fe               …..          26.05
S                  ……           33.17
AL,03            ……           2.70
CaO             …..           0.30

Gold = 0.018 ounces and silver 4.87 ounces per ton.
To this material is added 5 per cent. of its weight in limestone of which the largest piece does not exceed 1 inch dia.

Calcines :—
Cu            ……          9.32 per cent.
SiO,          , ……       28.07
FeO           ……       38.42
S                  ……     8.01
Al2O3           ……     4.38
CaO           ……        3.01

Ag per ton      4.92 oz.
Au         „         0.019 oz.

Total cost of roasting a ton of ore at these works, inclusive of supplies, renewals, management, and general expenses, is said to be a little less than 30 cents per ton (33 cents per met. ton). Average output per furnace per 24
houre 40.53 tons (36.76 metric tons).”

MR. J. WARNE CHENHALL : The substances causing the greater part, perhaps over 90 %, of the chemical action in furnaces include copper pyrites, bornite, covellite, iron pyrites, enargite, arsenical pyrites, sphalerite, hematite and other iron oxides, together with alumina, silica and limestone.  The Anaconda general practice is that all concentrates under 5 mm. go into the furnace, making about 90% of the charge, the remainder consisting of fine ore and fine limestone. The temperature is insufficient to break up the
limestone into CaO and CO2. The concentrates contain about 33% sulphur and it is necessary to roast out about 75% of the S to produce a suitable product for reverberatory smelting. The concentrates may be considered as a mixture of the gangue “with chalcopyrite, bornite, iron pyrites and enargite,
the latter containing the arsenic. On the first hearth the concentrates are dried. On the second the roast begins with the action upon iron pyrites.   The temperature is gradually increased oil the succeeding hearths. If too much air is admitted, the charge turns red due to oxidation of FeO into Fe2O3.
Enargite breaks up into CuO, As2O3 and SO2, but all the As is not expelled. A portion goes with copper to form an arsenite or arsenate which goes on to the reverberatory furnace where some of the remaining arsenic goes into slag and the remainder into matte. The following applies, more or less, to the roasting of flue dust in arsenic furnaces.

Arsenic or antimony may be present in ore to be roasted, often in the form of sulphides, or as arsenides or antimonides of other metals and sometimes in very complex mineral form. The sulphides of arsenic, As2S2 (realgar) and
As2S3 (orpiment), are readily volatile, but not to the same extent, as the antimony sulphides. Some of the arsenic and antimony is eliminated in the early stages of the roasting by the formation of these sulphides, aided by the distillation of sulphur from the pyrites present. The lower oxides of these elements form As2O3, which is readily volatile at 218°C or 424°F, and Sb2,O3 which is volatile only at a low red heat.
Considerable portions of the arsenic and antimony may be eliminated in this way. Further oxidation, however, changes these oxides into the higher oxides As2O5 and Sb205, which, when present alone, are again readily dissociated at a full red heat, but in the presence of certain base metal oxides, such as
iron and copper, are converted into very stable arseniates and antimoniates of these metals, which will persist as such in the roasted material. In the process of roasting, the ore particles are alternately on the surface of the ore bed subject to an oxidizing atmosphere, and submerged among partially
roasted particles from which sulphur is distilling. This sulphur vapour and SO2 gas will reduce arseniates and antimoniates and again cause the volatilization of sulphides of arsenic and antimony and their lower oxides. The arsenic and antimony are therefore more readily eliminated by alternate oxidation and reduction. A roasting furnace may be specially operated in this way, by methods of firing, when much arsenic or antimony is present.

The last paragraph, suggests the possible recovery of additional As2O3 by roasting with pyrites. Several roasts were made with varying amounts of pyrites to see if a better recovery could be obtained. Residues of high temperature roasts were mixed v/ith 20% of pyrites and samples taken of
mixture were roasted at various temperatures up to 1600°F. Samples were rabbled to reproduce furnace conditions. There was decrease in weight but the total amount of arsenic remained undiminished. The conclusion reached was that arsenic in flue dust, when once “burned in,” is not again broken up at temperatures below 1600°F ; also that about 30% of the arsenic is present as compounds stable at temperatures below 1600°F.

The best temperature indicated for furnace operations is 950°F. At this temperature the As2O5 and As2S5 break down and ASgOg is volatilized. Only a small amount of SO2 is liberated and the metallic oxides have a very low vapour pressure at this temperature.

Mr. HUTCHIN, acknowledging a hearty vote of thanks, said he was satisfied that he had been instrumental in initiating a very interesting and profitable discussion, in which those members most experienced in calcination operations had freely and fully related their experiences for the benefit
of one and all.