From this will be seen that cyanogen bromide can only be described as a rapid cyanicide, and great care must therefore be exercised in the quantity which is used, otherwise such a great proportion of the alkaline cyanide will be decomposed that the extraction will be poorer than could be obtained by using the same amount of alkaline cyanide alone.

The compound cyanogen bromide solvent is a very powerful one, but as we have shown it is a very short lived one even under the most favourable conditions, and in order to use it to the best advantage, the haloid compound should be added almost continuously in the same way as fresh oxygen bas to be added to plain cyanide solutions. For this reason when treating high grade ores it has become usual to divide the cyanogen bromide into several portions, which are added at intervals of three or four hours during the agitation.

Nature of Ore Treated.-Owing to the instability of the solvent it is also necessary that the gold which is to be dissolved must be thoroughly exposed to it; in other words, the ore must be ground so finely that the gold particles are exposed to the solvent on practically all sides. If we imagine a flake of gold enclosed between the cleavage planes of pyrites so that only one edge of it is exposed, the solvent will have such a small area for attack in comparison to the weight of the particle, that it will have lost its power before all the gold has dissolved, and in such an instance a solution of plain cyanide would give a much better ultimate result, although it would take very much longer to accomplish. This is why cyanogen bromide has never been successfully used in the treatment of sand or concentrate by percolation; and also why no ore has yet been successfully treated which has not been ground to slime. In this instance we would describe slime as a product passing through a 150 mesh sieve (I.M.M. Standard), the proportion of colloids present being quite

immaterial.

of

When the cyanogen bromide process was first introduced, a great deal was written about the possibilities treating refractory gold ores, such as those containing arsenic, antimony, and copper, but no records exist to show that these hopes were ever realised. As ores of this nature usually have their gold value contained in the mineralised portion it is easy to see now why so many failures took place. Ore which is to be prepared for treatment by this process must first have all gold which is too coarse to dissolve rapidly removed either by amalgamation or concentration, and it is also necessary to remove by concentration any mineral which contains gold in such a condition as not to be effectively exposed. With regard to the tellurides of gold these are dissolved much more rapidly than they are by the simple cyanide solution, the time of agitation required often being only hours as compared with days required for the same extraction with the simple solution. Fortunately, the tellurides are very friable, so that there is no difficulty in reducing them to the desired state of division in the process of crushing the ore.

In order to demonstrate practically the short life of the solvent and the impossibility of dissolving gold which is not thoroughly exposed, we have made percolation tests on sand charges comparing cyanogen bromide and plain cyanide. Two leaching vats were constructed which were each 5 ft. deep and 3 in. in diameter. They were built in sections, each section being I ft. long, and they were joined together with rubber joints in order that the vat could be readily taken to pieces and the sand from each foot of depth sampled and assayed. Sand was prepared by crushing sulpho-telluride ore in the stamp battery through a 50× 50-mesh screen, and the bulk of the slime

NEWS

and concentrate was subsequently removed by panning, the sand being dried in the sun. A grading analysis of it showed that 18 per cent would remain on an 80 mesh standard sieve; it could therefore be regarded as sand in good physical condition for treatment by percolation. Each vat had a capacity of 10,000 grms. of sand, which was first saturated with water, 2200 cc. being added before any commenced to leach through. This moistening of the sand contained in the vat was done in order to copy as closely as possible the manner in which sand vats are filled on a practical scale. The contents of the vats were then treated as follows, each receiving the same amount of solution :—

(a) Was treated with simple cyanide, of which two leaching solutions were employed of a volume of 2200 cc. each, and containing o'2 per cent KCN. Lime was mixed with the sand before charging the vat at the rate of 2 lbs. per ton. Each leaching solution was sufficient to displace the water which had been used to moisten the charge. After the first solution bad been put on 2200 cc. of displaced water were allowed to leach off slowly, and the leaching cock was then closed and twenty-four hours contact was given with the first solution. At the end of the time of contact, the charge was allowed to leach dry and stand for twenty-four hours to allow it to become thoroughly aërated. The second solution was then put on and the 2200 cc. of the first solution allowed to leach through. The leaching cock was then closed and twenty. four hours contact given with the second solution. The contents of the vat were finally drained dry, washed to remove soluble gold, dried, saippled and assayed. The total time of treatment was four days.

24 hours contact with first solution.
leaching dry.

24 24 24

contact with second solution.
draining and washing.

4 days total time of treatment.

bromide in the absence of protective-alkali. The reason (b) The treatment was with cyanide and cyanogen for using no alkali will be explained subsequently. Two solutions were used of the same volume as in (a), but containing o'2 per cent KCN and 0.05 per cent CNBr. The first solution was put on, and when the 2200 cc. of water used to moisten the contents of the vat had leached through, the leaching cock was closed and forty-eight hours contact given. The first solution was displaced with 2200 cc. of the second solution, the leaching cock being opened until 2200 cc. of the first solution had leached through. This second solution was given fortyeight hours contact with the sand, after which it was washed free from dissolved gold, dried, sampled and assayed. In this experiment, unlike the first one, the charge was not allowed to leach dry and stand, as it was assumed that aëration could be of no benefit when using cyanogen bromide. The total time of treatment was four days. 48 hours contact with first solution. second solution.

48

NEWS

bromide has. The value of the top layer is nearly the same in both experiments, but the lower layers in the cyanogen bromide experiments are all considerably higher than they are in the plain cyanide one, despite the fact that in the former more cyanide was consumed per ton. The explanation of this marked difference is due to the cyanogen bromide solvent being decomposed before it has travelled through 1 ft. of sand, and in decomposing it has destroyed so much of the alkaline cyanide as to make its strength inferior to that contained in the other vat; as a consequence the extraction of the gold has suffered considerably. We have abundantly proved, by an enormous number of experiments, that unless the ore be very finely ground so as to expose the gold particles and also to enable the solvent to be brought in intimate and rapid contact with them by agitation, there is no benefit, but an absolute detriment, in using cyanogen bromide, and prolonged treatment with plain cyanide will always give superior results. The experiment which has been cited may be criticised from the point of view that the extraction obtained by either method was not a commercial one, but we would point out that ore of a high grade and refractory nature was specially taken for this experiment in order to make the differences more striking. With ore of lower grade and of less refractory nature similar results have been obtained, but the differences are, of course, not so great.

From these experiments it may be inferred that cyanogen bromide can be used to advantage on two classes of ore. The first class includes ores which contain minerals such as the tellurides, which are more soluble in this solvent than they are in plain cyanide. The second class includes ores with which a rapid solution of the gold is necessary in order that other substances contained in the ore shall not have time to dissolve and foul the solution. (To be continued)

IN testing an ore for ore dressing possibilities there is probably nothing so enlightening as screen classification, assay of products, and experimentation upon them by different metallurgical processes. The use of the microscope in this connection also cannot be over-emphasised. (The use of the microscope for use in connection with an ore dressing problem is described in another paper, "Approximate Quantitative Microscopy of Pulverised Ores," by W. H. Coghill and J. P. Bonardi, Tech. Paper 211, Bureau of Mines).

Accordingly, the so-called middling material, as received, was first screened into six products as shown in Table I.

TABLE I.-Screen Analysis (Feed Assayed 5′50

-28+35

per cent Mo).

Per cent Wt. Per cent Mo.

Content Mo.

Grms.

O'344

0'432

9.67

0'473 1.678

1.665

- 65

16:36

5'42

0.887

under the microscope to ascertain the degree of locked wulfenite grains in order to determine whether the material would be classed as a true middling, and would need any further crushing to liberate the wulfenite before submitting to further concentration. The treatment of middlings has followed, as a rule, two well-beaten paths: reground and returned to ore stream, or returned to ore stream without regrinding. This results often in mixing two products of widely different grades; the ore itself being perhaps of low content and the middlings of relatively high content, as regards the mineral to be saved. This fact must be borne in mind in studying the treatment to be given the material under discussion. On examination under the microscope the plus 20-mesh was found to be the only screen product containing any locked material, and this was only small in amount. That the material as received was not a true middling was readily seen, as the term "middling" is applied to a crushed product, the individual grains of which are composite; that is, they contain, attached to or imbedded in the gangue, particles of valuable mineral. Theoretically, no free grains of valuable minerals should be present, but in practice there will always be a varying proportion of such grains. The proportion of locked grains in this so-called middling material, being of such small proportion to the free grains, would readily classify it as not a true middling.

Next, in order to try out the separation of the wulfenite from the barite in the various screen products, as shown by the screen analyses, as well as on the unclassified material as received, the different sized products and unclassified material were submitted separately to a laboratory experimental size Wilfley table. Since the screen products were dry-sized between small limits, the separation of the free mineral grains of wulfenite from the gangue barite, in which the difference of specific gravity exceeds two units, should be an easy matter. The results of a few simple tests made on the small experimental size Wilfley table are give in Table II. The material in each run was passed but once over the table, and only the concentrates of each were collected, weighed, and assayed.

TABLE II.-Concentration Tests on Small Experimental Size Wilfley Table.

Run. Mesh.

I Unclassified

Per cent Per cent Mo in

conc.

recovery Mo.

ཆེན་

Grms.

material

2

3

-20+28 (a) 500

-28+35 (a) 500

28.80 (c)

-35+48

450

9 67

7'40

-65

155 13.16 61.20

500 5:42 160 12.60

74'50

Jigging of these sizes was not attempted, since the company submitting the wulfenite ore had not installed

any jigs in the mill. Further, expert attention would, in all probability, be required in the successful operation of jigs treating material of such fineness; besides taking into consideration wulfenite which when occurring in an ore in square, tabular, and sometimes extremely thin crystals would cause trouble by not following the settling ratio of the specific gravity difference.

In the runs made, as indicated in Table II., the grade of concentrates in each case could have been bettered by cutting off a larger middling product, or by running the

concentrates that were collected over the table a second time. This was not deemed necessary when using the experimental machine, since only indicative results were desired.

Table II. clearly illustrates the fact that the large wulfenite crystals-greater than 35-mesh-in association with barite are not adapted to table work, since they have a tendency to roll along with the coarse barite gangue. Wulfenite grains or crystals in this case, being approxi mately the same weight, due to their thin and tabulai structure as barite grains, did not have an opportunity to arrange themselves on the table according to their specific gravity differences.

In order to make a comparison of the sized runs with an unclassified ground feed, a sample of the original material was ground dry in a ball mill to pass 80-mesh; 2000 grms. of this minus 80-mesh material was run over the experimental table once, as in runs made with screened sizes. Table III. represents the results obtained.

Table III. clearly illustrates the following points :1. That crushing in this manner does not entail any great loss of wulfenite in the slime going to waste, since only 3.80 per cent Mo was unaccounted for.

2. That 66 20 per cent of the wulfenite can be recovered in a concentrate of marketable grade.

3. That barite can be successfully concentrated from wulfenite, since only 10 28 per cent of the total remained with the concentrate.

4. That the above tails and middling product respec tively are of the same grade as the original milled ore and first middling produced, and, consequently, further recovery can be made by treatment.

The results of the experiments as performed and given in Tables II. and III. clearly illustrate the fact that in the wet separation of wulfenite from barite, when using tables, particular attention must be given to the oredressing method adopted, and to the degree of grinding. Grinding at least through 48 mesh would be recommended for this class of material as Table II. indicates.

In order to determine approximately the highest grade of concentrate that can be produced from this barite middling material, 500 grms. of the concentrates, assaying 14 32 per cent molybdenum produced in experiment as given in Table III., were carefully re-run several times over the Wilfley table with the results given in Table IV. No slime was produced in this treatment, so the products were easily collected and recovered.

wulfenite contains 39:23 per cent MoO3, so the above concentrate represents a grade of 68 per cent pure wulfenite. This would then in all probability represent the highest grade of concentrate that could be obtained by treating such an unclassified feed of this material. The chemical analysis of this concentrate (Table V.) will further emphasise this point.

TABLE V.—Analysis of Concentrate.

MoO3.. PbO Fe2O3.. Al2O3. SiO 2 V205 P205 A$205.

BaSO4 Cu

Total determined..

Per cent.

26 93

52.15

1903

1.19

330

065

0:17

o go

12.76

none

99'08

The above analysis indicates that theoretically further concentration could be performed by elimination of the 12.76 per cent barite and the little other gangue still remaining with the concentrate; but in practice, if this were attempted, close sizing of the feed would be required. Other lead minerals than wulfenite, vanadinite, mimetite, pyromorphite, are indicated by the excess amount of lead oxide that is required to combine with the above elements given in the analysis. The presence of both anglesite and cerussite in the concentrate was later confirmed, but only in traces.

Flotation Tests on Barite " Middling " Material.

On this particular ore, ground through 80-mesh, flotation was attempted after giving it a sulphidising treatment. The material was easily sulphidised by heating for a short period within a dilute sodium sulphide solution, in proportion of 10 pounds Na2S to 1 ton of ore with a pulp ratio of 1:1. In treating the sulphidised material in a flotation machine it was found on several trials with various oil mixtures and variations of factors, that the barite and wulfenite would come over with the froth in approximately the same proportion as contained in the feed. In tracing down this cause it was found that this was due to the fact that part of the barite and the siliceous gangue matter was iron stained, and as this iron-stained material was easily sulphidised it was carried along over with the sulphidised wulfenite, and consequently no marked separation could be made. Also, it was found in performing the sulphidising treatment that a considerable amount of wulfenite was rendered soluble, and, consequently, could not be accounted for in assaying the products. The results did not warrant doing any further work by flotation; therefore this was discontinued.

Conclusions as to Laboratory Tests.

As the experimental tests were of necessity made on a laboratory scale and on very small samples of ore, the results obtained on this particular ore must be considered

treatment of commercial quantities of ore on machines of standard size, where the proper conditions for separation can be better controlled. The results from the small laboratory machine are at the best only indicative, and should be regarded only in this light. Flotation was not a factor on the ore treated.

The results found with the experimental Wilfley table indicate that wulfenite can be concentrated from a barite gangue with very promising results, since on the small laboratory machine used a recovery of between 60 and 70 per cent of the molybdenum could be effected by passing the material only once over the table. It must be re

CHEMICAL NEWS,

March 12, 1919

125

In conclusion, the author wishes to express his appreciation and gratitude to Dr. R. B. Moore, Superintendent of the U.S. Bureau of Mines mining experiment station of Golden, Colorado, for his valuable suggestions throughout the progress of the work.

THE SEPARATION OF ZIRCONIUM AND TITANIUM AS THE PHOSPHATES.

By JAMES BROWN and H. T. MADDEN.

In a previous paper (Fourn. Am. Chem. Soc., 1917, xxxix., 2358) one of us made use of the Hillebrand method ("Analysis of Silicate and Carbonate Rocks," U.S. Geol. Survey, Bull. 422, 1910, 141) of separating zirconium from titanium by precipitation of the former by use of hydrogen peroxide and alkaline phosphate in a faintly acid solution. The titanium, when large in amount, was determined as a difference or by the permanganate method, and by the colorimetric method when small amounts were present. The work recorded in the present paper was undertaken with the object of making a direct gravimetric determination of the titanium in the filtrate from the zirconium phosphate precipitation.

The standard solutions employed were prepared from pure potassium zirconium fluoride (K2Z1F6), and from pure potassium titanium fluoride (K2TIF6), respectively. The pure salts of commerce were recrystallised from boiling water several times, and were then converted into the sulphates by evaporation with sulphuric acid in a platinum dish until all fluorine was expelled. The residues were diluted with water. Qualitative tests showed the absence of iron, aluminia, manganese, and rare earths from both

solutions.

Standardisation of Solutions.

The solutions of zirconium sulphate and of titanium sulphate thus prepared were standardised by precipitation with ammonia, and ignition to the dioxide in the usual way.

From other portions of the zirconium solutions the base was precipitated as the phosphate, by disodium or diammonium phosphate, the solution containing I-2 per cent by volume of sulphuric acid. The zirconium phosphate was filtered off, washed, ignited, and converted to the dioxide as follows:-It was fused with sodium carbonate, the melt leached with water, and the insoluble residue dissolved either in hydrochloric acid or by fusion with sodium bisulphate and subsequent treatment with a dilute solution of sulphuric acid. The zirconium phosphate showed marked decrepitation during ignition. The zirconium was precipitated with ammonia, and weighed as ZrO2 after ignition. The zirconium dioxide thus obtained from the phosphate corresponded in each case with the standard obtained by treating the original solution with ammonia. The amount of zirconium dioxide involved varied from 0.1608 to 0 3216 grm. In view of the qualitative and quantitative checks thus obtained the zirconium sulphate solution was considered free from interfering bases. The solution of titanium sulphate, standardised by precipitation with ammonia, and ignition to titanium dioxide, was further standardised as follows:

As a further check on the titanium solution titanium was determined by reduction with zinc and sulphuric acid, addition of an excess of ferric sulphate, and titration of the reduced iron salt by use of permanganate (Newton, Am. Journ. Sci., 1908, [4], xxv., 130). The results thus obtained agreed very well with the standard obtained by use of ammonia alone, and the titanium solution was judged free from interfering bases.

Separation of Zirconium and Titanium.

Measured amounts of the solutions of zirconium and titanium sulphates were mixed and made acid to the extent of 1-2 per cent of free sulphuric acid by volume. An excess of hydrogen peroxide was added. A solution of disodium or diammonium phosphate was added in excess, and the solution allowed to stand over-night. The precipitated zirconium phosphate was filtered off and washed with water containing a few drops of sulphuric acid and hydrogen peroxide. The zirconium precipitate is coloured yellow by titanium. To remove this titanium we prosodium carbonate, leaching, and redissolving by fusion ceeded as directed by Hillebrand (loc. cit.) by fusing with The zirconium was reprecipitated by hydrogen peroxide with sodium bisulphate and action of dilute sulphuric acid. and phosphate, allowed to stand, filtered, and washed. small amounts of titanium, which were removed by In some cases this second zirconium precipitate contained repeating the above treatment.

zirconium dioxide, as in the standardisation, and weighed The zirconium phosphate precipitated was converted to as such. The results for the zirconium are given in the table.

precipitation were heated to get rid of the hydrogen perThe combined filtrates from the zirconium phosphate oxide. The volume of the solution was then adjusted so that the content of sulphuric acid was about 2 per cent by volume, and disodium or diammonium phosphate solution The liquid was allowed

was added to assure an excess.

to stand about one hour, and the precipitate filtered and oxide as in the standardisation, and weighed as such. washed. The precipitate was converted to titanium diThe results are given in Table I.

tion it should be converted to the dioxide to obtain quantitative results. Journal of the American Chemical Society, xlii., No. 1.

THE

EXTRACTION AND PRODUCTION OF SUGAR
FROM DIFFERENT VEGETABLE SOURCES.*
By ARTHUR R. LING.
(Concluded from p. 116).

DISCUSSION AT THE MIDLAND COUNTIES SECTION. THE CHAIRMAN (Mr. G. R. Burdass) said that it seemed to him that cane produced twice the quantity to the acre as compared with beet. If they were to keep up their name as the producers of the greatest portion of sugar, they ought to encourage their Colonies in every possible way by giving them more up-to-date machinery and encourage them to produce the sugar required. Some small experiments had been made in the production of beet sugar, but he did not know of any instance where the farmer had thought fit to continue the production of beet; presumably, he had not found that it paid him so well as mangold wurzels, and, in view of the comparatively small acreage of the British Isles, he thought their land should be devoted to the production of food-stuffs; to a great extent they could do without sugar more easily than they could do without cereals. He did not think that there would ever be a large beet sugar industry in Great Britain, and if anything warranted their promoting sugar growing he thought should be done in the direction of encouraging the Colonies, where sugar could be so much more easily grown,

Mr. JOHN M. LONES said that he had been particularly interested in the early portion of the paper dealing with the synthetic formation of sugar in plant life; in other words, nature's efforts to manufacture our sugars. It carried his mind back many years to a paper read by the late Dr. Sykes-an old friend of Mr. Ling's-wherein he described the development and formation of starch in the barley corn on similar lines to those adopted by Mr. Ling to-night. That paper was written in his early days, but he always remembered the interest he had at the time in Dr. Sykes' attempt to unlock nature's secrets. When they realised the immense tracks of land within the British Empire suitable for cultivation of either cane or beet, it reflected no credit upon their commercial enterprise to learn that prior to the war 96 per cent of our sugar was imported from various countries, and of this total 80 per cent was sent to them by European countries. For obvious reasons, this high percentage had been reduced since 1914, but from past experience they must recognise that if this matter were not taken up by this country with sufficient energy and thoroughness, what they still called "enemy countries" would strive to restore their former rate of production in the hope of making Great Britain again dependent on them. The Empire Sugar Supply Committee, with Mr. Ling as its Chairman, was working hard to check this tendency, and in doing so both he and his colleagues were performing, in the truest sense of the word, work of the highest national importance in attempt: ing to make the Empire self-supporting regarding its sugar supplies. They were deserving of the strongest encouragement and support which could be given them, for their work, like all big undertakings, was crowded with difficulties. Mr. Ling had stated to-night, and had published the same evidence on more than one occasion, that this country could grow sugar beet equal in sugar-content and purity to that grown on the Continent, which made it the more disappointing to have to admit that so far the production of beet sugar had not met with the success it deserved here, chiefly owing to the want of technical

* From the Journal of the Institute of Brewing, February, 1920.

knowledge on the part of the agriculturist, the absence of organised co-operation between farmer and factory, and meagre support from the Government. The report of Mr. Ling's Committee showed that between the total quantity of sugar produced throughout the Empire and the total consumed, there was a deficiency of 2 million tons. This was serious and evidently under the most favourable conditions would take some years to make good. The conditions to-day were not favourable, there was a general scarcity of labour, new plant was costly, the investors required coaxing with ample security before parting with their money. Mr. Ling was evidently casting longing eyes upon India to increase the Empire's supply of sugar, and it was possible by improved methods of cultivation and recovery of the sugar and by making fuller and better use of Gur, a substantial increase might be effected, and it was to be hoped that in India, as well as in other parts of the Empire, something would be done in that direction. Mr. Ling had referred to Spain as the only European country growing cane, and having in mind the very serious shortage and the improbability of it being made good within the Empire within a reasonable time, caused one to enquire if efforts could not be made to induce a friendly nation like Spain to increase her cultivation of She was comparatively near to us, and being protected from cold winds on the north by the Pyrenees, and having the benefit of the warm winds from the Mediter ranean, her climate was a most favourable one. A fertile well-watered country with the sea washing three sides of its borders, her position gave her such natural advantages, that surely something could be done to encourage her to increase the world's sugar crop.

cane.

Mr. R. H. HOPKINS said that they had been reading a lot lately in the local Press about brewers and farmers. He suggested that now, indirectly, the brewer had a fresh grievance against the farmer who declined to produce beet and preferred to grow mangold wurzels. If the farmer would grow sugar beet, that might keep out the German beet sugar; the domestic consumer could then have our beet sugar, allowing the brewer to use cane sugar. He would like to ask Mr. Ling if he personally believed in the synthesis of cane sugar supposed to have been effected ty by the Russian chemist Marchlewski in 1899. This was effected by the condensation of potassium fructosate with chloracetyl-glucose. He believed that many chemists agreed that it was not a genuine synthesis. It was evident that a difference of opinion existed on the matter. He would like to know Mr. Ling's view on the subject.

Mr. W. SCOTT said that he had had some experience of sugar growing in the West Indies just about the time the Many planters gave up sugar planting in consequence, and German sugar bounty was killing the trade out there. devoted their land to planting cocoa, coffee, and other things, and no doubt at the present time the particular island in which he lived was practically doing nothing in the way of sugar growing. Mr. Ling had put on the screen some slides of very crude plant. He (the speaker) thought that in the factory he was connected with the three-roller mill, and had to carry the canes to the mill in plant was equally crude. They had a water-wheel and a bullock carts. Because of the poor price obtained for sugar when shipped to England many of the planters did form of syrup, for which there was a ready sale amongst not trouble to crystallise out their sugar but sold it in the the natives. Without better organisation it would be difficult for small planters of cane to hold their own against the well-organised beet industry, and it was pleasing to note from the paper that this matter was receiving the attention of the Government.

Mr. LING, replying to the discussion, thanked the meeting for their kind reception of his paper. He had brought with him a good deal of statistical matter, but he found that the shortness of time made it quite impossible to deal with it. With regard to the question of beet-growing, it had been suggested that the land should be used in England for the production of other foods and