US2357419A - Size band flotation of silica - Google Patents

Description

Patented Sept. 5, 1944 Barry L. Mead, Brewster, and Ernest Maust, Lakeland, Fla., assignors to American Cyanamid Company, New York, N. Y., a corporation of Maine No Drawing. Application August 11, 1942,

. Serial No. 454,432

. 1 Claim'.

, This invention relates to a process of floating silica away from other minerals and more particularly to the beneflciation of Florida pebble phosphate ore by froth flotation.

Beneflciation of phosphate ore has been efiected in general by two processes. The first is a froth flotation process in which the phosphate is preferentially floated, leaving the silica in the tail. Reagents employed in this process are primarily anionic collectors which may be used with various oils or, in certain recent processes, without other oleaginous material. The second method of beneflciation of phosphate ore is a froth flotation proces in the presence of a cationic collector in which the silica is preferentially floated, leaving the phosphate as a tail. It is to this latter type of flotation process that the present invenion beongs.

Beneficiation of phosphate rock presents in accentuated form the problem which is inherent in the beneflciation of non-metallic materials, namely that the final material is relatively low in cost and hence economy in the flotation procedure is of primary importance. The beneflciation of phosphate presents the further problem of grade,

because a considerable premium is paid for higher grade phosphate material. In'a sense the phosphate and silica flotation have complementary advantages and disadvantages. Relatively cheap reagents but somewhat inferior grade is encountered in the ordinary anionic flotation whereas the flotation of silica while involving more expensive reagent has the advantage of somewhat higher I that it is not sufiicient to deslime thoroughly the feed, suflicient secondary slime being formed in the flotation operation itself so that results are in many cases not commercial. This has ledto a procedure in whlch'the phosphate ore is deslimed, then subjected to a polishing operation and finally again deslimed to remove a secondary slime generated in'the-polishing operation. This processis described and claimed in our Patent No. 2,216,- 040 dated September 24, 1940. The present invention constitutes an improvement on this general procedure and is based ona sizing operation which is carried out on the feed at a particular point, followed by froth flotation under separately selected conditions of the different size bands. The sizing according to the present invention must be eilected prior to polishing as otherwise it is not fully eifective.

The flotation of phosphate ore by anionic means in a plurality of size bands is a known procedure used as standard in some of the largest phosphate operations in Florida. This is described and claimed in the patent to Mead and Weaver, No. 2,156,245 dated April 25, 1939. In general the flotation in a plurality of size bands results in a saving in reagent and improvement in grade or recovery or both.

When it is attempted to apply size band flotation to the removal of silica from phosphate ore in the saine manner as it is applied to the feed for anionic flotation as described in the Mead and Weaver patent referred to above; namely before introducing into the strings of the flotation cells, the desired results are not obtained, or rather the full improvement is not obtained. We have found according to the present invention that when dealing with size band flotation of silica, the sizing must precede polishing and desliming be cause the polishing conditions vary with the particle size range, a phenomenon which would not be ordinarily expected as it has no relation to the reagent economy which is the normal result from size band flotation. It is not knownfully why it is essential to eflect sizing before polishing and to use different polishing times for the different size bands, in general much longer times of polishing .for the fine sizes. One factor which may be of importance is the effect of line mud balls or clay which tend to segregate with the fines when the sizing is effected, as it should be for economical operation, by hydraulic classification, for example in a hindered settling classifier. It is probable that thepresence of these clay aggregates may affect polishing time and results, because according to the present invention a much longer polishing time is necessary with the flnes than with the coarse material which is economically no serious draw-back because of the relatively minor volume of flnes in ordinary feed. This factor may be the principle one or it is possible that other factors may contribute ormay even be of greater importance, and it is not desired to limit the invention to any theory of action. In general the coarser size bands which form the vast majority by weight of the ordinary feed can be effectively polished by employing a polishing time of from /2 to 1% the time used with the fines.

The degree of slimeremoval when sizing is effected prior to polishing and particularly when hydraulic sizing is employed which tends to segregate light mud balls with fines will vary with the coarse and fine material even when the fines are subjected to a relatively longer polishing time. The slimes however, are a much greater detriment in the flotation of coarser size bands and when an unsized ore is polished and deslimed it is diflicult to effect a sufllcient polishing and slime removal to permit froth flotation of all of the coarse silica.

The present invention overcomes the difficulty in polishing encountered when unsized feed is employed and at the same time this advantage is obtained without any compromise with metallurgical efllciency of the froth flotation operation or with reagent cost. In fact, the present process not only improves the polishing and desliming operation greatly but it also permits obtaining the advantages in saving of reagent and improvement the coarse bands also using some kerosene.

with polyethylene polyamine using a small amount of kerosene. The other portion was deslimed, sized in a hindered settling classifier into. three bands, 25 +48 mesh, 48+65 mesh and 65 mesh.

A screen analysis of the original feed and size bands is as follows:

28 +48 mesh feed- 22. 2 30. 2'32. 11. 4 3. 7 65. 5 48+66mesh iced. 73.5 23.8 2.7 23.9 65 mesh feed 78.0 19.7 2.1 .2 10.6

mesh feed 16.0i20.6l21 718.4 17.6 5.1 .6 .1 100.0

The different sizes were then separately highly polished using 3-4 minutes for the coarser bands 20 and 10-12 minutes for the 65 mesh material.

Following polishing all the size ranges were very thoroughly deslimed. The three size bands were then separately floated using varying amounts of the same cationic collector and in the case of The metallurgical results appear in the following table, the lime marked 20 mesh relating to the first test in which the feed was unsized.

I Table I Feed Concentrates Middlings Tailings Midds. loss Midds. rec. Lbs./ton couc.

Size Band P P P ererer- Per- B. P. L cent B. P. L. Insol. mm B. P. L cent B. P. L cent Ratio Rec Ratio Rec 3$ Egg 20 +48 mesh 43.12 65. 5 78. 51 2. 70.0 52. 11 51. 1 3. 39 62.0 1. 95 93. 4 1. 89 .3 1.7. 1 48 mesh 30.88 23. 9 78. 72 1.50 18.5 22. 52 33. 3 .81 28. 5 2. 68 95. 1 2. 59 4 l. 2? f 65 mesh. 4". 23 10.6 77. 1. 11. 5 32. 69 15. 6 1.14 9. 5 1. 94 94. 8 1. 86 98.9 .89

Composite. 40. 10 100. 0 78. 46 2. 19 100. O 39. 23 100. 0 2. 44 100. 0 2. 08 94. l 2. 02 96. 9 l. 55 11 20 mesh 42. 31 100.0 78. 95 2. 20 100. 0 58. 25 100. 0 7. 63 100. 0 2.17 86.0 2.06 90. 6 2. 34 .18

in metallurgical results which are noted in size 4.3

band flotation of unpolished feed with anionic collectors described in the Mead and Weaver patent.

It is an advantage of the present invention that the sizing operation when carried out at a particular point in the feed preparation, namely just before polishing, should combine all of the improvements normally associated with size band flotation and at the same time solve the slime problem which in its nature is entirely different from that solved by the size band flotation of phosphate described in the-Mead and Weaver patent.

The present invention isparticularly important in the beneflciation of phosphate ore, but is also of use in other flotation processes in which silica is removed from other similar non-metallic ores.

The invention will be'described in greater detail in conjunction with the following specific examples.

EXAMPLE 1 A typical example of Florida pebble phosphate ore having a B. P. L. content of about 42% was divided into two portions. One portion was deslimed, polished, again thoroughly deslimed and floated with the appropriate amount of a salt of a condensation product of coconut oil fatty acids EXAMPLE 2 A slightly higher grade pebble phosphate ore of +20 mesh was deslimed and classified in a hindered settling classifier and resulted in the following screen analysis:

Coarsebani. 7.9 22.0 42.2 22.8 4.7 .4 Intermediate pend 3.4 38.6 46.5 10.9 .0

65 Fineband .4 15.9 40.5 24.8 11.1 1.1

75 machine including a rougher float and a cleaner float. The metallurgical results and polishing time are shown in the following table:

tion machines as described in Example 1. employing a single'cleaning to produce one mid- Table If COABB E BAND Phosphate tails I silica C oiiector Polishing f i concentr. Ratio Recovery lbs./ton time,

' B. R L InsoL B. P. L. feed minutes IN TEBMEDIATE BAND 46. 12 76. 7. 00 10. 52 2. 01 1. 68 98. 2 73 0 43. 61 79. 42 2. 55 27. 06 1. 79 1. 86 97. 9 73 2 44.09 so. 95 1. 90 49. 27 2. 12 l. 88 97. 7 73 4 43. 15 80.98 2. 00 48. Q7 2. 01 1. ill 97. 7 73 6 FINE BAND 36.79 36.79 No float .77 0 36. 79 59. 87 23. 60 2. 12 1. 67 97. 4 77 2 37. 76. 78 3. 60 11. 90 2. 09 98. 6 77 4 36. 76 78. 22 2. 8O 12. 17 l. 01 2. 16 98. 5 77 6 36. 18 78. 95 2. 60 16. 82 1. 14 2. 22 98. 3 77 8 35. 25 79. 48 2. 50 9. 42 68 2. 28 '98. 9 58 12 From the above results it is apparent that a much longer polishing time is necessary with dling product. The metallurgical results are shown in the following table:

Table III COARSE BAND Phosphate tails Silica Amine Pine oil Polishing g' ig f concentr. Ratio Recovery lbs./ton lbs./ton time,

' Rh MEL B. P L. ieed -i'eed minutes MEDIUM BAND FINE BAN D the fines band, improvements including a, saving in the amount of collector required being noted even up to 12 minutes. But little improvement results in polishing the coarser band beyond 2 minutes or the intermediate band beyond 4 minutes. .With a powerful collector for coarse silica such as the'polyethylene polyamine coconut oil acid condensation product, good results are obtained even in the coarse band.

EXAMPLE 3 the polyethylene polyamine fatty acid condensation products should be employed. The coarse band, is however, included in the table so as to show that even with a. reagent which is not commercially suitable for floating such coarse material, eflects of the present invention are to be noted.

In the examples three size bands have been shown. This number is a desirable compromise ln the case of Florida ebble phosphate feed as it gives optimum results without undue complication. Markedly improved results over unsized fbed is obtained when the sizing is such that two size bands only. are produced and sizing may be carried beyond three size bands, but the urther sizing in such cases usually does not vsufliciently improve the results to warrant the additional equipment and complication of the process. Other non-metallic ores will in general require a comparable number of size bands, although the best sizes and the optimum number of bands will vary with each ore.

We claim:

A method of beneflciating phosphate ore by froth flotation of silica therefrom in the presence of a cationic collector which comprises desliming and sizing the ore into a plurality of size bands, separately polishing each size band until 'substantially all of the secondary slime is formed the polishing time for the fines size band being several times that of the coarser size bands thoroughly desliming each size band and floating the silica from each band in a separate flotation operation in the presence of an optimum quantity of cationic collector.

HARRY L. MEAD. ERNEST J. MAUST.