US2395475A - Beneficiation of beryllium ores - Google Patents

Description

Patented Feb. 26, 1946 2,395,475 BENEFICIATION OF BEB-YLLIUM ORES Harold L. Gibbs and Henry D. Snedden, Salt Lake to the Government of the United States, as represented by the Secretary of the Department of Interior No Drawing. Application March 2, 1945, Serial No. 580,578

(Granted under the act of March 3, 1883, as

amended April 80, 1928; 370 0. G. 757) City, Utah, assignors 4 Claims.

The invention described herein may be manufactured and used by or for the Government of the United States for governmental purposes without the payment to us of any royalty thereon in accordance with the provisions of the act of April 30, 1928 (Ch. 460. 45 Stat. L. 467).

This invention relates to the beneilciatlon of beryllium ores and has for its object the production of a high grade beryllium ore concentrate from low gradeberyllium deposits. Another obiect is the provision of a process whereby relatively coarse beryllium ore having an average particle size as large a plus loll-mesh or larger, can be beneficiated through a selective mineraliilming technique.

The foregoing and other objects hereinafter apparent are accomplished in accordance with this invention wherein a beryllium ore is beneilciated by subjecting such an ore to a mineralconcentrating treatment involving selective filming under acidic condition of a desired constituent in the presence of a soluble fluoride. a longchaln fatty acid and a long-chain alkyl amine salt. It has been found that beryllium ores can be beneflciated to produce high grade concentrates even though relatively coarse beryllium ore be employed having an average particle size at least as large as plus loo-mesh, for example, between 20 and loll-mesh as determined by standard Tyler screens, when the mineral-concentrating treatment is carried out in the presence of a long-chain aikyl amine salt, such as for example lauryl amine hydrochloride.

Suitable beryllium ores for employment in accordance with this invention include ores containing beryl, chrysoberyl and other beryllium minerals containing gems of widely varying composition including such constituents for example as mica, quartz, apatite, iron compounds, pegmatite, feldspar, tourmaline and other game materials normally associated with beryllium minerals. The ores may be treated in a wide range of particle size in accordance with this invention. such as for example in minus -mesh to minus BOO-mesh, and we have found that our invention provides a very high recovery in the larger particle sizes at least as large as plus 100- mesh not hitherto successfully concentrated.

It has been found that the presence of metallic iron or iron compounds exerts a detrimental influence on the beneiiclation of beryllium ores by mineral filming techniques such as for example agglomerative tabling or froth flotation, and therefore we prefer to subject the selected ore to an iron removal treatment, preferably after initial grinding. For example, we may remove the socalled grinding iron" introduced durin grinding in iron ball mill equipment together with magnetic iron compounds which may be present in the ore, by subjecting the comminuted ore to wet magnetic separation. This is accomplished in the usual way by passing the ore pulp over a suitable magnet such as for example an electromagnet. A preferred iron removal treatment.

which may be employed alone or in conjunction with magnetic separation, is the treatment of the comminuted ore pulp with a strong mineral acid such as for example hydrochloric acid, nitric acid, hydrofluoric acid, sulfuric acid, and preferably a mixture of'a soluble fluoride with sulfuric in accordance with acid. After the acid treatment, the soluble iron compounds produced are removed together with slimes by washing the ore sands with a large excess of water. The foregoing acid treatment in conjunction with a. soluble fluoride constitutes a preferred conditioning treatment or pretreatment this invention and not only serves the function of removing iron, but also renders the desired beryllium minerals peculiarly receptive to selective filming with the reagents hereinafter to be described.

Following the pro-treatment or conditioning, the beryllium ore can be concentrated by asglomerative tabling employing preferably oleic acid, No. 1 or similar fuel oil, and a soluble salt of a long-chain alkyl amine. This is accomplished in the usual manner by passing an aqueous pulp of the ore and reagents across a reciprocating table such as for example a Wiliicy tabie provided with a cross wash of water. The

beryllium agglomerates are carried over the side ofthetableandthegangueparticiesproceed over the end of the table. Flotation, involving agitating and aerating the ore pulp with reagents and separating the mineral-bearing froth. is a preferred concentrating treatment involving selective filming in our invention.

Regardless of the mineral concentrating treatment employed. it has been found that a desirable recovery can be made when the treatment is carried out in the presence of a long-chain alkyl amine salt. Suitable long-chain alkyl amine salts are the formates, acetates, sulfates, and preferably the hydrochlorides of aliphatic amines having from 8 to 18 carbon atoms in the alkyl radical. The commercial mixture of the long-chain aliphatic amine hydrochlorides derived from the fatty acids present in coconut oil, sold under the trade name D. P. 243 and the mixture believed to contain largely lauryl amine hydrochloride sold under the trade name "E. P. 487," have been found to be especially suitable long-chain alkyl amine salts for employment in accordance with this invention although salts of long-chain aliphatic amines derived from or corresponding to the fatty acids produced from other vegetable, animal, marine, and synthetic sources can be employed. Generally, the long-chain alkyl amine salt is employed in proportions of from 0.1 to pounds of the commercially sold mixture per ton of ore and is preferably employed in the somewhat more restricted proportions of between 0.3 and 3.5 pounds of the commercial mixture per ton of ore. Greater or lesser amounts can also be used.

Suitable soluble fluorides for employment in the conditioning or pre-treatment step and also in the flotation or agglomeration stage, are the alkali metal fluorides including sodium and potassium fluoride, and other compounds yielding a fluoride ion under acidic aqueous conditions, such as for example, sodium silicoiluoride. Generally sodium fluoride or other soluble fluoride is employed in proportions of from 0.2 to 10 pounds of soluble fluoride per ton of ore and preferably within the more restricted range of between 2 and 5 pounds of soluble fluoride per ton of ore being treated. Hydrofluoric acid is a very desirable soluble fluoride and in addition provides the necessary acidic conditions for carrying out this invention, but in general, because of handling difflculties, we prefer to employ sodium fluoride in conjunction with sulfuric acid.

Both the pro-treatment or conditioning step and the flotation or tabling step in our inventiim are carried out under acidic conditions. by which is meant under a pulp acidity at least as acid as pH 6.9. Generally, the pH of the pulp during conditioning and tabling or flotation is maintained within a pH between about pH 1.5 and pH 6.5, while preferably the acidity is maintained between pH 2 and pH 6.

The selective filming treatment of this invention whether it be agglomerative tabling or flotation, is carried out in the presence of a long-chain fatty acid having from 8 to 18 carbon atoms, as a mineral-filming reagent. Representative longchain fatty acids are stearic acid, palmitic acid, lauric acid and preferably oleic acid. Other suitable iatty acids include the mixture of fatty acids derived from the oils of coconut, tallow, sardine, menhaden. whale, cotton seed, peanut, and similar vegetable. animal and marine oils. The fatty acid is admixed with the pre-treated ore pulp in controlled amounts cu. cient to float the desired assume constituent, in flotation or sumcient to agglomerate the desired constituent in the case of the agglomerative tabling. A suitable amount of oleic or other fatty acid to employ is from 0.2 to 10 pounds of long-chain fatty acid per ton of ore undergoing treatment. Since the fatty acid probably exists largely in the free state in our acidic ore pulp, it is desirable to introduce it in the form of an aqueous emulsion to provide for its better dispersion in the ore pulp. A suitable emulsifying agent comprises the ammonium salt of lauryl diethylene glycol sulfate sold under the trade name "Emulsol X-l, although other emulsifying agents can be used. It is preferably employed in amounts of from 0.1 to 2.0 pounds per ton of ore undergoing treatment.

In carrying out our invention, it is often desirable to subject the ore pulp to a preliminary flotation employing very restricted proportions of reagents to selectively remove the mica contained in the ore. Likewise, various other mineral beneflciating treatments known to the art can be employed in conjunction with our process.

The following illustrative examples show how the invention may be carried out but it is not restricted thereto. Parts and percentage compositions are by weight unless otherwise designated.

DESCRIPTION OF ORES TESTED Much of the test work was done on an 800- pound sample of ore from the Wisdom Ranch, Larimer County, Colorado. This sample assayed 0.94 percent B120 and consisted of beryl and chrysoberyl associated with olmoclase feldspar. quartz, muscovite, and garnet. The bulk of the beryl occurs in plus ill-mesh sizes; however, enough beryl and chrysoberyl occur flnely intergrown at contact areas with gangue rock to necessltate flne grinding of the ore for optimum beryllium recovery.

Other test work was done on two samples of low-grade beryl ore from the Mica Mountain District, Latah County. Idaho. One of these samples assayed 0.39 percent BeO and represented the minus V -inch rejects from a screening and hand sorting operation. The lower grade sample assayed only 0.17 percent BeO. Mineralization of both samples was similar. Principal minerals were beryl, muscovite, feidspars, quartz, tourmaline, magnetite, and apatite. Beryl occurred as fragments ranging in size from about 54-inch to loll-mesh; most of the beryl was liberated by grinding the are through 65-mesh.

Commercial grade products, approximately 10 percent BeO, were made from all three ores. Several products obtained assayed between 12 and 13 percent neo which is approximately the mo content of pure beryl mineral.

, TEST PROCEDURES 00m Guvrrr Where part of the beryl in an ore occurred in coarse crystals, it was partially concentrated by gravity at coarse sizes by sink-and-float methods. Standard laboratory procedures were used and both galena suspensions and heavy liquids were found to be satisfactory. Test results indicate that the method would be of most value when used to reject coarse waste rock with a minimum of crushing rather than as a. complete concentration process used to make finished beryl products.

Aootomsrrou Tamra Beryl was concentrated at intermediate using agglomeration tabling. Commercial grade concentrates were made. The beryl was not readily agglomerated until the ore had been given a special acid-fluoride conditioning treatment as described later under flotation (Step 3). Following this conditioning treatment, the ore was washed. agglomerated with oleic acid and No. 1 fuel oil, and tabled.

Ftorlirron Preliminary flotation tests iollowing published flotation procedures and reagent combinations were not satisfactory. A successful procedure was finally developed that involved flotation of deslimed sands after a special conditioning treatment which is described as follows:

BASIC PROCEDURE FOR BE'RYL FLOTATION pH of 2.0 to 2.5 with Na! (3 to 6 pounds per ton) and H2804 ('7 to 15 pounds per ton) filtered and washed.

4. Filter cake is repulped in flotation machine with fresh water, NaF (0.2 pound per ton) H2804 (for a pH of 6.0 to 6.9), conditioned for 5 minutes with a. water emulsion of oleic acid stabilized with Emulsol X-1 and the beryl is floated.

Satisfactory recovery of beryl was obtained in Y a high-grade product when HF and HCl were substituted for Na! and H2504 in conditioning step 3 of basic procedure. Substitution oi CaFa for Na! was not satisfactory as the calcium ion consumed oleic acid and probably activated gangue minerals. Small additions of fluoride to grinding and beryl flotation circuits were beneflcial but not essential. Using the above procedure, flotation of minus loo-mesh beryl was rapid and the products were clean, but flotation of plus loll-mesh beryl was slow and the products were not clean. Washing sands by decantation made. Micawauelectivelyfloateduslnganalhlamine'saltsuchasnamormlm as collector and The mica could be iloatedeitherfromanacidcircuitpriortotbe acid-fluoride eondltioningstcpor fromthe acidfluoride circuit itself. Theaminesaltusedto floatmicawasalsoacollectorforchrysoberyl andiithismineralwerepresentintheorathe micafloathadtobemadeverycareiullyorsome chrysoberyl was lost in the mica concentrate. Berylwasactivatedbytheaminesaltiaftcr activation, coarse beryl was more easily floated with fatty acid. The activation of coarse beryl by anaminesaltinthe acid-fluoride conditioner was eflective even though a mica float was not actually made. when a separate mica float was not made, the amine salt caused some mica to float with the beryl. but concentrates plus lilpercent BeO were still readily obtained.

(1'. Apatite and bowl were separated by selective flotation. Apatite floats more readily than beryl and by careful additions of oleic acid was floated ahead of beryl to make a separate rougher concentrate. If this separate rougher concentrate was not made, the selective flotation of apatite from beryl was diflicult although sometimes still posslble. The apatite rougher concentrate containing some beryl was conditioned with NaOH at a pH of 11.5 filtered. washed, re-

pulped with water, and the apatite selectively floated at a pH of 8.4 using small additions of 2-1. Any apatite remaining in the beryl concentrate could be leached out with dilute acid.

D. The substitution of caustic for acid-fluoride 86 in the conditioning stage was partially successful provided a pebble mill was used to grind the ore. Results were interior to those obtained with acidfluoride and no separation of beryl could be made if the ore had been ground in an iron ball mill.

0 Results of typical tests, making use of the was as efl'ective as filterin and washing on the a The following modifications were made on the basic procedure to improve separation of beryl from certain minerals.

A. Dry beryl concentrates were treated on a high intensity, this type. magnetic separator to remove garnet and other magnetic minerals. In addition to floating beryl. oleic acid also collected minerals such as garnet. tourmaline, and biotite. No attempt was made to selectively float beryl from these minerals as the separation was readily made magnetically and the quantity of such impurities was small.

B. In general, mica was eflectively depressed when using the previously outlined basic procedure for beryl flotation. However. .where the mica' to beryl ratio of low-grade ores was high. it was advantageous to remove mica before the beryl was floated. It was possible tofloat mica at coarse 81268,'Wh10h would aid in reducing grinding costs, and. from suitable ores. clean commercial grade mica by-products could be concentration procedures just discussed, are given for three western beryl ores.

EXAMPLES NOS. 1 AND 2.WISDOM BERYL rator, and the ore pulped with water and deslimed. In both tests concentration procedures were the same; mica and beryl were selectively floated and the dried beryl concentrates were treated on highintensity magnetic separator to remove garnet.

Reagent consumptions and test results are tabulated for both testsinTables 1 and 1A.

'l'ssanL-Rnsmr ConsmflOrImPounns Pn'l'onorou Pcbblemlllgrind an 4.8 0.05. so as Ballmillgrind as u an 1.0 a:

TAN-I 1A Weight my per Dietrib. percent cent he!) cent fieO Test products 5 Pebble Ball Pebble Ball Pebble Ball mill mill mill mill mill mill No.1berylconct... 0.0 1.1 10.25 10.46 66.1 71.8 No.2herylconct... 2.0 1.0 5.75 0.70 12.4 1.0 R r flotation 1 02 5 47. 2 0. 01 0. 01 0. 0. Miss concentrate" 18.0 25. 1 0. 28 0. 53 4. 4 l2 9 21.0 18. 6 0. 73 0. 75 10. 4 l3. 6 Magnetic product 0. 6 0. 4 0. 3i 0. 50 0. 2 0. 2

Total head- 100. 0 1M. 0 0. 93 l. 03 11!). 0 100. 0

By flotation oi minus loo-mesh ore, 66.1 and 71.8 percent of the beryllium were concentrated in products assaying 10.25 and 10.45 percent BeO, respectively. Loss or chrysoberyl in mica concentrate was roughly proportional to the weight floated and the amount of D. P. 243 used. Loss 01' B60 in the slime fraction was 13 to 16 percent as compared with 20 to 22 percent when conventional web ball mill grinding procedures were employed.

EXAMPLE NO. 3.WISDOM BERYL A locked roughing test was run, in six cycles, to determine the efl'ect of returning middlings to the flotation circuit. Each sample was stage ground wet to minus loo-mesh in a laboratory ball mill. In this test the basic flotation procedure was employed except that the second rougher concentrate or middling from each flotation stage was retreated with new are in the next flotation stage. All concentrates and the small amount 01' final middling were combined and cleaned by flotation, using the same procedure as in the rougher flotation circuit. Cleaner flotation concentrate was dried and treated on a magnetic separator to remov garnet. Reagent consumptions and test resuits tabulated in Tables 2 and 2A.

TABLI 2.Ruem Consmnon m Pouuns Pn By flotation or minus loo-mesh ore, '16 percent 01 the Boo was recovered in a product assaying 11.8 percent BeO. Mica was effectively depressed and no special mica separation was needed. Test results indicate-that the process can be operated on a continuous basis and middlings re-treated to recover the beryl they contain. It should be noted that loss or BeO in slime is 20.! percent as compared with only 18.4 and 13.8 percent when special grinding procedures, previously described, were used.

assume EXAMPLE NO. 4.-WIBDOM BERYL In an attempt to reduce loss of beryl in slimes. a test was run in which coarse ore was treated by sink-and-float and hand sorting while fine ore was treated by flotation. Ore, as received, was screened on 1-inch, i -inch. B-mesh, and 20-mesh screens. The minus 20-mesh material, consisting largely of top soil and roots with very little beryl, was discarded without treatment. The plus 3- mesh portions were treated separately by sinkand-fioat using galena suspensions of 2.60 and 2.66 specific gravities to reject coarse feldspar and quartz, respectively. Sink products were hand sorted to remove free beryl crystals. Hand-sorted rejects were combined with the original minus 3 plus 20-mesh material and the composite was stage ground to minus 100-mesh and floated to recover beryl. Dry beryl flotation concentrate was treated in the high intensity magnetic separator to remove garnet.

Reagent consumptions and test results are tabulated in Tables 3 and 3A.

Tests 3.--Rmcrur Couslmr'rlou m Poms Pea Ton H150 NaF 253 x-i Flotation feed 22 0 8.0 2 0 0. 0 Original ore 8. 4 3. 3 0.8 0. 18

Tara: an

BeO, percent Weight, percent Assay'ed Distribution Sorted plus 1-inch beryl 0. 32 11.9 4.3 Sorted minus 1 plus -inch beryl 1. 99 i9. 6 28. 2 Sorted minus 54-inch plus 3 mesh beryl l. 20 i2. 45 16. 7 Beryl flotation concentrate. 2. 69 9. 52 7 Slimes 10. 43 l. 00 11. 7 Quartz "float" at 2.66 sp. gravity 12. 32 0. ll 1. 5 Feldspar "float" at 2.60 sp.

gravity 33. 54 0. 05 2. 0 Re her flotation tailing 24. 0. ll 3. 0 Or minus ail-mesh discard. 12. 30 0. 26 3. 6 Magnetic product 0. 56 0. 61 0. 3

Total head 100. (I) 0. 89 100. 0

Composite beryl concentrate. ti. 2) l1. 2 77. 9

Using a combination method, treating coarse ore by sink-and-float and hand sorting and flne ore by flotation, 77.9 percent of the Eco was concentrated in a product assaying 11.2 percent BeO. Principal beryl loss is still in the slime fraction.

' However, this loss is lower than in tests in which all of the ore was ground to flotation sizes.

EXAMPLE NO. 5.WISDOM BERYL Treatment of coarse sizes of ore by agglomeration tablins also was tried to reduce slime losses. A sample of ore was roll crushed dry through 20- mesh and screened on 35-mesh and lib-mesh screens. The plus 65-mesh sizes were conditioned with acid-fluoride, as in step 3 of the basic flotation procedure, beryl was agglomerated with oleic acid and No. 1 fuel oil and the ore tabled. Minus 65-inesh ore was ground through loo-mesh and the beryl was floated. Mica that contaminated all beryl concentrates was removed by elutriatlon. Dry beryl concentrates were treated on a high intensity magnetic separator to remove garnet.

Reagent consumptions and test results are tabulated in Tables 4 and 4A.

Tun: 41-423mm column at Poms Pu T! or Oar mso. mar 25:, X-l g g-g gfglomeration tabllng 1.8 0.4 0.28 0.23 otation.. 2.3 0.7 0.25 0.05 Total 4.1 1.1 0.53 0.05 0.23

Tam 4A BeO percent warm,

Assay Dbtribution #fglmnerate concentrate 5. 46 0. 5 ea otation eonoentrato.... 1.08 10. 2 11. Table rejects. 60. 77 0. 31 10. Flotation tailing... ill. 30 0.065 1. Elutriated mice. 3. 84 2. 34 0. Blimes 0. It 0. 06 4. Magnetic product 2. N 1. l3 3.

Total head 1m. 00 0. 99 100. 0 Composite beryl concentrate- 0. M 9. 6 03. 0

Treatment of ore at intermediate sizes. by agglomeration tabiing with only the minus 65- mesh ore ground through 100-mesh for flotation, resulted in a loss of only 4.0 percent of beryl in the slimes. Considerable beryl was lost in table rejects and only 63.6 percent of the Be() was recovered in a product assaying 9.6 percent .BeO. The small loss of beryl in the slimes was oilset by a large loss in the table rejects so that the greatest advantage of agglomeration tabling was the reduction in grinding cost made by treatment of are at coarse sizes. Recovery could be raised and slime still kept at a minimum by grinding and floating beryl from the table rejects.

NOS. 6, I, AND B.MICA MOUNTAIN BERYL l. The sample was roll crushed and screened on -mesh and 48-mesh Clean mica.

rejected as plus 20-mesh material.

2. The mica was floated as previously described in the preceding examples, from the plus til-mesh portion and the tailings were ground through 48-mesh and combined with original minus 48- mesh ore.

3. The minus ia-mesh ore selectively floated to remove beryl.

Final products were marketable mica and beryl concentrates and low-grade apatite products 5 to 15 percent PaOs.

The two samples on which these three tests were made represent types or are not amenable to concentration by any previously shown method. The higher grade sample, assaying 0.39 percent BeO. is the fine screen reject from a hand sorting operation. The lower grade sample. assaying only 0.1! percent 3420, is a mica schist with beryl diswas deslinled and mica, apatite, and

tributed in small veinlets. Considering the extremely low grade of the samples and number of mineral separations made, results obtained are remarkably good.

Reagent consumptions and test results are tabulated in Tables 5 and 5A, 6 and 6A, and 'I and 'IA.

TABLE fie-RIMIIH'I' 008M011 IR POUNDS PI! mac. NaF g- 232; x-: mon

Mica flotation.. 15.60 4.2) 0.07

Beryl rougher float 0. 92 0.25 2.4 0.46

Beryl cleaner float 0. 70 0. 25 0. 0 0. 00 Apatita flotatiom. 0. t3 1. 1 'lotal--- 17.22 4.70 0.07 8.8 0.55 1.!

Taste 5A B00 percent Weight, petoen Assay Distribution Beryl concentrate 2. 78 i0. 5 Ti. 8 Mica product 25. 84 0. 05 3. 5 Rougher flotation tailing e0. 55 0. 0i i. 5 Slimes 6. 0. 4-! 7. 4 Apatite product 1 so 0. 40 2. 0 Beryl cleaner flotation tailing. 2 6i 0. 04 0. 7 Magnetic product, 0. l0 0. 30 0. 1 Total head 1(1). 00 0. 38 1m. 0

TABLI: 6.-Ruomr CONSUMPTION 1R POUNDS Plli Ton or On:

' E. P Oleio H.801 Na! M 3-! Mlcaflotation l0.8 2.5 8.2 Beryl flotation 3.2 0.4 0.08 Apatita flotation 0. 2 0. 12 Total 10. B 5. 8 I. 2 0. 0 0. 2)

Tm: 6A

B00 percent lt'ii' Assay Distribution 21. 29 01 l. 4 0. B2 0. 5B 37. l 4. as 0. l8 s 0 C one! 0. I) 0. 9 l. 5 Rougher flotation tailing......- 50. 44 0. on 30. 3 Blimes 13. 59 0. l6 l4. 8 Magnetic rs'oduct........----.- 0.10 0.08 0.0 Total lead 100.00 0. 14 100. 0

By flotation oi minus -mesh ore, 87.1 percent oi the BeO was concentrated in a product assaymg 9.55 percent BeO. Oleic acid consumption was small and principal loss 0! beryl was in rougher flotation tailing. The beryl-apatite separation was good, apatite was floated before beryl with small reagent additions, and the apatite product without cleaning assayed 4.88 percent Pros and only 0.18 percent BeO.

g ish'ibu- Unit! Oleic Emulsol acid Pu'eent, wt.

moon

5 selective method for bene- Amine E. P. No!

lteoaentapoaadspertonotore Army, Grams,

Use oi an amine in the conditioning step inescription and examples provides an ce- Metolluroioel data Olelo acid 34:0, percent Assay Distribution 08 mwmmmmm t...

weight.

=Tos or on:

Tests 75 Tara: is-Rum: Consumes m Poms Pm 'lotcl-- Mia iiotstlom. Beryl detailed.-. Apotltc iiotatlom oi the BeO was concentrated in two products, a beryl concentrate assaying 9.30 percent BeO and a beryl-apatite product assaying 4.30 percent BeO and 18.20 percent PaOs. Rouaher beryl flotation creased the recovery oi beryl very markedly. The 45 plus IOU-mesh tailing shows only a small Perccntage oi the Be!) as compared with the test in which no amine was used and the plus loo-mesh concentrate shows a corresponding increase in IBcO content.

amples we have provided a very desirable method for the beneflciation oi beryllium ores. particularly applicable to the heneiiciation of coarse sizes of ore as large asplus 65-mesh or larger. These 65 larger sizes oi ore have never successfully heretofore been concentrated by flotation techniques or other techniques involving selective filming. and we believe that our combination or treatment steps and reagents as taught in the fore- 00 going :1

peciaily valuable an ilciating all beryllium ores. A particular advantage oiourinventionisthat chrysoberylisre- Oleio Emulsol acid Imam data No! HIBOs EXAMPLE NO. 9.WIBDOM BERYL CussFBsaYL-Plomml Theore was stage groundin anironball mill to pass through a ell-mesh screen. The pulp was 50 As shown by the foregoing description and exdeslimed, abraided iron removed magnetically.

Reagents, pounds per ton of ore lm-mesh flotation--.

was satisfactory; the tailingmade contained only 4.5 percent of the BeO as compared to 39.3 percent in the previous test.

and the sands were screened on loo-mesh into two sizes. Each size was conditioned. filtered. and floated separately.

+i00-mesh conditioning. 3.0

W anew M. MM was m mm. m a m mmm m m... WWW s M mm c; Wm a new swam m ascetic covered along with beryl and other beryllium minerals which largely increases the recovery 0! the beneflciation process.

Various changes can be made in the invention as illustrated and described since many apparently widely diflering embodiments will occur to one skilled in the art.

'What is claimed is:

1. In a process for the production oi a beryllium ore concentrate, the improvement which comprises subjecting such an ore to a pretreatment to remove iron and then to a mineral concentrating treatment involving selective filming under acidic conditions of a desired constituent in the presence oi a soluble fluoride. a lone-chain i'atty acid and a long-chain alkyl amine salt.

2. In a process for the production or a beryllium ore concentrate, the improvement which comprises removing iron lmm such an ore, desliming said ore, then agitating and aeratins'the low-.

iron comminuted ore in an aqueous pulp at pH 2 to pH 6 in the presence or per ton of ore, from 0.2 to 10 pounds of a soluble fluoride, from 0.2 to 10 pounds of a ions-chain fatty acid, and from 0.1 to 5.0 pounds 01' commercial long-chain alkyi u amineaait.'

, 3. In a process for the production oi a beryllium ore concentrate, the improvement which comprises treating such a comminuted ore with a soluble fluoride and a strong mineral acid in an aqueous Pulp. than washing the treated sands with excess water, and thereafter agitating and aerating an aqueous pul of the thus-treated ore in the presence or a soluble fluoride. a strong mineral acid, oleic acid, and a mixture of longchain alkyl amine salts corresponding to the coconut oil fatty acids.

4. In a process for the production of a beryllium ore concentrate, the improvement which comprises treating such a comminuted Ore with a soluble fluoride, a strong mineral acid. and a iong-cl1ain alkyl amine salt in an aqueous pulp, then washing the treated sands with excess water, and thereafter agitating and aerating an aqueous pulp of the thus treated ore in the presence or a soluble fluoride, a strong mineral acid, and oleic acid.

HAROID L. GIBBS. HENRY D. SNEDDEN.

Certificate of Correction Patent N 0. 2,395,475.

HAROLD L. GIBBS ET AL.

February 26, 1946.

It is hereby certified that errors appear in the rinted s ecifieation of the above numbered patent requiring correction as follows:

the word shown read a e 5, t column, line 71, for

known; and second column, e 65, Table 6A, first column thereof, for Total lead read Total head; and that the said Letters Patent should be read with these corrections therein that the same may conform to the record of the case in the Patent Oflide.

Signed and sealed this 30th day of April, A. D. 1946;

LESLIE First Assistant Commissioner of Patents.

ascetic covered along with beryl and other beryllium minerals which largely increases the recovery 0! the beneflciation process.

Various changes can be made in the invention as illustrated and described since many apparently widely diflering embodiments will occur to one skilled in the art.

'What is claimed is:

1. In a process for the production oi a beryllium ore concentrate, the improvement which comprises subjecting such an ore to a pretreatment to remove iron and then to a mineral concentrating treatment involving selective filming under acidic conditions of a desired constituent in the presence oi a soluble fluoride. a lone-chain i'atty acid and a long-chain alkyl amine salt.

2. In a process for the production or a beryllium ore concentrate, the improvement which comprises removing iron lmm such an ore, desliming said ore, then agitating and aeratins'the low-.

iron comminuted ore in an aqueous pulp at pH 2 to pH 6 in the presence or per ton of ore, from 0.2 to 10 pounds of a soluble fluoride, from 0.2 to 10 pounds of a ions-chain fatty acid, and from 0.1 to 5.0 pounds 01' commercial long-chain alkyi u amineaait.'

, 3. In a process for the production oi a beryllium ore concentrate, the improvement which comprises treating such a comminuted ore with a soluble fluoride and a strong mineral acid in an aqueous Pulp. than washing the treated sands with excess water, and thereafter agitating and aerating an aqueous pul of the thus-treated ore in the presence or a soluble fluoride. a strong mineral acid, oleic acid, and a mixture of longchain alkyl amine salts corresponding to the coconut oil fatty acids.

4. In a process for the production of a beryllium ore concentrate, the improvement which comprises treating such a comminuted Ore with a soluble fluoride, a strong mineral acid. and a iong-cl1ain alkyl amine salt in an aqueous pulp, then washing the treated sands with excess water, and thereafter agitating and aerating an aqueous pulp of the thus treated ore in the presence or a soluble fluoride, a strong mineral acid, and oleic acid.

HAROID L. GIBBS. HENRY D. SNEDDEN.

Certificate of Correction Patent N 0. 2,395,475.

HAROLD L. GIBBS ET AL.

February 26, 1946.

It is hereby certified that errors appear in the rinted s ecifieation of the above numbered patent requiring correction as follows:

the word shown read a e 5, t column, line 71, for

known; and second column, e 65, Table 6A, first column thereof, for Total lead read Total head; and that the said Letters Patent should be read with these corrections therein that the same may conform to the record of the case in the Patent Oflide.

Signed and sealed this 30th day of April, A. D. 1946;

LESLIE First Assistant Commissioner of Patents.