FROTH FLOTATION PROCESS

CROSS REFERENCE TO RELATED APPLICATION

This application is a continuation in part of copending ap¬ plication Serial No. 934, 132 filed August 15, 1978.

BACKGROUND OF THE INVENTION ^' This invention relates to an improvement in a froth flotation process for concentration and separation of metallic sulfide mineral ores . The improved process is directed to separations wherein a mercaptan is

WIPO

utilized as- a collector in an earlier flotation stage. The improved method of this invention includes the addition of activated carbon to achieve de- acttvatiαn of the mercaptan prior to a mineral separation stage and to achieve enhanced separation of the metallic elements desired. Froth flotation is a process commonly employed for separating, collecting, and, hence, concentrating valuable minerals, particularly sul¬ fide and oxide ores, from -the gangue minerals associated with these min¬ erals in their ores. The usual steps are as follows :

(a) The ore is crushed and subjected to wet grinding to provide a pulp wherein the ore particles are typically reduced to minus

48 mesh with about 50% of the particles being in the minus 200 mesh fractions.

(b) The ore pulp is generally diluted with water to approx¬ imately 30% solids by weight. (c) Various conditioning, collecting, and frothing agents are then added to the mineral pulp .

(d) The pulp is then aerated to produce air bubbles that rise to the surface of the pulp and to which the desired mineral particles selectively attach themselves by virtue of the characteristics of the col- lectors employed, thereby permitting removal of these minerals in a con¬ centrated form.

There are, of course, numerous patents on processes for froth flotation concentration and separation of minerals. One such patent is U .S . Patent 2, 559, 104 (issued July 3, 1951) to Arbiter et al which relates to a flotation recovery method for molybdenite. Arbiter et al teaches a specific system in which a collector is oxidized prior to sub¬ sequent separation stages. The problem addressed in the Arbiter et al

patent involves reducing excess frother and excess collector in the sub¬ sequent cleaning stage. They tend tα collect by virtue of the fact that the bulk of the collector and. frother are carried forward into the next cleaning stage. In the Arbiter et al patent, reduction of the excess frother is accomplished by the addition of activated carbon as required. U .S . application serial number 852, 413, filed February 17, 1977 by Adriaan Wiechers, (the specification and claims of which are specifically incorporated herein by reference) . teaches an improved pro¬ cess utilizing a mercaptan as a collector, the preferred mercaptan being normal dodecyl mercaptan ("DDM") . As will be seen hereinafter, the use of DDM increases the overall copper recovery from the ore, but at the same time can make separation of the copper from the -molybdenite more difficult.

DRAWINGS . Figures I and II are general flowsheets illustrating treat¬ ment of ores from two different sources , Ore A in Figure I and Ore B in Figure II. In each figure, the flowsheets compare the treatment steps and recovery percentages for a standard plant process of concen¬ tration and separation, a process employing DDM concentration and stan- dard separation, and a process employing DDM concentration and the novel separation procedures of the present invention.

SUMMARY OF INVENTION The improved process of this invention relates to the spe¬ cific separation of metallic sulfide mineral ores comprising copper and molybdenum through flotation wherein an alkyl mercaptan has been used as a collector in an earlier flotation stage to provide a cleaner ^" concentrate

OMPI

having the mercaptan present. The improvement in the process comprise deactivating the mercaptan, whereby the subsequent separation flotation stage is improved. The deactivation of the mercaptan is achieved by the addition of an effective amount of powdered activated carbon. - From the drawings, it is clear that an improvement in the overall yield of copper can be achieved by employing an alkyl mercaptan collector, 91.5% as compared to 90% in treatment of Ore A, and 89 to 89.7% as compared to 86.6% in treatment of Ore B . Unfortunately, 33.4% of. the copper from Ore A and .67.4% of the copper from Ore B are car¬ ried into the molybdenum circuit when DDM is employed, as compared to 18. 7% and -42.9%, respectively, for the previously employed standard plant procedure. Using the separation procedure of the present inven¬ tion to deactivate the DDM prior to separation, only 8.2% of the copper in Ore A and ^' 11. 0% of the copper in Ore B are carried into the molyb- denum circuit, providing a copper concentrate of 91.8% for Ore A and and 89% for Ore B as compared to 81.3% and 57. 1% for the standard plant process .

More specifically, the improved process is a method for re¬ covery of metal values by froth flotation from metallic sulfide mineral ores comprising copper and molybdenum, including the steps of :

(A) forming an aqueous mineral pulp from the ore-:

(B) subjecting the pulp to rougher flotation to provide a scavenger feed and a rougher concentrate :

(C) adding an effective amount of an alkyl mercaptan of the formula C H, ₊ ι SH in which n is at least 12 to the primary flotation stages as a collector and subjecting the scavenger concentrate;

(D) combining, regrinding, and cleaning the concentrates

frora the primary flotation stages (B) and (C) to provide a copper molybdenum cleaner concentrate; and then

(E) subjecting the cleaner concentrate of step (D) to com¬ ponent mineral stage flotation separation; the improvement which com- prises deactivating substantial amount of the mercaptan collector on the mineral of the ore in the cleaner concentrate of step (D) prior to the component mineral stage flotation separation in step (E) , said deactivat¬ ing comprising adding a deactivating effective amount of activated carbon to the cleaner concentrate prior to flotation in step (E) ; to provide more effective mineral separation.

• It is preferred that the activated carbon be added within the range of about 0.25 to about 1.0 pound of activated carbon per ton of initial ore feed and t at it be added to the cleaner concentrate for a sufficient time interval prior to step (E) to provide substantial deactiva- tion of the mercaptan prior to commencement of step (E) . Such time interval is preferably within the range of about 5 to about- 30 minutes .

The invention is particularly applicable to copper-molybdenum sulfide containing mineral ores and is quite suited to the typical type of Arizona porphyry ores.

Description of the Preferred Embodiments

The process of this invention involves subjecting the ore feed to primary grinding and then rougher flotation, including the ad¬ dition of the appropriate reagents, to provide a feed to the scavenger flotation stage after which the rougher concentrate and the scavenger concentrate are combined, subjected to a regrinding, and then subjected to a number of cleaner flotation stages. Prior to commencement of the scavenger flotation stage, from about 0.005 to about 0.02 pounds per ton

ore of a mercaptan ( such as normal dodecyl mercaptan, ^" "DDM") is adde as an auxiliary collector or promoter to provide increased metals recove during the primary flotation stages. With certain sulfide minerals such as copper and molybdenum containing ores, the DDM produces undesir- able effects in the subsequent separation stage. The process of this invention involves substantially deactivating the DDM prior to the min¬ eral separation stage .

Ore Sample A A representative ore sample which is the feed to a concen- trator is obtained from a typical producing copper-molybdenum concen¬ trator located in Arizona. Copper occurs predominately as chalcopyrite and molybdenum occurs primarily as molybdenite .

Distribution data for the ore sample show that copper value are approximately equally distributed on all size fractions from 65- to plus 400-mesh with a high distribution of copper ( 47%) in the minus 400 mesh ( 37 micrometers) . A relatively constant distribution of molybdenu occures in the coarser size fractions while 67% reports to the minus 400 mesh fraction. The copper and molybdenum minerals are liberated at a relatively coarse mesh of grind. The assays of the three concentrator cyclone overflow samp utilized in the examples are as follows :

Table 1

Assay , %

Direct Calculated ¹

Cu Mo Cu Mo

Sample 3 0. 39 0. Q14 0. 38 0. 01 Sample 4 0. 37 0.018 0. 38 0.01 Sample 5 0. 35 0.003 - 0. 34 0.00 calculated from i tests

Standard conditions and reagent balance is shown in Table 2. The reagent balance is substantially identical to that of current conven¬ tional plant practice. •

Table 2 Test Conditions and Reagent Balance

Feed - 4000 grams dry solids cyclone overflow pulp sample

Reagents Added, lb /Ton of Ore ¹ Time, Shell Minute !S

Stage CaO Z-6 ³ AF-238* 1638 ⁵ Cond 1 Froth PH

Condition 1.0 ^" 1 11.0

Rougher . 0.01 0.005 0.03 1 5

Scavenger 0.01 1 5 10. 7

Thicken ²

Regrind 0.25 10

1st cleaner 0.005 1 3 11.2

2nd cleaner 0.10 1 3 11.2

3rd cleaner 0. 10 1 2 11.2

NaCN / (NHk S ₂ NaSH ZnSO t _*

Condition 1 . 11.0 10

Condition 2 25.0 5

Mo rougher 3 9. 3

Mq 1st cleaner 5. 0 5 3

Mo 2nd cleaner 2.0 3 2 9.0

Re gent additions based on lb /ton of ore with exception of (NH^ ) ₂ S , NaSH , and NaCN /ZnSO additions which are based on lb /ton Cu-Mo cleaner concentrate .

² Combine rougher and scavenger concentrates . Thicken to approximately 60% solids . _ ³ Potassium amyl xanthate

^Sodium di secondary butyl dithiophosphate ^s 85% methyl isobutyl carbinol, 15% distillate bottoms

The most desirable, readily available activated carbon usef in deactivating the mercaptan collector is of a relatively high pore sur face area of about 0.95 ml per gram and is a lignite-based powdered a tivated carbon. ICI type GFP is particularly useful. Activated carbon addition is made prior to the sulfidizing reagent addition in the copper-molybdenum separation and about 10 minutes allowed for conditioning.

Summarized in Table 3 -are the comparative results illustrat the significant improvement in deactivating the mercaptan collector (DD with the addition of activated carbon, while the effect of varying level ■ of activated carbon is illustrated by the results shown in Table 4.

Table 3 \

Comparison of Effect of General Ci u-Mo Separation Processes

Feed Weight _, Distribution, Sample Percent Assay, , % % Overall

No Process Product Overall Cu Mo Cu Mo

2 Standard-plant Mo Ro Cone 0.20 27.9 1.48 18.7 38.8 (no DDM) Cu Cone 0.75 26.2 0.07 66.0 7.0 Cu+Mo Cl Cone 0.95 26.6 0.37 84.7 45.7

2 Standard-plant* Mo Ro Cone 0.40 25.1 0.86 33.4 43.9

10 Cu Cone 0.65 24.3 0.06 52.6 4.9 Cu+Mo Cl Cone 1.05 24.6 0.36 86.0 48.8

4 Standard-plant* Mo Ro Cone 0.37 25.7 1.19 26.3 36.0 1

1 Cu Cone 0.74 23.8- 0.04 54.2 2.3 Cu+Mo Cl Cone- 1.11 26.2 0.40 80.5 38.3

15 3 Activated carbon* Mo Ro Cone 0.20 19.5 2.23 10.0 32.9 (1.0 lb /ton ore) Cu Cone 1.06 26.0 0.05 71.3 3.9 Cu+Mo Cl Cone 1.26 25.0 0.40 81.3 36.8

Table 4

Effect of Varying Level of Activated Carbon on Cu-Mo Separation

Activated Distribution,

Sample Carbon Weight Assay, , % % Overall

No lb/Ton Ore Product Percent Cu Mo Cu Mo

2 Mo Ro Cone 0.40 25.1 0.86 33.4 43.9

- Cu Cone 0.65 24.3 0.059 52.6 4.9

Cu+Mo Cl Cone 1.05 .24.6 0.36 86.0 48.8

4 - Mo Ro Cone 0.37 25.7 1.19 26.3 36.0

10 Cu Cone 0.74 23.8 0.035 54.2 2.3

Cu+Mo Cl Cone 1.11 26.2 0.40 80.5 38.3

4 0.25 Mo Ro Cone 0.23 19.9 1.84 13.6 23.3

Cu Cone 0.88 26.0 0.041 67.6 2.0

Cu+Mo Cl Cone 1.11 24.7 0.41 81.2 25.3

15 3 0.50 Mo Ro Cone 0.22 24.0 2.27 13.9 35.4 1

4-4

Cu Cone 0.94 27.0 0.060 67.1 4.0 o

1

Cu+Mo Cl Cone 1.15 26.7 0.48 81.0 39.4

3 1.0 Mo Ro Cone 0.20 19.5 2.23 10.0 32.9

Cu Cone 1.06 26.0 0.050 71.3 3.9

20 Cu+Mo Cl Cone 1.26 , 25.0 0.40 81.3 36.8

4 1.35 Mo Ro Cone 0.20 15.7 2.06 10.9 24.3

Cu Cone 0.86 24.4 0.14 73.3 7.1

Cu+Mo Cl Cone 1.06 22.8 0.50 84.2 31.4 '

4 2.0 Mo Ro Cone 0.18 17.7 1.24 8.9 12.2

25 Cu Cone 1.07 24.2 0.31 72.7 18.2

... The results "indicate that 0.25 to 0.50 pound activated carbon per ton ore is sufficient to reduce the copper displacement in the molyb- denum circuit to approximately 13% from approximately 30% without acti¬ vated carbon. Increasing the activated carbon level to one pound per ton ore result in only a marginal further decrease of copper loss in the molybdenum circuit to about 10%.

Increasing the activated carbon level to greater, than one pound per ton of ore does not appear, to significantly reduce copper loss to the molybdenum circuit, but it may result in reduced molybdenum recovery to the molybdenum rougher concentrate .

A similar series of experiments were conducted on another typical copper molybdenum ore from a different location in Arizona, designated for convenience, as Ore B . These experiments developed the data for Tables 5 through 9. Table 5 contains the head assay, Table 6 sets forth the " reagent balance , and Table 7 the copper-molybdenum separation reagent balance for the Ore B experiments. Table 8 shows that using activated carbon in the process of the present invention, the copper concentrate contains 92, 5% of the copper as compared with 57. 1% for the standard plant process and 32. 6% for DDM with the standard separation process . Table 9 shows the effect of varying levels of activated carbon, while Table 10 illustrates the wise variety of activated carbons which can be employed.

Table 5

Head Assays - Ore B

Assay, %

Direct Calculated ¹

^T Mo Cu Mo

Sample 1

(HRI No. T-229) 0.70 ovors 0. 69 0.015

Sample 2

(HRI No. T-236) 0.72 0.018 0.73 0.018

^L Average head assays as calculated from all tests

Additional assays were performed on the Sample 1 head sample.' The results are shown below. Assay, % on- Non- Sulfide Sulfide Cu ¹ Mo Fe S (Total)

Sample 1 0.060 <0.001 3.05 1. 77

^x Assay confirmed by two analysts

Λ,

Table 6

Reagent Balance - Ore B

Reagents Added , lb /Ton Ore Time,

Fuel Minutes

Stage CaO Sm-8 ¹ Oil ² Z-ll ³ MIBC Cond Froth pH

1.2 _, 0.015 0.025 0.05

Rougher 6 10.0

Scavenger 0.003 0.01 1 ^' 6 9.7

Thicken ⁵ . -

Regrind 0. 2 0.01 -

1st cleaner 0.005 1 4 10. 0

2nd cleaner 1 . 3 9. 2

Stage Rougher-scavenger 1st, 2nd cleaner

Equipment Denver D-l, 1000 g cell Denver D-l, 250 g cell Speed , rpm 1900 1200

Airflow, 1/min L6 % solids 35 15 .

^inerec Sm-8

² Fuel oil - 50 : 50 mixture No. 2 diesel oil/kerosene ³ Sodium ethyl xanthate

^MIBC - 85% methyl iosbutyl carbinol /15% MIBC distillation bottoms ⁵ Thickened rougher-scavenger concentrate to approximately 60% solids - decanted (reclaim) water used as makeup in cleaner stages

Table 7

Copper-Molybdenum Separation Reagent Balance

Reagents Added, lb /Ton Concentrate Feed Time,

NaCN ₃ Na-Ferro K-Ferri Minutes

Stage H.. SO., ¹ ZnO ² H ₂ O ₂ CN CN NaOCl MIBC Cond Froth EH

Condition 1 0. 50 0. 46 - - . - - 20 - 8.7-6. 7

Condition 2 0.20 3.75 - - - 20 - - 6.9-6. 6

Mo rougher 0. 20 _ 2.0 - 0.004 4 7.0

Mo 1st cleaner - - 1.0 - 0.003 3 7.4

10 Mo 2nd cleaner - _ - 0. 20 1.0 - 3 7.6

Mo 3rd cleaner - - - 0. 10 0.02 2 7.7 .

1

Mo 4th cleaner - - - 0.10 0.02 2 7.8 •-•

Mo 5th cleaner - - - 0. 10 0.01 2 8.0

Mo 6th cleaner - - - 0. 10 0.01 π 8. 1

15 Condition 1, 2 - pulp density 50% solids Mo rougher - pulp density 20% solids

Table 8

Comparing Cu/Mo Separation With and Without DDM and Activated Carbon

Weight Assay, % Distribution, %

Cu Mo Cu Mo

Conditions Product %

13.3 19.6 0.8 51.5

Standard separation Mo Cl Cone 1.68 on concentrate with¬ Mo Ro Cone ^• 35.81 31.6 1.61 42.9 90.7 out DDM Cu Cone 64.19 23.4 0.09 57.1 9.3

Head (calc) 100.00 26.3 0,64 100. 0 100.0

9. 5 82.8

67. 4 95.4 32. 6 4.6 Head (calc) 100.00 26.7 0.61 100. ,0 100.0 7 0. .5 57.8

DDM plus 0.6 lbs/ Mo Cl Cone 0.91 13.8 33. ton ore activated Mo Ro Cone 6.78 28.0 7.04 7 .5 89.8 carbon 93.22 25.1 0.058 92 .5 10.2

15 Cu Cone Head (calc) 100.00 25.3 0.53 100.0 100.0

Table 9

Effect of Varying Level of Activated Carbon In Ore B Experiments

Weight Assay , * Dlstribi it ion, %

Conditions Product % ^" Cu Mo Cu Mo

5 Standard, no acti- Mo 3rd Cl cone 8.74 28.9 5.80 9.5 82.8

^■ vated carbon Mo Ro cone 57.38 31.3 1.02 67.4 95.4

Cu cone 42.61 20.4 0.067 32.6 4.6

Head (calc) 100.00 26.7 0.61 100.0 100.0

0.075 lb activated Mo 3rd Cl cone 9.23 29.2 ^• 5.30 10.8 81.3

10 carbon /ton ore Mo Ro cone 38.21 30.8 1.46 47.5 93.8

(1.37 lb/ton cone) Cu cone 61.79 21.0 0.059 52.5 6.2

Head (calc) 100.00 24.8 0.60 100.0 100.0

0.15 lb activated Mo 3rd Cl cone 3.66 • 26.4 12.0 3.9 76.5 carbon /ton ore Mo Ro cone 23.84 30.6 . 2.20 29.5 91.5

15 (2.73 lb /ton cone) Cu cone 76.16 22.8 0.064 70.5 8.5 1

Head (calc) 100.00 24.7 0.57 100.0 100.0 σt

0.30 lb activated Mo 3rd Cl cone 2.74 21.2 16.1 2.4 74.7 carbon /ton ore Mo Ro cone 16.80 28.9 3.18 19.8 90.5

(5.45 lb /ton cone) Cu cone 83.20 23.7 0.068 80.2 9.5

20 , Head (calc) 100.00 24.6 0.59 100.0 100.0

0.60 lb activated Mo 3rd Cl cone 1.77 13.7 23.5 1.0 69.8 carbon/ton ore Mo Ro cone 10.73 26.6 4.98 11.6 89.5

(10.91 lb/ton cone) Cu cone 89.27 24.4 ^' 0.070 88.4 10.5

Head (calc) 100.00 24.6 0.60 100.0 iόό.o ^•

25 0.90 lb activated Mo 3rd Cl cone. 2.60 18.5 15.5 2.0 75.1

_' carbon /ton ore Mo Ro cone 11.47 26.9 4.17 12.6 89.2

(16.38 lb/toii cone) ' Cu cone 88.53 24.2 0.066 87.4 10.8

. / Head (calc) 100.00 24.5 0.54 100.0 ' 100.0

Table 10

Effect of Type of Activated Carbon (0.6 Pounds Per Ton Ore)

Weight Assay, % Distribu ition, %

Activated Carbon Product % Cu Mo Cu Mo

Darco-GFP Mo 2nd Cl cone 0.91 13.8 33.7 0.5 57.8

Mo Ro cone ^• 6.78 28.0 7.04 7.5 89.8

Cu cone 93.22 25.1 0.058 92.5 10.2

Head (calc) 100.00 25.3 0.53 100.0 100.0

Darco-FM-1 Mo 3rd Cl cone 1.16 10.5 28.4 0.5 67.1

10 Mo Ro cone 7.30 26.2 6.17 7.5 91.7

Cu cone 92.70 25.4 0.044 92.5 8.3

Head (calc) 100.00 25.5 0.49 100.0 100.0

Calgon-PCB Mo 3rd Cl cone 2.57 18.3 17.0 1.9 78.4

Mo Ro cone 13.13 28.6 3.99 15.0 93.9 1

15 Cu cone 86.87 24.8 0.039 85.0 6.1

Head (calc) 100.00 25.3 0.56 100.0 100.0

Union Carbide-LC Mo 3rd Cl cone 2.40 14.0 18.5 1.3 74.0

Mo Ro cone 11,70 27.6 ^• 4.75 12.8 92.6

Cu cone 88.30 25.0 0.050 87.2 7.4

20 Head (calc) 100.00 25.3 0.60 100.0 100.0

Norit-RO 0.8 Mo 3rd Cl cone 1.33 5.52 31.3 0.3 67.5

Mo Ro cone 10.80 26.4 5.35 11.2 93.7

Cu cone 89.20 25.4 0.043 88.8 6.3

Head (calc) 100.00 25.5 0.61 100.0 100.0

Sethco-powdered Mo 3rd Cl cone 4.30 23.1 9.92 3.9 76.9

Mo Ro cone 18.00 29. ^' 8 2.85 21.0 92.3 •

Reference was made hereinbefore to. United States Patent No. 2, 559, 104 ^' to Arbiter et al which teaches the oxidizing of a collector prio to the subsequent separation stages, and the use of activated carbon to reduce excess frother and excess collector in the subsequent cleaning 5 stages. While apparently similar to the process of the present invention the chemical route taught by Arbiter et al is, in fact, exactly opposite to that employed in the process of the present invention. Thus while Arbiter et al teaches the use of an oxidizing agent to deactivate the collector, the process of the present invention employes activated car¬ lo bon to deactivate the collector, and there is strong evidence that in so doing, the activated carbon acts as a reducing agent.

Measurements were made of the oxidiation-reduction potential (emf) of the pulp just prior to molybdenum rougher flotation. These measurements were made at various levels of activated carbon and the 15 results are set forth in Table 11.

Table 11

Pounds Activated Carbon Pulp emf,

Per Ton Ore -mv

0.00 380

20 0.075 360

1.15 ^' 300-

0.30 260

0.60 190

Θ.90 180

25 1.25 170

In addition, it has been found that sodium . zinc cyanide^ whic was heretofore considered to be an essential reagent to the process, can be omitted. A further series of tests were conducted in which the emf was measured on a series of pulps wherein the sodium zinc cyanide was

^' 30 omitted, the level of activated carbon was maintained constant, and only the conditioning time was varied. The data developed in these further

tests are set forth in Table 12, while the distribution of copper and molybdenum is described in Table 13.

Table 12

0..60 lb Activated Carbon Pulp emf,

/Ton Ore -rav

(20 minute A.C . cond time) 160

(10 minute A.C . cond time) 190

( 5 minute A .C . cond time) 230

OMPI

Table 13

Effect f Elimination of Sodium Zinc Cyanide

^'

Weight Assay, % Distribi ition,

Condition Product % Cu Mo Cu Mo

Standard, with Mo 3rd Cl cone 1.77 13.7 23.5 1.0 69.8 NaZnCN 0.60 lb A.C. Mo Ro cone 10.73 26.6 ^• 4.98 11.6 89.5 /ton ore to Cond 1 Cu cone 89.27 24.4 0.070 88.4 10.5

Head (calc) 100.00 24.6 0.60- 100.0 100.0

0.60 lb A.C. /ton ore Mo 3rd Cl cone 1.13 12.7 29.7 0.6 65.8

10 No NaZnCN Mo Ro cone 9.03 ^' 29.7 5.10 10.4 90.2

Cu cone 90.97 25.3 0.055 89.6 9.8

Head (calc) 100.00 25.7 0.51 100.0 100.0 I t o

No activated carbon Mo Ro cone 45.97 30.2 1.38 54.3 96.4 No NaZnCN Cu cone 54.03 21.6 ^• 0.044 45.7 3.6

15 Head (calc) 100.00 25.6 0.66 100.0 100.0

The data in Tables 11 and 12 clearly indicate that as the level of activated carbon increased, and/or as the conditioning time in¬ creased for a fixed level of carbon, the emf pf the pulp decreased. In other words, the net effect of the treatment with activated carbon was to achieve a reduction reaction as evidenced by these substantially lower emf measurements.

Though not willing to be bound by any one theory by which the ^' functioning of the activated carbon might be explained, at lease one possible mechanism is that the activated carbon functions by desorption of oxygen from the collector-mineral surface bond to render a given sulfide mineral hydrophillic. Desorption of the oxygen from the sulfide minerals surface would render collector inactive, and therefore, the mineral particle hydrophillic. In a copper molybdenum separation, the action of the activated carbon is apparently specific to copper and iron sulfide minerals rendering these less floatable than the molybdenite , while it very surprisingly does not appear to cause desorption of oxygen and/or collector from the molybdenite surface and the molybdenite, therefore, continues to be hydrophobic.

It will, of course, be obvious to those skilled in the art,. that many changes and substitutions can be made in the specific mate-, rials, reactants, and procedural steps set forth hereinbefore, without departing from the scope of the present invention, and it is my inten¬ tion to be limited only by the appended claims. As my invention, .1 claim :