GB2114023A - Froth flotation mineral recovery process - Google Patents

Abstract

Mineral values are recovered from ground ore (or concentrate thereof) in a plurality of froth flotation steps in a commercial-scale vessel, the improvement being that agitation is effected by an impeller at a high rate, such that the Froude number is at least 2.0 (Froude number is DN<2>/g, where D is the diameter of the impeller in feet, N is the number of revolutions of the impeller per second and g is the gravitational constant in feed per square second). The method involves using, on a commercial scale, a Froude number at least as high as the Froude number used in optimum conditions on a laboratory scale.

Description

SPECIFICATION Froth flotation mineral recovery process This invention relates to a mineral recovery system employing froth flotation.

Froth flotation has been widely practised in the recovery of many different mineral values and the basic concepts of froth flotation have been known for many years.

The purpose of froth flotation, like all mineral recovery steps, is to effect an efficient separation of the desired mineral from the undesired, or gangue. First, the mineral ore is ground to suitably small dimensions, then the ground material is treated with a flotation aid often called a collector, usually a compound having hydrophobic and hydrophilic components, the hydrophilic one typically having an affinity for the material to be floated and removed in the froth. Thus, the hydrophobic component has the effect of making the particles of material hydrophobic which causes it to go to the froth. The material to be removed in or with the froth is not necessarily, but frequently is, a concentrated form of the desired mineral.

In some processes, the froth flotation step is immediately preceded by classifiers or other separating devices or steps, which may employ centrifugal force, magnetic attraction, or other means which act in various ways to take advantage of differences in the ground ore or mineral in such physical attributes as size, shape, or specific gravity, and their relation to the viscosity, solids content, etc., of the medium in which it is suspended. Generally speaking, however, froth flotation, when used, is the last step and is almost always preceded by one or more "conditioner" steps in which a flotation aid is added.

The flotation aid added in the "conditioner" step may be cationic or anionic, meaning the hydrophilic portion has a cationic or anionic site from which an anion or a cation, respectively, will dissociate in water, but most of the commercial ones are anionic, such as xanthates and fatty acids.

After treatment with a flotation aid, the material may be passed to a "rougher" flotation series. The rougher froth may be upgraded and refloating in a "cleaner" flotation series of cells. The refloatation may be repeated in yet another series of cells. The flotation cells will typically include means for introducing air such as through a pipe concentric with the impeller drive shaft. Air may be introduced in any manner from near the bottom of the tank or cell, and is very finely dispersed by the agitation of the impeller. Recovery of material with the froth produced on the surface is partly a function of the froth's stability, which in turn is controlled to a great degree by the efficiency of the frothing agent.

Workers throughout the years have studied the physical and chemical variables of froth flotation in great detail, changing the power, shape, rpm, speed, size and other characteristics of the impeller, the shape and volume of the cell itself, the means for introducing feed material and removing froth and non-froth material, changing the frothing agent, the way of handling air, the solids content, and so forth. A review of power requirements for mixing may be found in Chapter 2 of Holland and Chapman, "Liquid Mixing and Processing in Stirred Tanks" (Reinhold, 1 965). In Chapter 3 of the same book, the problem of "scaling up" i.e. of designing commercial systems based upon laboratory results, is described in detail and the authors explain the derivation of the Reynolds number, the Froude number, and the Weber number of any given system.All of these are described as "possible scale-up criteria" (P 55), but it is stated that they are incompatible with each other, no doubt since they are expressed in different exponential functions of the same variables. None of the prior art of which we are aware suggests that the Froude number may be used alone for scale-up to reproduce laboratory recovery results in commercial vessels geometrically similar to the laboratory vessels.

Special attention should be paid, however, to a paper by E. F. Young, Jr. and John W. H. Chi, presented at the 1961 AIME meeting and incorporated herein by reference, entitled "Conditioning Rate Studies of an Iron Ore". In this paper, the authors present data which support our observation that when scaling-up a conditioner process, laboratory results may be simuiated in large scale units geometrically similar to the laboratory vessels by simply keeping the Froude number constant, varying either the RPM or the impeller diameter accordingly. However, the authors did not appreciate that the use of a high Froude number (typical for good results in a laboratory vessel) in both the conditioner cells and the flotation cells, is essential for optimum results. The art has not had, until now, a consistent and predictable method of scaling-up a froth flotation system.

According to the present invention there is provided a method of recovering mineral values from a ground ore or a concentrate thereof comprising subjecting said ore or concentrate to a plurality of froth flotation steps in series under agitation by an impeller, each of said froth floatation steps being conducted in a commercial-scale vessel under conditions such that the Froude number, DN2/g, is at least 2.0, where D is the diameter of the impeller in feet, N is the number of revolutions of the impeller per second, and g is the gravitational constant.

The method according to the invention enables improved results to be obtained in a consistent and predictable manner.

The Froude number is high, i.e. at least 2.0, and preferably as high as 4; the high Froude number is preferably maintained at all times in all conditioners (if they are used at all) in series and all flotation cells in series.

The present invention also comprises a method of scaling-up laboratory results in a commercialscale ore conditioning and froth flotation facility having a plurality of conditioning cells in series and a plurality of froth flotation cells in series, the method comprising the steps of (a) providing laboratory vessels geometrically similar to the commercial-scale vessels, (b) running a series of tests on a subject ore in the laboratory vessels to determine optimum value recoveries under various conditions, and (c) operating the commercial-scale conditioning and froth-flotation cells for said ore at optimum conditions and at ratios of DN2/g, where D is the impeller diameter in feet, N is the number of revolutions per second of the impeller, and g is the gravitational constant, at least as high as the same ratio in the laboratory at optimum recoveries.

The invention is applicable to the recovery or beneficiation of any mineral value including coal, but we have illustrated it primarily with respect to fluorspar.

First, to illustrate the problem, it had been found that in a particular commercial concentrator prior to the discovery, recovery of fluorspar (CaF2) at acid grade (97.5% CaF2) by fatty acid flotation was only about 30P/o. In the pilot plant and laboratory, recoveries had been as shown below: Recovery, percent, as acid grade Plant concentrator 30 Pilot plant 75 Laboratory 85 The conditioning and flotation steps in the laboratory, pilot plant, and plant concentrator were carried out in equipment of conventional design at operating conditions normally used.

For reasons that were previously unknown, results achieved in the pilot plant were poorer than in the smaller scale laboratory equipment and the plant concentrator, and where still larger equipment was used, results were much poorer still. In other words, some condition or conditions that prevailed in the laboratory were not being fully duplicated in the pilot plant and to an even lesser degree in the commercial plant.

The present invention resolves this problem by use of the Froude number (NFr=DN2/g where D is impeller diameter in feet, N is impeller revolutions per second and g is 32 feet/sec2) to characterize the degree of violence of agitation in conditioning and in flotation and to facilitate selection or convenient combinations of impeller speed and impeller diameter to achieve the desired degree of agitation.

Our invention will be illustrated with reference to the accompanying drawings, in which: Figure 1 shows the relationship between NFr in the flotation cells and CaF2 recovery at acid grade for the commercial ore used in the laboratory.

Figure 2 shows the relationship for NFr in conditioning and CaF2 recovery. It is observed that CaF2 recovery increases as N in flotation is increased. NFr in conditioning has a similar but lesser effect if the NFr in flotation is high but if NFr in flotation is low, increasing NFr in conditioning has little or no effect. That is, to achieve best results, it is necessary to have high NFr in both flotation and conditioning.

Figure 3 shows the relationship between NFr (conditioning and flotation) and CaF2 recovery at acid grade in the laboratory, pilot plant, and commercial plant as originally operated and designed.

Figure 4 is a more or less diagrammatic flow sheet of a commercial froth flotation plant.

Figure 5 shows the effects of varying the Froude number at two different slurry pulp densities, and Figure 6 shows the effects of varying the Froude number in conditioning of the material upstream of the froth flotation step.

Most commercially available flotation machines of 100 ft3 volume or larger are operated at NFr of about 1 or less compared with an NFr of 4 or more in most laboratory cells. Commerciai scale and laboratory conditioning shows similar values for NFr, respectively.

The data for Figure 1 were obtained under laboratory conditions wherein only the Froude number was varied, ail other conditions being kept constant.

Figure 1 suggests that if results equivalent to the laboratory are to be achieved in the plant, the NFr should be 4 or higher. One combination that could be used to achieve this would be a 22-inch impeller rotating at about 8.2 rps. Because of limitations of the installed drive motors at the commercial installation, an NFr of only about 2.4 was obtainable. This was achieved using a 22-inch impeller rotating at about 6.5 rps. Using this NFr in flotation with a similar NFr in commercial conditioners. CaF2 recovery was increased from 30% to about 65 to 70% as acid grade. This is in good agreement with values predicted from laboratory work as shown in Figure 1 and also with the curve in Figure 3.

Similar but less pronounced relationships have been observed for amine flotation of silica and silicates from iron oxides. Such effects of NFr are apparently much less in sulfide flotation, and in the flotation of coal from ash, probably because of the high degree of flotability of coal and of most sulfides when properly conditioned and activated.

The flotation testing of ores is commonly performed in a standard laboratory batch machine, which is usually a prototype of the larger cells available in a particular manufacturer's family of flotation machines. Most of the previous testing of the commercial fluorspar ores used throughout this specification was done in a standard Denver laboratory machine using a 3-liter cell. All parts of the laboratory cell have approximately the same geometric shape as the corresponding parts of the production cells; the ratios of corresponding linear dimensions between laboratory and production cells tend to be constant in the smaller sizes but deviate somewhat in the larger sizes. The various dimensions and scale ratios of the Denver Sub-A family of flotation machines are shown in Table I. The scale ratios of the laboratory, 50-ft3, and 1 00-ft3 cells satisfy geometric similarity.Production of acidgrade and metallurgical-grade concentrates had been accomplished in 1 00-ft3 and 50-ft3 Denver Sub A flotation machines, respectively, using the same impeller-diffuser design as the laboratory machine but recovery was much lower.

The impeller peripheral speed appears to be a key factor used by the manufacturer (Denver) in flotation machine scale-up. For example, it is essentially constant at about 1 500 minute for the entire family of Denver machines, as shown in Table I. The same is true of many other brands of flotation machines.

While in the past it has been generally true, at least partly because the peripheral speed of the impeller is kept relatively constant, that the Froude numbers of laboratory machines have been considerably higher than "scaled-up" commercial units, our approach in scaling up is to maintain the Froude number relatively constant, i.e. at least as high as the laboratory optimum. Since an increase in the rotational speed has a greater effect on DN2/g than on increase in the diameter of the impeller, the size of the impeller need not be increased in proportion to the other dimensions of the cell, so long as the speed of the impeller is increased.

Table 1 Effect of size on characteristics of flotation cells Tankvolume-ft3 0.125 10 50 100 300 500 1275 Depth-ft-H 0.46 3.00 3.90 4.60 5.00 6.50 8.5 Length-ft-L 0.47 1.83 3.58 4.66 7.33 8.83 12.5 H/L 0.98 1.64 1.09 0.99 0.82 0.74 0.68 lmpellerdiameter-ft-D 0.234 0.92 1.83 2.25 2.75 2.75 4.17 -rev/sec-N 26.7 8.94 4.41 3.58 2.78 2.78 1.98 -peripheralfI/min 1550 1540 1520 1518 1441 1441 1556 D/L 0.50 0.50 0.51 0.48 0.38 0.33 0.33 NRex 103 * 0.29 1.51 2.95 3.62 4.20 4.20 6.89 NFr 5.18 2.28 1.11 0.90 0.66 0.66 0.51 Theoretical power--HP 0.059 2.07 7.75 11.7 14.9 14.9 43.1 HP/ft3 0.47 0.21 0.16 0.12 0.05 0.03 0.03 * Based on 7816r/ft3 density p and viscosity of 0.39 Ib/ft-secx 10-3 Denver cells The ratios of the applied forces between systems must be constant in order to satisfy dynamic similarity. The forces that may exist in a fluid system are pressure, gravity, viscous, elastic, and surfacetension forces. Some of the forces are relatively insignificant in a particular system and can be neglected.

Dimensional analysis can be used to represent the ratio of the applied to the predominant opposing forces in a system. Pressure forces and elastic forces are not predominant in the flotation process; however, viscous forces, surface tension forces, and gravity forces may be significant. The dimensionless groups such as Reynolds number, Weber number, and Froude number are often used to describe the ratios of the applied to resisting viscous, surface tension, and gravitational forces, respectively.Reynolds number NAa is D2Np/y, Weber number Nwe is D3N2p a, and Froude number NFr is DN2/g where D is impeller diameter in feet, N is impeller speed in revolutions per second, p is slurry density in pounds per cubic foot, u is slurry viscosity in pounds per foot-second, a is surface tension in poundals per foot, and g is gravitational acceleration of 32.2 feet per second.

The previously mentioned study on fatty-acid conditioning for iron ore flotation, (E. F. Young, Jr.

and John W. H. Chi, Jones s Laughlin Steel Corporation, "Conditioning Rate Studies of an Iron Ore", January 27, 1 961), showed a strong correspondence between iron recovery and Froude number for different size conditioners used independently. An examination of data on recovery in the laboratory, pilot plant, and production plant showed a similar correlation with the commercial fluorspar studied and related in this disclosure. Consequently, the Froude number in conditioning and flotation was chosen for investigation to explain the difference between the high CaF2 recovery consistently achieved in laboratory flotation testing and the low CaF2 recovery experienced in the plant.In the standard laboratory flotation test, conditioning was performed at N Fr 5.34 and 60% solids, and flotation was performed at NFr 3.80 or higher; in the commercial plant, conditioning was performed at NFr 0.78 and 35% solids, and flotation was performed at NFr 1.22.

The configuration of the commercial plant was as illustrated in Figure 4. The effect of NFr in flotation on the CaF2 recovery normalized to acid-grade concentrate is shown in Figure 5. The flotation tests were performed at NFr 1.22, 1.50, 2.40, and 3.80 on ore conditioned at 35% solids and 60% solids and conditioning NFr 0.91,2.00, and 3.05. A sharp increase in CaF2 recovery occurred as flotation NFr was increased up to 2.40; only a small increase in recovery occurred above flotation NFr 2.40. For example, on ore conditioned at 60% solids, the normalized CaF2 recovery averaged 51, 67, 82, and 84% for flotation NF, 1.22, 1.50, 2.40, and 3.80, respectively.Ore conditioned at 60% solids resulted in a normalized CaF2 recovery of about 10 percentage points higher than ore conditioned at 35% solids; also, ore conditioned at high NFr generally gave higher CaF2 recovery, but an insufficient number of tests were performed to develop a family of curves.

The effect of NFr in conditioning on the CaF2 recovery normalized to acid-grade concentrate is shown in Figure 6. Conditioning for the results shown in Figure 5 involved a plant configuration as in Figure 4 wherein small amounts of wattle bark extract (a lignin derivative) and sodium carbonate were added in an upstrqam conditioner, followed by small amounts of fatty acid. As is known in the art, the object of "conditioning" is to prepare the material, usually in the absence of aeration, for treatment in the froth flotation units.Dispersants such as wattle bark are employed to achieve a high degree of water-wetness, i.e. an enhancement of the hydrophilic tendencies, in this case, of the carbonate component of the fluorspar, which will ultimately "suppress" the tendency of it to enter the froth; the fatty acid, through agitation in the conditioner, tends to associate with an enhance the relative hydrophobic tendencies of the fluorspar, thus increasing the likelihood it will report to the froth.

Typically, nothing is discarded in a conditioning step. Nevertheless, we have found that a high NFr in the conditioning step will enhance recoveries in the froth flotation step. Flotation tests were performed on ore conditioned at 35% solids and at 60% solids, with the flotation NFr held constant at 2.40, which appeared to be near optimum based on the previous section. The NFr in conditioning had a significant effect on recovery, especially at 35% solids. At NFr 0.91 and 35% solids, similar to current plant conditioning, the CaF2 recovery was 57%; recovery increased rapidly to 72% at NFr 2.00 and 79% at NFr 3.05. The effect of NFr on increased CaF2 recovery at 60% solids conditioning is less than at 35% solids; recovery increased from 80% at NFr 0.91 to a maximum of 86% at NFr 2.00, and did not increase beyond NFr 2.00.The flotation response of ore conditioned at 60% solids was better than that of ore conditioned at 35% solids, especially at low NFr; normalized recovery was about 23 percentage points higher at NFr 0.91, 14 percentage points higher at NFr 2.00, and 7 percentage points higher at NFr 3.05.

Thus, it can be seen that the Froude number appears to be a reliable scale-up factor between laboratory and commericial-sized conditioning and flotation machines; this conclusion is based on the close agreement between CaF2 recovery of acid-grade concentrate achieved in the laboratory and that achieved at the commercial installation, and on the work of others with iron ore conditioning.

The plant had been achieving a CaF2 recovery of acid-grade concentrate of about 30% when conditioning the ore at 35% solids and NFr 0.78, and floating at NFr 1.22; laboratory flotation under the same conditions resulted in about the same CaF2 recovery. The following changes listed in order of magnitude in improvement of CaF2 recovery, were made in the laboratory and resulted in increasing the CaF2 recovery to 86%.

1. Increase flotation N,,--as flotation NFr was increased from 1.22 to 2.40, the CaF2 recovery of acid-grade concentrate improved from about 30 to 57% (shown in Figure 1).

2. Increase conditioning N,i--as conditioning NFr was increased from 0.91 to 3.05, the CaF2 recovery of acid-grade concentrate improved from about 57 to 79%.

3. Increase conditioning density-as conditioning density was incrased from 35 to 60% solids, the CaF2 recovery of acid grade improved from about 79 to 86%.

Our observations from the above-described data is that the Froude number must be high, i.e. over about 2.0, in both flotation cells and conditioning cells.

Claims (13)

Claims

1. A method of recovering mineral values from a ground ore or a concentrate thereof comprising subjecting said ore or concentrate to a plurality of froth flotation steps in series under agitation by an impeller, each of said froth flotation steps being conducted in a commercial-scale vessel under conditions such that the Froude number, DN2/g, is at least 2.0, where D is the diameter of the impeller in feet, N is the number of revolutions of the impeller per second, and g is the gravitational constant.

2. A method according to claim 1 wherein the ore treated is fluorspar.

3. A method according to claim 1 or 2 wherein the ore or concentrate is treated with a flotation agent prior to the froth flotation steps.

4. A method according to claim 3 wherein the flotation agent is a fatty acid.

5. A method according to claim 3 or 4, wherein the flotation agent is added in a conditioning step under agitation and a Froude number of at least 2.0 is employed in both the conditioning steps and the flotation steps.

6. A method according to claim 5, wherein the Froude number of all the conditioning and flotation steps is at least 2.4.

7. A method according to claim 6, wherein the Froude number of all the conditioning and flotation steps is at least 3.8.

8. A method according to any one of claims 1 to 5, wherein the Froude number in all the froth flotation steps is at least 2.4.

9. A method according to claim 8, wherein the Froude number in all the froth flotation steps is at least 4.

10. A method according to any of claims 1 to 9, wherein the commercial-scale vessel has a capacity of at least 50 cubic feet.

11. A method according to claim 10, wherein the capacity is at least 100 cubic feet.

12. A method of scaling-up laboratory results in a commercial-scale ore conditioning and froth flotation facility having a plurality of conditioning cells in series and a plurality of froth flotation cells in series, the method comprising the steps of (a) providing laboratory vessels geometrically similar to the commercial-scale vessels, (b) running a series of tests on a subject ore in the laboratory vessels to determine optimum value recoveries under various conditions, and (c) operating the commercial-scale conditioning and froth-flotation cells for said ore at optimum conditions and at ratios of DN2/g, where D is the impeller diameter in feet, N is the number of revolutions per second of the impeller, and g is the gravitational constant, at least as high as the same ratio in the laboratory at optimum recoveries.

13. A method of recovering mineral values from a ground ore or a concentrate thereof, substantially as herein described.