- Email: [email protected]
Factors affecting the floto-elutriation process efficiency of a copper sulfide mineral
Minerals Engineering 86 (2016) 59–65
Contents lists available at ScienceDirect
Minerals Engineering journal homepage: www.elsevier.com/locate/mineng
Factors affecting the floto-elutriation process efficiency of a copper sulfide mineral M. Paiva, J. Rubio ⇑ Minerals Engineering Department, PPGE3M – Universidade Federal do Rio Grande do Sul, Av. Bento Gonçalves, 9500, Prédio 43819 – Setor 6, 91501-970 Porto Alegre, RS, Brazil1
a r t i c l e
i n f o
Article history: Received 17 September 2015 Revised 2 November 2015 Accepted 25 November 2015
Keywords: Floto-elutriation Particle size distribution Recovery Grades
a b s t r a c t Coarse mineral particles exhibit poor conventional flotation efficiency because of many factors, including the low carrying capacity of bubbles, bubble/particle adhesion problems due to cell turbulence, and low degrees of liberation (low hydrophobicity). Many attempts to improve the recovery of coarse fractions have been explored, such as floto-elutriation operating at a high solid content while dispersed in a fluidized (or expanded) bed formed with a continuous injection of compressed air and an uprising water flow. This work analyzed the comparative performances of floto-elutriation (FE) and conventional flotation (CF) on a classified copper sulfide mineral feed as an example of a difficult-to-liberate low-grade ore. Contrary to expectations, CF and FE (Hydrofloat) displayed similar particle recovery rates with feed size distributions for P80s of 130, 240 and 280 lm. However, metallurgical recoveries from classified fractions of 297+210 lm were 25% higher in FE than in CF and as expected, coarse (+297 lm) particles were not recovered in the CF, but in the FE. The recovery of fine fractions in the FE process was due to high hydraulic entrainment and surprisingly the recovery of intermediate and liberated fractions (+74 149 lm) was very low, due to its low air hold-up. However, the enhancement of the holdup in FE increased the recovery of these mid-sized fractions. Because of the hydraulic carryover caused by the bubbles and water elutriation, the metallurgical grades obtained in all cases were very low compared to conventional bench flotation. It is believed that this FE equipment works better with coarse, narrowly classified particles and high-grade feeds and that performance decreases for low-grade ores requiring high liberation. Certain features of these findings are visualized. Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction The low flotation recovery of coarse particles is known to be caused by a high mass (which reduces the load carried by air bubbles), a low degree of liberation, a low collector adsorption (low hydrophobicity) compared to fine particles, and a low degree of dispersion in the pulp, thus allowing insufficient time for the capture by bubbles (Dunne, 2012; King, 1982; Schubert and Bischofberger, 1979; Gontijo et al., 2007; Ata and Jameson, 2013; Tabosa et al., 2013). Some cited alternatives for recovering these fractions in rougher flotation units include i. feed splitting into coarse and fine fractions to treat them separately (Kohmuench et al., 2010); ii. increasing the particle residence time in flotation units (Dunne, 2012; Froth Flotation: A Century of Innovation, Part 4, 2007); iii. flotation with a negative bias (Barbery, 1989; Soto and Barbery, 1991; Soto, 1992; Brum, 2004); iv. performing a staged addition of the collector in ⇑ Corresponding author. 1
E-mail address: [email protected] (J. Rubio). http://www.ufrgs.br/ltm.
http://dx.doi.org/10.1016/j.mineng.2015.11.012 0892-6875/Ó 2015 Elsevier Ltd. All rights reserved.
rougher circuits (Reyes Bahena et al., 2006; Bazin and Proulx, 2001); and v. the use of new cells combining flotation and elutriation (Fosu et al., 2015; Awatey et al., 2014; Jameson, 2010; Kohmuench et al., 2007, 2010, 2013). A Hydrofloat (a floto-elutriator) is a separator that combines both flotation and gravity separation. Because it employs fluidization, coarse particles are better dispersed than in turbulent mechanical flotation machines where centrifugal force pulls the coarse particles away. As a result, the equipment allows for a better capture of coarser fractions by the bubbles, it can decrease the probability of bubble/particle detachment, thus enhancing the recovery of difficult-totreat larger particles. In recent years, this technology has been applied to industrial minerals, with several full-scale units installed to recover particles up to and exceeding 3 mm in diameter within the industrial mineral sector. Many recent papers have shown examples of the applications, advantages and future trends of floto-elutriation cells and their similarities with elutriation columns or columns with negative bias.
60
M. Paiva, J. Rubio / Minerals Engineering 86 (2016) 59–65
Table 1 Recent studies on mineral floto-elutriation (FE); some comparing to conventional flotation (CF). Author (year)
Process scale
Mineral/ore system
Separation parameters/general comments
Awatey et al. (2013, 2014)
Laboratory (FE and FC)
Sphalerite (+0.250–1.180 mm)
The contact angle required for flotation of sphalerite coarse particles in the CF was higher than in EF, and increased as particle size increased The coarse sphalerite recovery increased with increasing the bed height, the superficial velocity of gas and water flow. The recovery was higher in the FE for particles larger than 0.450 mm
Quartz particles (+0.250–1.180 mm) were used to build the fluidised-bed Kohmuench et al. (2013)
Laboratory (FE) (£-cell diameter = 150 mm)
Copper/lead/zinc ore (0.800 0.200 mm)
Laboratory (FE and CF)
Copper ore (1.200 0.600 mm and 0.600 0.125 mm) The feed grade was not given Gold (0.500 0.100 mm)
Laboratory
Testa et al. (2011) Kohmuench et al. (2007, 2010)
Pilot (£ = 400 mm) Pilot (FE)
Gold (0.500 0.100 mm) Phosphate (+0.150 mm)
Laboratory (FE) (£ = 100 mm)
Phosphate (Coarse: 0.710 0.425 mm; Ultracoarse: 1.20 0.710 mm)
Pilot (FE) (£ = 300 mm)
Feed: coarse and ultracoarse
Full-Scale (Coarse: £ = 2.5 m) (Ultracoarse: £ = 1.2 m)
Feed: coarse and ultracoarse
Table 1, shows some of the recent studies on coarse mineral flotation by FE. This work measures the efficiency of a Hydrofloat on typical copper sulfide ore, where only intense grinding permits the liberation of valuables. A comparison with conventional flotation is conducted and some operating parameters are studied. 2. Experimental 2.1. Materials 2.1.1. Copper sulfide ore The copper ore used in the study corresponded to a sample composed primarily of chalcopyrite (45%) and bornite (55%). Feed grade varied between 0.9% and 1.1% Cu. The sample was ground and wet-sieved at different size fractions and composites of different P80s (80% passing product in a given mesh) were prepared as described in Table 2. For flotation studies with very coarse particles, samples of +297 lm and 297 +210 lm were used. 2.2. Methods 2.2.1. Particle size and copper content distribution Particle size analyses of the feed and products of lab conventional flotation (CF) and floto-elutriation (FE) were performed by wet screening (duplicates) at 297 lm, 210 lm, 149 lm, 74 lm and 37 lm. The material retained on each sieve was dried in an oven at 60 °C for 24 h before weighing. A copper chemical analysis was performed after acid sample digestion and analysis in a flame atomic absorption spectroscope (Varian, model AA110).
Mass recovery = 13–26% Recovery Cu = 70%; Enrichment ratio Cu = 3 Recovery Zn = 90%; Enrichment ratio Zn = 5 Recovery Pb = 90%; Enrichment ratio Pb = 4.8 FE (0.600 0.125 mm): Recovery = 93.5–99% Cu FE (1.20 0.600 mm): Recovery = 70–78% CF: Recovery = 2.0% Feed Grade = 3.6 g t 1 Mass recovery = 0.7–2.0% Gold recovery = 98% Gold concentrate Grade = 400 g t 1 Gold Recovery = 96–99% Gold concentrate Grade = 175–500 g t 1 Concentrate Grade = 30% P2O5 Apatite recovery = 90% Coarse: BPL Recovery = 98%; BPL Grade = 66.2%; Insolubles Grade = 8.1% Ultracoarse: BPL Recovery = 96%; BPL Grade = 66.9%; Insolubles Grade = 10% Coarse: BPL Recovery = 90–98%; BPL Grade = 60%; Insolubles Grade = 20% Ultracoarse: BPL Recovery = 88–98%; BPL Grade = 64–69%; Insolubles Grade = 5–13% Coarse: FE-BPL Recovery = 90%; CF-BPL Recovery = 80%; Ultracoarse: FE-BPL Recovery = 97%; BPL Recovery = 64%.
Table 2 Particle size distribution of the copper ore at different P80’s composition. Particles size fraction (lm)
+297 297+210 210+149 149+74 74+37 37
Mass % in each P80’s composition
Grade Cu (%)
130 lm
240 lm
280 lm
1.7 3.2 10.8 27.8 21.9 34.6
7.3 11.5 24.1 27.5 12.5 17.1
10.4 15.4 31.1 23.1 9.3 10.8
0.6 0.8 0.9 1.0 1.5 1.7
Grades Cu P80’s composition: 130 lm = 1.1%; 240 lm = 1.0%; 280 lm = 0.9%.
2.2.2. Holdup measurements Holdup (air % by volume) measurements were performed for a two-phase system (air/liquid) with different superficial air (0.2, 0.27, and 0.33 cm s 1) and water (0.42, 0.63, and 0.84 cm s 1) velocities. Determinations based on a bed expansion technique (Bahri et al., 2013) were performed with a sampler consisting of an acrylic cylinder endowed with two pistons spaced 120 mm from each other and fixed by a central axis (Fig. 1). These pistons, actioned by a rod, were inserted into the cylinder to trap a volume of the mixture (water + air). The trapped mixture between the pistons was weighed and the air volume calculated by the volume difference (cylinder and the collected volume). 2.2.3. Conventional flotation studies – CF Tests were performed in a 1.5 L capacity laboratory mechanical cell (EdemetÒ) with a pulp solution containing 30% solids (by weight) at a pH of 10.5, adjusted with lime (Ca (OH)2). The pulp was conditioned under agitation (750 rpm) for two min with
61
M. Paiva, J. Rubio / Minerals Engineering 86 (2016) 59–65
Open
Closed
Cylinder
Piston
(b) Water+Air
Axis Piston
(a) Fig. 1. (a) Sampler used for the air holdup measurements; (b) top view of Hydrofloat, showing the sampling points.
28 g t 1 of Aero Promoter-AP 3477 (sodium diisobutyl dithiophosphate) and 20 g t 1 of MIBC (methyl isobutyl carbinol) as a frother. Air was then progressively injected, between 2.5 and 4.5 L min 1 for the first three min, and increased up to 7 L min 1 for the final flotation times ending at 9 min. The concentrates were collected at flotation intervals of 0–0.5, 0.5–1, 1–2, 2–3, 3–6 and 6–9 min. The froth collector employed an automatic scraper aid at a fixed scraping speed of 10 collections per minute while maintaining a constant cell volume with extra water. 2.2.4. Floto-elutriation studies in the Hydrofloat The floto-elutriation (FE) experiments were performed in a 5 10 cm Hydrofloat (Eriez-Brazil) (Fig. 2) with a pulp solution of 50% solids (by weight) while maintaining all other conditions (pH and reagent concentration) constant as determined in the CF studies. The mechanical agitation was kept constant at 2300 rpm
in a conical bottom tank with a 15 cm diameter in the circular section. The pulp was then fed by gravity to the Hydrofloat filled with the frother solution and dosed with a peristaltic pump (50 ppm Dowfroth 250). This solution was previously prepared in the elutriation water reservoir tank and pumped at a rate of 0.63 cm s 1 (Jw) as controlled by a flowmeter. Floto-elutriation was initiated and lasted for 9 min, as in the conventional flotation process. 2.2.5. Floto-elutriation studies with injection of mid-sized particle bubbles (MSB) Studies in the Hydrofloat with the injection of MSB were carried out following the floto-elutriation bench procedure with the addition of an MSB generation system (Fig. 3). The generation of bubbles with median sizes between 60 and 600 lm (following Rubio et al., 2003) is possible through aeration solutions containing low concentrations of foaming (40 ppm) and discharged in the cell along with the bubbles produced in the actual cell. The particle size (P80) of the copper ore samples fed to the floto-elutriation systems with and without the MSB injection was comparatively evaluated. 2.2.6. Liberation studies The method employed was based on quantitative image analysis. An MLA (Mineral Liberation Analyzer) was used with backscattered electron images to define and categorize the particles. Their composition was then measured by an EDS detector. The characterization analysis was carried out by the CETEM-Center for Mineral Technology-MST at Rio de Janeiro with a scanning electron microscope equipped with the MLA system developed by JKMRC (Gu, 2003). The modal analysis and associated phases measured mineral-grade content classes from particle cross-sections; images were generated and the phases were identified. Fig. 4 shows, in a typical flotation feed, the distribution of copper species (chalcopyrite and bornite) (grade-liberation) by size between 37 and 297 lm. 3. Results and discussion
Fig. 2. Hydrofloat employed for flotation of the copper sulfide.
Fig. 5(a) shows that the metallurgical flotation recoveries of the CF and FE processes were fairly close (50–75% Cu) for the various P80s. However, in the classified fractions (FCL) of +297 and 297 +210 lm, the values were 25% higher in FE than in CF. This verifies the superior coarse particle recovery of the FE elutriation–flotation
62
M. Paiva, J. Rubio / Minerals Engineering 86 (2016) 59–65
1. Mechanical stirrer 2. Pulp conditioning tank 3. Floto-elutriator (HydroFloat) 4. Water tank elutriation 5. Centrifugal pump 6. Rotameter water 7. Rotameter (air) 8. Solution tank for generating mid-sized bubbles (MSB) 9. Peristaltic pump 10. Venturi tube 11. Rotameter (air) 12. Manometer
Surface Exposure Distribution (%)
Fig. 3. Schematic HydroFloat Rig with the mid-sized bubbles injection system.
-74+37 µm -149+74 µm -210+149 µm -297+210 µm +297 µm
15
12
9
6
3
0
Grade (% weight) Fig. 4. Copper species (chalcopyrite + Bornite) grades by size (37–297 lm).
equipment - a 36% copper content in this particular case. As expected, the coarse copper bearing particles (>297 lm) did not float at all in the CF (Fig. 5(c)). Due to the high degree of entrainment in the FE process caused by water elutriation and very large bubbles (>1 mm), the concentrate grades remained as low as 1.2–2.5% Cu, for enrichment ratios of approximately 1–2 (Fig. 5 and Table 3). Conversely, the enrichment ratios reached 13% Cu in the CF for P80: 280 lm, increasing the concentrate grade (from 5% to 12% Cu) with increasing P80. Fig. 6(a) shows comparative copper flotation recoveries for each P80 feed size. Both CF and FE (Fig. 6(a–d)) recoveries were higher than 80% Cu for very fine fractions (<37 lm) but higher copper recoveries in mid-sized fractions (>37 149 lm) were obtained in the CF. Again, the entrainment of very fine particles (due to greater buoyancy) might explain the results obtained in the FE. The recovery of intermediate-sized fractions (not locked) by floto-elutriation appears to require a higher air holdup because
96
18
FE
CF
80 64 48 32 16
12 9 6 3 0
0 130
240
280
130
240
P80 (µm)
P80 (µm)
(a)
(b)
280
12
30
CF
CF
FE
25
FE
10
Grade Cu (%)
Recovery Cu (%)
FE
15
Grade Cu (%)
Recovery Cu (%)
CF
20 15 10 5
8 6 4 2
0
0 -297
+210
+297
Size (µm)
(c)
-297
+210
+297
Size (µm)
(d)
Fig. 5. Comparative results of copper recoveries and grades for different (a, b) P80 and evaluated (c, d) coarse particles (conventional flotation – CF; floto-elutriation – FE).
63
M. Paiva, J. Rubio / Minerals Engineering 86 (2016) 59–65
recoveries ranging from 93.5% to almost 99% and the coarser material showed recoveries ranging from 70% to 78%, depending on operating conditions. The authors claimed that the higher recoveries for the finer material were likely due to more favorable fluidization when teetering the finer particles. Moreover, the authors believed that the successful attachment of an air bubble can overcome the size effects that could otherwise naturally control the hindered settling velocity of the coarse material. Unfortunately, the copper feed grade was not provided to verify the enrichment rate and selectivity obtained at that particle size distribution.
Table 3 Flotation enrichment ratios with different techniques and P80’s values. Samples composition (lm)
Enrichment ratio CF
FE
FE-MSB
P80 = 130 P80 = 240 P80 = 280 297+210 +297
4.5 7.6 13.3 11.2 0
1.2 1.2 1.5 3.2 2.7
1.0 1.5 1.5 – –
3.1. Final remarks the flotation of these particles is enabled by bubble capture (collision + adhesion) mechanisms. Thus, the following conditions appear to be necessary: i. the enhancement of air dispersion values (Sb, Jg or air holdup); ii. a suitable residence time; and iii. a minimum collision energy for the water film thinning and rupture. The effectiveness of the latter appears to be fairly low in an upward flow without turbulence. Fig. 7 shows the results for two P80 values (130 and 240 lm) enhancing the air holdup to approximately 12% while employing mid-sized bubbles (100–600 lm) to avoid turbulence and bed rupture. Accordingly, both the concentrate recoveries and the grades increased after the injection of small bubbles. It should also be noted that this did not occur while increasing the air holdup with large bubbles (1–2 mm in diameter), where destabilization of the bed and turbulence was observed. Kohmuench et al. (2013) presented preliminary FE results in a copper ore underflow sample classified as 1.20 0.600 and 0.600 0.125 mm. The finer grind size resulted in copper
The Hydrofloat can recover coarse particles (>0.2 mm), particularly in high-grade ores where liberation is not a significant problem. Pulp fluidization or expansion is possible through proper water supply (containing frother) and with an aerated teeter bed injecting compressed air. The bubble size distribution, which determines the particle capture capacity, depends on the frother concentration; unfortunately, these values are not well known. Hydrofloat performance primarily depends on aspects of the mineral/ore system such as high feed grades, particle density, and degree of liberation versus particle size distribution. Recovery rates will be higher for high liberation of coarse sizes; the particles will exhibit high acceleration (on settling) in an expanded bed and thus facilitate the capture of coarse particles without reaching their terminal velocity. Performance increases as the pulp reaches the optimal boundary conditions for an expanded dense bed and a feed composed of a narrow distribution of coarse particles, which requires a quiescent system for minimizing detachment. Under
90
90 130 µm 240 µm 280 µm
FE 75
Recovery Cu (%)
Recovery Cu (%)
75 60 45 30
60 45 30 15
15 0
FE-MSB
0
5
10
15
20
0
25
0
1
2
Grade Cu (%)
Grade Cu (%)
(a)
(b)
90 75
FE 75
FE-MSB
Recovery Cu (%)
Recovery Cu (%)
4
90 FE
60 45 30 15 0
3
FE-MSB
60 45 30 15
0
1
2
Grade Cu (%)
(c)
3
4
0
0
1
2
3
4
Grade Cu (%)
(d)
Fig. 6. Copper grade-recoveries curves for different P80 feed values: (a) conventional flotation (CF); floto-elutriation (FE) and floto-elutriation with injection of mid-sized bubbles (FE-MSB) (200–600 lm); (b) P80: 130 lm; (c) P80: 240 lm; (d) P80: 280 lm.
M. Paiva, J. Rubio / Minerals Engineering 86 (2016) 59–65
Size by size recovery Cu (%)
64
100
130 µm 240 µm 280 µm
80 60 40 20 0 -37
37
74
149
210
297
Size (µm)
100
130 µm 240 µm
80
280 µm 60 40 20 0
-37
37
74
149
210
297
Size by size recovery Cu (%)
Size by size recovery Cu (%)
(a) 100 130 µm 240 µm 280 µm
80 60 40 20 0 -37
37
74
149
Size (µm)
Size (µm)
(b)
(c)
210
297
Fig. 7. Size by size recoveries for different P80 feed values: (a) conventional flotation; (b) floto-elutriation and (c) floto-elutriation with injection of mid-sized bubbles. Standard deviations were, in all cases <5%.
these optimal conditions, the maximum floatable particle size can increase to several millimeters when conditions are quiescent (Schulze, 1984; Soto, 1988). If the liberation process is complex, as in metal sulfides, the particle size distribution will be wider and the presence of mid-size (locked or not) of approximately 149 lm or less, combined with the liberated <74 lm particles, will decrease overall process efficiency and recovery of the coarser fractions. The presence of liberated particles and middlings (semi-locked particles) appear to hinder bed fluidization and bubble capture capacity. Thus, the separation of these particles is determined by physicochemical– interfacial phenomena and entrainment of the fine particles. The optimal quiescence conditions for bubbles attaching to coarse particles are presumably lost and it would function only in a deslimed feed. Thus, the probabilities of collision and attachment in a Hydrofloat are low when compared with conventional flotation due to the low collision energy (required for film thinning) and the need for high bubble concentrations (Sb or air holdup). However, if valuables accumulate in P80s <240 lm, the Hydrofloat is not the best choice.
4. Conclusions A Hydrofloat was found to behave as a non-selective elutriator for fine copper sulfide mineral particles and as a flotation unit for coarse particles (>250 lm) only. Recoveries were very close to conventional flotation for all P80 fractions (130, 240 and 280 lm) with very low concentrate grades or enrichment ratios (<2%). Fine (<74 lm) particles were found to be entrained and mid-sized particles (74–149 lm), although liberated, were only collected by enhancing the air holdup (from 3% to 12%) with mid-sized bubbles (100–600 lm). The results appear to show that the particle size
distribution of copper sulfides feeding a Hydrofloat must be either deslimed before recovering the coarse particles. Should enhancing the air hold-up is required best alternative is with mid-sized bubbles, to avoid coarse particles bed rupture.
Acknowledgments The authors would like to thank all the Brazilian Institutes supporting this research, namely CNPq, Fapergs and UFRGS. Thanks are due to MEC–CAPES sponsoring the doctorate of Meise Paiva and to Reiner Neumann for the MLA analysis. Special thanks to Jaqueline Mohr, Douglas Alegre, Christian Dall’Agnol Boni; Andressa Ikeda; Mateus Lottermann, all undergraduate students at our laboratory.
References Ata, S., Jameson, G.J., 2013. Recovery of coarse particles in the froth phase – a case study. Miner. Eng. 45, 121–127. Awatey, B., Thanasekaran, H., Kohmuench, J.N., Skinner, W., Zanin, M., 2014. Critical contact angle for coarse sphalerite flotation in a fluidised-bed separator vs. a mechanically agitated cell. Miner. Eng. 60, 51–59. Awatey, B., Thanasekaran, H., Kohmuench, J., Skinner, W., Zanin, M., 2013. Optimization of operating parameters for coarse sphalerite flotation in the HydroFloatÒ fluidised-bed separator. Miner. Eng. 50–51, 99–105. Bahri, Z., Shafaei, S.Z., Karamoozian, M., 2013. Investigation of effective parameters on the gas holdup in column flotation of a coal tailing sample. Int. J. Coal Preparation Utilization 33 (2), 47–58. Barbery, G., 1989. Method for separation of coarse particles. U.S. Patent No. 4,822,493. 8p. Bazin, C., Proulx, M., 2001. Distribution of reagentes down a flotation bank to improve the recovery of coarse particles. Int. J. Miner. Process. 61 (1), 1–12. Brum, I.A.S., 2004. Concentração de partículas minerais grossas de fluorita em coluna de flotação. [Tese de doutorado], PPGE3M–UFRGS, Porto Alegre, 134p (in portuguese).
M. Paiva, J. Rubio / Minerals Engineering 86 (2016) 59–65 Dunne, R.C., 2012. The journey of the coarse particle through the pulp and froth in flotation. In: Proceedings XXVI International Mineral Processing Congress, New Delhi, India, pp. 1259–1268. Fosu, S., Awatey, B., Skinner, W., Zanin, M., 2015. Flotation of coarse composite particles in mechanical cell vs. the fluidised-bed separator (The HydroFloatTM). Miner. Eng. 77, 137–149. Froth Flotation: A Century of Innovation, Part 4, 2007. Society for Mining, Metallurgy, and Exploration, (Jan. 1 2007), 891p. Gontijo, C.de.F., Fornasiero, D., Ralston, J., 2007. The limits of fine and coarse particle flotation. Can. J. Chem. Eng. 85 (5), 739–747. Gu, Y., 2003. Automated scanning electron microscope based mineral liberation analysis, an introduction to JKMRC/FEI mineral liberation analyser. J. Miner. Mater. Charact. Eng. 2 (1), 33–41. Jameson, G.J., 2010. New directions in flotation machine design. Miner. Eng. 23, 835–841. King, R.P., 1982. Principles of Flotation, vol. 3. South African Institute of Mining and Metallurgy, Johannesburg, South Africa, p. 268. Kohmuench, J., Thanasekaran, H., Seaman, B., 2013. Advances in coarse particle flotation-copper and gold. In: Proceedings of the MetPlant Conference, 2013, Perth, Australia (AusIMM, Publication series 5/2013), pp. 378–386. Kohmuench, J.N., Mankosa, M.J., Yan, E.S., Wyslouzil, H., Christodoulou, L., 2010. Advances in coarse particle flotation industrial minerals. In: Proceedings 25th International Mineral Processing Congress, 2010, pp. 2065–2076. Kohmuench, J.N., Mankosa, M.J., Kennedy, D.G., Yasalonis, J.L., Taylor, G.B., Luttrell, G.H., 2007. Implementation of the HydroFloat technology at the South Fort Meade Mine. Miner. Metall. Process. 24 (4), 264–270.
65
Reyes Bahena, J.L., Franzidis, J.P., Manglapig, E.V., López Valdivieso, A., Ojeda Escamilla, M.C., 2006. Assessment of reagent and regrinding in a flotation circuit. In: Memoria del XVI Congreso Internacional de Metalurgia Extractiva, México, pp. 195–205. Rubio, J., Capponi, F., Matiolo, E., Nunes, D., Guerrero, C. P., Berkowitz, G., 2003. Advances in flotation of mineral fines. In Proceedings XXII International Mineral Processing Congress, Cape-Town, South Africa, pp. 1014–1020. Schubert, H., Bischofberger, C., 1979. On the optimization of hydrodynamics in flotation processes. In: XIII International Mineral Processing Congress, Part B, pp. 353–375. Schulze, H.J., 1984. Physico-Chemical Elementary Processes in Flotation. Elsevier, Amsterdam, 348p. Soto, H., 1992. Development of novel flotation–elutriation method for coarse phosphate beneficiation. Final Report. FIPR Publication. Florida Institute Phosphate Research. 77p. Soto, H., 1988. Private Report to the Florida Institute of Phosphate Research. Bartow, Florida, 35p. Soto, H., Barbery, G., 1991. Flotation of coarse particles in a counter-current column cell. Miner. Metall. Process. 8 (1), 16–21. Tabosa, E., Runge, K., Duffy, K., 2013. Strategies for increasing coarse particle flotation in conventional flotation cells. Flotation 13, 18–21 November, 15p. Testa, F.G., Nascimento, A.L., Duarte, F.E., Yan, E., 2011. Estudo piloto de flotação de grossos utilizando hydrofloat com minério fosfático de Cajati-SP. In: XXIV Encontro Nacional de Tratamento de Minérios e Metalurgia Extrativa, SalvadorBA, pp. 152–158 (in Portuguese).
Contents lists available at ScienceDirect
Minerals Engineering journal homepage: www.elsevier.com/locate/mineng
Factors affecting the floto-elutriation process efficiency of a copper sulfide mineral M. Paiva, J. Rubio ⇑ Minerals Engineering Department, PPGE3M – Universidade Federal do Rio Grande do Sul, Av. Bento Gonçalves, 9500, Prédio 43819 – Setor 6, 91501-970 Porto Alegre, RS, Brazil1
a r t i c l e
i n f o
Article history: Received 17 September 2015 Revised 2 November 2015 Accepted 25 November 2015
Keywords: Floto-elutriation Particle size distribution Recovery Grades
a b s t r a c t Coarse mineral particles exhibit poor conventional flotation efficiency because of many factors, including the low carrying capacity of bubbles, bubble/particle adhesion problems due to cell turbulence, and low degrees of liberation (low hydrophobicity). Many attempts to improve the recovery of coarse fractions have been explored, such as floto-elutriation operating at a high solid content while dispersed in a fluidized (or expanded) bed formed with a continuous injection of compressed air and an uprising water flow. This work analyzed the comparative performances of floto-elutriation (FE) and conventional flotation (CF) on a classified copper sulfide mineral feed as an example of a difficult-to-liberate low-grade ore. Contrary to expectations, CF and FE (Hydrofloat) displayed similar particle recovery rates with feed size distributions for P80s of 130, 240 and 280 lm. However, metallurgical recoveries from classified fractions of 297+210 lm were 25% higher in FE than in CF and as expected, coarse (+297 lm) particles were not recovered in the CF, but in the FE. The recovery of fine fractions in the FE process was due to high hydraulic entrainment and surprisingly the recovery of intermediate and liberated fractions (+74 149 lm) was very low, due to its low air hold-up. However, the enhancement of the holdup in FE increased the recovery of these mid-sized fractions. Because of the hydraulic carryover caused by the bubbles and water elutriation, the metallurgical grades obtained in all cases were very low compared to conventional bench flotation. It is believed that this FE equipment works better with coarse, narrowly classified particles and high-grade feeds and that performance decreases for low-grade ores requiring high liberation. Certain features of these findings are visualized. Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction The low flotation recovery of coarse particles is known to be caused by a high mass (which reduces the load carried by air bubbles), a low degree of liberation, a low collector adsorption (low hydrophobicity) compared to fine particles, and a low degree of dispersion in the pulp, thus allowing insufficient time for the capture by bubbles (Dunne, 2012; King, 1982; Schubert and Bischofberger, 1979; Gontijo et al., 2007; Ata and Jameson, 2013; Tabosa et al., 2013). Some cited alternatives for recovering these fractions in rougher flotation units include i. feed splitting into coarse and fine fractions to treat them separately (Kohmuench et al., 2010); ii. increasing the particle residence time in flotation units (Dunne, 2012; Froth Flotation: A Century of Innovation, Part 4, 2007); iii. flotation with a negative bias (Barbery, 1989; Soto and Barbery, 1991; Soto, 1992; Brum, 2004); iv. performing a staged addition of the collector in ⇑ Corresponding author. 1
E-mail address: [email protected] (J. Rubio). http://www.ufrgs.br/ltm.
http://dx.doi.org/10.1016/j.mineng.2015.11.012 0892-6875/Ó 2015 Elsevier Ltd. All rights reserved.
rougher circuits (Reyes Bahena et al., 2006; Bazin and Proulx, 2001); and v. the use of new cells combining flotation and elutriation (Fosu et al., 2015; Awatey et al., 2014; Jameson, 2010; Kohmuench et al., 2007, 2010, 2013). A Hydrofloat (a floto-elutriator) is a separator that combines both flotation and gravity separation. Because it employs fluidization, coarse particles are better dispersed than in turbulent mechanical flotation machines where centrifugal force pulls the coarse particles away. As a result, the equipment allows for a better capture of coarser fractions by the bubbles, it can decrease the probability of bubble/particle detachment, thus enhancing the recovery of difficult-totreat larger particles. In recent years, this technology has been applied to industrial minerals, with several full-scale units installed to recover particles up to and exceeding 3 mm in diameter within the industrial mineral sector. Many recent papers have shown examples of the applications, advantages and future trends of floto-elutriation cells and their similarities with elutriation columns or columns with negative bias.
60
M. Paiva, J. Rubio / Minerals Engineering 86 (2016) 59–65
Table 1 Recent studies on mineral floto-elutriation (FE); some comparing to conventional flotation (CF). Author (year)
Process scale
Mineral/ore system
Separation parameters/general comments
Awatey et al. (2013, 2014)
Laboratory (FE and FC)
Sphalerite (+0.250–1.180 mm)
The contact angle required for flotation of sphalerite coarse particles in the CF was higher than in EF, and increased as particle size increased The coarse sphalerite recovery increased with increasing the bed height, the superficial velocity of gas and water flow. The recovery was higher in the FE for particles larger than 0.450 mm
Quartz particles (+0.250–1.180 mm) were used to build the fluidised-bed Kohmuench et al. (2013)
Laboratory (FE) (£-cell diameter = 150 mm)
Copper/lead/zinc ore (0.800 0.200 mm)
Laboratory (FE and CF)
Copper ore (1.200 0.600 mm and 0.600 0.125 mm) The feed grade was not given Gold (0.500 0.100 mm)
Laboratory
Testa et al. (2011) Kohmuench et al. (2007, 2010)
Pilot (£ = 400 mm) Pilot (FE)
Gold (0.500 0.100 mm) Phosphate (+0.150 mm)
Laboratory (FE) (£ = 100 mm)
Phosphate (Coarse: 0.710 0.425 mm; Ultracoarse: 1.20 0.710 mm)
Pilot (FE) (£ = 300 mm)
Feed: coarse and ultracoarse
Full-Scale (Coarse: £ = 2.5 m) (Ultracoarse: £ = 1.2 m)
Feed: coarse and ultracoarse
Table 1, shows some of the recent studies on coarse mineral flotation by FE. This work measures the efficiency of a Hydrofloat on typical copper sulfide ore, where only intense grinding permits the liberation of valuables. A comparison with conventional flotation is conducted and some operating parameters are studied. 2. Experimental 2.1. Materials 2.1.1. Copper sulfide ore The copper ore used in the study corresponded to a sample composed primarily of chalcopyrite (45%) and bornite (55%). Feed grade varied between 0.9% and 1.1% Cu. The sample was ground and wet-sieved at different size fractions and composites of different P80s (80% passing product in a given mesh) were prepared as described in Table 2. For flotation studies with very coarse particles, samples of +297 lm and 297 +210 lm were used. 2.2. Methods 2.2.1. Particle size and copper content distribution Particle size analyses of the feed and products of lab conventional flotation (CF) and floto-elutriation (FE) were performed by wet screening (duplicates) at 297 lm, 210 lm, 149 lm, 74 lm and 37 lm. The material retained on each sieve was dried in an oven at 60 °C for 24 h before weighing. A copper chemical analysis was performed after acid sample digestion and analysis in a flame atomic absorption spectroscope (Varian, model AA110).
Mass recovery = 13–26% Recovery Cu = 70%; Enrichment ratio Cu = 3 Recovery Zn = 90%; Enrichment ratio Zn = 5 Recovery Pb = 90%; Enrichment ratio Pb = 4.8 FE (0.600 0.125 mm): Recovery = 93.5–99% Cu FE (1.20 0.600 mm): Recovery = 70–78% CF: Recovery = 2.0% Feed Grade = 3.6 g t 1 Mass recovery = 0.7–2.0% Gold recovery = 98% Gold concentrate Grade = 400 g t 1 Gold Recovery = 96–99% Gold concentrate Grade = 175–500 g t 1 Concentrate Grade = 30% P2O5 Apatite recovery = 90% Coarse: BPL Recovery = 98%; BPL Grade = 66.2%; Insolubles Grade = 8.1% Ultracoarse: BPL Recovery = 96%; BPL Grade = 66.9%; Insolubles Grade = 10% Coarse: BPL Recovery = 90–98%; BPL Grade = 60%; Insolubles Grade = 20% Ultracoarse: BPL Recovery = 88–98%; BPL Grade = 64–69%; Insolubles Grade = 5–13% Coarse: FE-BPL Recovery = 90%; CF-BPL Recovery = 80%; Ultracoarse: FE-BPL Recovery = 97%; BPL Recovery = 64%.
Table 2 Particle size distribution of the copper ore at different P80’s composition. Particles size fraction (lm)
+297 297+210 210+149 149+74 74+37 37
Mass % in each P80’s composition
Grade Cu (%)
130 lm
240 lm
280 lm
1.7 3.2 10.8 27.8 21.9 34.6
7.3 11.5 24.1 27.5 12.5 17.1
10.4 15.4 31.1 23.1 9.3 10.8
0.6 0.8 0.9 1.0 1.5 1.7
Grades Cu P80’s composition: 130 lm = 1.1%; 240 lm = 1.0%; 280 lm = 0.9%.
2.2.2. Holdup measurements Holdup (air % by volume) measurements were performed for a two-phase system (air/liquid) with different superficial air (0.2, 0.27, and 0.33 cm s 1) and water (0.42, 0.63, and 0.84 cm s 1) velocities. Determinations based on a bed expansion technique (Bahri et al., 2013) were performed with a sampler consisting of an acrylic cylinder endowed with two pistons spaced 120 mm from each other and fixed by a central axis (Fig. 1). These pistons, actioned by a rod, were inserted into the cylinder to trap a volume of the mixture (water + air). The trapped mixture between the pistons was weighed and the air volume calculated by the volume difference (cylinder and the collected volume). 2.2.3. Conventional flotation studies – CF Tests were performed in a 1.5 L capacity laboratory mechanical cell (EdemetÒ) with a pulp solution containing 30% solids (by weight) at a pH of 10.5, adjusted with lime (Ca (OH)2). The pulp was conditioned under agitation (750 rpm) for two min with
61
M. Paiva, J. Rubio / Minerals Engineering 86 (2016) 59–65
Open
Closed
Cylinder
Piston
(b) Water+Air
Axis Piston
(a) Fig. 1. (a) Sampler used for the air holdup measurements; (b) top view of Hydrofloat, showing the sampling points.
28 g t 1 of Aero Promoter-AP 3477 (sodium diisobutyl dithiophosphate) and 20 g t 1 of MIBC (methyl isobutyl carbinol) as a frother. Air was then progressively injected, between 2.5 and 4.5 L min 1 for the first three min, and increased up to 7 L min 1 for the final flotation times ending at 9 min. The concentrates were collected at flotation intervals of 0–0.5, 0.5–1, 1–2, 2–3, 3–6 and 6–9 min. The froth collector employed an automatic scraper aid at a fixed scraping speed of 10 collections per minute while maintaining a constant cell volume with extra water. 2.2.4. Floto-elutriation studies in the Hydrofloat The floto-elutriation (FE) experiments were performed in a 5 10 cm Hydrofloat (Eriez-Brazil) (Fig. 2) with a pulp solution of 50% solids (by weight) while maintaining all other conditions (pH and reagent concentration) constant as determined in the CF studies. The mechanical agitation was kept constant at 2300 rpm
in a conical bottom tank with a 15 cm diameter in the circular section. The pulp was then fed by gravity to the Hydrofloat filled with the frother solution and dosed with a peristaltic pump (50 ppm Dowfroth 250). This solution was previously prepared in the elutriation water reservoir tank and pumped at a rate of 0.63 cm s 1 (Jw) as controlled by a flowmeter. Floto-elutriation was initiated and lasted for 9 min, as in the conventional flotation process. 2.2.5. Floto-elutriation studies with injection of mid-sized particle bubbles (MSB) Studies in the Hydrofloat with the injection of MSB were carried out following the floto-elutriation bench procedure with the addition of an MSB generation system (Fig. 3). The generation of bubbles with median sizes between 60 and 600 lm (following Rubio et al., 2003) is possible through aeration solutions containing low concentrations of foaming (40 ppm) and discharged in the cell along with the bubbles produced in the actual cell. The particle size (P80) of the copper ore samples fed to the floto-elutriation systems with and without the MSB injection was comparatively evaluated. 2.2.6. Liberation studies The method employed was based on quantitative image analysis. An MLA (Mineral Liberation Analyzer) was used with backscattered electron images to define and categorize the particles. Their composition was then measured by an EDS detector. The characterization analysis was carried out by the CETEM-Center for Mineral Technology-MST at Rio de Janeiro with a scanning electron microscope equipped with the MLA system developed by JKMRC (Gu, 2003). The modal analysis and associated phases measured mineral-grade content classes from particle cross-sections; images were generated and the phases were identified. Fig. 4 shows, in a typical flotation feed, the distribution of copper species (chalcopyrite and bornite) (grade-liberation) by size between 37 and 297 lm. 3. Results and discussion
Fig. 2. Hydrofloat employed for flotation of the copper sulfide.
Fig. 5(a) shows that the metallurgical flotation recoveries of the CF and FE processes were fairly close (50–75% Cu) for the various P80s. However, in the classified fractions (FCL) of +297 and 297 +210 lm, the values were 25% higher in FE than in CF. This verifies the superior coarse particle recovery of the FE elutriation–flotation
62
M. Paiva, J. Rubio / Minerals Engineering 86 (2016) 59–65
1. Mechanical stirrer 2. Pulp conditioning tank 3. Floto-elutriator (HydroFloat) 4. Water tank elutriation 5. Centrifugal pump 6. Rotameter water 7. Rotameter (air) 8. Solution tank for generating mid-sized bubbles (MSB) 9. Peristaltic pump 10. Venturi tube 11. Rotameter (air) 12. Manometer
Surface Exposure Distribution (%)
Fig. 3. Schematic HydroFloat Rig with the mid-sized bubbles injection system.
-74+37 µm -149+74 µm -210+149 µm -297+210 µm +297 µm
15
12
9
6
3
0
Grade (% weight) Fig. 4. Copper species (chalcopyrite + Bornite) grades by size (37–297 lm).
equipment - a 36% copper content in this particular case. As expected, the coarse copper bearing particles (>297 lm) did not float at all in the CF (Fig. 5(c)). Due to the high degree of entrainment in the FE process caused by water elutriation and very large bubbles (>1 mm), the concentrate grades remained as low as 1.2–2.5% Cu, for enrichment ratios of approximately 1–2 (Fig. 5 and Table 3). Conversely, the enrichment ratios reached 13% Cu in the CF for P80: 280 lm, increasing the concentrate grade (from 5% to 12% Cu) with increasing P80. Fig. 6(a) shows comparative copper flotation recoveries for each P80 feed size. Both CF and FE (Fig. 6(a–d)) recoveries were higher than 80% Cu for very fine fractions (<37 lm) but higher copper recoveries in mid-sized fractions (>37 149 lm) were obtained in the CF. Again, the entrainment of very fine particles (due to greater buoyancy) might explain the results obtained in the FE. The recovery of intermediate-sized fractions (not locked) by floto-elutriation appears to require a higher air holdup because
96
18
FE
CF
80 64 48 32 16
12 9 6 3 0
0 130
240
280
130
240
P80 (µm)
P80 (µm)
(a)
(b)
280
12
30
CF
CF
FE
25
FE
10
Grade Cu (%)
Recovery Cu (%)
FE
15
Grade Cu (%)
Recovery Cu (%)
CF
20 15 10 5
8 6 4 2
0
0 -297
+210
+297
Size (µm)
(c)
-297
+210
+297
Size (µm)
(d)
Fig. 5. Comparative results of copper recoveries and grades for different (a, b) P80 and evaluated (c, d) coarse particles (conventional flotation – CF; floto-elutriation – FE).
63
M. Paiva, J. Rubio / Minerals Engineering 86 (2016) 59–65
recoveries ranging from 93.5% to almost 99% and the coarser material showed recoveries ranging from 70% to 78%, depending on operating conditions. The authors claimed that the higher recoveries for the finer material were likely due to more favorable fluidization when teetering the finer particles. Moreover, the authors believed that the successful attachment of an air bubble can overcome the size effects that could otherwise naturally control the hindered settling velocity of the coarse material. Unfortunately, the copper feed grade was not provided to verify the enrichment rate and selectivity obtained at that particle size distribution.
Table 3 Flotation enrichment ratios with different techniques and P80’s values. Samples composition (lm)
Enrichment ratio CF
FE
FE-MSB
P80 = 130 P80 = 240 P80 = 280 297+210 +297
4.5 7.6 13.3 11.2 0
1.2 1.2 1.5 3.2 2.7
1.0 1.5 1.5 – –
3.1. Final remarks the flotation of these particles is enabled by bubble capture (collision + adhesion) mechanisms. Thus, the following conditions appear to be necessary: i. the enhancement of air dispersion values (Sb, Jg or air holdup); ii. a suitable residence time; and iii. a minimum collision energy for the water film thinning and rupture. The effectiveness of the latter appears to be fairly low in an upward flow without turbulence. Fig. 7 shows the results for two P80 values (130 and 240 lm) enhancing the air holdup to approximately 12% while employing mid-sized bubbles (100–600 lm) to avoid turbulence and bed rupture. Accordingly, both the concentrate recoveries and the grades increased after the injection of small bubbles. It should also be noted that this did not occur while increasing the air holdup with large bubbles (1–2 mm in diameter), where destabilization of the bed and turbulence was observed. Kohmuench et al. (2013) presented preliminary FE results in a copper ore underflow sample classified as 1.20 0.600 and 0.600 0.125 mm. The finer grind size resulted in copper
The Hydrofloat can recover coarse particles (>0.2 mm), particularly in high-grade ores where liberation is not a significant problem. Pulp fluidization or expansion is possible through proper water supply (containing frother) and with an aerated teeter bed injecting compressed air. The bubble size distribution, which determines the particle capture capacity, depends on the frother concentration; unfortunately, these values are not well known. Hydrofloat performance primarily depends on aspects of the mineral/ore system such as high feed grades, particle density, and degree of liberation versus particle size distribution. Recovery rates will be higher for high liberation of coarse sizes; the particles will exhibit high acceleration (on settling) in an expanded bed and thus facilitate the capture of coarse particles without reaching their terminal velocity. Performance increases as the pulp reaches the optimal boundary conditions for an expanded dense bed and a feed composed of a narrow distribution of coarse particles, which requires a quiescent system for minimizing detachment. Under
90
90 130 µm 240 µm 280 µm
FE 75
Recovery Cu (%)
Recovery Cu (%)
75 60 45 30
60 45 30 15
15 0
FE-MSB
0
5
10
15
20
0
25
0
1
2
Grade Cu (%)
Grade Cu (%)
(a)
(b)
90 75
FE 75
FE-MSB
Recovery Cu (%)
Recovery Cu (%)
4
90 FE
60 45 30 15 0
3
FE-MSB
60 45 30 15
0
1
2
Grade Cu (%)
(c)
3
4
0
0
1
2
3
4
Grade Cu (%)
(d)
Fig. 6. Copper grade-recoveries curves for different P80 feed values: (a) conventional flotation (CF); floto-elutriation (FE) and floto-elutriation with injection of mid-sized bubbles (FE-MSB) (200–600 lm); (b) P80: 130 lm; (c) P80: 240 lm; (d) P80: 280 lm.
M. Paiva, J. Rubio / Minerals Engineering 86 (2016) 59–65
Size by size recovery Cu (%)
64
100
130 µm 240 µm 280 µm
80 60 40 20 0 -37
37
74
149
210
297
Size (µm)
100
130 µm 240 µm
80
280 µm 60 40 20 0
-37
37
74
149
210
297
Size by size recovery Cu (%)
Size by size recovery Cu (%)
(a) 100 130 µm 240 µm 280 µm
80 60 40 20 0 -37
37
74
149
Size (µm)
Size (µm)
(b)
(c)
210
297
Fig. 7. Size by size recoveries for different P80 feed values: (a) conventional flotation; (b) floto-elutriation and (c) floto-elutriation with injection of mid-sized bubbles. Standard deviations were, in all cases <5%.
these optimal conditions, the maximum floatable particle size can increase to several millimeters when conditions are quiescent (Schulze, 1984; Soto, 1988). If the liberation process is complex, as in metal sulfides, the particle size distribution will be wider and the presence of mid-size (locked or not) of approximately 149 lm or less, combined with the liberated <74 lm particles, will decrease overall process efficiency and recovery of the coarser fractions. The presence of liberated particles and middlings (semi-locked particles) appear to hinder bed fluidization and bubble capture capacity. Thus, the separation of these particles is determined by physicochemical– interfacial phenomena and entrainment of the fine particles. The optimal quiescence conditions for bubbles attaching to coarse particles are presumably lost and it would function only in a deslimed feed. Thus, the probabilities of collision and attachment in a Hydrofloat are low when compared with conventional flotation due to the low collision energy (required for film thinning) and the need for high bubble concentrations (Sb or air holdup). However, if valuables accumulate in P80s <240 lm, the Hydrofloat is not the best choice.
4. Conclusions A Hydrofloat was found to behave as a non-selective elutriator for fine copper sulfide mineral particles and as a flotation unit for coarse particles (>250 lm) only. Recoveries were very close to conventional flotation for all P80 fractions (130, 240 and 280 lm) with very low concentrate grades or enrichment ratios (<2%). Fine (<74 lm) particles were found to be entrained and mid-sized particles (74–149 lm), although liberated, were only collected by enhancing the air holdup (from 3% to 12%) with mid-sized bubbles (100–600 lm). The results appear to show that the particle size
distribution of copper sulfides feeding a Hydrofloat must be either deslimed before recovering the coarse particles. Should enhancing the air hold-up is required best alternative is with mid-sized bubbles, to avoid coarse particles bed rupture.
Acknowledgments The authors would like to thank all the Brazilian Institutes supporting this research, namely CNPq, Fapergs and UFRGS. Thanks are due to MEC–CAPES sponsoring the doctorate of Meise Paiva and to Reiner Neumann for the MLA analysis. Special thanks to Jaqueline Mohr, Douglas Alegre, Christian Dall’Agnol Boni; Andressa Ikeda; Mateus Lottermann, all undergraduate students at our laboratory.
References Ata, S., Jameson, G.J., 2013. Recovery of coarse particles in the froth phase – a case study. Miner. Eng. 45, 121–127. Awatey, B., Thanasekaran, H., Kohmuench, J.N., Skinner, W., Zanin, M., 2014. Critical contact angle for coarse sphalerite flotation in a fluidised-bed separator vs. a mechanically agitated cell. Miner. Eng. 60, 51–59. Awatey, B., Thanasekaran, H., Kohmuench, J., Skinner, W., Zanin, M., 2013. Optimization of operating parameters for coarse sphalerite flotation in the HydroFloatÒ fluidised-bed separator. Miner. Eng. 50–51, 99–105. Bahri, Z., Shafaei, S.Z., Karamoozian, M., 2013. Investigation of effective parameters on the gas holdup in column flotation of a coal tailing sample. Int. J. Coal Preparation Utilization 33 (2), 47–58. Barbery, G., 1989. Method for separation of coarse particles. U.S. Patent No. 4,822,493. 8p. Bazin, C., Proulx, M., 2001. Distribution of reagentes down a flotation bank to improve the recovery of coarse particles. Int. J. Miner. Process. 61 (1), 1–12. Brum, I.A.S., 2004. Concentração de partículas minerais grossas de fluorita em coluna de flotação. [Tese de doutorado], PPGE3M–UFRGS, Porto Alegre, 134p (in portuguese).
M. Paiva, J. Rubio / Minerals Engineering 86 (2016) 59–65 Dunne, R.C., 2012. The journey of the coarse particle through the pulp and froth in flotation. In: Proceedings XXVI International Mineral Processing Congress, New Delhi, India, pp. 1259–1268. Fosu, S., Awatey, B., Skinner, W., Zanin, M., 2015. Flotation of coarse composite particles in mechanical cell vs. the fluidised-bed separator (The HydroFloatTM). Miner. Eng. 77, 137–149. Froth Flotation: A Century of Innovation, Part 4, 2007. Society for Mining, Metallurgy, and Exploration, (Jan. 1 2007), 891p. Gontijo, C.de.F., Fornasiero, D., Ralston, J., 2007. The limits of fine and coarse particle flotation. Can. J. Chem. Eng. 85 (5), 739–747. Gu, Y., 2003. Automated scanning electron microscope based mineral liberation analysis, an introduction to JKMRC/FEI mineral liberation analyser. J. Miner. Mater. Charact. Eng. 2 (1), 33–41. Jameson, G.J., 2010. New directions in flotation machine design. Miner. Eng. 23, 835–841. King, R.P., 1982. Principles of Flotation, vol. 3. South African Institute of Mining and Metallurgy, Johannesburg, South Africa, p. 268. Kohmuench, J., Thanasekaran, H., Seaman, B., 2013. Advances in coarse particle flotation-copper and gold. In: Proceedings of the MetPlant Conference, 2013, Perth, Australia (AusIMM, Publication series 5/2013), pp. 378–386. Kohmuench, J.N., Mankosa, M.J., Yan, E.S., Wyslouzil, H., Christodoulou, L., 2010. Advances in coarse particle flotation industrial minerals. In: Proceedings 25th International Mineral Processing Congress, 2010, pp. 2065–2076. Kohmuench, J.N., Mankosa, M.J., Kennedy, D.G., Yasalonis, J.L., Taylor, G.B., Luttrell, G.H., 2007. Implementation of the HydroFloat technology at the South Fort Meade Mine. Miner. Metall. Process. 24 (4), 264–270.
65
Reyes Bahena, J.L., Franzidis, J.P., Manglapig, E.V., López Valdivieso, A., Ojeda Escamilla, M.C., 2006. Assessment of reagent and regrinding in a flotation circuit. In: Memoria del XVI Congreso Internacional de Metalurgia Extractiva, México, pp. 195–205. Rubio, J., Capponi, F., Matiolo, E., Nunes, D., Guerrero, C. P., Berkowitz, G., 2003. Advances in flotation of mineral fines. In Proceedings XXII International Mineral Processing Congress, Cape-Town, South Africa, pp. 1014–1020. Schubert, H., Bischofberger, C., 1979. On the optimization of hydrodynamics in flotation processes. In: XIII International Mineral Processing Congress, Part B, pp. 353–375. Schulze, H.J., 1984. Physico-Chemical Elementary Processes in Flotation. Elsevier, Amsterdam, 348p. Soto, H., 1992. Development of novel flotation–elutriation method for coarse phosphate beneficiation. Final Report. FIPR Publication. Florida Institute Phosphate Research. 77p. Soto, H., 1988. Private Report to the Florida Institute of Phosphate Research. Bartow, Florida, 35p. Soto, H., Barbery, G., 1991. Flotation of coarse particles in a counter-current column cell. Miner. Metall. Process. 8 (1), 16–21. Tabosa, E., Runge, K., Duffy, K., 2013. Strategies for increasing coarse particle flotation in conventional flotation cells. Flotation 13, 18–21 November, 15p. Testa, F.G., Nascimento, A.L., Duarte, F.E., Yan, E., 2011. Estudo piloto de flotação de grossos utilizando hydrofloat com minério fosfático de Cajati-SP. In: XXIV Encontro Nacional de Tratamento de Minérios e Metalurgia Extrativa, SalvadorBA, pp. 152–158 (in Portuguese).