- Email: [email protected]
Effect of energy input on flocculation process and flotation performance of fine scheelite using sodium oleate
Minerals Engineering 112 (2017) 27–35
Contents lists available at ScienceDirect
Minerals Engineering journal homepage: www.elsevier.com/locate/mineng
Effect of energy input on flocculation process and flotation performance of fine scheelite using sodium oleate
MARK
⁎
Wei Chen, Qiming Feng, Guofan Zhang , Longfei Li, Saizhen Jin School of Mineral Processing and Bioengineering, Central South University, Changsha 410083, China
A R T I C L E I N F O
A B S T R A C T
Keywords: Energy input Fine scheelite Flotation rate Shear flocculation Sodium oleate
We have investigated the effect of energy input on the flocculation process and flotation behavior of fine scheelite of less than 10 µm size. Sodium oleate was used as a dual-function reagent, acting as flocculant and collector. Energy input in shear flocculation was controlled in a four-baffle cylindrical tank with a four-blade impeller by changing the agitation speed. The flocculation process was investigated by measuring continuous transformations in size distribution and observing floc shape. The results show that with increasing energy input, the size distribution of fine scheelite transforms from unimodal to bimodal. The flocs produced tend to possess more branches with low energy input and tends to become globule-like with high energy input. A parameter termed the flocculation degree was introduced to quantify the flocculation process as a function of energy input. The flocculation degree with increasing energy input reveals the aggregation order of different size fractions (all less than 10 µm) when forming flocs. The flotation rate of flocs formed with different energy input was studied. The results demonstrate that the flotation rate is closely related to energy input and also, exhibits an intimate correlation with flocculation degree. These results could potentially be used to routinely monitor the flotation performance of fine particles in operating plants when shear flocculation is used.
1. Introduction Scheelite is the principle mineral of tungsten and is usually recovered by froth flotation. Froth flotation is a physico-chemical separation process that utilizes the difference in surface properties of the valuable minerals and the unwanted gangue minerals (Wills and Napier-Munn, 2005). Due to the brittle nature of scheelite and/or mineral liberation of low grade, finely disseminated scheelite ore, slimes (particles less than 10 µm) containing high grade of scheelite are often generated in the size reduction process. The fine scheelite in flotation feeds is difficult to recover and the loss of fine scheelite is generally regarded as a main reason for low flotation efficiency in flotation plants (Sivamohan, 1990; Ralston, 1992). It is reported that more than 1/5 of total tungsten in flotation feeds is presented as slimes (Gaudin, 1995). Therefore, promoting the recovery of fine scheelite is of great importance. Common collectors for scheelite, such as fatty acids and their derivatives, have been widely demonstrated to be adsorbed onto scheelite through chemisorption, and this adsorption makes its surface hydrophobic (Atademir et al., 1981; Rao and Forssberg, 1991). The apparent size of particles is a significant factor during the process, as the adsorption exploits the interaction between collector and exposed particle
⁎
surface. Fine particle characteristics, such as large specific surface area, often leads to high reagent consumption and low flotation rate in flotation (Sivamohan, 1990). Calculations on efficiencies of particle–bubble collision have demonstrated that the low flotation rate of fine particles are attributed to the low efficiency of collision between fine particles and gas bubbles in flotation cells (Pyke et al., 2003). Increasing the apparent size of fine particles can improve the efficiency. The size increasing process is known as flocculation, coagulation, or aggregation and was firstly introduced in pre-treatment of fine scheelite flotation as “shear flocculation” in 1975 (Warren, 1975). Shear flocculation is the process where charged hydrophobic fine particles are aggregated under high shear conditions. It has been proven that surface of scheelite is negatively charged in dilute alkaline medium and becomes more negative when sodium oleate is added (Arnold and Warren, 1974). Shear flocculation utilizes the energy of hydrophobic association that takes place when the hydrocarbon chains of the collector on particle surface come in contact (Subrahmanyam and Forssberg, 1990), but energy is required to be input into the system to allow the hydrophobic association to occur, which has electrostatic repulsion as a barrier to the process. In previous published research, shear flocculation for fine scheelite was performed in a cylindrical stirred tank with a one-blade agitator. At sufficient high agitating
Corresponding author. E-mail address: [email protected] (G. Zhang).
http://dx.doi.org/10.1016/j.mineng.2017.07.002 Received 25 September 2016; Received in revised form 10 June 2017; Accepted 11 July 2017 Available online 17 July 2017 0892-6875/ © 2017 Elsevier Ltd. All rights reserved.
Minerals Engineering 112 (2017) 27–35
W. Chen et al.
the flotation rate of shear flocculated pulps containing flocs and discrete particles is still unclear and has rarely been reported. The main objective of this paper is to define the effect of energy input (EI) on the flocculation process and flotation rate of fine scheelite. A reagent scheme for shear flocculation was determined by basic flotation behavior of fine scheelite (less than 10 µm) in contrast to that of coarse scheelite (larger than 10 µm). A four-baffle cylindrical tank with a four-blade impeller was used to facilitate EI by changing agitation speed. Particle size measurement, optical observation and flotation tests with different EI were conducted. To quantify the flocculation process of coarse fines (CF, +7.5–10 μm), intermediate fines (IF, +2.5–7.5 μm) and fine fines (FF, −2.5 μm) of scheelite, a parameter termed the flocculation degree, was introduced. On the basis of calculating the flocculation degree of these size fractions with different EI, the flocculation process was described by an aggregation and growth model. Finally, the correlation between the flotation rate and flocculation degree as a function of EI was discussed.
speed, hydrophobic, negatively charged fine scheelite particles (about 1 µm) overcame their energy barrier and aggregated into flocs (larger than 4 µm), which were easy to be recovered by conventional froth flotation (Warren, 1975). It was reported that the pre-treatment via shear flocculation on a fine scheelite ore (70 wt.% less than 15 μm) could achieve a 9% improvement in recovery and a 5–6% improvement in scheelite grade (Koh and Warren, 1980). Based on flocculation and flotation of fine scheelite by offering high shear conditions, shear flocculation can potentially upgrade deposits of finely-grained minerals, and, more generally, for separating different types of fine particles suspended in a fluid (Warren, 1992). Numerous papers about shear flocculation have been published, for example, aggregation of chromite fines using sodium oleate (Akdemir and Hiçyilmaz, 1996) and its separation from serpentine (Akdemir and Hiçyilmaz, 1998), flocculation-flotation of hematite (Akdemir, 1997; Yin et al., 2011) and its separation from quartz (Pascoe and Doherty, 1997), floc flotation of galena-sphalerite fines using potassium amyl xanthate (Song et al., 2001), shear flocculation of celestite using anionic surfactants (Ozkan et al., 2006), shear-induced flocculation of kaolinite (Mietta et al., 2009) and shear flocculation of colemanite (Ucbeyiay and Ozkan, 2011, 2014). Though shear flocculation has always appeared promising for recovering valuable mineral particles that are not floatable due to their fine size, there are little large-scale applications of shear flocculation in flotation practice. However, the energy input range for forming optimal flocs with no destruction or re-dispersion is hard to determine (Bakker et al., 2002). It is thought that the limited industrial uptake of shear flocculation is attributed to lack of full understanding and characterization of shear flocculation (Forbes, 2011). So far, how to determine an appropriate energy input for realizing effective shear-flocculation hasn't been studied. Unfortunately, published papers related to flocs and energy input mainly focused on the size, shape and strength of flocs. It has been reported that size distribution of flocculated scheelite pulp is bimodal. The flocculation behavior of flocs and discrete particles can be described by mathematical models (Koh et al., 1986). The size distribution of flocculated suspensions in larger tanks and the growth of flocs during flocculation process can be predicated by population-balance model (Koh et al., 1989, 1987; Heath, 2006). The floc shape of fine particles can be characterized by parameters such as convexity, circularity (Vlieghe, 2014) and roundness (Koivuranta et al., 2013). The floc structure has been reported to be described by aspect ratio (Koivuranta et al., 2014), area fractal dimension (Gruy, 2011) and mass fractal dimension (Zhou and Franks, 2006). The strength of flocs induced by shear flocculation can be measured by its shear resistance capacity and interaction forces (Zhou et al., 2008; Liang et al., 2015). In addition, the process variables influencing the aggregation and breakage kinetics of flocculation has also been studied (Heath et al., 2009). In summary, the published work to date has put emphasis on the characterization techniques for flocs induced by shear flocculation, rather than the method of selecting the suitable energy input to achieve the most effective improvement in recovery of fine particles and the flotation rate. It should be also noted that energy input on mineral pulp has considerable influence on its separation process and flotation rate in flotation. Energy input (represented by agitation intensity) in pilot-scale mechanical flotation cells has been reported to be a significant factor for the separation effect of platinum ores and gangue minerals (Deglon, 2005). Similarly, energy input in coal flotation (measured by shaft power and flotation time in cell) is also found to have a remarkable effect on the flotation grade and recovery of coal (Gui et al., 2014, 2013). Studies on the flotation kinetics of minerals using different energy input devices, such as oscillating grid flotation cell (Changunda et al., 2008; Massey et al., 2012) and a “Rushton turbine cell” (Newell and Grano, 2006), have demonstrated that flotation rates of galena, pyrite, pentlandite, apatite, hematite and quartz (Safari et al., 2016) are all energy input dependent. However, the influence of energy input on
2. Experimental 2.1. Materials and reagents Hand-picked pure scheelite crystals were crushed to less than 1 mm by a laboratory jaw crusher and a laboratory roll crusher. The crushed products were concentrated on a concentrating table to remove the heavy particles and on a high-intensity magnetic separator several times to discard the magnetic minerals. The non-magnetic products were ground in a porcelain mill and classified into five narrow size fractions: −10, +10–38, +38–55, +55–74 and +74–106 μm. Sodium oleate (C18H33O2Na) used in this study was purchased from Tianjing Kermil Chemical Reagents Development Centre, Tianjin, China. Its molecular structure consists of a hydrocarbon chain and a carboxyl head group, as shown in Fig. 1. It is selected as the flocculant in shear flocculation operation because it is a traditional collector in flotation practice of scheelite. Hydrochloric acid (HCl) and sodium carbonate (Na2CO3) were used as pH regulators. Distilled water was used for all tests. The effect of sodium oleate on the zeta potential of scheelite has been measured (Arnold and Warren, 1974). At the vicinity of pH 10 adjusted by Na2CO3, the zeta potential of scheelite is about −36 mV. Conditioned by 1 × 10−4 mol/L sodium oleate, the zeta potential becomes −42 mV. The decrease in zeta potential is attributed to the adsorption of the sodium oleate on the surface of scheelite as introduced in Section 1. 2.2. EI and its measurement EI was realized using a four-baffle agitation tank with a four-blade impeller, as shown in Fig. 2. The tank was a cylinder with a height of 100 mm and a diameter of 60 mm respectively. There were 4 baffles (80 mm × 5 mm) in the tank. The whole diameter of the impeller was 30 mm and its blade (10 mm × 10 mm) was perpendicular to the axis. The distance between impeller and tank bottom was 30 mm. The impeller was driven by an IKA Eurostar power control-visc6000 agitator, which could control and adjust the rotational speed in a stepless manner in range of 150–6000 rpm. The torque to overcome the fluid resistance was recorded by a torque meter installed inside the agitator. When agitating with a slurry volume of 200 mL, it was found that air entrainment in the liquid was minimal even at 1500 rpm. Mineral pulp was prepared by adding 10 g fine scheelite to 200 mL
Fig. 1. Molecule structure of sodium oleate.
28
Minerals Engineering 112 (2017) 27–35
W. Chen et al.
1000 μm × 1000 μm region of slurry can be videoed by a digital camera, so the slurry images at any recorded time can be analyzed. 2.5. Flotation tests Flotation tests were carried out in a XFG flotation machine with a 100 mL cell at 1500 rpm. For flotation tests on scheelite in different size, pulp was prepared by adding 5 g solid into the cell with 100 mL distilled water. The pH was adjusted to a desired value by adding Na2CO3 or HCl stock solutions. Sodium oleate was added at a desired concentration. The conditioning time was 3 min and the flotation time was 6 min for each flotation test. For shear flocculation flotation tests of fine scheelite (−10 μm), conditioning operation was conducted in the agitation tank (shown in Fig. 2). After conditioning, the slurry was transferred into the flotation cell and mixed for 10 s at 1500 rpm until the flotation began. For each shear flocculation test with certain EI, the total flotation time was 6 min and the concentrate was collected in batches as a flotation time of 20 s, 20 s, 20 s, 20 s, 30 s, 40 s, 50 s, 60 s and 100 s. For fine scheelite (−10 μm) flotation tests, the concentrate and tails were collected and dried in a watch glass. Flotation recovery was calculated based on solid weight distributions among total weight of all products. A first-order model was used to describe the flotation kinetics of shear flocculation pre-treated pulp. The recovery of scheelite at flotation time t to a first approximation is given by the following expression:
Fig. 2. Schematic of the agitation tank used for EI on slurry.
distilled water in the tank. The liquid-solid ratio by weight was kept at 20:1. The temperature was controlled by a water bath, and was set at 25 °C. The agitation started 1 min before pH regulator was added. Agitation time was recorded after sodium oleate was added. Injection syringes were used to take samples for particle size measurements and flotation tests. EI on slurry was calculated by following expression:
W=
T ·n·t 9549V
where W is EI on slurry (J/m3), T is the torque to overcome the fluid resistance (N·m), n is agitating speed (rpm), t is agitating time (s) and V is the slurry volume (m3). EI is a term that describes the agitation intensity on the slurry. In this study, the agitation time and slurry volume were controlled at 15 min and 200 mL.
R (t ) = Rmax (1−e−kt )
where k is the flotation rate constant and Rmax is their flotation recovery at an infinite time (Bulatovic, 2007). A nonlinear least square regression was used to calculate k and Rmax from the best fit of the curve of experimental flotation recoveries versus flotation time using Eq. (2). These flotation rate constants are referred in the text as experimental flotation rate constants.
2.3. Particle size measurements Particle size distribution of the mineral slurry sample was measured using a Malvern Mastersizer 2000, by light scattering. Each pulp sample was sub-sampled twice, with the particle size distribution of each subsample measured in triplicate. The aggregate of six resulting measurements was used as an average particle size distribution curve. The apparent size of fine scheelite with certain EI was evaluated by D50 (the maximum particle diameter below which 50% of the sample volume exists) and D90 (the maximum particle diameter below which 90% of the sample volume exists) of the size distribution, which were directly offered by the Mastersizer. The flocculation behavior of different size fractions in the fine scheelite was described by flocculation degree, which was calculated by following expression:
Rfloc (i) =
αi−βi × 100% αi
(2)
3. Results and discussion 3.1. Basic flotation performance of fine scheelite (−10 μm) The effect of pH on the flotation recovery of fine and coarse scheelite was studied by single mineral flotation tests and the results are shown in Fig. 3a. Flotability of scheelite in different size ranges seldom changed with increasing pH but the recovery of fine scheelite (−10 μm) (about 50%) was much lower than that of coarse scheelite (+10–106 μm) (about 90%) at the same pH range. The results show that adjusting the pH of mineral slurry doesn't help to improve the recovery of fine scheelite. Fig. 3b shows the effect of sodium oleate dosage on flotation behavior of scheelite in different size fractions. Recoveries of all size fractions increased with the increase of sodium oleate dosage. When sodium oleate dosage increased to 2 × 10−4 mol/L, the recoveries of the coarse size fractions (+38–55 μm, +55–74 μm, +74–106 μm) exceeded 90%, while that of fine scheelite (−10 μm) reached only 54%. For fine scheelite, a high sodium oleate dosage up to 8 × 10−4 mol/L was required in order to get a recovery higher than 90%, which was four times of that for coarse fractions. The reason can be its higher surface area than coarse scheelite, as stated in Section 1. The aforementioned flotation results indicate that sodium oleate has a good collecting ability for scheelite, albeit with reduced recovery of fine particles. The collecting ability derives from the chemisorption, i.e., the interaction between –COO– in sodium oleate and Ca2+ on scheelite surface (Rao and Forssberg, 1991; Bo et al., 2015). Affected by the eCH2e groups in sodium oleate on particle surface, scheelite becomes hydrophobic and floatable. Increasing the apparent size by shear flocculation offers a way to improve the recovery of fine scheelite. In this study, pH 10 and sodium oleate dosage of 2 × 10−4 mol/L were used as reagent scheme in shear flocculation to investigate the
(1)
where Rfloc(i) is the flocculation degree of i size fraction, αi is the average content of i size fraction without addition of sodium oleate, βi is the average content of i size fraction after sodium oleate was added and agitated with certain EI. Flocculation degree is a term that evaluates the relative change in content of i size fraction with certain EI. For a specified size fraction, its flocculation degree increases with the decreases of the content of this size fraction. Since the samples in this study are less than 10 μm, i in Eq. (1) is sub-divided in CF (+7.5–10 μm), IF (+2.5–7.5 μm) and FF (−2.5 μm) via their different variation tendency verses EI. 2.4. Observation of flocs The observation of scheelite suspensions was conducted using a transmission light microscope (OLYMPUS-CX31RTSF). 1 mL of suspension was sampled by a syringe and diluted 10 times in a beaker by distilled water. Then dilute solution was dropped on a microslide and was flatted by a coverslip on the objective table. Around 29
Minerals Engineering 112 (2017) 27–35
W. Chen et al.
Fig. 3. Effects of pH and sodium oleate dosage on the flotation behavior of scheelite.
flocculation process and flotation behavior of fine scheelite with different EI.
bimodal, which agreed well with other reports (Koh et al., 1986). Flotation recovery also witnessed a significant increase from 60.84% to 76.83% (Fig. 5b), showing that the growth of flocs remarkably promoted the flotation performance of fine scheelite. Fig. 6 shows the formation of big flocs and corresponding promoted flotation performance. When EI increased from 6.41 kJ/m3 to 9.55 kJ/ m3, the bimodal peak was replaced by a new, high unimodal peak that located in +10 μm area. In this EI range, D50 increased from 8.67 μm to 29.87 μm and D90 increased from 46.02 μm to 70.41 μm. Accordingly, the formation of the big flocs was the result of the FF flocculation as the arrows pointed (Fig. 6a). Flotation recovery met a big increase (from 76.83% to 84.38%) as shown in Fig. 6b. Since large-scale flocculation of IF has occurred in stage 2, there is no doubt that the improved flotation derives from the flocculation of FF and formation of big flocs in the EI range. Fig. 7 shows the stabilization of flocs and promoted flotation when EI continued to increase from 9.55 kJ/m3 to 12.06 kJ/m3. The unimodal peak shifted to left, suggesting a certain degree of decrease in apparent size of the flocs. In the meantime, the flocculation of FF was also observed in Fig. 7a. Taken the decrease and the flocculation together, the apparent size met drops both in D50 (from 29.87 μm to 23.65 μm) and D90 (from 70.41 μm to 53.48 μm), however, flotation recovery witnessed a slight increase from 84.38% to 86.50%, as shown in Fig. 7b. The results demonstrate that the flocculation of the FF has outweighed the negative effect of flocs breakage thus increasing the flotation recovery in the stabilization stage. Fig. 8 shows the breakage of flocs and decreased flotation when excessive EI is applied. The dominant peak in Fig. 8a lowered and created a new peak located on the left, indicating that big flocs had broken down into discrete particles or small flocs. The variation of D90 (from 53.48 μm to 54.84 μm) and D50 (from 23.65 μm to 20.48 μm) also
3.2. Changes in size distribution and flotation performance of fine scheelite versus EI The size distribution and flotation behavior of fine scheelite with different EI were studied and the results are shown in Figs. 4–8. According to the formation, growth, stabilization and breakage of flocs with different EI, the whole flocculation process as well as corresponding flotation performance was discussed in stages. Fig. 4 shows the transformations in size distribution and flotation behavior of fine scheelite in the beginning of floc formation stage. With increasing EI, unimodal distribution in Fig. 4a started to fissure. The high peak of fine scheelite lowered and formed a new peak that located in the +10 μm area. Apparent size data showed that D90 increased from 10.29 μm to 23.22 μm and D50 varied between 4.76 μm and 4.26 μm when EI increased from 0.19 kJ/m3 to 2.64 kJ/m3. The results suggested a slight increase in the apparent size of the fine scheelite (−10 μm). The increase in apparent size resulted in a slightly promoted flotation, with flotation recovery increasing from 54.28% to 60.84% as shown in Fig. 4b. Therefore, in this EI range, the dispersed fine scheelite began to aggregate into flocs and lead to a slightly promoted flotation. Fig. 5 exhibits the growth of the flocs and the remarkably improved flotation behavior of the scheelite fines. The dominant peak decreased further as a new peak generated and the whole peak shifted to right when EI increased more, as shown in Fig. 5a. In this stage, D50 increased from 4.26 μm to 8.67 μm and D90 increased from 23.22 μm to 46.02 μm. The results indicated an obvious growth in apparent size of the flocs, which were mainly derived from the flocculation of IF and FF. The unimodal size distribution of fine scheelite totally changed to
Fig. 4. Stage 1: Beginning of floc formation and slightly promoted flotation.
30
Minerals Engineering 112 (2017) 27–35
W. Chen et al.
Fig. 5. Stage 2: Growth of flocs and remarkably improved flotation.
confirmed the breakage of flocs. As a result, the flotation recovery met a decrease from 86.50% to 77.28%, as shown in Fig. 8b. The size distribution of fine scheelite changed from unimodal to bimodal and then unimodal, with increasing EI and increasing apparent size of flocs. However, the phenomena of bimodal size distributions indicated that finer fractions in scheelite fines behaved differently in the flocculation process. For example, peak of the FF was the last to disappear during the formation stage of flocs but the first to appear during the breakage stage of flocs. The different behaviors of finer fractions in flocculation process with the same EI are very important for selecting the suitable EI to realize the flocculation of all finer fractions in the maximum extent.
hydrophobicity was induced by sodium oleate and EI was offered by the agitation tank. The discrete fine particles could overcome the energy barrier when applied EI exceeds 2.04 kJ/m3, and aggregated into flocs. The flocs experienced small, branchy and globule-like as EI increased. On the other hand, the different floc shape formed with different EI illustrates its formation, growth, stabilization and breakage qualitatively. As a result, the flotation performance of fine scheelite is promoted to different degree in different flocculation stage. Therefore, an optimal value of EI to achieve the most effective flocculation should be determined by quantitative analysis on flocculation behavior of each fractions less than 10 μm. 3.4. Flotation rate and flocculation degree as a function of EI
3.3. Changes in floc shape in different flocculation stage Flotation rate was calculated by fitting the cumulative recovery curves. The fitting results are shown in Appendix A. The flocculation degree of finer size fractions in the scheelite fines (CF, IF and FF) was also calculated using the size distribution data in Section 3.2. The flotation rate constant and flocculation degree as a function of EI was presented in Fig. 10. The flotation rate constant was observed to increase as EI increased with a maximum at 12.06 kJ/m3. The variation trend was in accord with the flotation performance in Section 3.2. For flocculation stage 1–3, or EI less than 9.55 kJ/m3, rate constant was approximately proportional to the 3.07 power of EI, as fitted by an allometric model in Appendix B. For the rest stages, the rate constant decreased slowly with excessive EI. The results indicate that the flotation rate constant of flocculated scheelite pulp is EI dependent and can be selected as an evaluation standard in determining the suitable EI in shear flocculation. The flocculation degree of CF witnessed a stepped increase with increasing EI. It nearly kept a constant of 20% when EI was less than
Fig. 9 presented a summary of the observations for flocs with different EI, and, at different stages in Figs. 4–8. It can be seen in Fig. 9a that in the absence of sodium oleate, the fine scheelite was well-dispersed in slurry. With the addition of sodium oleate and increasing EI, the particles tended to agglomerate, forming flocs (Fig. 9b) and possessing more branches (Fig. 9c–e). When EI exceeded the critical value for floc breakage, the flocs with branches fell apart and formed globulelike aggregations. At the same time, the branches were re-dispersed and became discrete particles, as shown in Fig. 9f. In fact, the changes in flocs shape reflected the effect of EI on the movement and aggregation behaviors of scheelite particles in the formation, growth, stabilization and breakage of flocs. The mechanisms for the movement and aggregation of fine scheelite have been widely studied. Hydrophobic particles can aggregate if the energy barrier between the particles was overcame by enough kinetic produced by agitation (Bilgen and Wills, 1991). In this research,
Fig. 6. Stage 3: Formation of big flocs and promoted flotation.
31
Minerals Engineering 112 (2017) 27–35
W. Chen et al.
Fig. 7. Stage 4: Stabilization of flocs and slightly promoted flotation.
flocculation, CF in the fine scheelite (−10 μm) agglomerates and generates floc core; with increasing EI, IF and FF adhere to the floc core, acting as branches and coatings, respectively; with suitable EI, all fractions gather into a big floc; with excessive EI, FF and IF are peeled off one and re-dispersed in the flocculated suspensions. On the basis of the above flocculation process, suitable EI to achieve the most effective flocculation in shear flocculation can be selected directly by monitoring the changing process of size distribution (represented by flocculation degree) and by calculating the flotation rate constant. In shear flocculation-flotation of fine scheelite, flotation recovery as well as flotation rate should be taken into consideration. Therefore, a compromise between the flocculation degree and flotation rate can be used to determine the EI range. In this mixing cell and flotation cell, the suitable EI lies in 9.55–12.06 kJ/m3, with flotation rate constant ranging from 0.0369 s−1 to 0.0418 s−1.
8.29 kJ/m3, implying that the content of CF in the flocculated pulp didn't change in this EI range. When EI increased further, it increased to around 40% and hardly decreased with excessive EI, suggesting that CF existed stably in flocs with high EI. For IF, its flocculation degree showed a trend of rise, then fall, with a turning point at 9.55 kJ/m3. The trend was consistent with the aggregation and breakage behaviors of flocs with increasing EI as described in Sections 3.2 and 3.3. For FF, its flocculation degree first dropped, and increased, and then decreased with increasing EI. When EI was less than 4.52 kJ/m3, the flocculation degree of FF was negative, indicating a relative increase of FF content in the flocculated pulp due to the flocculation of CF and IF in this EI range. The flocculation degree results provide a good understanding of the roles played by CF, IF and FF during the flocculation process. In the formation stage of flocs, the sequence of the three fractions entering the flocs is ranked as CF, IF, FF. FF starts to aggregate only when EI has exceeded 4.52 kJ/m3, in which range IF and CF have agglomerated to a certain level. Therefore, FF gets into flocs by its adhesion on the already formed flocs by CF and IF. The conclusion can also be supported by the phenomenon of flocs that possess many branches in the growth stage. In the breakage of flocs, the stability of the three fractions in the flocs with increasing EI is ranked as CF > IF > FF. FF has the least stability in flocs and is prior to get re-dispersed in pulp with excessive EI. In summary, CF acts as the core of flocs, and FF forms the branches of flocs, and IF plays both roles in the flocculation process of fine scheelite (−10 μm). Combining the formation and breakage of flocs, the flocculation process of fine scheelite with increasing EI can be described by an aggregation and growth model, as shown in Fig. 11. In the model, CF, IF and FF are represented by balls in different diameters, respectively. The flocculation process with different EI was simulated by the aggregation and dispersion of these balls. With basic EI just able to produce shear
4. Conclusions Given the aforementioned results, we arrived at the following primary conclusions: (1) Shear flocculation induced by suitable energy input can significantly improve the flotability of fine scheelite (−10 μm) using sodium oleate as collector. (2) Energy input influences the apparent size and flotation rate of fine scheelite. Apparent size of fine scheelite and its flotation rate increase as energy input increases in limited range; the size distribution of fine scheelite transforms from unimodal to bimodal as energy input changes. (3) Low energy input tends to induce large flocs with more branches but weak resistance to breakage; high energy input tends to form
Fig. 8. Stage 5: Breakage of flocs and decreased flotation.
32
Minerals Engineering 112 (2017) 27–35
W. Chen et al.
Fig. 9. Floc shape with different EI.
a. Well-dispersed fine scheelite
c. Growth of flocs (6.41 kJ/m3)
e. Breakage of flocs with branches(12.06 kJ/m3)
b. Formation of flocs (2.04 kJ/m3)
d. Flocs with branches (9.55 kJ/m3)
f. Globule-like flocs (17.09 kJ/m3) (4) The flocculation process of fine scheelite can be described by an aggregation and growth model. The process starts with aggregation of coarse particles (+7.5–10 μm) as core of floc, and develops by involving intermediate particles (+2.5–7.5 μm), and ends with rolling the fine particles (−2.5 μm) onto surface of floc with increasing energy input.
Acknowledgement The authors acknowledge the support of the Major State Basic Research Development Program of China (973program) (2014CB643402). Special thanks to Mr. David Beattie for his valuable discussions and the final language adjustment. Fig. 10. Flotation rate constant and flocculation degree (Rfloc) of fine scheelite versus EI.
globule-like flocs with high stability but inevitable re-dispersion. Appendix A. Rate constant calculation The flotation rate constant was obtained from nonlinear regression based on the first-order model equation: R(t) = Rmax(1 − e−kt) (see Section 2.5). The results are shown in Table A1.
33
Minerals Engineering 112 (2017) 27–35
W. Chen et al.
Fig. 11. Aggregation and growth model: formation and breakage of scheelite floc. Table A1 Fitting results of flotation recovery curves with different EI. EI (kJ/ m3)
1.38 2.64 4.52 6.41 8.29 9.55 12.06 14.57 17.09
R (t = 6 min)
54.28 60.84 70.33 76.83 81.01 84.38 86.5 83.63 77.09
Fitting results Rmax (%)
k (s−1)
Adj. Rsquare
Reduced ChiSqr
56.69 58.54 67.98 74.44 79.12 81.24 83.86 78.60 71.95
0.0137 0.0160 0.0171 0.0210 0.0287 0.0369 0.0418 0.0399 0.0348
0.9971 0.9937 0.9864 0.9916 0.9958 0.9896 0.9959 0.9872 0.9720
1.0358 2.3680 6.8030 4.9988 2.7884 7.0390 2.9069 8.0506 14.8158
Table B1 Fitting results of flotation rate as a function of EI. Parameter
Value
Standard error
Adj. R-square
Reduced Chi-Sqr
a b c
0.01463 2.16E−05 3.07139
6.19E−04 1.67E−05 0.3403
0.9915
6.77E−07
Appendix B. Fitting result of flotation rate constant as s function of EI An allometric model equation: k = a + b∗ρc, where k is the flotation rate, ρ is EI and a, b, c are parameters, was used to describe the relationship between EI and flotation rate constant. The fitting result is shown in Table B1.
Gui, X., Cheng, G., Liu, J., Cao, Y., Li, S., He, Q., 2013. Effects of energy consumption on the separation performance of fine coal flotation. Fuel Process. Technol. 115, 192–200. Gui, X., Liu, J., Cao, Y., Cheng, G., Li, S., Wu, L., 2014. Flotation process design based on energy input and distribution. Fuel Process. Technol. 120, 61–70. Heath, A.R., 2006. Polymer flocculation of calcite: population balance model. FLUID Mech. Transp. Phenom. 52, 1641–1653. Heath, A.R., Bahri, P.A., Farrow, P.D.F., Farrow, J.B., 2009. Polymer flocculation of calcite: experimental results from turbulent pipe flow. IFAC Proc. 7, 405–410. Warren, J.L., 1975. Shear-flocculation of ultrafine scheelite in sodium oleate solutions. J. Colloid Interface Sci. 50, 307–318. Koh, P.T.L., Andrews, J.R.G., Uhlherr, P.H.T., 1987. Modelling shear-flocculation by population balances. Chem. Eng. Sci. 42, 353–362. Koh, P.T.L., Andrews, J.R.G., Uhlherr, P.H.T., 1986. Floc-size distribution of scheelite treated by shear-flocculation. Int. J. Miner. Process. 17, 45–65. Koh, P.T.L., Lwin, T., Albrecht, D., 1989. Analysis of weight frequency particle size distributions: with special application to bimodal floc size distributions in shear-flocculation. Powder Technol. 59, 87–95. Koh, P.T.L., Warren, L., 1980. A pilot plant test of the shear-flocculation of ultrafine scheelite. In: Eighth Australian Chemical Engineering Conference at Melbourne, Australia, pp. 2–6. Koivuranta, E., Keskitalo, J., Haapala, A., Stoor, T., Sarén, M., Niinimäki, J., 2013. Optical monitoring of activated sludge flocs in bulking and non-bulking conditions. Environ. Technol. 34, 679–686. Koivuranta, E., Keskitalo, J., Stoor, T., Hattuniemi, J., Sarén, M., Niinimäki, J., 2014. A comparison between floc morphology and the effluent clarity at a full-scale activated sludge plant using optical monitoring. Environ. Technol. 35, 1605–1610. Liang, L., Peng, Y., Tan, J., Xie, G., 2015. A review of the modern characterization
References Akdemir, U., 1997. Shear flocculation of fine hematite particles and correlation between flocculation, flotation and contact angle. Powder Technol. 94, 1–4. Akdemir, Ü., Hiçyilmaz, C., 1998. Selective shear flocculation of chromite from chromiteserpentine mixtures. Colloids Surf. A: Physicochem. Eng. Asp. 132, 75–79. Akdemir, Ü., Hiçyilmaz, C., 1996. Shear flocculation of chromite fines in sodium oleate solutions. Colloids Surf. A: Physicochem. Eng. Asp. 110, 87–93. Arnold, R., Warren, L.J., 1974. Electrokinetic properties of scheelite. J. Colloid Interface Sci. 47, 134–144. Atademir, M.R., Kitchener, J.A., Shergold, H.L., 1981. The surface chemistry and flotation of scheelite, II. Flotation “collectors”. Int. J. Miner. Process. 8, 9–16. Bakker, M.G., Spears, D.R., Murphy, D.D., Turner, G.L., 2002. Whatever happened to shear flocculation. Fluid – Part. Sep. J. 14, 155–168. Bo, F., Xianping, L., Jinqing, W., Pengcheng, W., 2015. The flotation separation of scheelite from calcite using acidified sodium silicate as depressant. Miner. Eng. 80, 45–49. Bulatovic, S.M., 2007. Handbook of Flotation Reagents. Changunda, K., Harris, M., Deglon, D.A., 2008. Investigating the effect of energy input on flotation kinetics in an oscillating grid flotation cell. Miner. Eng. 21, 924–929. Deglon, D.A., 2005. The effect of agitation on the flotation of platinum ores. Miner. Eng. 18, 839–844. Forbes, E., 2011. Shear, selective and temperature responsive flocculation: a comparison of fine particle flotation techniques. Int. J. Miner. Process. 99, 1–10. Gaudin, A.M., 1995. Flotation. McGraw-hill Book Company. Gruy, F., 2011. Population balance for aggregation coupled with morphology changes. Colloids Surf. A: Physicochem. Eng. Asp. 374, 69–76.
34
Minerals Engineering 112 (2017) 27–35
W. Chen et al. techniques for flocs in mineral processing. Miner. Eng. Massey, W.T., Harris, M.C., Deglon, D.A., 2012. The effect of energy input on the flotation of quartz in an oscillating grid flotation cell. Miner. Eng. 36–38, 145–151. Mietta, F., Chassagne, C., Winterwerp, J.C., 2009. Shear-induced flocculation of a suspension of kaolinite as function of pH and salt concentration. J. Colloid Interface Sci. 336, 134–141. Newell, R., Grano, S., 2006. Hydrodynamics and scale up in Rushton turbine flotation cells: part 1 – cell hydrodynamics. Int. J. Miner. Process. 81, 65–78. Ozkan, A., Ucbeyiay, H., Aydogan, S., 2006. Shear flocculation of celestite with anionic surfactants and effects of some inorganic dispersants. Colloids Surf. A: Physicochem. Eng. Asp. 281, 92–98. Pascoe, R.D., Doherty, E., 1997. Shear flocculation and flotation of hematite using sodium oleate. Int. J. Miner. Process. 51, 269–282. Pyke, B., Fornasiero, D., Ralston, J., 2003. Bubble particle heterocoagulation under turbulent conditions. J. Colloid Interface Sci. 265, 141–151. Ralston, J., 1992. The influence of particle size and contact angle in flotation. Dev. Miner. Process. 12, 203–224 (Chapter 6). Rao, K.H., Forssberg, K.S.E., 1991. Mechanism of oleate interaction on salt-type minerals. Part III. Adsorption, zeta potential and diffuse reflectance FT-IR studies of scheelite in the presence of sodium oleate. Colloids Surf. 54, 161–187. Bilgen, S., Wills, B.A., 1991. Shear flocculation – a review. Miner. Eng. 4, 483–487. Safari, M., Harris, M., Deglon, D., Filho, L.L., Testa, F., 2016. The effect of energy input on flotation kinetics. Int. J. Miner. Process. 1–8. Sivamohan, R., 1990. The problem of recovering very fine particles in mineral
processing—a review. Int. J. Miner. Process. 28, 247–288. Song, S., Lopez-Valdivieso, A., Reyes-Bahena, J.L., Lara-Valenzuela, C., 2001. Floc flotation of galena and sphalerite fines. Miner. Eng. 14, 87–98. Subrahmanyam, T.V., Forssberg, K.S.E., 1990. Fine particles processing: shear-flocculation and carrier flotation – a review. Int. J. Miner. Process. 30, 265–286. Ucbeyiay, H., Ozkan, A., 2014. Two-stage shear flocculation for enrichment of fine boron ore containing colemanite. Sep. Purif. Technol. 132, 302–308. Ucbeyiay Sahinkaya, H., Ozkan, A., 2011. Investigation of shear flocculation behaviors of colemanite with some anionic surfactants and inorganic salts. Sep. Purif. Technol. 80, 131–139. Vlieghe, M., 2014. In situ characterization of floc morphology by image analysis in a turbulent Taylor-Couette reactor. Am. Inst. Chem. Eng. 60, 2389–2403. Warren, L.J., 1992. Shear-flocculation. Develop. Miner. Process. 309–329. Wills, B.A., Napier-Munn, T., 2005. 12 – Froth flotation BT – Wills’ mineral processing technology. In: Wills’ Mineral Processing Technology, seventh ed. ButterworthHeinemann, Oxford, pp. 267–352. Yin, W.Z., Yang, X.S., Zhou, D.P., Li, Y.J., Lü, Z.F., 2011. Shear hydrophobic flocculation and flotation of ultrafine Anshan hematite using sodium oleate. Trans. Nonferrous Met. Soc. China 21, 652–664. Zhou, Y., Franks, G.V., 2006. Flocculation mechanism induced by cationic polymers investigated by light scattering. Langmuir 22, 6775–6786. Zhou, Y., Gan, Y., Wanless, E.J., Jameson, G.J., Franks, G.V., 2008. Interaction forces between silica surfaces in aqueous solutions of cationic polymeric flocculants: effect of polymer charge. Langmuir 10920–10928.
35
Contents lists available at ScienceDirect
Minerals Engineering journal homepage: www.elsevier.com/locate/mineng
Effect of energy input on flocculation process and flotation performance of fine scheelite using sodium oleate
MARK
⁎
Wei Chen, Qiming Feng, Guofan Zhang , Longfei Li, Saizhen Jin School of Mineral Processing and Bioengineering, Central South University, Changsha 410083, China
A R T I C L E I N F O
A B S T R A C T
Keywords: Energy input Fine scheelite Flotation rate Shear flocculation Sodium oleate
We have investigated the effect of energy input on the flocculation process and flotation behavior of fine scheelite of less than 10 µm size. Sodium oleate was used as a dual-function reagent, acting as flocculant and collector. Energy input in shear flocculation was controlled in a four-baffle cylindrical tank with a four-blade impeller by changing the agitation speed. The flocculation process was investigated by measuring continuous transformations in size distribution and observing floc shape. The results show that with increasing energy input, the size distribution of fine scheelite transforms from unimodal to bimodal. The flocs produced tend to possess more branches with low energy input and tends to become globule-like with high energy input. A parameter termed the flocculation degree was introduced to quantify the flocculation process as a function of energy input. The flocculation degree with increasing energy input reveals the aggregation order of different size fractions (all less than 10 µm) when forming flocs. The flotation rate of flocs formed with different energy input was studied. The results demonstrate that the flotation rate is closely related to energy input and also, exhibits an intimate correlation with flocculation degree. These results could potentially be used to routinely monitor the flotation performance of fine particles in operating plants when shear flocculation is used.
1. Introduction Scheelite is the principle mineral of tungsten and is usually recovered by froth flotation. Froth flotation is a physico-chemical separation process that utilizes the difference in surface properties of the valuable minerals and the unwanted gangue minerals (Wills and Napier-Munn, 2005). Due to the brittle nature of scheelite and/or mineral liberation of low grade, finely disseminated scheelite ore, slimes (particles less than 10 µm) containing high grade of scheelite are often generated in the size reduction process. The fine scheelite in flotation feeds is difficult to recover and the loss of fine scheelite is generally regarded as a main reason for low flotation efficiency in flotation plants (Sivamohan, 1990; Ralston, 1992). It is reported that more than 1/5 of total tungsten in flotation feeds is presented as slimes (Gaudin, 1995). Therefore, promoting the recovery of fine scheelite is of great importance. Common collectors for scheelite, such as fatty acids and their derivatives, have been widely demonstrated to be adsorbed onto scheelite through chemisorption, and this adsorption makes its surface hydrophobic (Atademir et al., 1981; Rao and Forssberg, 1991). The apparent size of particles is a significant factor during the process, as the adsorption exploits the interaction between collector and exposed particle
⁎
surface. Fine particle characteristics, such as large specific surface area, often leads to high reagent consumption and low flotation rate in flotation (Sivamohan, 1990). Calculations on efficiencies of particle–bubble collision have demonstrated that the low flotation rate of fine particles are attributed to the low efficiency of collision between fine particles and gas bubbles in flotation cells (Pyke et al., 2003). Increasing the apparent size of fine particles can improve the efficiency. The size increasing process is known as flocculation, coagulation, or aggregation and was firstly introduced in pre-treatment of fine scheelite flotation as “shear flocculation” in 1975 (Warren, 1975). Shear flocculation is the process where charged hydrophobic fine particles are aggregated under high shear conditions. It has been proven that surface of scheelite is negatively charged in dilute alkaline medium and becomes more negative when sodium oleate is added (Arnold and Warren, 1974). Shear flocculation utilizes the energy of hydrophobic association that takes place when the hydrocarbon chains of the collector on particle surface come in contact (Subrahmanyam and Forssberg, 1990), but energy is required to be input into the system to allow the hydrophobic association to occur, which has electrostatic repulsion as a barrier to the process. In previous published research, shear flocculation for fine scheelite was performed in a cylindrical stirred tank with a one-blade agitator. At sufficient high agitating
Corresponding author. E-mail address: [email protected] (G. Zhang).
http://dx.doi.org/10.1016/j.mineng.2017.07.002 Received 25 September 2016; Received in revised form 10 June 2017; Accepted 11 July 2017 Available online 17 July 2017 0892-6875/ © 2017 Elsevier Ltd. All rights reserved.
Minerals Engineering 112 (2017) 27–35
W. Chen et al.
the flotation rate of shear flocculated pulps containing flocs and discrete particles is still unclear and has rarely been reported. The main objective of this paper is to define the effect of energy input (EI) on the flocculation process and flotation rate of fine scheelite. A reagent scheme for shear flocculation was determined by basic flotation behavior of fine scheelite (less than 10 µm) in contrast to that of coarse scheelite (larger than 10 µm). A four-baffle cylindrical tank with a four-blade impeller was used to facilitate EI by changing agitation speed. Particle size measurement, optical observation and flotation tests with different EI were conducted. To quantify the flocculation process of coarse fines (CF, +7.5–10 μm), intermediate fines (IF, +2.5–7.5 μm) and fine fines (FF, −2.5 μm) of scheelite, a parameter termed the flocculation degree, was introduced. On the basis of calculating the flocculation degree of these size fractions with different EI, the flocculation process was described by an aggregation and growth model. Finally, the correlation between the flotation rate and flocculation degree as a function of EI was discussed.
speed, hydrophobic, negatively charged fine scheelite particles (about 1 µm) overcame their energy barrier and aggregated into flocs (larger than 4 µm), which were easy to be recovered by conventional froth flotation (Warren, 1975). It was reported that the pre-treatment via shear flocculation on a fine scheelite ore (70 wt.% less than 15 μm) could achieve a 9% improvement in recovery and a 5–6% improvement in scheelite grade (Koh and Warren, 1980). Based on flocculation and flotation of fine scheelite by offering high shear conditions, shear flocculation can potentially upgrade deposits of finely-grained minerals, and, more generally, for separating different types of fine particles suspended in a fluid (Warren, 1992). Numerous papers about shear flocculation have been published, for example, aggregation of chromite fines using sodium oleate (Akdemir and Hiçyilmaz, 1996) and its separation from serpentine (Akdemir and Hiçyilmaz, 1998), flocculation-flotation of hematite (Akdemir, 1997; Yin et al., 2011) and its separation from quartz (Pascoe and Doherty, 1997), floc flotation of galena-sphalerite fines using potassium amyl xanthate (Song et al., 2001), shear flocculation of celestite using anionic surfactants (Ozkan et al., 2006), shear-induced flocculation of kaolinite (Mietta et al., 2009) and shear flocculation of colemanite (Ucbeyiay and Ozkan, 2011, 2014). Though shear flocculation has always appeared promising for recovering valuable mineral particles that are not floatable due to their fine size, there are little large-scale applications of shear flocculation in flotation practice. However, the energy input range for forming optimal flocs with no destruction or re-dispersion is hard to determine (Bakker et al., 2002). It is thought that the limited industrial uptake of shear flocculation is attributed to lack of full understanding and characterization of shear flocculation (Forbes, 2011). So far, how to determine an appropriate energy input for realizing effective shear-flocculation hasn't been studied. Unfortunately, published papers related to flocs and energy input mainly focused on the size, shape and strength of flocs. It has been reported that size distribution of flocculated scheelite pulp is bimodal. The flocculation behavior of flocs and discrete particles can be described by mathematical models (Koh et al., 1986). The size distribution of flocculated suspensions in larger tanks and the growth of flocs during flocculation process can be predicated by population-balance model (Koh et al., 1989, 1987; Heath, 2006). The floc shape of fine particles can be characterized by parameters such as convexity, circularity (Vlieghe, 2014) and roundness (Koivuranta et al., 2013). The floc structure has been reported to be described by aspect ratio (Koivuranta et al., 2014), area fractal dimension (Gruy, 2011) and mass fractal dimension (Zhou and Franks, 2006). The strength of flocs induced by shear flocculation can be measured by its shear resistance capacity and interaction forces (Zhou et al., 2008; Liang et al., 2015). In addition, the process variables influencing the aggregation and breakage kinetics of flocculation has also been studied (Heath et al., 2009). In summary, the published work to date has put emphasis on the characterization techniques for flocs induced by shear flocculation, rather than the method of selecting the suitable energy input to achieve the most effective improvement in recovery of fine particles and the flotation rate. It should be also noted that energy input on mineral pulp has considerable influence on its separation process and flotation rate in flotation. Energy input (represented by agitation intensity) in pilot-scale mechanical flotation cells has been reported to be a significant factor for the separation effect of platinum ores and gangue minerals (Deglon, 2005). Similarly, energy input in coal flotation (measured by shaft power and flotation time in cell) is also found to have a remarkable effect on the flotation grade and recovery of coal (Gui et al., 2014, 2013). Studies on the flotation kinetics of minerals using different energy input devices, such as oscillating grid flotation cell (Changunda et al., 2008; Massey et al., 2012) and a “Rushton turbine cell” (Newell and Grano, 2006), have demonstrated that flotation rates of galena, pyrite, pentlandite, apatite, hematite and quartz (Safari et al., 2016) are all energy input dependent. However, the influence of energy input on
2. Experimental 2.1. Materials and reagents Hand-picked pure scheelite crystals were crushed to less than 1 mm by a laboratory jaw crusher and a laboratory roll crusher. The crushed products were concentrated on a concentrating table to remove the heavy particles and on a high-intensity magnetic separator several times to discard the magnetic minerals. The non-magnetic products were ground in a porcelain mill and classified into five narrow size fractions: −10, +10–38, +38–55, +55–74 and +74–106 μm. Sodium oleate (C18H33O2Na) used in this study was purchased from Tianjing Kermil Chemical Reagents Development Centre, Tianjin, China. Its molecular structure consists of a hydrocarbon chain and a carboxyl head group, as shown in Fig. 1. It is selected as the flocculant in shear flocculation operation because it is a traditional collector in flotation practice of scheelite. Hydrochloric acid (HCl) and sodium carbonate (Na2CO3) were used as pH regulators. Distilled water was used for all tests. The effect of sodium oleate on the zeta potential of scheelite has been measured (Arnold and Warren, 1974). At the vicinity of pH 10 adjusted by Na2CO3, the zeta potential of scheelite is about −36 mV. Conditioned by 1 × 10−4 mol/L sodium oleate, the zeta potential becomes −42 mV. The decrease in zeta potential is attributed to the adsorption of the sodium oleate on the surface of scheelite as introduced in Section 1. 2.2. EI and its measurement EI was realized using a four-baffle agitation tank with a four-blade impeller, as shown in Fig. 2. The tank was a cylinder with a height of 100 mm and a diameter of 60 mm respectively. There were 4 baffles (80 mm × 5 mm) in the tank. The whole diameter of the impeller was 30 mm and its blade (10 mm × 10 mm) was perpendicular to the axis. The distance between impeller and tank bottom was 30 mm. The impeller was driven by an IKA Eurostar power control-visc6000 agitator, which could control and adjust the rotational speed in a stepless manner in range of 150–6000 rpm. The torque to overcome the fluid resistance was recorded by a torque meter installed inside the agitator. When agitating with a slurry volume of 200 mL, it was found that air entrainment in the liquid was minimal even at 1500 rpm. Mineral pulp was prepared by adding 10 g fine scheelite to 200 mL
Fig. 1. Molecule structure of sodium oleate.
28
Minerals Engineering 112 (2017) 27–35
W. Chen et al.
1000 μm × 1000 μm region of slurry can be videoed by a digital camera, so the slurry images at any recorded time can be analyzed. 2.5. Flotation tests Flotation tests were carried out in a XFG flotation machine with a 100 mL cell at 1500 rpm. For flotation tests on scheelite in different size, pulp was prepared by adding 5 g solid into the cell with 100 mL distilled water. The pH was adjusted to a desired value by adding Na2CO3 or HCl stock solutions. Sodium oleate was added at a desired concentration. The conditioning time was 3 min and the flotation time was 6 min for each flotation test. For shear flocculation flotation tests of fine scheelite (−10 μm), conditioning operation was conducted in the agitation tank (shown in Fig. 2). After conditioning, the slurry was transferred into the flotation cell and mixed for 10 s at 1500 rpm until the flotation began. For each shear flocculation test with certain EI, the total flotation time was 6 min and the concentrate was collected in batches as a flotation time of 20 s, 20 s, 20 s, 20 s, 30 s, 40 s, 50 s, 60 s and 100 s. For fine scheelite (−10 μm) flotation tests, the concentrate and tails were collected and dried in a watch glass. Flotation recovery was calculated based on solid weight distributions among total weight of all products. A first-order model was used to describe the flotation kinetics of shear flocculation pre-treated pulp. The recovery of scheelite at flotation time t to a first approximation is given by the following expression:
Fig. 2. Schematic of the agitation tank used for EI on slurry.
distilled water in the tank. The liquid-solid ratio by weight was kept at 20:1. The temperature was controlled by a water bath, and was set at 25 °C. The agitation started 1 min before pH regulator was added. Agitation time was recorded after sodium oleate was added. Injection syringes were used to take samples for particle size measurements and flotation tests. EI on slurry was calculated by following expression:
W=
T ·n·t 9549V
where W is EI on slurry (J/m3), T is the torque to overcome the fluid resistance (N·m), n is agitating speed (rpm), t is agitating time (s) and V is the slurry volume (m3). EI is a term that describes the agitation intensity on the slurry. In this study, the agitation time and slurry volume were controlled at 15 min and 200 mL.
R (t ) = Rmax (1−e−kt )
where k is the flotation rate constant and Rmax is their flotation recovery at an infinite time (Bulatovic, 2007). A nonlinear least square regression was used to calculate k and Rmax from the best fit of the curve of experimental flotation recoveries versus flotation time using Eq. (2). These flotation rate constants are referred in the text as experimental flotation rate constants.
2.3. Particle size measurements Particle size distribution of the mineral slurry sample was measured using a Malvern Mastersizer 2000, by light scattering. Each pulp sample was sub-sampled twice, with the particle size distribution of each subsample measured in triplicate. The aggregate of six resulting measurements was used as an average particle size distribution curve. The apparent size of fine scheelite with certain EI was evaluated by D50 (the maximum particle diameter below which 50% of the sample volume exists) and D90 (the maximum particle diameter below which 90% of the sample volume exists) of the size distribution, which were directly offered by the Mastersizer. The flocculation behavior of different size fractions in the fine scheelite was described by flocculation degree, which was calculated by following expression:
Rfloc (i) =
αi−βi × 100% αi
(2)
3. Results and discussion 3.1. Basic flotation performance of fine scheelite (−10 μm) The effect of pH on the flotation recovery of fine and coarse scheelite was studied by single mineral flotation tests and the results are shown in Fig. 3a. Flotability of scheelite in different size ranges seldom changed with increasing pH but the recovery of fine scheelite (−10 μm) (about 50%) was much lower than that of coarse scheelite (+10–106 μm) (about 90%) at the same pH range. The results show that adjusting the pH of mineral slurry doesn't help to improve the recovery of fine scheelite. Fig. 3b shows the effect of sodium oleate dosage on flotation behavior of scheelite in different size fractions. Recoveries of all size fractions increased with the increase of sodium oleate dosage. When sodium oleate dosage increased to 2 × 10−4 mol/L, the recoveries of the coarse size fractions (+38–55 μm, +55–74 μm, +74–106 μm) exceeded 90%, while that of fine scheelite (−10 μm) reached only 54%. For fine scheelite, a high sodium oleate dosage up to 8 × 10−4 mol/L was required in order to get a recovery higher than 90%, which was four times of that for coarse fractions. The reason can be its higher surface area than coarse scheelite, as stated in Section 1. The aforementioned flotation results indicate that sodium oleate has a good collecting ability for scheelite, albeit with reduced recovery of fine particles. The collecting ability derives from the chemisorption, i.e., the interaction between –COO– in sodium oleate and Ca2+ on scheelite surface (Rao and Forssberg, 1991; Bo et al., 2015). Affected by the eCH2e groups in sodium oleate on particle surface, scheelite becomes hydrophobic and floatable. Increasing the apparent size by shear flocculation offers a way to improve the recovery of fine scheelite. In this study, pH 10 and sodium oleate dosage of 2 × 10−4 mol/L were used as reagent scheme in shear flocculation to investigate the
(1)
where Rfloc(i) is the flocculation degree of i size fraction, αi is the average content of i size fraction without addition of sodium oleate, βi is the average content of i size fraction after sodium oleate was added and agitated with certain EI. Flocculation degree is a term that evaluates the relative change in content of i size fraction with certain EI. For a specified size fraction, its flocculation degree increases with the decreases of the content of this size fraction. Since the samples in this study are less than 10 μm, i in Eq. (1) is sub-divided in CF (+7.5–10 μm), IF (+2.5–7.5 μm) and FF (−2.5 μm) via their different variation tendency verses EI. 2.4. Observation of flocs The observation of scheelite suspensions was conducted using a transmission light microscope (OLYMPUS-CX31RTSF). 1 mL of suspension was sampled by a syringe and diluted 10 times in a beaker by distilled water. Then dilute solution was dropped on a microslide and was flatted by a coverslip on the objective table. Around 29
Minerals Engineering 112 (2017) 27–35
W. Chen et al.
Fig. 3. Effects of pH and sodium oleate dosage on the flotation behavior of scheelite.
flocculation process and flotation behavior of fine scheelite with different EI.
bimodal, which agreed well with other reports (Koh et al., 1986). Flotation recovery also witnessed a significant increase from 60.84% to 76.83% (Fig. 5b), showing that the growth of flocs remarkably promoted the flotation performance of fine scheelite. Fig. 6 shows the formation of big flocs and corresponding promoted flotation performance. When EI increased from 6.41 kJ/m3 to 9.55 kJ/ m3, the bimodal peak was replaced by a new, high unimodal peak that located in +10 μm area. In this EI range, D50 increased from 8.67 μm to 29.87 μm and D90 increased from 46.02 μm to 70.41 μm. Accordingly, the formation of the big flocs was the result of the FF flocculation as the arrows pointed (Fig. 6a). Flotation recovery met a big increase (from 76.83% to 84.38%) as shown in Fig. 6b. Since large-scale flocculation of IF has occurred in stage 2, there is no doubt that the improved flotation derives from the flocculation of FF and formation of big flocs in the EI range. Fig. 7 shows the stabilization of flocs and promoted flotation when EI continued to increase from 9.55 kJ/m3 to 12.06 kJ/m3. The unimodal peak shifted to left, suggesting a certain degree of decrease in apparent size of the flocs. In the meantime, the flocculation of FF was also observed in Fig. 7a. Taken the decrease and the flocculation together, the apparent size met drops both in D50 (from 29.87 μm to 23.65 μm) and D90 (from 70.41 μm to 53.48 μm), however, flotation recovery witnessed a slight increase from 84.38% to 86.50%, as shown in Fig. 7b. The results demonstrate that the flocculation of the FF has outweighed the negative effect of flocs breakage thus increasing the flotation recovery in the stabilization stage. Fig. 8 shows the breakage of flocs and decreased flotation when excessive EI is applied. The dominant peak in Fig. 8a lowered and created a new peak located on the left, indicating that big flocs had broken down into discrete particles or small flocs. The variation of D90 (from 53.48 μm to 54.84 μm) and D50 (from 23.65 μm to 20.48 μm) also
3.2. Changes in size distribution and flotation performance of fine scheelite versus EI The size distribution and flotation behavior of fine scheelite with different EI were studied and the results are shown in Figs. 4–8. According to the formation, growth, stabilization and breakage of flocs with different EI, the whole flocculation process as well as corresponding flotation performance was discussed in stages. Fig. 4 shows the transformations in size distribution and flotation behavior of fine scheelite in the beginning of floc formation stage. With increasing EI, unimodal distribution in Fig. 4a started to fissure. The high peak of fine scheelite lowered and formed a new peak that located in the +10 μm area. Apparent size data showed that D90 increased from 10.29 μm to 23.22 μm and D50 varied between 4.76 μm and 4.26 μm when EI increased from 0.19 kJ/m3 to 2.64 kJ/m3. The results suggested a slight increase in the apparent size of the fine scheelite (−10 μm). The increase in apparent size resulted in a slightly promoted flotation, with flotation recovery increasing from 54.28% to 60.84% as shown in Fig. 4b. Therefore, in this EI range, the dispersed fine scheelite began to aggregate into flocs and lead to a slightly promoted flotation. Fig. 5 exhibits the growth of the flocs and the remarkably improved flotation behavior of the scheelite fines. The dominant peak decreased further as a new peak generated and the whole peak shifted to right when EI increased more, as shown in Fig. 5a. In this stage, D50 increased from 4.26 μm to 8.67 μm and D90 increased from 23.22 μm to 46.02 μm. The results indicated an obvious growth in apparent size of the flocs, which were mainly derived from the flocculation of IF and FF. The unimodal size distribution of fine scheelite totally changed to
Fig. 4. Stage 1: Beginning of floc formation and slightly promoted flotation.
30
Minerals Engineering 112 (2017) 27–35
W. Chen et al.
Fig. 5. Stage 2: Growth of flocs and remarkably improved flotation.
confirmed the breakage of flocs. As a result, the flotation recovery met a decrease from 86.50% to 77.28%, as shown in Fig. 8b. The size distribution of fine scheelite changed from unimodal to bimodal and then unimodal, with increasing EI and increasing apparent size of flocs. However, the phenomena of bimodal size distributions indicated that finer fractions in scheelite fines behaved differently in the flocculation process. For example, peak of the FF was the last to disappear during the formation stage of flocs but the first to appear during the breakage stage of flocs. The different behaviors of finer fractions in flocculation process with the same EI are very important for selecting the suitable EI to realize the flocculation of all finer fractions in the maximum extent.
hydrophobicity was induced by sodium oleate and EI was offered by the agitation tank. The discrete fine particles could overcome the energy barrier when applied EI exceeds 2.04 kJ/m3, and aggregated into flocs. The flocs experienced small, branchy and globule-like as EI increased. On the other hand, the different floc shape formed with different EI illustrates its formation, growth, stabilization and breakage qualitatively. As a result, the flotation performance of fine scheelite is promoted to different degree in different flocculation stage. Therefore, an optimal value of EI to achieve the most effective flocculation should be determined by quantitative analysis on flocculation behavior of each fractions less than 10 μm. 3.4. Flotation rate and flocculation degree as a function of EI
3.3. Changes in floc shape in different flocculation stage Flotation rate was calculated by fitting the cumulative recovery curves. The fitting results are shown in Appendix A. The flocculation degree of finer size fractions in the scheelite fines (CF, IF and FF) was also calculated using the size distribution data in Section 3.2. The flotation rate constant and flocculation degree as a function of EI was presented in Fig. 10. The flotation rate constant was observed to increase as EI increased with a maximum at 12.06 kJ/m3. The variation trend was in accord with the flotation performance in Section 3.2. For flocculation stage 1–3, or EI less than 9.55 kJ/m3, rate constant was approximately proportional to the 3.07 power of EI, as fitted by an allometric model in Appendix B. For the rest stages, the rate constant decreased slowly with excessive EI. The results indicate that the flotation rate constant of flocculated scheelite pulp is EI dependent and can be selected as an evaluation standard in determining the suitable EI in shear flocculation. The flocculation degree of CF witnessed a stepped increase with increasing EI. It nearly kept a constant of 20% when EI was less than
Fig. 9 presented a summary of the observations for flocs with different EI, and, at different stages in Figs. 4–8. It can be seen in Fig. 9a that in the absence of sodium oleate, the fine scheelite was well-dispersed in slurry. With the addition of sodium oleate and increasing EI, the particles tended to agglomerate, forming flocs (Fig. 9b) and possessing more branches (Fig. 9c–e). When EI exceeded the critical value for floc breakage, the flocs with branches fell apart and formed globulelike aggregations. At the same time, the branches were re-dispersed and became discrete particles, as shown in Fig. 9f. In fact, the changes in flocs shape reflected the effect of EI on the movement and aggregation behaviors of scheelite particles in the formation, growth, stabilization and breakage of flocs. The mechanisms for the movement and aggregation of fine scheelite have been widely studied. Hydrophobic particles can aggregate if the energy barrier between the particles was overcame by enough kinetic produced by agitation (Bilgen and Wills, 1991). In this research,
Fig. 6. Stage 3: Formation of big flocs and promoted flotation.
31
Minerals Engineering 112 (2017) 27–35
W. Chen et al.
Fig. 7. Stage 4: Stabilization of flocs and slightly promoted flotation.
flocculation, CF in the fine scheelite (−10 μm) agglomerates and generates floc core; with increasing EI, IF and FF adhere to the floc core, acting as branches and coatings, respectively; with suitable EI, all fractions gather into a big floc; with excessive EI, FF and IF are peeled off one and re-dispersed in the flocculated suspensions. On the basis of the above flocculation process, suitable EI to achieve the most effective flocculation in shear flocculation can be selected directly by monitoring the changing process of size distribution (represented by flocculation degree) and by calculating the flotation rate constant. In shear flocculation-flotation of fine scheelite, flotation recovery as well as flotation rate should be taken into consideration. Therefore, a compromise between the flocculation degree and flotation rate can be used to determine the EI range. In this mixing cell and flotation cell, the suitable EI lies in 9.55–12.06 kJ/m3, with flotation rate constant ranging from 0.0369 s−1 to 0.0418 s−1.
8.29 kJ/m3, implying that the content of CF in the flocculated pulp didn't change in this EI range. When EI increased further, it increased to around 40% and hardly decreased with excessive EI, suggesting that CF existed stably in flocs with high EI. For IF, its flocculation degree showed a trend of rise, then fall, with a turning point at 9.55 kJ/m3. The trend was consistent with the aggregation and breakage behaviors of flocs with increasing EI as described in Sections 3.2 and 3.3. For FF, its flocculation degree first dropped, and increased, and then decreased with increasing EI. When EI was less than 4.52 kJ/m3, the flocculation degree of FF was negative, indicating a relative increase of FF content in the flocculated pulp due to the flocculation of CF and IF in this EI range. The flocculation degree results provide a good understanding of the roles played by CF, IF and FF during the flocculation process. In the formation stage of flocs, the sequence of the three fractions entering the flocs is ranked as CF, IF, FF. FF starts to aggregate only when EI has exceeded 4.52 kJ/m3, in which range IF and CF have agglomerated to a certain level. Therefore, FF gets into flocs by its adhesion on the already formed flocs by CF and IF. The conclusion can also be supported by the phenomenon of flocs that possess many branches in the growth stage. In the breakage of flocs, the stability of the three fractions in the flocs with increasing EI is ranked as CF > IF > FF. FF has the least stability in flocs and is prior to get re-dispersed in pulp with excessive EI. In summary, CF acts as the core of flocs, and FF forms the branches of flocs, and IF plays both roles in the flocculation process of fine scheelite (−10 μm). Combining the formation and breakage of flocs, the flocculation process of fine scheelite with increasing EI can be described by an aggregation and growth model, as shown in Fig. 11. In the model, CF, IF and FF are represented by balls in different diameters, respectively. The flocculation process with different EI was simulated by the aggregation and dispersion of these balls. With basic EI just able to produce shear
4. Conclusions Given the aforementioned results, we arrived at the following primary conclusions: (1) Shear flocculation induced by suitable energy input can significantly improve the flotability of fine scheelite (−10 μm) using sodium oleate as collector. (2) Energy input influences the apparent size and flotation rate of fine scheelite. Apparent size of fine scheelite and its flotation rate increase as energy input increases in limited range; the size distribution of fine scheelite transforms from unimodal to bimodal as energy input changes. (3) Low energy input tends to induce large flocs with more branches but weak resistance to breakage; high energy input tends to form
Fig. 8. Stage 5: Breakage of flocs and decreased flotation.
32
Minerals Engineering 112 (2017) 27–35
W. Chen et al.
Fig. 9. Floc shape with different EI.
a. Well-dispersed fine scheelite
c. Growth of flocs (6.41 kJ/m3)
e. Breakage of flocs with branches(12.06 kJ/m3)
b. Formation of flocs (2.04 kJ/m3)
d. Flocs with branches (9.55 kJ/m3)
f. Globule-like flocs (17.09 kJ/m3) (4) The flocculation process of fine scheelite can be described by an aggregation and growth model. The process starts with aggregation of coarse particles (+7.5–10 μm) as core of floc, and develops by involving intermediate particles (+2.5–7.5 μm), and ends with rolling the fine particles (−2.5 μm) onto surface of floc with increasing energy input.
Acknowledgement The authors acknowledge the support of the Major State Basic Research Development Program of China (973program) (2014CB643402). Special thanks to Mr. David Beattie for his valuable discussions and the final language adjustment. Fig. 10. Flotation rate constant and flocculation degree (Rfloc) of fine scheelite versus EI.
globule-like flocs with high stability but inevitable re-dispersion. Appendix A. Rate constant calculation The flotation rate constant was obtained from nonlinear regression based on the first-order model equation: R(t) = Rmax(1 − e−kt) (see Section 2.5). The results are shown in Table A1.
33
Minerals Engineering 112 (2017) 27–35
W. Chen et al.
Fig. 11. Aggregation and growth model: formation and breakage of scheelite floc. Table A1 Fitting results of flotation recovery curves with different EI. EI (kJ/ m3)
1.38 2.64 4.52 6.41 8.29 9.55 12.06 14.57 17.09
R (t = 6 min)
54.28 60.84 70.33 76.83 81.01 84.38 86.5 83.63 77.09
Fitting results Rmax (%)
k (s−1)
Adj. Rsquare
Reduced ChiSqr
56.69 58.54 67.98 74.44 79.12 81.24 83.86 78.60 71.95
0.0137 0.0160 0.0171 0.0210 0.0287 0.0369 0.0418 0.0399 0.0348
0.9971 0.9937 0.9864 0.9916 0.9958 0.9896 0.9959 0.9872 0.9720
1.0358 2.3680 6.8030 4.9988 2.7884 7.0390 2.9069 8.0506 14.8158
Table B1 Fitting results of flotation rate as a function of EI. Parameter
Value
Standard error
Adj. R-square
Reduced Chi-Sqr
a b c
0.01463 2.16E−05 3.07139
6.19E−04 1.67E−05 0.3403
0.9915
6.77E−07
Appendix B. Fitting result of flotation rate constant as s function of EI An allometric model equation: k = a + b∗ρc, where k is the flotation rate, ρ is EI and a, b, c are parameters, was used to describe the relationship between EI and flotation rate constant. The fitting result is shown in Table B1.
Gui, X., Cheng, G., Liu, J., Cao, Y., Li, S., He, Q., 2013. Effects of energy consumption on the separation performance of fine coal flotation. Fuel Process. Technol. 115, 192–200. Gui, X., Liu, J., Cao, Y., Cheng, G., Li, S., Wu, L., 2014. Flotation process design based on energy input and distribution. Fuel Process. Technol. 120, 61–70. Heath, A.R., 2006. Polymer flocculation of calcite: population balance model. FLUID Mech. Transp. Phenom. 52, 1641–1653. Heath, A.R., Bahri, P.A., Farrow, P.D.F., Farrow, J.B., 2009. Polymer flocculation of calcite: experimental results from turbulent pipe flow. IFAC Proc. 7, 405–410. Warren, J.L., 1975. Shear-flocculation of ultrafine scheelite in sodium oleate solutions. J. Colloid Interface Sci. 50, 307–318. Koh, P.T.L., Andrews, J.R.G., Uhlherr, P.H.T., 1987. Modelling shear-flocculation by population balances. Chem. Eng. Sci. 42, 353–362. Koh, P.T.L., Andrews, J.R.G., Uhlherr, P.H.T., 1986. Floc-size distribution of scheelite treated by shear-flocculation. Int. J. Miner. Process. 17, 45–65. Koh, P.T.L., Lwin, T., Albrecht, D., 1989. Analysis of weight frequency particle size distributions: with special application to bimodal floc size distributions in shear-flocculation. Powder Technol. 59, 87–95. Koh, P.T.L., Warren, L., 1980. A pilot plant test of the shear-flocculation of ultrafine scheelite. In: Eighth Australian Chemical Engineering Conference at Melbourne, Australia, pp. 2–6. Koivuranta, E., Keskitalo, J., Haapala, A., Stoor, T., Sarén, M., Niinimäki, J., 2013. Optical monitoring of activated sludge flocs in bulking and non-bulking conditions. Environ. Technol. 34, 679–686. Koivuranta, E., Keskitalo, J., Stoor, T., Hattuniemi, J., Sarén, M., Niinimäki, J., 2014. A comparison between floc morphology and the effluent clarity at a full-scale activated sludge plant using optical monitoring. Environ. Technol. 35, 1605–1610. Liang, L., Peng, Y., Tan, J., Xie, G., 2015. A review of the modern characterization
References Akdemir, U., 1997. Shear flocculation of fine hematite particles and correlation between flocculation, flotation and contact angle. Powder Technol. 94, 1–4. Akdemir, Ü., Hiçyilmaz, C., 1998. Selective shear flocculation of chromite from chromiteserpentine mixtures. Colloids Surf. A: Physicochem. Eng. Asp. 132, 75–79. Akdemir, Ü., Hiçyilmaz, C., 1996. Shear flocculation of chromite fines in sodium oleate solutions. Colloids Surf. A: Physicochem. Eng. Asp. 110, 87–93. Arnold, R., Warren, L.J., 1974. Electrokinetic properties of scheelite. J. Colloid Interface Sci. 47, 134–144. Atademir, M.R., Kitchener, J.A., Shergold, H.L., 1981. The surface chemistry and flotation of scheelite, II. Flotation “collectors”. Int. J. Miner. Process. 8, 9–16. Bakker, M.G., Spears, D.R., Murphy, D.D., Turner, G.L., 2002. Whatever happened to shear flocculation. Fluid – Part. Sep. J. 14, 155–168. Bo, F., Xianping, L., Jinqing, W., Pengcheng, W., 2015. The flotation separation of scheelite from calcite using acidified sodium silicate as depressant. Miner. Eng. 80, 45–49. Bulatovic, S.M., 2007. Handbook of Flotation Reagents. Changunda, K., Harris, M., Deglon, D.A., 2008. Investigating the effect of energy input on flotation kinetics in an oscillating grid flotation cell. Miner. Eng. 21, 924–929. Deglon, D.A., 2005. The effect of agitation on the flotation of platinum ores. Miner. Eng. 18, 839–844. Forbes, E., 2011. Shear, selective and temperature responsive flocculation: a comparison of fine particle flotation techniques. Int. J. Miner. Process. 99, 1–10. Gaudin, A.M., 1995. Flotation. McGraw-hill Book Company. Gruy, F., 2011. Population balance for aggregation coupled with morphology changes. Colloids Surf. A: Physicochem. Eng. Asp. 374, 69–76.
34
Minerals Engineering 112 (2017) 27–35
W. Chen et al. techniques for flocs in mineral processing. Miner. Eng. Massey, W.T., Harris, M.C., Deglon, D.A., 2012. The effect of energy input on the flotation of quartz in an oscillating grid flotation cell. Miner. Eng. 36–38, 145–151. Mietta, F., Chassagne, C., Winterwerp, J.C., 2009. Shear-induced flocculation of a suspension of kaolinite as function of pH and salt concentration. J. Colloid Interface Sci. 336, 134–141. Newell, R., Grano, S., 2006. Hydrodynamics and scale up in Rushton turbine flotation cells: part 1 – cell hydrodynamics. Int. J. Miner. Process. 81, 65–78. Ozkan, A., Ucbeyiay, H., Aydogan, S., 2006. Shear flocculation of celestite with anionic surfactants and effects of some inorganic dispersants. Colloids Surf. A: Physicochem. Eng. Asp. 281, 92–98. Pascoe, R.D., Doherty, E., 1997. Shear flocculation and flotation of hematite using sodium oleate. Int. J. Miner. Process. 51, 269–282. Pyke, B., Fornasiero, D., Ralston, J., 2003. Bubble particle heterocoagulation under turbulent conditions. J. Colloid Interface Sci. 265, 141–151. Ralston, J., 1992. The influence of particle size and contact angle in flotation. Dev. Miner. Process. 12, 203–224 (Chapter 6). Rao, K.H., Forssberg, K.S.E., 1991. Mechanism of oleate interaction on salt-type minerals. Part III. Adsorption, zeta potential and diffuse reflectance FT-IR studies of scheelite in the presence of sodium oleate. Colloids Surf. 54, 161–187. Bilgen, S., Wills, B.A., 1991. Shear flocculation – a review. Miner. Eng. 4, 483–487. Safari, M., Harris, M., Deglon, D., Filho, L.L., Testa, F., 2016. The effect of energy input on flotation kinetics. Int. J. Miner. Process. 1–8. Sivamohan, R., 1990. The problem of recovering very fine particles in mineral
processing—a review. Int. J. Miner. Process. 28, 247–288. Song, S., Lopez-Valdivieso, A., Reyes-Bahena, J.L., Lara-Valenzuela, C., 2001. Floc flotation of galena and sphalerite fines. Miner. Eng. 14, 87–98. Subrahmanyam, T.V., Forssberg, K.S.E., 1990. Fine particles processing: shear-flocculation and carrier flotation – a review. Int. J. Miner. Process. 30, 265–286. Ucbeyiay, H., Ozkan, A., 2014. Two-stage shear flocculation for enrichment of fine boron ore containing colemanite. Sep. Purif. Technol. 132, 302–308. Ucbeyiay Sahinkaya, H., Ozkan, A., 2011. Investigation of shear flocculation behaviors of colemanite with some anionic surfactants and inorganic salts. Sep. Purif. Technol. 80, 131–139. Vlieghe, M., 2014. In situ characterization of floc morphology by image analysis in a turbulent Taylor-Couette reactor. Am. Inst. Chem. Eng. 60, 2389–2403. Warren, L.J., 1992. Shear-flocculation. Develop. Miner. Process. 309–329. Wills, B.A., Napier-Munn, T., 2005. 12 – Froth flotation BT – Wills’ mineral processing technology. In: Wills’ Mineral Processing Technology, seventh ed. ButterworthHeinemann, Oxford, pp. 267–352. Yin, W.Z., Yang, X.S., Zhou, D.P., Li, Y.J., Lü, Z.F., 2011. Shear hydrophobic flocculation and flotation of ultrafine Anshan hematite using sodium oleate. Trans. Nonferrous Met. Soc. China 21, 652–664. Zhou, Y., Franks, G.V., 2006. Flocculation mechanism induced by cationic polymers investigated by light scattering. Langmuir 22, 6775–6786. Zhou, Y., Gan, Y., Wanless, E.J., Jameson, G.J., Franks, G.V., 2008. Interaction forces between silica surfaces in aqueous solutions of cationic polymeric flocculants: effect of polymer charge. Langmuir 10920–10928.
35