An investigation of the flotation minimum in the oleate flotation of hematite under alkaline conditions

Minerals Engineering 113 (2017) 71–82

Contents lists available at ScienceDirect

Minerals Engineering journal homepage: www.elsevier.com/locate/mineng

An investigation of the flotation minimum in the oleate flotation of hematite under alkaline conditions

MARK

Keith Quast Future Industries Institute, University of South Australia, Mawson Lakes, South Australia 5095, Australia

A R T I C L E I N F O

A B S T R A C T

Keywords: Oleate flotation Hematite Alkaline pH

Over the years there have been a number of instances describing a minimum in the flotation recovery of hematite in the alkaline pH region when oleate is used as the collector. These cases usually coincide with the use of a commercial oleate sample containing other fatty acids and/or salts as the collector. It is the purpose of this paper to provide experimental data involving the effects of pH on the flotation of a natural hematite sample using two collectors containing oleate and compare these data to changes in surface tension and bubble characteristics of aqueous oleate systems also as a function of pH. The main aim of this paper is to highlight and address this difference in hematite flotation behaviour with oleate solution characteristics and develop a better understanding of these interactions. It is proposed that this flotation minimum is associated with a reduction in both contact angle and bubble-particle flotation rate constant under these conditions, even though the surfactant solution chemistries would suggest that flotation should still be strong in this region.

1. Introduction The author has published the results of a number of studies on the oleate flotation of hematite as a function of pH (Quast, 1999, 2012a, 2016a; Quast and Quast, 2010; Joseph-Soly et al., 2015). Conditioning times and zeta potentials have also been used to investigate the hematite-oleate system (Quast, 2015, 2016b). These findings have shown that conditioning at acidic pH values involves the interactions of waterinsoluble droplets or colloids with the mineral surface, whereas under alkaline pH conditions, soluble oleate species are adsorbed. One of the commercial reagents used in this study was very effective in floating hematite after short conditioning times, but very susceptible to the presence of fines (Quast, 2015). For this reason the optimum conditioning time determined previously was used in this study to mitigate the effect of conditioning time on hematite flotation. Suggested mechanisms include physical adsorption or precipitation of oleate colloids/micelles (Quast, 2016c) and chemisorption of oleate forming a ferric oleate complex as demonstrated by Joseph-Soly et al. (2015). In this study, the oleate flotation as a function of pH for a sample of naturally occurring hematite ore from the Iron Prince deposit in the Middleback Ranges of South Australia using two surfactants containing oleate will be reported. These data, together with information derived from other literature sources, will be used as the basis for the discussions on the hematite-oleate-aqueous system. This study will therefore address the conundrum of why hematite flotation using these two collectors is depressed under pH conditions where the surfactant

E-mail address: [email protected]. http://dx.doi.org/10.1016/j.mineng.2017.08.002 Received 8 February 2017; Received in revised form 13 June 2017; Accepted 1 August 2017 0892-6875/ © 2017 Elsevier Ltd. All rights reserved.

activity of aqueous oleate is at a maximum as indicated by a minimum in surface tension/maximum in surface pressure. 2. Literature review 2.1. Adsorption studies A maximum in the oleate adsorption density or flotation of hematite at its isoelectric point, usually close to neutral pH, has been observed by a number of researchers (e.g. Peck et al. 1966, Gutierrez and Iskra, 1977, Ofor, 1995). For fine mineral hematite in contact with sodium oleate, the contact angle and aggregation efficiency were maxima at the isoelectric point of hematite as reported by Song and Lu, (1994). Quast (1999) showed how the maximum in hematite flotation with oleate usually occurred under neutral conditions, where the isoelectric points of hematites are normally reported (see data of Kosmulski (2009) and Quast (2016a)). 2.2. Surface tension (and surface pressure) studies It is not the purpose of this paper to discuss the phenomenon of the surface tension of aqueous surfactant solutions, as this has been reported in standard texts (e.g. Davies and Rideal, 1961; Gutierrez and Iskra, 1977; Ofor, 1995). The measurement of surface tension of aqueous surfactant systems using ring tensiometry has been criticised by various workers (e.g. Padday, 1969 and Boucher et al. 1967) although

Minerals Engineering 113 (2017) 71–82

K. Quast

Mankovich (1968) claimed extraordinary accuracy by taking “wellknown precautions” into account. Lukenheimer and Wantke (1981) have suggested ways of eliminating errors involved in testing under conventional conditions, thus allowing an accurate measurement of the surface tension of surfactant solutions, and these were incorporated in the experimental procedure. Over the years many researchers have recorded minima in the surface tension of oleate solutions in alkaline solutions. These include Bottazzi, 1913; Mikumo, 1927; Mullin, 1928; Powney, 1935; Cupples, 1935; Mankovich, 1953; Kulkarni, 1976; Williams, 1980; Wakamatsu et al. 1980; Pugh and Stenius, 1985; Pugh, 1986; Ananthapadmanabhan and Somasundaran, 1988; Muratoglu et al. 2000; Hernainz and Caro, 2001; Yu et al. 2015 and Atrafi and Pawlik, 2016. Kulkarni’s data were analysed by Beunen et al. (1978) who found that the observed minimum in surface tension could be explained without resorting to the use of species at very low concentrations (acid soap complex) postulated without experimental justification. They assumed that at low pH, oleic acid will be mostly undissolved and that this precipitate acted as a reservoir of surfactant molecules which entered the solution in the dissociated form as the pH was increased. This increase in solution concentration resulted in increased adsorption at the interface with a consequent lowering of the surface tension. At the so-called “solubility edge”, the surfactant became completely soluble and the solution concentration of the surfactant became constant, hence there was no further tendency for the surface tension to decrease as the pH was increased. Indeed, the conversion of neutral to charged surfactant species caused the monolayer to charge up. Thus, for pH values higher than the solubility edge, the increasing electrostatic repulsion of oleate ions from the interface caused the surface tension to increase. A minimum in surface tension exists, therefore, at the solubility edge. Ananthapadmanabhan et al. (1978) reported a minimum in the surface tension of potassium oleate at pH 8 for measurements conducted from pH 4–12. This was ascribed to the maximum concentration of the acid soap at this pH. Ananthapadmanabhan and Somasundaran (1981) reported that the surface tension minimum for potassium oleate solutions shifted to higher values of pH as the oleate concentration was raised. Also, de Castro and Borrego (1995) found that the surface tension minimum for aqueous solutions of sodium oleate shifted from pH 7 at low oleate additions to almost pH 10 at higher oleate conditions. This is in keeping with the position of the solubility edge of the oleate as shown in diagrams of its aqueous equilibria (see Quast, 2016a, 2016a, 2016c). Theander and Pugh (2001) reported values of the surface tension of oleate solutions for concentrations between 10 -6 molar and 10 -5 molar. They reported two minima, between pH 8 and 9, and at about pH 11. The values of the surface tension also decreased rapidly beyond pH 12. Recent data by Atrafi and Pawlik (2016) showed that the pH corresponding to the minimum in the surface tension of aqueous solutions of sodium oleate increased as the addition of oleate increased from 10 -5 molar to 10 -2 molar. The pH corresponding to the precipitation of oleic acid also increased from 7.5 to 10.3 as the oleate addition was increased from 10 -5 molar to 10 -2 molar. According to Pugh and Stenius (1985), the minimum in surface tension corresponds to the maximum concentration of an acid-soap species. According to the speciation diagrams reported by Pugh and Stenius (1985), the onset of precipitation also occurs at or near the pH of minimum surface tension. The data reported by Atrafi and Pawlik (2016) showed that the pH values corresponding to the precipitation of the oleic acid occurred at lower values than the observed minimum in surface tension at each oleate addition. The values of surface tension at pH 4 for all the oleate additions used were close to 40 mN/m and may correspond to an insoluble film of oleic acid at the air-water interface, since pH 4 is 1 pH unit below the traditional value of the pKa for oleic acid. The fact that it is this relatively low value at pH 4 could also mean the presence of some surface-active species in the solution. Surface tension values at low values of pH are lower than at high pH, supporting the view that the oleic acid form is

quite surface active compared to the soluble oleate anion. It also suggests that the adsorption of anions at the air-water interface could be inhibited by lateral electrostatic repulsive forces, whereas electrostatic repulsion forces between adsorbed oleic acid molecules could be lower although Laskowski (1988) measured an isoelectric point for oleic acid colloids around 2.5. One of the few articles to actually report surface pressure of oleate as a function of pH was written by Kulkarni and Somasundaran (1975). They examined the effects of ionic strength and temperature on final surface pressure and showed that the maximum shifted to higher values of pH with increasing ionic strength and increasing temperature. Surface pressure maxima were in the pH range 7–9, decreasing rapidly at pH 10. 2.3. Bubble size and frothing Iwasaki et al. (1960) and Cooke et al. (1960) measured the frothing characteristics of oleic acid as a function of temperature and pH. The concentration of oleic acid used was 10 -4 molar, and the froth height was determined by a manual agitation method. Maximum frothing occurred at pH 11, with little frothing noted at pH values below 6. The decrease in frothing at lower pH values was attributed to the precipitation of the free acid, whereas the decrease at higher pH values was thought to be due to the “salting out” of soaps from solution. The froth height at pH 10.7 with additions of sodium chloride showed increased turbidity and reduced froth height with increasing additions of salt. Titration of a sample of BDH “technical grade” sodium oleate after filtration to remove any insoluble material was conducted by Agars (1976). The titration curve suggested a pKa value of approximately 5.5. This same reagent was used to float manganese ions using a 2 m froth column at pH between its initial value of 10.9 down to 3.5. Observations included a decrease in frothiness as the pH was lowered, with a limit where there was no froth at pH 8. The addition of oleate corresponded to 3.64 mM. High purity nitrogen was used to generate the froth. Nunez and Yalkowsky (1997) measured the foaming activity of potassium oleate as a function of pH and compared the rapid change in foam height to the reported values of pKa. They found that the rapid increase in foam height occurred at pH ∼ 5, and compared this with literature pKa values of 4.8 and 5.0 as reported by Somasundaran and Ananthapadmanabhan (1978). Nunez and Yalkowsky (1997) used this method of determining the onset of foaming activity to estimate the value of the pKa for a number of surfactants. Atrafi et al. (2012) measured the frothing behaviour of oleic acid as a function of pH. At pH 7.1, no foaming occurred for the addition of 20 mg/L sodium oleate. As the pH was raised, the foaming of the oleic acid solutions was significantly enhanced, although a small decrease was observed between pH 9 and 10. The authors reported very high foam growth rates and volumes at pH values below the pH of oleic acid precipitation (pH 8.4), which was interpreted that small to moderate amounts of precipitates played a synergistic role in the foam flotation process. More recent data on the foaming characteristics of oleate have been reported by Shu et al. (2014). They reported a maximum in foam volume at pH 10.02 for a solution containing 8 g/L oleic acid. The foam volume was still high at pH 12.19. These authors postulated that the pKa of oleic acid was 9.89, much higher than that measured for saturated fatty acids by the author (Quast, 2016d). Atrafi and Pawlik (2016) measured the bubble size distributions as a function of oleate concentration and its associated pH. At pH 10, bubble size was small (∼1.5 mm) and independent of pH or oleate concentration. At lower oleate additions and lower pH values, the average bubble sizes were larger (∼3 mm) corresponding to the dominance of colloidal species resulting from the lower solubility of the oleate. Since the concentration of insoluble precipitate increases at lower pH at the expense of other species, the authors concluded that oleic acid 72

Minerals Engineering 113 (2017) 71–82

K. Quast

precipitates are inactive in terms of their ability to reduce bubble sizes. Since pine oil was used as the frother in the flotation testing, its frothing characteristics as a function of pH must be taken into consideration. According to Veras et al. (2014), average bubble sizes for Hercules Yarmor F pine oil (similar to the frother used in the current test work) increased from 2.4 mm to 2.65 mm as the pH was raised from 2 to 10. Mean bubble size in the absence of frother was 2.7 mm. The critical coalescence concentration for pine oil was reported as 30 ppm, similar to the concentration used in the current test work.

Table 1 Chemical constitutions (in%) of the surfactants used in this study.

2.4. Contact angles The contact angle for hematite across a wide pH range in the absence of surfactants is often virtually constant at about 45–55 ° with a value of 47° reported by the author (Quast, 2012a) and Abaka-Wood et al. (2016, 2017). Other values for contact angles on unactivated hematite include 30–70° (Iveson et al. 2000, 2004), 54–79.5° (Mao et al. 2013), 46.5° (Shang et al. 2008) and approximately 50° for pH values < 6, (Shrimali et al. 2016). For pH values > 8.5, Shrimali et al. (2016) measured a zero contact angle on the hematite (0 0 1) surface. Hematite contacted with sodium oleate increased the contact angle to 88° (Iwasaki et al., 1960; Cooke et al., 1960; Oko and Salman, 1966 (for pH 10); Yap et al., 1981 (for oleate additions corresponding to 10−5 molar for pH values between 7 and 9), Rath et al., 2014 and AbakaWood et al., 2016, 2017) showing how sodium oleate increased the modest inherent surface hydrophobicity of hematite under the conditions of measurement of the contact angle. Yap et al. (1981) reported how the contact angle of sodium oleate at 10 -6 molar on hematite decreased dramatically as the pH was raised above 8. At higher concentrations of sodium oleate the pH corresponding to the decrease in contact angle occurred at higher values. Similar trends were noted for the oleate adsorption density on the hematite. High contact angles (74–85 °) have also been measured on mineral hematite (Lourenco et al. 2015), but these high values usually coincide with high values of surface roughness (Maeda et al. 2005). Depending on particle size, Drzymala et al. (1992) showed significant collectorless hematite flotation using 0.5 g samples in a Hallimond Tube at the natural pH of the suspension. According to Gontigo et al. (2007) flotation should be possible at the particle sizes in the ore for contact angles > 40°. Iskra (1997) compared the adsorption of oleate and contact angle on silicon carbide as functions of pH. He reported that as the contact angle decreased from pH 2–10, oleate adsorption decreased in a similar mode, suggesting that contact angles are a good measure of the degree of surfactant adsorption on a surface.

Cn designation

SC Oleic acid

C10 C12 C14 C14.1 C15 C16 C16.1 C17 C17.1 C18 C18.1 C18.2 C18.3 Iodine Value Colour/form

Trace 0.9 5.1 3.0 0.8 4.9 8.0 0.9 1.6 1.8 66.9 5.8 0.3 Approx. 90 Pale yellow liquid

Ajax Sodium stearate

3.2 4.5 2.2 25.6 4.5 1.6 0.4 19.8 36.8 1.3 42.2 Brown granules

by electrostatic attraction (Toikka et al. 1996). The true condition of zero surface charge is represented by the zero point of charge (6.7), which was measured using the modified titration technique of Mular and Roberts (1966). 3.2. Samples of oleate Two oleate samples were used in this study. One sample was oleic acid distributed by Steetley Chemicals (SC) and the other sample was labelled “sodium stearate” marketed by Ajax Chemicals. The chemical constitutions of these reagents are shown in Table 1 where the Cn designation shows the number of carbon atoms and the number of double bonds e.g. C18.1 represents oleic acid. The iodine value is a measure of the degree of unsaturation of the reagent. The iodine value corresponding to the presence of a single double bond per molecule is 90, and for two double bonds per molecule it is 180. 4. Experimental procedure 4.1. Flotation testing 4.1.1. Multistage additions at constant pH The ground ore was washed into a 3 L capacity flotation cell fitted to a Galigher Agitair laboratory flotation machine, model LA 500. The surface area of the rod milled product, as measured using a Fisher Subsieve sizer Model 95 was 1550 cm2/g. The pH of the pulp was adjusted to the desired value using dilute solutions of either hydrochloric acid or sodium hydroxide. The pulp was agitated for 30 s, allowed to settle for 2 min, and the supernatant liquid containing minus 2 µm particles siphoned off. The benefits of this “partial deslime” were twofold:

3. Materials examined 3.1. Hematite

• Most of the ultrafines were removed, since they consume large amounts of reagent, and • The pulp density was raised from 21% solids to 42% solids by

The hematite sample was from Iron Prince, a deposit in the Middleback Range area of South Australia. The ore sample used contained 69% Fe corresponding to 98.5% Fe2O3. The main impurity in this sample was fine silica. This is the same material that was used in the study involving the flotation of hematite as a function of pH using C6C18 saturated fatty acids (Quast, 2006). This sample was crushed to all passing 1.7 mm and riffled into 500 g charges. Prior to flotation testing, these charges were ground in a Linatex lined rod mill at 50% solids by weight for 10 min using 25 mm stainless steel rods. The ore, after grinding, was all finer than 150 µm. Demineralised water with a conductivity of approximately 0.5 mS/m was used throughout all the test work. The surface chemistry of this hematite has been established and reported earlier (Quast, 2006, 2012b). The isoelectric point of this hematite was determined as 2.7 and the zero point of charge was 6.7. The low value of the isoelectric point was due to the presence of fine silica in the ore. Any liberated silica will adsorb onto hematite surfaces

weight for the conditioning stage.

An initial reagent addition was made, and the thickened pulp conditioned for 15 min at 1000 rpm with frequent pH checks and adjustment. At the end of this time, the pH was checked, adjusted if necessary and one drop of Yarmour 302 pine oil frother (corresponding to 50 g/t), a product of Hercules Inc., was added. (Fuerstenau et al. (1970) reported that pine oil was the preferred frother when using oleate to float hematite). Conditioning continued for a further 5 min. Demineralised water at the pH of the flotation test was added to bring the pulp level to within 20 mm of the overflow weir of the flotation cell. This returned the pulp density to approximately 21% solids by weight. The pulp was aerated at 3 L/min for 3 min, during which time the 73

Minerals Engineering 113 (2017) 71–82

K. Quast

100

50 g/t 100 g/t 150 g/t 200 g/t

Recovery (%)

80

60

40

20

0 0

2

4

6

8

10

12

pH Fig. 1. Hematite flotation using SC oleic acid at low addition rates as a function of pH. An addition of 50 g/t corresponds to a solution concentration of 3 × 10−5 M. Pine oil addition was 50 g/t.

first concentrate. The results for the flotation of hematite using this procedure are given in Figs. 1–3.

froth layer was removed using a spatula. Agitation speed remained constant at 1000 rpm for the duration of the test. Another equal addition of collector was made to the pulp. Conditioning was conducted for 10 min at 20% solids, and the pulp aerated for 3 min with another concentrate being removed. This process was repeated three more times. Strict pH control was exercised throughout the test by small additions of either dilute HCl or NaOH as required. The five concentrates and tailing were filtered, dried and weighed, and the recovery of hematite calculated as a weight%. The recovery reported is the cumulative weight% floated up to that stage. The weight% of the decanted slimes was included in the weight% of the

4.1.2. Single additions at constant pH These flotation tests were conducted on 100 g dried samples of the rod milled ore in a 1 L flotation cell attached to the Galigher Agitair laboratory flotation machine Model LA 500. (This material was not deslimed as reported above). Demineralised water was added to bring the slurry level up to 20 mm from the overflow weir. The pH was adjusted to the desired value using dilute solutions of either HCl or NaOH, and the appropriate oleate addition made. The slurry was conditioned

100 200 g/t 400 g/t 600 g/t

80

800 g/t

Recovery (%)

1000 g/t

60

40

20

0 0

2

4

6

8

10

12

pH Fig. 2. Hematite flotation using SC oleic acid at high addition rates as a function of pH. An addition of 200 g/t corresponds to a solution concentration of 1.2 × 10−4 M. Pine oil addition was 50 g/t.

74

Minerals Engineering 113 (2017) 71–82

K. Quast

100 200 g/t 400 g/t 600 g/t 800 g/t

80

Recovery (%)

1000 g/t

60

40

20

0 0

2

4

6

8

10

12

pH Fig. 3. Hematite flotation using sodium stearate as a function of pH. An addition of 200 g/t corresponds to a solution concentration of 1.1 × 10−4 M based on the chemical analysis of the reagent. Pine oil addition was 50 g/t.

100

10 -5 M

80

Recovery (%)

10 -4 M

60

40

20

0 0

2

4

6

8

10

12

pH Fig. 4. Hematite flotation using SC oleic acid as a function of pH at the same additions used in the surface tension measurements. Pine oil concentration was 25 mg/L.

during flotation, they were agitated in the 1 L flotation cell at the same speed as during the flotation. The reagent was conditioned with a small amount of demineralised water for a short time to disperse it, then diluted to the final volume, a similar procedure to that used in the flotation using single additions at constant pH (see Section 4.1.2), but no mineral was present. At the end of the conditioning period, the solution was sampled while still being agitated, and the sub-sample poured into a clean petri dish and placed under the ring of a du Nouy Tensiometer. Readings of surface tension were taken every 5 min until the values were stable, up to a maximum of 1 h, using the procedure of Kolthoff et al. (1967). The diameters of the ring and wire were specified as 525/1000 in. and 9/1000 in. respectively, giving a ratio of ring: wire

for 15 min at 1000 rpm, with frequent pH checks and adjustment if necessary. One drop of pine oil frother was added, the pulp conditioned for a further 2 min, and aerated at 2 L/min for 3 min. The concentrate was removed using a spatula, then filtered, dried and weighed. This weight represented the percentage recovered by flotation. The flotation recovery of hematite was determined at oleate additions of 10 -5 molar and 10 -4 molar, corresponding to the additions made in the determination of the surface tension of the solutions. Results are given in Fig. 4. 4.2. Measurement of surface tension In order to simulate the dispersion of the surfactants in solution 75

Minerals Engineering 113 (2017) 71–82

K. Quast

35 10 -5 M 10 -4 M

Surface Pressure (mN/m)

30

25

20

15

10

5

0 0

2

4

6

8

10

12

pH Fig. 5. Surface pressures of SC oleic acid aqueous mixtures as a function of pH.

and the surfactants at the same pH and plotted as Figs. 5 and 6.The surface pressure thus represents the lowering of the surface tension due to the addition of the surfactant.

diameters of 58.3. The Harkins and Jordan (1930) correction factor for this ring in water was 1.0 i.e. no correction factor was required. Rigorous analyses of this method have been reported by Zuidema and Waters (1941), Huh and Mason (1975) and Furlong et al. (1983). Readings were taken at reagent concentrations of 10−5 and 10−4 molar for the oleic acid and 10−4 and 10−3 molar for the sodium stearate, with dilute HCl and NaOH solutions used for pH adjustment. The readings were taken at an ambient temperature of 15–20 °C, and the platinum ring was cleaned by flaming prior to each sample change to minimise contamination. The surface tension of the water used was also determined as a function of pH. Values of surface pressure were calculated as the difference between the surface tension of the water

4.3. Measurement of bubble characteristics In order to further investigate the frothing characteristics of the frother and the two surfactants used in this study, a similar procedure to that described above regarding the measurements of the surface tension of aqueous solutions was used. Since pine oil additions were 25 mg/L (see Section 4.1.2), the same addition was added both in isolation and together with the collectors to understand the relative contributions of

45 10 -4 M 10 -3 M

40

Surface Pressure (mN/m)

35 30 25 20 15 10 5

0 0

2

4

6

8

pH Fig. 6. Surface pressures of sodium stearate aqueous mixtures as a function of pH.

76

10

12

Minerals Engineering 113 (2017) 71–82

K. Quast

Table 2 Bubble and froth characteristics for the addition of 10 added).

-4

Table 5 Bubble and froth characteristics for the addition of 10−4 molar sodium stearate (pine oil added).

molar oleic acid (no pine oil

Approximate pH

Approximate dominant bubble size (mm)

Bubble and froth characteristics

Approximate pH

Approximate dominant bubble size (mm)

Bubble and froth characteristics

2.5 3.0 4.0 5.0 6 7 8 9 10 11 11.8

5–6 5–6 5–6 4–6 4–5 3–4 3–4 2–4 2–3 2–3 1–2

Very few bubbles Few bubbles More bubbles Good bubble swarm Good bubble swarm Good bubble swarm Good bubble swarm Good bubble swarm Dense bubble swarm Dense bubble swarm Very dense bubble swarm, froth at air–water interface

2.5

<1

3.1

<1

3.9

<1

5.0

<1

6.7

<1

7.7 8.7 9.4 10.0 11.0

2–5 3–5 5–10 5–10 5–10

Unstable, very fine mist spray above surface Unstable, very fine mist spray above surface Unstable, very fine mist spray above surface Unstable, very fine mist spray above surface Unstable, very fine mist spray above surface More stable foam, less fine mist Larger bubbles, less fine mist More stable foam Stable foam, depth 20 mm Very stable foam, 25 mm depth, no mist spray

Table 3 Bubble and froth characteristics for the addition of 10 oil added).

-4

molar sodium stearate (no pine

Approximate pH

Approximate dominant bubble size (mm)

Bubble and froth characteristics

2.3 3.0 4.1 5.5 6.6 7

5–8 5–8 5–8 5–6 4–5 3–4

8

2–3

8.3

1–2

8.8

0.5–2

9.0

0.5–1

Hardly any bubbles Hardly any bubbles Hardly any bubbles Slightly more bubbles More bubbles, no foam Fine bubble swarm, small amount of foam Fine bubble swarm, foam over half the surface Fine bubble swarm, 20 mm foam on surface Fine bubble swarm, 50 mm foam on surface Dense fine bubble swarm, foam overflowing cell

5. Results and discussion 5.1. Flotation testing The flotation recovery of the Iron Prince hematite sample using sequential, multistage additions of oleic acid is shown in Fig. 1. At the strongly acidic pH region (pH 2), the flotation recovery is very low. This is in accord with the fact that the oleic acid is present as an insoluble oil, and the hematite surface is positively charged. The oleic acidaqueous system has been described in recent publications by the author (e.g. Quast, 2016a, 2016a, 2016c) which can be consulted for information. In summary, insoluble oleic acid would be the main oleate phase present at pH values less than 8. Drzymala (1985) has reported a major increase in transmittance of an aqueous emulsion and a significant decrease in the average mean diameter of oleic acid droplets at this pH. In the alkaline region, oleate ions, ionic oleate dimers and an ionomolecular acid soap complex would be present in varying concentrations depending on pH. Atrafi et al. (2012) reported a small decrease in foaming of oleic acid solutions between pH 9 and 10 (see literature survey). At pH 8.5 there was a sudden drop in flotation recovery, and recovery increased again as the pH was increased to 10 (see Fig. 1). Rath et al. (2014) also showed a large drop in hematite recovery with oleate at pH > 7.5, and this remained very low at pH 9. Flotation testing was not continued above pH 9 by Rath et al. (2014). Under the lowest oleate addition, there appeared to be a minor maximum in flotation recovery at pH 10, but at higher values of pH, recovery increased as the pH was increased to pH 12 (see Fig. 1). A maximum in flotation of hematite has been observed close to the zero point of charge of the Iron Prince hematite at 6.7 shown in Figs. 1, 3 and 4 for low additions of oleate, as reported in the Literature review. Increasing the addition of oleic acid gave the more conventional

the frother and the collectors to the frothing and bubble characteristics. Additions of collectors corresponding to 10−4 molar either with or without pine oil were made to a 2 L perspex flotation cell with a bottom driven Agitair mechanism. The initial surfactant addition corresponding to a final concentration of 10−4 molar was made to a volume of 1 L demineralised water, and this was agitated at 900 rpm for 15 min to disperse it. Small amounts of dilute solutions of either NaOH or HCl were made and the pH monitored. After this time, an additional 1 L of demineralised water was added and the final pH recorded. The aqueous solution was aerated at 3 L/min, corresponding to a superficial gas velocity, Jg, of 0.22 cm/s, and the bubble sizes and froth character observed visually and recorded. The results are given in Tables 2–5.

Table 4 Bubble and froth characteristics for the addition of 10

-4

molar oleic acid (pine oil added).

Approximate pH

Approximate dominant bubble size (mm)

Bubble and froth characteristics

2.2 3.2 4.1 5.2 6.3 7.2 8.7 9.6 10.4 11.0

< 1 mm < 1 mm < 1 mm < 1 mm < 1 mm < 1 mm < 1 mm < 1 mm < 1 mm 1–5 mm

Unstable, very fine mist spray Unstable, very fine mist spray Unstable, very fine mist spray Unstable, very fine mist spray Unstable, very fine mist spray Unstable, very fine mist spray Unstable, very fine mist spray Unstable, very fine mist spray Unstable, very fine mist spray Larger, more stable bubbles

77

above above above above above above above above above

surface surface surface surface surface surface surface surface surface

Minerals Engineering 113 (2017) 71–82

K. Quast

about 2.5 molecules deep was reached, when another change occurred and further addition produced only a slight effect. The variation of surface tension/surface pressure of aqueous oleate solutions with pH follows the same pattern as reported by other researchers, with a characteristic minimum in surface tension (maximum in surface pressure for an addition of 10−5 molar oleic acid) at approximately pH 7, corresponding closely to the solubility edge of the oleate solutions at the respective concentrations (see Fig. 5). The surface tension minimum is shifted to a higher pH value (8.5) when the oleate concentration is increased, as predicted by solution equilibria (Quast, 2016a, 2016a, 2016c) and noted by Muratoglu et al. (2000). In the acidic region, where insoluble oleic acid is the dominant phase, one would expect the surface tension to be similar to that of the equilibrium spreading pressure, measured at 33 mN/m. Actual values are closer to 45 mN/m, indicating incomplete spreading at the aqueous/air interface. For the sodium stearate tests, the surface tensions in the acidic pH region were close to 40 mN/m, which corresponded to a higher addition of a surfactant containing a lower concentration of oleate. Values of surface pressure were plotted in Figs. 5 and 6 because higher surface activity corresponds to a higher surface pressure which should lead to a higher flotation recovery. This means that maxima in surface pressure should correspond to maxima in flotation recovery at that pH. Since the sodium stearate contains almost 20% stearate (see Table 1) the surface tension or surface pressure of this moiety must also be taken into account. Surface pressure of stearic acid as a function of pH has been measured previously by the author (Quast, 2006) and showed maximum surface pressures of 20 mN/m at concentrations of 10−3 molar. Surface pressures of the sodium stearate shown in Fig. 5 are almost double this value, hence the main contribution to the surfactant activity is the oleate contained in the sodium stearate. At pH values above the solubility limit, the increasing electrostatic repulsion of oleate ions at the aqueous/air interface would cause the surface pressure to fall, as shown in Figs. 5 and 6. It does not, however, explain the rapid increase in surface pressures at high values of pH as shown in Fig. 5.

recovery vs. pH plots as shown in Fig. 2. The curves for the same additions of sodium stearate as for the oleic acid shown in Fig. 2 are shown in Fig. 3, which also display the same flotation minimum at pH 8.5 as shown in Fig. 1. This is understandable since the sodium stearate only contains 36.8% oleic acid, approximately half that of the oleic acid (see Table 1). Finally, flotation recovery vs. pH curves for the oleic acid additions corresponding to the oleate additions used to measure the surface tension (and hence surface pressure) are shown in Fig. 4. The lowest oleate addition (10−5 molar) also displays the flotation recovery minimum at pH 8.5 as shown in Figs. 1 and 3. The fact that recovery did not diminish at pH values equal to or higher than the zero point of charge of the mineral is conclusive evidence of strong chemisorption of the collector onto the mineral surface at alkaline pH values. Even at pH 12 flotation recovery was not adversely affected, in fact in most instances raising the pH from 10 to 12 actually caused an increase in flotation recovery. In contrast to the flotation recovery data shown in Figs. 1, 3 and 4, Fuerstenau et al. (1970) reported a minimum for the oleate flotation of hematite at pH 5–6, with a maximum in recovery at pH 7.5–8.5. Natarajan (1978) also reported a minimum in hematite flotation from a low grade iron ore at pH 5 using sodium oleate and pine oil, and a maximum in hematite recovery at pH 9. Peck et al. (1966) reported a flotation maximum at pH 7, and a flotation minimum at pH 8 for the flotation of a synthetic hematite with a high surface area (76 m2/g) using sodium oleate. From the Literature Review, Section 2.4, contact angles can be used as a measure of the hydrophobicity of a mineral surface. Yap et al. (1981) reported how the contact angle on hematite in the presence of oleate decreased to very low values as the pH was raised above 8 for low additions of oleate. The direct measurement of the adsorption of oleate onto hematite as a function of pH is the subject of another publication by the author [in progress], which will address this particular issue. The change in hematite recovery with oleate as a function of pH as reported in the current data can therefore possibly be attributed to the change in hydrophobicity as the pH of the suspension is raised above pH 8. Before concluding the discussion on the flotation recovery vs. pH curves, it is instructive to investigate the changes in surface pressure with pH for the oleic acid additions corresponding to the additions in Fig. 4. These are given in Fig. 5. Surface pressure vs. pH data for the sodium stearate at additions of 10−4 molar and 10−3 molar are given in Fig. 6. When the results from Fig. 1 are compared to those reported in Fig. 5, oleic acid frothing ability as predicted by surface pressure should be a minimum at pH 11, whereas the flotation recovery of the hematite increases as the pH is raised from 9 to 12 except at the lowest oleic acid addition. The high froth height at pH 11 corresponds to a maximum in hematite recovery at this pH for an oleate addition of 10 -5 molar (see Fig. 4). The low froth height for pH values below 6 reported by Iwasaki et al. (1960) correspond to lower flotation recoveries, however there is still high recovery of hematite, in fact a local maximum in recovery around pH 6, at oleate additions of 10−5 molar (see Fig. 4).

5.3. Comparison of surface pressure and flotation recovery Examination of the flotation data in Figs. 1 and 3 show flotation recovery minima at pH 8.5, which corresponds to a maximum in surface pressure at this pH. Flotation recoveries steadily increase as the pH is raised from 9 to 12 (Figs. 1 and 3), but the surface pressures show minima at pH 11 and either increase or remain approximately constant as the pH is increased from 11 to 12. There have been a number of instances where the recovery of various minerals with oleate has shown several maxima and minima, and some of these are described briefly below. Kivalo and Lehmusvaara (1958) reported two maxima in flotation of magnetite at a low addition (150 g/t) of sodium oleate. These were at pH 7 and 11. Adding Dowfroth frother increased the flotation recovery, especially at pH 7 and shifted the second recovery maximum to pH 10. At higher additions of oleate (500 g/t), three maxima in the flotation of magnetite were reported. These were observed at pH values of ∼3, 7 and 9.5, with minima at pH 4 and 8. Apart from noting the various flotation maxima, no interpretation for the changes in flotation behaviour was given. Choi and Whang (1963) reported two maxima for the oleate flotation of zircon at pH values of 4 and 6–9. The isoelectric point of the zircon was measured as 5. The increase in the zircon recovery at pH > 6 using oleate was attributed to its activation by Fe 2+and Ti 4+ ions associated with the zircon. The existence of two values of maximum flotation recovery using oleic acid to float oxide and silicate minerals has been well documented by the late Professor M.C. Fuerstenau, who reported this for hematite (Fuerstenau et al. 1970), pyrolusite (Fuerstenau and Rice, 1968), two samples of chromite (Palmer et al.1975) and diopside and augite

5.2. Surface tension and surface pressure The surface tension at the oleic acid/air interface for the sample of oleic acid used in these experiments was measured as 33 mN/m using the du Nuoy ring, identical to the value reported by Tomoaia-Cotisel et al. (1978). Reported values of the equilibrium spreading pressure of oleic acid on water were approximately 30 mN/m (Blodgett, 1935; Donnison and Heymann, 1946; Kivalo and Lehmusvaara, 1958; Jalal et al. 1980 and Rakshit et al. 1981). An early study of the effect of adding oleic acid to water was published by Taggart and Gaudin (1923). They found that the addition of oleic acid only had a slight effect on the surface tension of water until the amount present was sufficient to form a monomolecular film. Beyond this point, surface tension dropped rapidly until an amount of acid equivalent to a film 78

Minerals Engineering 113 (2017) 71–82

K. Quast

Recent work flotation conducted at the University of South Australia and reported by Abaka-Wood et al. (2016) using a different sample of natural hematite and a sample of sodium oleate from a different supplier (Sun Ace) also showed multiple flotation maxima. A minor maximum was observed at pH 5 and a major maximum at pH 9 for oleate additions of 1000 g/t. These authors also showed that the presence of oleate reduced the isoelectric point of hematite from 6.9 to 5.5, in agreement with the data of Quast (2016b). The maximum flotation recovery at pH 7 (Figs. 1, 3 and 4) corresponds to the minimum in surface tension of oleate solutions reported by other workers, and coincided with the so-called solubility edge (see above). Ananthapadmanabhan and Somasundaran (1981) reported that the pH of maximum hematite flotation with potassium oleate shifted with the pH of its minimum surface tension as its concentration was increased. They showed a single linear correlation for the pH of the minimum in surface tension of the oleate and the pH of maximum hematite flotation as a function of the log of the concentration of potassium oleate. Yousef, (1967) showed that the changes in surface tension of oleate solutions were identical to the trends in flotation recovery of manganese ore as a function of pH. A surface tension argument cannot be applied to explain the flotation maximum obtained at pH 11. The reduction in flotation recovery at pH > 11 for low additions of oleate can easily be explained in terms of the competition between oleate ions and hydroxyl ions on the hematite surface. Higher additions of oleate cause sufficient oleate ions to be available to render the hematite surface hydrophobic, causing strong flotation. In order to investigate the changes in the surfactant activity of oleate with pH, it is instructive to examine its frothing characteristics in aqueous solutions.

(Fuerstenau et al. 1977). In the cases of oxides, the lower pH value for maximum flotation was approximately 4, and was interpreted as due to physical adsorption of oleate. For the iron-bearing silicates, the flotation at pH 2.5 was thought to be due to hydrolysis of the ferric ions on the surface to FeOH++. Ananthapadmanabhan et al. (1978) reported a minimum in hematite flotation with 3 × 10−5 molar potassium oleate at pH 6 and a maximum recovery at pH 8. Pavez and Peres (1993) also reported two flotation maxima, at pH 3 and 7, for oleate flotation of monazite, zircon and rutile. Fuerstenau and Shibata (1999) reported the flotation recovery of electrolytic manganese dioxide using sodium oleate as a function of pH. They found maxima at pH values of 3 and between 7 and 10, with the flotation at pH 3 the result of the co-adsorption of oleate ions and oleic acid molecules onto a positively charged colloid. The flotation at alkaline pH was the result of a strong chemical interaction of oleate with a negatively charged surface. Moon and Fuerstenau (2003) reported two flotation maxima at pH 4 and 8 for the flotation of spodumene with sodium oleate. These corresponded to maxima in contact angles, with the major peak observed at pH 8 attributed to the chemisorption of oleate on surface aluminium ions on the spodumene and the minor peak associated with the presence of a ferric ion impurity. The major flotation peak at pH 8 corresponded to the major oleate adsorption, major infrared absorption and the maximum negative zeta potential for the oleate-spodumene system. This mismatch between the pH values of surface pressure and flotation recovery have also been reported for this ore using dodecanoic acid (Quast, 2006). The flotation of hematite using dodecanoic acid along with other surfactants containing 12 carbon atoms has been discussed previously by Quast, (2000), which can be consulted for further information. It is not the purpose of this paper to revisit the flotation of hematite using dodecanoic acid, but the flotation recovery and surface pressure curves for the two surfactants are remarkably similar with flotation minima at pH 8.5 and surface pressure minima at pH 11. For both cases, the simplest explanation would involve the physical adsorption of undissociated fatty acids in the acidic pH region, and chemisorption of dissociated fatty acid anions under alkaline pH conditions. The recovery minimum at pH ∼ 9 using sodium dodecanoate is not unique to hematite. Laskowski and Sobieraj (1969) reported the same thing for the flotation of Albanian chromite using sodium dodecanoate. Fan et al. (2009) reported two maxima for the flotation of ilmenite using 1.64 × 10−4 molar sodium oleate. These were at pH 2.5 and between pH 5 and 8. The isoelectric point of the ilmenite was measured at approximately 5.5, and its isoelectric point and zeta potential were both significantly reduced in the presence of sodium oleate. The reduction in flotation at high values of pH was attributed to the competition between oleate ions and hydroxyl ions, whereas the reduction in strong acid solutions was attributed to the dissolution of metallic ions from the surface of the ilmenite. Liu et al. (2015a) reported the electrokinetic and flotation behaviours of hemimorphite (zinc silicate) in the presence of sodium oleate. They reported a flotation maximum at pH 8, a minimum at pH 9 and another maximum at pH 11. The flotation maximum at pH 8 was attributed to the presence of oleate and acid-soap ions. Liu et al. (2015b) found flotation minima and maxima for the oleate flotation of ilmenite, titanaugite and forsterite using sodium oleate as collector as a function of pH. They reported flotation maxima around pH 5, minima around pH 9 and maxima around pH 11. Under the oleate concentration used (2 × 10−4 molar), the authors highlighted the existence of oleate ions and dimers at pH 6 and the acid-soap complex at pH 8.4. Flotation was attributed to the interactions of these species with the metal cations on the surfaces of the minerals, with the higher adsorption of surfactant on the surface of the ilmenite. This was supported by the results of zeta potential and FT-IR analyses of the mineral-oleate systems.

5.4. Frothing characteristics of oleic acid and pine oil as a function of pH According to Atrafi and Pawlik (2016), the typical range for superficial gas velocity, Jg, in froth flotation is 0.5–2.5 cm/s. For the frothing tests used, this was measured at 0.22 cm/s. The low value should generate the lowest bubble sizes achievable, with high values increasing the probability of bubble collisions which would enhance bubble coalescence and lead to increased bubble sizes. With the addition of only pine oil frother to the cell, a very unstable froth was observed for all pH values up to 9.1 when air was added. The solubility of pine oil at 25 °C is 2.5 g/L (Booth and Freyberger, 1962). According to Booth and Freyberger (1962), pine oil generates a small bubble froth of closely knit texture, and excessive quantities of pine oil tend to flatten froth, decrease its volume and cause effervescence at the surface. Only very fine bubbles (< 1 mm) were observed that continued to burst on the surface creating a very fine mist spray above the surface to a height of about 30 mm. The froth appeared similar to that of a highly carbonated soft drink poured into a glass. Above pH 9.1, more stable froths were observed, with depths up to 10 mm. These were not tabulated but just described since the observations were virtually identical up to pH 9. From Tables 2 and 3 it can be seen that the two surfactants in the absence of pine oil have different frothing characteristics. For the oleic acid, agitation caused the formation of only a few large bubbles at low values of pH, with bubbles becoming finer as the pH was raised. It was only at pH values above 10 that dense bubble swarms were observed. Unfortunately, the available photographic equipment was unable to provide clear evidence of bubble characteristics. The frothing characteristics observed for a similar concentration of the sodium stearate were different to those for the oleic acid. Under acidic conditions, only a few large bubbles were observed. Once the pH was neutral, fine bubbles and the onset of foaming was noted. Above pH 8, a fine bubble swarm and foaming was noted, with the foaming increasing rapidly as the pH was raised to 9. This behaviour has been reported by various authors including Atrafi and Pawlik (2016) who reported Sauter mean bubble diameters around 2 mm for oleate additions of 10−4 molar and 79

Minerals Engineering 113 (2017) 71–82

K. Quast

developed by Pyke et al. (2003). This has been tested extensively by and reported by Pyke et al. (2003) and Duan et al. (2003) and validated as a very instructive approach in plant trials by Ralston et al. (2007). Pyke et al. (2003) investigated bubble-particle heterocoagulation under turbulent conditions in a Rushton turbine cell applicable to those operating in a flotation cell. The flotation rate constant was linked directly with hydrodynamic variables such as bubble size and bubble velocity, to which the rate constant is inversely proportional, turbulent dissipation energy, gas flow rate, cell volume, kinematic viscosity as well as to the multiple of the collision, attachment and stability efficiencies. During the flotation experiments conducted in the presence of pine oil and oleic acid (Figs. 1 and 2) or pine oil and sodium stearate (Fig. 3) the gas flow rate, cell volume, turbulent dissipation energy and kinematic viscosity can be taken as essentially constant. The rate constant and thus recovery will be dominated by bubble size and the collection efficiency dominated by contact angle (Pyke et al. (2003), Duan et al. (2003)). From Table 5 it can be seen that there is a marked increase in bubble size above pH 8, and an increase in bubble velocity is expected. The changes in bubble characteristics noted in Table 4 may be modified by the presence of the oleic acid droplets below pH 8. At the same time Yap et al. (1981) have reported a sharp decrease in contact angle above pH 8, so a marked reduction in rate constant and thus recovery is expected based on a decrease in attachment efficiency (Dai et al. 1999). Contact angles can be used as a measure of the hydrophobicity of a mineral surface, and Yap et al. (1981) reported how the contact angle on hematite in the presence of oleate dropped to very low values as the pH was raised above 8 for low additions of oleate. The reduction in contact angle as the pH is raised will affect the attachment efficiency as noted above. Flotation recovery increases with increasing contact angles of particles in the pulp (Muganda et al. 2011; Chipfunhu et al. 2012), hence by definition flotation recovery will decrease when the contact angle decreases. The combination of effects of increase in bubble size and a reduction in collection efficiency reflected through a decrease in contact angle at pH values around 8.5 is a reasonable explanation of the recovery minimum noted at these elevated pH values. It would be very useful indeed to acquire contact angle data as a function of pH and inert electrolyte concentration from pH 8–11; however this is outside the scope of this present study. The changes in contact angle with pH and ionic strength for a titania surface partially covered with surfactant has been reported by Hanly et al. (2011), and can serve as a template for investigating the changes in wettability on either side of the zero point of charge of the hematite for the hematiteoleate system.

pH values between 7.5 and 10.0. At pH 6.5, the Sauter mean bubble diameter was approximately 3.5 mm at 10−4 molar oleate. A later publication by these authors (Atrafi and Pawlik, 2017), showed that the total gas volume in the foam increased from virtually zero at pH 7.2 to ∼150 mL at pH 7.5 to > 700 mL at pH values > 8 at a time of 2 min. Unfortunately the onset of foaming activity at pH 5 did not correspond to rapid changes in surface pressure reported in Figs. 5 and 6. Miles and Ross (1944) found that the maximum foam stability of sodium oleate solutions occurred at pH values > 9. In summary, the frothing characteristics of the two surfactants in the absence of added frother are very different. The frothing data for the oleic acid followed the data reported by Nunez and Yalkowsky (1997), who correlated this increase at the pKa of the oleic acid with the presence of soluble oleate species at pH values above the pKa. The foaming characteristics for the sodium stearate corresponded closely to the data reported by Atrafi et al. (2012), where the foaming began at pH 7. In the current work, the pH was not increased above pH 9 for the sodium stearate because there was foam overflowing the cell, leading to a depletion in solution volume in the cell. When pine oil was added together with the collectors, and the solution aerated, the frothing characteristics were dominated by the pine oil (see Tables 4 and 5). There is also a difference in the bubble size above pH 8. The character of the bubbles in terms of size is very different for sodium oleate and oleic acid and seems to mirror the results of Atrafi et al. (2012). Under acidic pH conditions froths were very unstable, with a fine mist spray above the surface, described as effervescence by Booth and Freyberger (1962). When oleic acid was used, fine droplets of undissolved oleic acid were visible in the froths for pH values less than 9. This was not observed when the sodium stearate was used, resulting in the changes in bubble size, which may be due to the greater solubility of the sample of sodium stearate compared to oleic acid. It must be remembered that all these measurements and observations were made in the absence of hydrophobic mineral particles, which would stabilise the froths in the case of the flotation testing but not influence bubble size in the pulp. The bubble and froth conditions corresponding to the addition of pine oil to the aqueous solutions of the surfactants were dominated by those of pine oil alone, even in the alkaline pH regions, showing how the addition of a frother has a major influence on the foams generated by surfactants with known frothing characteristics. Although contact angles can significantly affect the flotation recovery of minerals in the presence of collectors, it was thought that there were other factors causing the observed flotation minimum of hematite in contact with oleate at pH 8–8.5. Since none of the above correlations based on surface and solution chemistry totally explained the flotation minimum observed at pH values around 8.5, although the reduction in contact angle may give us an indication, other factors will now be considered. Perez and Aplan (1975) floated iron hydroxide with a commercial oleic acid collector and reported their data as a function of pH. As the pH was raised from 5 to 9, mineral recovery increased linearly, the ratio of collapsed froth volume/volume feed to the cell increased, but the concentration factor, corresponding to the coefficient of mineralisation or enrichment ratio (concentration of metal ion in the froth/concentration in the bulk) fell after pH 7, showing a change in the latter parameter with pH above pH 7. The recovery of particles by air bubbles during flotation occurs as a result of three sub-processes: collision, attachment and stability (Dai et al., 1998, 1999; Duan et al. 2003; Pyke et al., 2003). The overall collection efficiency, E, is the multiplication of the efficiencies of collision, attachment and stability. Collision is mainly controlled by hydrodynamics (e. g bubble size) and attachment efficiency is dominated by hydrophobicity (e. g. contact angle). These two are very strongly linked (Dai et al., 1999). Stability depends on both hydrodynamics and interfacial behaviour. A general rate equation relating a mechanical term, primary turbulence term and elementary processes has been

6. Conclusions The data reported here are the results of carefully conducted experiments to provide more information on the hematite-oleate system. The author has previously investigated the effects of pH on the flotation of various samples of hematite and showed how knowledge of the solution chemistry of the aqueous oleate system is imperative in our understanding of this system. In this paper, additional information involving the variation of surface tension (and surface pressure) of aqueous oleate has been reported. A mismatch between the surfactant activity of the oleates and their flotation activity has been reported here. The frothing characteristics as determined by surface tension for the two surfactants investigated show results similar to those reported by other researchers. For the hematite-oleate system, the dominant factor leading to the flotation recovery minimum at pH values around 8.5–9.0 has been investigated in this study. There is obviously a reduction in the contact angle for the interaction of oleate with hematite in this region, but there is insufficient reported data to show that the contact angle increases at pH values above 10. The flotation minimum does not appear to coincide with any reduction in surface pressure or foaming ability. It is proposed that this recovery minimum is associated with an 80

Minerals Engineering 113 (2017) 71–82

K. Quast

Drzymala, J., Chmielewski, T., Wolters, K.L., Birlingmair, D.H., Wheelock, T.D., 1992. Microflotation measurement based on modified Hallimond tube. Trans., Inst. Min. Metallur., Section C 101, C17–C24. Duan, J., Fornasiero, D., Ralston, J., 2003. Calculation of the flotation rate constant of chalcopyrite particles in an ore. Int. J. Miner. Proc. 72, 227–237. Fan, X., Waters, K.E., Rowson, N.A., Parker, D.J., 2009. Modification of ilmenite surface chemistry for enhancing surfactants adsorption and bubble attachment. J. Colloid and Interface Sci. 329, 167–172. Fuerstenau, D.W., Shibata, J., 1999. On using electrokinetics to interpret the flotation and interfacial behaviour of manganese dioxide. Int. J. Miner. Proc. 57, 205–217. Fuerstenau, M.C., Harper, R.W., Miller, J.D., 1970. Hydroxamate vs. fatty acid flotation of iron oxide. Trans. Soc. Min. Engineers of AIME 247, 69–73. Fuerstenau, M.C., Palmer, B.R., Gutierrez, G., 1977. Mechanisms of flotation of selected iron-bearing silicates. Trans. Soc. Min. Engineers of AIME 262, 234–236. Fuerstenau, M.C., Rice, D.A., 1968. Flotation characteristics of pyrolusite. Trans. Soc. Min. Engineers of AIME 241, 453–457. Furlong, D.N., Freeman, P.A., Metcalfe, I.M., White, L.R., 1983. Wall effects in du Nouy ring tensiometry. Theory and experiment. J. Am. Chem. Soc., Faraday Trans. I 79, 1701–1719. Gontigo, C.de F., Fornasiero, D., Ralston, J., 2007. The limits of fine and coarse particle flotation. Canadian J. Chem. Eng. 85 (5), 739–747. Gutierrez, C., Iskra, J., 1977. The action of neutral oleic acid in the flotation of hematite. Int. J. Miner. Proc. 4, 163–171. Hanly, G., Fornasiero, D., Ralston, J., Sedev, R., 2011. Electrostatics and metal oxide wettability. J. Phys. Chem., Section C 115, 14914–14921. Harkins, W.D., Jordan, H.F., 1930. A method for the determination of surface and interfacial tension from the maximum pull on a ring. J. Am. Chem. Soc. 52, 1751–1772. Hernainz, F., Caro, A., 2001. Contact angle and surface tension in the celestite+sodium oleate aqueous solutions+air system. J. Chem. Eng. Data 46, 107–109. Huh, C., Mason, S.G., 1975. A rigorous theory of ring tensiometry. Colloid and Polymer Science 253, 566–580. Iskra, J., 1997. Flotation properties of silicon carbide 1 flotation of silicon carbide with anionic and cationic collector. Ceram. Int. 23, 337–342. Iveson, S.M., Holt, S., Biggs, S., 2000. Contact angle measurements of iron ore powders. Colloids and Surf. A: Physicochem. Eng. Aspects 166, 203–214. Iveson, S.M., Holt, S., Biggs, S., 2004. Advancing contact angle of iron ores as a function of their hematite and goethite content: implications for pelletising and sintering. Int. J. Miner. Proc. 74, 281–287. Iwasaki, I., Cooke, S.R.B., Choi, H.S., 1960. Flotation characteristics of hematite, goethite and activated quartz with 18-carbon aliphatic acids and related compounds. Trans. AIME 217, 237–244. Jalal, I.M., Zografi, G., Rakshit, A.K., Gunstone, F.D., 1980. Monolayer properties of fatty acids I. Thermodynamics of spreading. J. Colloid and Interface Sci. 76 (1), 146–156. Joseph-Soly, S., Quast, K., Connor, J.N., 2015. Effects of Eh and pH on the oleate flotation of iron oxides. Miner. Eng. 83, 97–104. Kivalo, P., Lehmusvaara, E., 1958. An investigation into the collecting properties of some of the components of tall oil, Proceedings, progress in mineral dressing. Transactions of the International Mineral Dressing Congress, Stockholm, Almqvist and Wiksell, Stockholm, pp. 577–586. Kolthoff, I.M., Elving, P.J., Sandell, E.B., 1967. Treatise on Anal. Chem., Volume 7, 4656–4661 Part 1. Kosmulski, M., 2009. Compilation of PZC and IEP of sparingly soluble metal oxides and hydroxides from literature. Adv. Colloid and Interface Sci. 152, 14–25. Kulkarni, R.D., 1976. Flotation properties of the hematite-oleate system and their dependence on the interfacial adsorption. Columbia University, New York Eng. Sc. D. thesis. Kulkarni, R.D. Somasundaran, P., 1975, Kinetics of oleate adsorption at the liquid/air interface and its role in hematite flotation, in: P. Somasundaran. R.B. Grieves (Eds.), Proceedings, advances in interfacial phenomena of particulate/solution/gas systems; applications to flotation research, American institute of chemical engineers symposium series, vol. 71, number 150, pp. 124–133. Laskowski, J., Sobieraj, S., 1969. Zero points of charge of spinel minerals. Trans. Inst. Min. Metallur. 78, C161–C162. Laskowski, J.S., 1988, Weak-electrolyte type collectors, in: A. Mular, G. Gonzalez, C. Barahona (Eds.), proceedings, “Copper 87”, volume 2, mineral processing and process control, pp. 137–154. Liu, C., Feng, Q., Zhang, G., 2015a. Electrokinetic and flotation behaviours of hemimorphite in the presence of sodium oleate. Miner. Eng. 84, 74–76. Liu, W., Zhang, J., Wang, W., Deng, J., Chen, B., Yan, W., Xiong, S., Huang, Y., Liu, J., 2015b. Flotation behaviours of ilmenite, titanaugite and forsterite using sodium oleate as collector. Miner. Eng. 72, 1–9. Lourenco, S.D.N., Woche, S.K., Bachmann, J., Saulick, Y., 2015. Wettability of crushed air-dried minerals. Geotechnique Letters 5, 173–177. Lukenheimer, K., Wantke, K.-D., 1981. Determination of the surface tension of surfactant solutions applying the method of Lecomte du Nuoy (ring tensiometer). Colloid and Polym. Sci. 259, 354–366. Maeda, T., Fukumoto, C., Matsumura, T., Nishioka, K., Shimizu, M., 2005. Effect of adding moisture and wettability on granulation of iron ore. ISIJ Int. 45 (4), 477–484. Mankovich, A.M., 1953. Physicochemical properties of surfactants. Indus. Eng. Chem. 45 (12), 2759–2766. Mankovich, A.M., 1968. Capacities of the du Nuoy tensiometer. J. Colloid and Interface Sci. 25, 590. Mao, H., Zhang, R., Lv, X., Bai, C., Huang, X., 2013. Effect of surface properties of iron ores on their granulation behaviour. ISIJ Int. 53 (9), 1491–1496. Mikumo, J., 1927. Studies on soap solutions. III. Effects of acids, bases and salts upon the surface tension of aqueous sodium oleate solutions. J. Soc. Chem. Indus., Japan 30,

increase in bubble size and velocity and a reduction in bubble-particle collision efficiency. These contributions outweigh the anticipated flotation recovery based on mineral and surfactant solution chemistries. The quantification of the effects of bubble-particle efficiency and extended contact angles on oleate flotation as a function of pH would be a very informative area for future research leading to a better understanding of the solution chemistries of both oleic acid and its interaction with oxide minerals. Acknowledgement The author wishes to acknowledge the GLC analyses conducted on the surfactant samples by Unichema, Port Melbourne which allowed a better understanding of their chemical properties. The author also wishes to acknowledge financial support through an Australian Government Research Training Program Scholarship scheme. Informative discussions with Emeritus Laureate Professor John Ralston, University of South Australia, are also gratefully acknowledged. References Abaka-Wood, G.B., Addai-Mensah, J., Skinner, W., 2016. Physicochemical characterisation of monazite, hematite and quartz minerals for flotation, in: Proceedings, 4th UMaT Biennial international mining and mineral conference. University of Mines and Technology, Tarkwa, Ghana, pp. 7p. Abaka-Wood, G.B., Addai-Mensah, J., Skinner, W., 2017. A study of flotation characteristics of monazite, hematite and quartz using anionic collectors. Int. J. Miner. Proc. 158, 55–62. Agars, K.M., 1976. Ion flotation of manganese. South Australian Institute of Technology Final year project report, Applied Chemistry undergraduate degree. Ananthapadmanabhan, K., Somasundaran, P., Healy, T.W., 1978, The chemistry of oleate and amine solutions in relation to flotation, in: Preprint number 78-B-67, 1978 AIME annual Meeting, Denver, Colorado, 14p, published in Transactions, Society of Mining Engineers of AIME, 266, 2003–2009. Ananthapadmanabhan, K.P., Somasundaran, P., 1981. Oleate chemistry and hematite flotation. In: Yarar, B., Spottiswood, D.J. (Eds.), Proceedings, “Interfacial Phenomena in Mineral Processing”, Engineering Foundation Conference, Rindge, New Hampshire. Engineering Foundation, pp. 207–227. Ananthapadmanabhan, K.P., Somasundaran, P., 1988. Acid-soap formation in aqueous oleate solutions. J. Colloid and Interface Sci. 122 (1), 104–109. Atrafi, A., Gomez, C.O., Finch, J.A., Pawlik, M., 2012. Frothing behaviour of aqueous solutions of oleic acid. Miner. Eng. 36–38, 138–144. Atrafi, A., Pawlik, M., 2016. Surface tension and gas dispersion properties of fatty acid solutions. Miner. Eng. 85, 138–147. Atrafi, A., Pawlik, M., 2017. Foamability of fatty acid solutions and surfactant transfer between foam and solution phases. Miner. Eng. 100, 99–108. Beunen, J.A., Mitchell, D.J., White, L.R., 1978. Surface tension minimum in ionic surfactant systems. Journal of the Chemical Society, Faraday Transactions I 74, 2501–2517. Blodgett, K.B., 1935. Films built by depositing successive unimolecular layers on a solid surface. J. Am. Chem. Soc. 57, 1008–1022. Booth, R.B., Freyberger, W.L., 1962. Froths and frothing agents. In: Fuerstenau, D.W. (Ed.), Froth Flotation 50th Anniversary Volume. Society of Mining Engineers of AIME, New York, pp. 258–276 Chapter 10. Bottazzi, F., 1913, Surface tension of solutions and suspensions of soaps, Atti Accad. Lincei, 21, II, 365-78, CA 7, 3041. Boucher, E.A., Grinchuk, T.M., Zettlemoyer, A.C., 1967. Measurement of surface tensions of surfactant solutions. J. Colloid and Interface Sci. 23, 600–603. Chipfunhu, D., Zanin, M., Grano, S., 2012. Flotation behaviour of fine particles with respect to contact angle. Chem. Eng. Res. Des. 90, 26–32. Choi, H.S., Whang, K.U., 1963. Surface Properties and floatability of zircon. Canadian Min. Metallur. Bulletin for June 466–468. Cooke, S.R.B., Iwasaki, I., Choi, H.S., 1960. Effect of temperature on soap flotation of iron ore. Trans. AIME 217, 76–83. Cupples, H.L., 1935. Wetting and spreading properties of aqueous solutions. Indus. Eng. Chem. 27 (10), 1219–1222. Dai, Z., Dukhin, S., Fornasiero, D., Ralston, J., 1998. The inertial hydrodynamic interaction of particles and rising bubbles with mobile surfaces. J. Colloid and Interface Sci. 197, 275–292. Dai, Z., Fornasiero, D., Ralston, J., 1999. Particle-bubble attachment in mineral flotation. J. Colloid and Interface Sci. 217, 70–76. Davies, J.T., Rideal, E.K., 1961. Interfacial Phenomena. Academic Press, New York. de Castro, F.H.B., Borrego, A.G., 1995. Modification of surface tension in aqueous solutions of sodium oleate according to temperature and pH in the flotation bath. J. Colloid and Interface Sci. 173, 8–15. Donnison, J.A., Heymann, E., 1946. The equilibrium spreading pressure of oleic acid and of ethyl sebacate on concentrated salt solutions. Trans. Faraday Soc. 42, 1–5. Drzymala, J., 1985. Potentiometric titration of sodium oleate in dilute aqueous solutions. J. Colloid and Interface Sci. 107 (2), 442–445.

81

Minerals Engineering 113 (2017) 71–82

K. Quast

selected fatty acids on hematite Part II laboratory investigations. Miner. Eng. 79, 301–305. Quast, K., 2016a. Surface chemistry and oleate flotation of three South Australian micaceous hematites. Miner. Eng. 85, 123–129. Quast, K., 2016b. The use of zeta potential to investigate the interaction of oleate on hematite. Miner. Eng. 85, 130–137. Quast, K., 2016c. Literature review on the interaction of oleate with non-sulphide minerals using zeta potential. Miner. Eng. 94, 10–20. Quast, K., 2016d. The use of zeta potential to investigate the pKa of saturated fatty acids. Adv. Powder Technol. 27, 207–214. Rakshit, A.K., Zografi, G., Jalal, I.M., Gunstone, F.D., 1981. Monolayer properties of fatty acids II. Surface vapour pressure and the free energy of compression. J. Colloid and Interface Sci. 80 (2), 466–473. Ralston, J., Fornasiero, D., Grano, S., Duan, J., Akroyd, T., 2007. Reducing uncertainty in mineral flotation-flotation rate constant prediction for particles in an operating plant ore. Int. J. Miner. Proc. 84, 89–98. Rath, S.S., Sinha, N., Sahoo, H., Das, B., 2014. Molecular modelling studies of oleate adsorption on iron oxides. Appl. Surf. Sci. 295, 115–122. Shang, J.S., Flury, M., Harsh, J.B., Zollars, R.L., 2008. Comparison of different methods to measure contact angles of soil colloids. J. Colloid and Interface Sci. 328, 299–307. Shrimali, K., Jin, J., Hassas, B.V., Wang, X., Miller, J.D., 2016. The surface state of hematite and its wetting characteristics. J. Colloid and Interface Sci. 477, 16–24. Shu, X., Meng, Y., Wan, L., Li, G., Yang, M., Jin, W., 2014. pH responsive aqueous foams of oleic acid/oleate solution. J. Dispersion Sci. Technol. 35 (2), 293–300. Somasundaran, P., Ananthapadmanabhan, K.P., 1978. Solution chemistry of surfactants and the role of it in adsorption and froth flotation in mineral-water systems, in: K.L. Mittal (Ed.), Solution chemistry of surfactants, vol. 2, New York. Song, S., Lu, S., 1994. Hydrophobic flocculation of fine hematite, siderite and rhodochrosite particles in aqueous solution. J. Colloid and Interface Sci. 166, 35–42. Taggart, A.F., Gaudin, A.M., 1923. Surface tension and adsorption phenomena in flotation. Trans. AIME 68, 479–535. Theander, K., Pugh, R.J., 2001. The influence of pH and temperature on the equilibrium and dynamic surface tension of aqueous solutions of sodium oleate. J. Colloid and Interface Sci. 239, 209–216. Toikka, G., Hayes, R.A., Ralston, J., 1996. Adhesion of iron oxide to silica studied by atomic force microscopy. J. Colloid and Interface Sci. 180, 329–338. Tomoaia-Cotisel, M., Zsako, J., Mocanu, A., Lupea, M., Chifu, E., 1978. Insoluble mixed monolayers III. The ionization characteristics of some fatty acids at the air/water interface. J. Colloid and Interface Sci. 117 (2), 464–476. Veras, M.M., Baltar, C.A.M., Paulo, J.B.de A., Leite, J.Y.P., 2014. Comparative study of the main flotation frothers using a new HYDROMESS adapted technique. Revista Escola de Minas de Ouro Petro 67 (1), 87–92. Wakamatsu, T., Numata, Y., Park, C.H., 1980. Fundamental study on the flotation of minerals using two kinds of collectors. In: In: Somasundaran, P. (Ed.), Fine Particles Processing Volume 1. Society of Mining Engineers of AIME, pp. 787–801 Chapter 40. Williams, D.R., 1980. The surface tension behaviour of aqueous solutions of oleic acid. University of Melbourne B.Sc. (Hons.) thesis. Yap, S.N., Mishra, R.K., Raghavan, S., Fuerstenau, D.W., 1981. The adsorption of oleate from aqueous solution onto hematite. In: Tewari, P.H. (Ed.), Proceedings, Adsorption from Aqueous Solutions. Plenum Press, New York, pp. 119–142. Yousef, A.A., 1967. Collector adsorption and recovery in the flotation of manganese ores. Min. Miner. Eng. June, 229–232. Yu, F., Wang, Y., Zhang, L., Zhu, G., 2015. Role of oleic acid ionic-molecular complexes in the flotation of spodumene. Miner. Eng. 71, 7–12. Zuidema, H.H., Waters, G.W., 1941. Ring method for the determination of interfacial tension. Indus. Eng. Chem. 13 (5), 312–313.

75–80 CA 21,1041. Miles, G.D., Ross, J., 1944. Foam stability of solutions of soaps of pure fatty acids. J. Phys. Chem. 48, 280–290. Moon, K.S., Fuerstenau, D.W., 2003. Surface crystal chemistry in selective flotation of spodumene (LiAl[SiO3]2) from other aluminosilicates. Int. J. Mineral Processing 72, 11–24. Muganda, S., Zanin, M., Grano, S.R., 2011. Influence of particle size and contact angle on the flotation of chalcopyrite in a laboratory batch flotation cell. Int. J. Miner. Proc. 98, 150–162. Mular, A.L., Roberts, R.B., 1966. A simplified method to determine the isoelectric points of oxides. Canadian Inst. Min. Bulletin. 59 (655), 1329–1330. Mullin, C.E., 1928. The pH and detergent value of soap solutions. The Textile Recorder 46, 65–66 (June 15, 1928) and 59-61 (July 14, 1928). Muratoglu, R.A., Hosten, C., Topkaya, Y. and Akser, M., 2000, Floatability of zircon with sodium oleate, in: P. Massacci (Ed.), Proceedings, XXI international mineral processing congress, Rome, Italy, vol. B. pp. B8b109-115. Natarajan, K.A., 1978. Statistical design and analysis of a flotation process. Trans. Indian Inst. Metals 31 (3), 169–176. Nunez, F.A.A., Yalkowsky, S.H., 1997. Foaming activity and pKa of some surface active compounds. Int. J. Pharmaceut. 151, 193–199. Ofor, O., 1995. Oleate adsorption at a Nigerian hematite-water interface: effect of concentration, temperature and pH on adsorption density. J. Colloid and Interface Sci. 174, 345–350. Oko, M.U., Salman, T., 1966. Adsorption of fatty acid soaps on hematite. Canadian Min. J. March, 61–66 72. Padday, J.F., 1969. Surface tension. Part II. Measurement of surface tension. In: In: Matijevic, E. (Ed.), Surface and Colloid Sci. Vol. 1. Wiley, pp. 101–149. Palmer, B.R., Fuerstenau, M.C., Aplan, F.F., 1975. Mechanisms involved in the flotation of oxides and silicates with anionic collectors. Part 2. Trans. Soc. Min. Engineers of AIME 258, 261–263. Pavez, O., Peres, A.E.C., 1993. Flotation of monazite-zircon-rutile with sodium oleate and hydroxamates. Proceedings, XVIII international mineral processing congress, Sydney, Australia 4, 1007–1012. Peck, A.S., Raby, L.H., Wadsworth, M.E., 1966. An infrared study of the flotation of hematite with oleic acid and sodium oleate. Trans. Soc. Min. Engineers of AIME 235, 301–307. Perez, J.W. and Aplan, F.F., 1975, Ion and precipitate flotation of metal ions from solution, in: P. Somasundaran, R.B. Grieves (Eds.), Advances in interfacial phenomena of particulate/solution/gas systems; applications to flotation research, American Institute of Chemical Engineers Symposium Series, 71, No. 150, pp. 34–39. Powney, J., 1935. The properties of detergent solutions. Part 1- The influence of hydrogen ion concentration on the surface tension of soap solutions. Trans. Faraday Soc. 31, 1510–1521. Pugh, R., Stenius, P., 1985. Solution chemistry studies and flotation behaviour of apatite, calcite and fluorite minerals with sodium oleate collector. Int. J. Min. Proc. 15, 193–218. Pugh, R.J., 1986. The role of the solution chemistry of dodecylamine and oleic acid collectors in the flotation of fluorite. Colloids and Surfaces 18, 19–41. Pyke, B., Fornasiero, D., Ralston, J., 2003. Bubble particle heterocoagulation under turbulent conditions. J. Colloid and Interface Sci. 265, 141–151. Quast, K.B., 1999. Flotation of hematite using oleates as collectors. Proceedings of the AusIMM 304 (1), 7–13. Quast, K., 2000. A review of hematite flotation using 12-carbon chain collectors. Miner. Eng. 13 (13), 1361–1376. Quast, K., 2006. Flotation of hematite using C6–C18 saturated fatty acids. Miner. Eng. 19, 582–597. Quast, K. and Quast, B., 2010, Haematite recovery from a tailings stream, in: Proceedings, XXV international mineral processing congress, Brisbane, Queensland, Australia, pp. 1155–1164. Quast, K., 2012a. Effect of 25% goethite on the hydrophobicity and oleate flotation of hematite. Int. J. Min. Eng. Min. Proc. 1 (2), 31–37. Quast, K., 2012b. Some surface characteristics of six hematite ores from the Middleback range area South Australia. Int. J. Min. Eng. Miner. Proc. 1 (2), 73–83. Quast, K., 2015. Use of conditioning time to investigate the mechanism of interactions of

Further reading Shaw, D.J., 1992. Introduction to Colloid and Surface Chemistry, 4th. Edition. Butterworth-Heinemann, Oxford. Adamson, A.W., Gast, A.P., 1997. Physical chemistry of surfaces. Wiley-Interscience, New York.

82