The interaction of clay minerals and saline water in coarse coal flotation

Fuel 134 (2014) 326–332

Contents lists available at ScienceDirect

Fuel journal homepage: www.elsevier.com/locate/fuel

The interaction of clay minerals and saline water in coarse coal flotation Bo Wang, Yongjun Peng ⇑ School of Chemical Engineering, The University of Queensland, St Lucia, 4072 QLD, Australia

h i g h l i g h t s  A synergistic interaction occurred between saline water and clay minerals in coal flotation.  Saline water promoted the formation of aggregates of clay platelets.  These aggregates sustained in the dynamic flotation condition and entered the concentrate.  They altered the froth property and recovery of coarse coal by true flotation.  Clay minerals may have a beneficial effect in some flotation scenarios.

a r t i c l e

i n f o

Article history: Received 2 April 2014 Received in revised form 27 May 2014 Accepted 28 May 2014 Available online 11 June 2014 Keywords: Saline water Coal Clay mineral Flotation Froth stability

a b s t r a c t In this study the fundamental science underpinning the interaction of saline water and clay minerals was studied under flotation contexts to address important problems confronting the coal industry. For the first time, the beneficial effect of clay minerals in some flotation scenarios was revealed after testing a typical coal sample having a low content of clay minerals and its mixture with another coal sample having a high content of clay minerals in de-ionised water and saline water with medium conductivity. Equipped with a range of techniques including froth stability measurements, modelling true flotation, settling tests and Cryo-SEM analyses, it was found that froth stability was higher in the flotation with a higher concentration of clay minerals in both de-ionised water and saline water, corresponding to increased combustible matter recovery and ash recovery. A synergistic interaction was evident between saline water and clay minerals in stabilising the froth and recovering combustible matter by true flotation. This is because saline water promoted the formation of association of clay platelets that sustained in the dynamic flotation condition, entered the flotation concentrate and altered the froth property and coarse coal flotation behaviour. Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction Clay minerals often associate with valuable minerals in a great range of ore bodies and have a substantial impact on the efficiency of many processes including grinding, pumping and dewatering, etc. when valuable minerals are extracted [1–3]. In mineral flotation, a number of studies have demonstrated that clay slime coatings can occur on mineral surfaces through the electrostatic attraction, reduce surface hydrophobicity and then depress mineral flotation [3]. These studies explore the anisotropic charges on edges and basal faces of clay minerals which are attracted to either negatively or positively charged minerals. In the previous study, we investigated the behaviour of clay minerals in fine coal ⇑ Corresponding author. Address: School of Chemical Engineering, The University of Queensland, St Lucia, Brisbane, QLD 4072, Australia. Tel.: +61 7 3365 7156. E-mail address: [email protected] (Y. Peng). http://dx.doi.org/10.1016/j.fuel.2014.05.085 0016-2361/Ó 2014 Elsevier Ltd. All rights reserved.

flotation and identified their entrainment and entrapment in the flotation process [4]. This study explores the nature of colloid size of clay minerals which promotes their entrainment through water films between air bubbles and entrapment through the aggregation of fine coal particles. However, all previous studies did not take into account another important nature of clay minerals, the formation of network structures of clay platelets in aqueous solutions. Clay minerals are anisotropic phyllosilicates and have a layer structure comprised of edges (E) and faces (F), and therefore clay platelets can associate in edge–edge (E–E), edge–face (E–F), and face–face (F–F) configurations. It is known that clay minerals have a basal permanent negative charge caused by isomorphous substitution, whilst the charge on the edges is either positive or negative depending on the pH [5–8]. Recent work has reported that the two basal plane surfaces of kaolinite can have positive or negative charges as well [9]. The anisotropic structure and charge properties

B. Wang, Y. Peng / Fuel 134 (2014) 326–332

lead to the association of clay platelets. The easy formation of the different types of association depends on the balance of electrostatic interactions (attractive or repulsive), which are controlled by the chemistry of the suspension, and the attractive van der Waals forces between the particles [8,10]. In dispersions with high clay mineral contents and a high negative edge charge density of the particles, the strong repulsion between the faces can also force the particles into adopting a certain parallel orientation promoting edge(–)/edge(–) aggregation and thereby altering the rheology behaviour [9]. In dilute clay mineral suspensions, the viscosity increases with E–E and E–F association whilst the viscosity decreases when the particles become thicker by F–F association. In concentrated clay mineral suspensions, E–E or E–F association leads to the formation of continuous, linked, card-house structures characterized by non-Newtonian flow. If F–F association occurs simultaneously, the number of units building the card-house structures is reduced [6]. The flotation of clayey ores is also complicated by the use of saline water. In Australian, all coal sites have introduced water reuse as a conventional practice due to scarcity of fresh water and stringent policy on the quality of water that can be discharged into local river systems. One of the consequences of increased water reuse is a concomitant increase in salinity on sites and subsequently in flotation. A number of studies have been conducted to investigate the effect of saline water on coal flotation. In general, saline water increases combustible recovery compared to fresh water and three mechanisms have been proposed to explain this phenomenon. Klassen and Mokrousov [11] and Blake and Kitchener [12] proposed that the lowering of electrokinetic potential of coal surfaces by the adsorbed ions reduced the stability of the hydration layers facilitating their disruption. The destabilisation of hydration layers increased the surface hydrophobicity and therefore coal flotation. Secondly, it was proposed that the electrolytes increased the electrical repulsive forces at bubble surfaces, inhibiting bubble coalescence [13–18]. The reduced bubbles increase bubble–particle collision efficiency and the overall flotation efficiency. Thirdly, the compression of the electrical double layer in saline water was attributed to the opening of hydrophobic surface sites enhancing the thinning and rupture of the wetting film between the particles and bubbles which is a critical step in the formation of a stable particle–bubble aggregate leading to an increase in flotation [19]. In the previous study addressing fine coal flotation using saline water, we found that saline water reduced the bubble size increasing the entrainment of clay minerals, and also enhanced the aggregation of fine coal particles exacerbating the entrapment of clay minerals. It is interesting to know that saline water alters the association of clay platelets. Unlike other colloidal dispersions, well-dispersed clay minerals may be coagulated by very low concentrations of inorganic salts and the critical coagulation concentration varies with clay mineral, electrolyte and pH [20]. It has been found that the edge of clay minerals (e.g., montmorillonite) is small relative to the Debye length at the critical salt concentration and the negative double layer extending from the basal plane surfaces spills over into the edge region [21]. The influence of the negative face charges is significant at the critical salt concentration (e.g., less than 103 M sodium salt concentration) and coagulation therefore occurs between edges(–) and faces(–) [22]. When the increased salt concentration required for edge(–)/face(–) coagulation approximates the salt concentration for face(–)/face(–) aggregation, the dispersion coagulates face-to-face because the area between two faces is larger than between an edge and a face [20]. Face/face aggregation between two layers or particles might be initiated at surface regions with lower than average charge density because of layer charge heterogeneity [23,24]. The objective of this study is to understand the interaction of clay minerals and saline water and its effect on coarse coal

327

flotation in parallel with the previous study on the effect of clay minerals on fine coal flotation [4]. The difficulty in floating coarse particles is another challenge confronting the coal and minerals industry. In coal flotation, there is a linear decrease in combustible recovery with increasing the particle size in the range of 127– 505 lm [25]. Two main reasons have been attributed to the poor recovery of coarse particles, inefficient collision of coarse particles and bubbles failed to generate stable aggregates, and insufficient buoyancy failed to float the heavy particle aggregates [26,27]. During flotation, effective collecting of the particles occurs when the collision of particles and bubbles generates a stable bubble– particle aggregate that rises to the surface of the pulp. However, for coarse particles, the process of attachment to an air bubble is limited by the acceleration of the bubble–particle complex, because if it exceeds a critical value, the particle will detach [28,29]. The probability of detachment increases as the particle size increases [30]. Besides, coarse particles are too heavy. The downward gravity force component dominates the whole force balance, leading to the settlement of coarse particles through the froth [31]. 2. Materials and methods 2.1. Raw materials A low-clay-content coal (LCC) and a high-clay-content coal (HCC) were supplied from a coal mine at Central Queensland, Australia, and examined in this study. LCC is currently processed in the plant to produce high quality hard coking coal for use in the international and domestic steel industry, whilst HCC cannot be processed economically due to the high clay content. The quantitative XRD analysis of these two samples as flotation feeds is indicated in Table 1. The two samples contain the same clay minerals, kaolinite and illite/smectite which are the major mineral matter. The LCC contains about 83% combustible matter and 7% clay minerals, while HCC contains about 61% combustible matter and 33% clay minerals. Artificial saline water was prepared and used in this study following the study by Ofori et al. [32] who conducted a survey of Australian coal preparation plants regarding the quality of process and make-up water. They summarised the data by giving the minimum, medium and maximum values of the concentration of major ions to give an indication of variation in composition of process water. In this study saline water with the medium value of the concentration of major ions was made and its compositions are shown   in Table 2. The major ions are Na+, Mg2+, Ca+, K+, SO2+ 4 , Cl and HCO3 . They were dissolved in de-ionised water. For a comparison, de-ionised water was examined as well throughout the study. MIBC (Methyl Isobutyl Carbinol) and Diesel, industrial grade, were used as frother and collector, respectively, in coal flotation. They are normally used in coal preparation plants in Australia. 2.2. Flotation Both LCC and HCC were screened to 710 lm following the bore core procedure of the coal mine. In the coal plant, the particles greater than 710 lm are subjected to gravity separation, while the particles smaller than 710 lm are floated. Previous studies in the plant indicated that minerals were well liberated in both feeds to gravity and flotation circuits. The cut-off size for gravity and flotation separation was chosen primarily based on the size limit of gravity devices. The particle size distribution of the flotation feed of LCC with 7% clay minerals and HCC with 33% clay minerals is showed in Fig. 1. In the plant, LCC is referred to as coarse coal with about 51% particles greater than 150 lm, while HCC as ‘‘clay’’ due

328

B. Wang, Y. Peng / Fuel 134 (2014) 326–332

Table 1 Mineral concentrations of LCC and HCC samples analysed by XRD (wt.%). Sample

Amorphous

Quartz

Kaolinite

Calcite

Dolomite

Siderite

Pyrite

Mixed clay illite/smectite

LCC HCC

84.7 63.1

5.1 1

3.8 18

1 –

0.1 –

1.5 2.6

0.3 0.8

3.6 14.5

Table 2 The composition of the saline water (mg/L). Conductivity (mS/cm)

Ca+

Mg2+

Na+

K+

HCO 3

SO2 4

Cl

6.04

93.0

126.8

1136.8

20.9

726.2

1177.8

1020.3

to the high clay content. In this study, besides floating LCC and HCC separately, 80% LCC and 20% HCC were mixed to make a coal sample having 12% clay minerals (medium-clay-content-coal, MCC) and floated to study the effect of a higher proportion of clay minerals on coarse coal flotation and whether the two clayey samples can be blended and then floated. The current view is that clay minerals have a deleterious effect on the flotation and any higher clay mineral content will exacerbate the flotation. After screening, the pulp was transferred to a 2.5 L JK batch flotation cell and then conditioned with collector (90 g/t) and frother (130 g/t) at an agitation speed of 700 rpm. The solid percentage in the flotation cell was about 15% which is consistent with the plant condition. The flotation pH was constant about 9.0 when saline water was used due to the buffer effect of saline water. When de-ionised water was used, NaOH solution was used to justify the flotation pH at 9.0. When flotation commenced, four concentrates were collected after cumulative times of 1, 2.5, 5, and 10 min with an air flow rate of 2.0 L/min. The flotation froth was scraped every 15 s. Flotation products were combusted at 815 °C for 2.5 h to obtain the ash and combustible contents [33]. 2.3. Froth measurement Froth images and froth stability was obtained by using VisioFroth software developed by Metso Minerals Cisa. The camera was set up above the flotation cell and connected with a computer. The measurement area was 15.9 cm  15.9 cm. The camera signals were relayed back via an ethernet cable and an optical fibre to the central processing computer. When flotation commenced, the camera recorded the froth every second automatically. The recorded images were then processed manually for calculating the froth stability. Froth stability calculated by VisoFroth is the % similarity between successive frames indicated by correlation peaks which are affected by bubble coalescence. High froth stability is indicated by high % similarity of successive frames. The detailed procedure has been described elsewhere [34]. 100

Volume (%)

80 60 40 20

100

200

300

400

500

600

Settling tests were conducted with 15 wt.% density of LCC, MCC and HCC using both de-ionised water and saline water in line with the flotation. Settling tests provide indirect information on the association of clay platelets and the rheology property of clay suspensions. In this study, the slurry was transferred to a 500 mL graduated cylinder after conditioning at pH 9.0. The cylinder was then stoppered and inverted for 5 times to ensure that the slurry was well mixed. Long-term settling tests were carried out and the thickness of the interface was recorded. 2.5. Cyro-SEM analysis Traditional SEM analysis involves drying samples which may alter the aggregate structures. In this study, Cryo-SEM, an in situ technique was used to directly detect the association of clay platelets that entered flotation concentrates. The cryo-transfer method of sample preparation was used to avoid a structural change caused by surface tension during oven or freeze drying. In the cryo-vitrification SEM analysis, the sample was taken by a largeaperture (>2 mm) pipette from flotation concentrates. The samples were then mounted onto the top of a 3 mm long brass rivet with outer-diameter 2.4 mm and inner-diameter 1.7 mm. This brass rivet was fixed on a sample holder and plunged into liquid nitrogen of the cryo-vitrification unit, which reduces the temperature at >800 °C min1 freezing the water without allowing crystallization to ice structures, i.e. vitrifying. The small volume of the sample (about 0.01 cm3) and high heat conductivity of brass minimize shrinkage and distortion of the sample during freezing. The very fast vitrification process avoids crystallization of water to ice and associated volume changes that can alter structures [35]. The sample was then transferred under vacuum to the sample preparation chamber equipped with an Oxford Instrument where the frozen sample was fractured to expose a fresh surface. Then the sample temperature was raised to 173 K (100 °C) to sublimate vitrified water for 8 min. This sublimation process removes fine vitrified water slivers generated during fracture and allows mineral structures to stand out above the glassy background. The sample was eventually coated with gold plasma for 1.5 min to avoid charging during the imaging process by a PHILIPS XL30 field emission gun scanning electron microscope (FESEM) normally operated at 15 kV. The sample was then examined in the SEM. Images were taken in backscattered electron mode (BSE), while elemental analysis was performed by energy dispersive spectrometry (EDS).

LCC

3. Results and discussion

HCC

3.1. Flotation

0 0

2.4. Settling tests

700

800

Particle size (µm) Fig. 1. The cumulative size distribution of LCC and HCC samples.

The flotation of LCC, MCC and HCC was conducted first in deionised water. Combustible recovery and ash recovery as a function of flotation time are shown in Fig. 2. The flotation of HCC was very

329

B. Wang, Y. Peng / Fuel 134 (2014) 326–332

80

LCC-Ash

LCC-Comb.

MCC-Ash

MCC-Comb.

HCC-Ash

HCC-Comb.

100

Combustible or ash recovery (%)

60

40

20

80

60

40

LCC-Ash

LCC-Comb.

MCC-Ash

MCC-Comb.

HCC-Ash

HCC-Comb.

20

0

0 0

2

4

6

8

0

10

2

poor with only 8% mass recovery at the end of flotation. The corresponding combustible recovery and ash recovery were 22% and 5%, respectively. This is consistent with the observation in the plant that the flotation of this high-clay-content coal is not effective with existing flotation conditions. It seems that the high clay mineral content caused high pulp viscosity inhibiting the mobility of bubbles and particles in the flotation cell and therefore limiting both true flotation and gangue mineral entrainment. The flotation of LCC was much better with 61% mass recovery at the end of flotation. The combustible recovery and ash recovery were 70% and 13%, respectively, in line with flotation results in the plant. The flotation of MCC was further increased with 72% combustible recovery and 14.5% ash recovery at the end of flotation. Then, the flotation of LCC, MCC and HCC was conducted in saline water. Combustible recovery and ash recovery as a function of flotation time are shown in Fig. 3. Again, the flotation of HCC was very poor in saline water although the combustible recovery was increased to 28%. However, saline water significantly increased the flotation of LCC and MCC with combustible recovery being more than 80%. A comparison of Figs. 2 and 3 indicates that a synergy occurred between saline water and clay minerals in increasing both combustible recovery and ash recovery. As shown in these figures, when the clay content was increased from 7% in LCC to 12% in MCC, combustible recovery increased by 2% and ash recovery by 1.8% when de-ionised water was used. When saline water was used, combustible recovery increased by 6% and ash recovery increased by 7%. In fact, the additional clay minerals facilitated the recovery of both combustible and mineral matter in particular in saline water. Flotation results were strongly related to the froth stability which is shown in Fig. 4. The higher the froth stability, the higher the combustible recovery and ash recovery. Froth stability was higher in saline water than in de-ionised water when the same coal sample was floated. This is consistent with previous studies. As discussed earlier, saline water may inhibit bubble coalescence [13– 18]. The reduced bubbles intent to produce more stable froth. Kurniawan et al. [36] studied the froth properties in coal flotation in MgCl2, NaCl, and NaClO3 solutions in the absence and presence of frother. They found that all the three salts produced more stable froth and smaller bubbles compared to fresh water and MgCl2 had the most pronounced effect. A synergy between saline water and clay minerals was also observed in the froth stability in flotation. As shown in Fig. 4, when the clay content was increased from 7% in LCC to 12% in MCC, froth stability increased by 2.3% when deionised water was used. When saline water was used, froth stability increased by 5.3%. Bulatovic et al. [37] found that the presence of clay minerals in the flotation of a copper ore was the main reason for the formation

6

8

10

Fig. 3. Combustible recovery and ash recovery as a function of flotation time from the flotation of LCC, MCC and HCC samples using saline water.

30

Stability-DI

100

Stability-SW

Dynamic froth stability (%)

Fig. 2. Combustible recovery and ash recovery as a function of flotation time from the flotation of LCC, MCC and HCC samples using de-ionised water.

4

Flotation time (min)

Flotation time (min)

25

Comb.-DI

80

Ash-DI

20

Comb.-SW Ash-SW

60

15 40 10 20

5

Combustible or ash recovery (%)

Combustible or ash recovery (%)

100

0

0 LCC

MCC

HCC

Fig. 4. Froth stability, combustible recovery and ash recovery from the flotation of LCC, MCC and HCC samples using de-ionised and saline water.

of unstable froth and reduced frothing power. The unstable froth with about 4% stability was observed in this study when HCC was floated. However, stable froth occurred when LCC was floated and additional clay minerals enhanced froth stability as seen in the flotation of MCC. In fact, clay minerals are considered to be responsible for the overly stable froth in the flotation of bitumen since the hydrophilic clay mineral particles together with flotation surfactants form a rigid film at the water–oil interface [38]. There is a great similarity in the flotation of coal and bitumen in that aliphatic surfactants are added in the flotation and the valuable matter is strongly hydrophobic. It is therefore possible that clay minerals promote the formation of a rigid film at the water–coal interface resulting in higher froth stability in coal flotation. Froth stability plays an important role in coarse particle flotation. Higher froth stability offsets the downward gravity force of coarse particles and reduces their drainage from froth thereby increasing coarse particle flotation. However, higher froth stability also increases mechanical entrainment which is a transfer process by which mineral particles suspended in water enter the flotation froth, move upwards, and finally leave the flotation cell [39]. In mineral flotation, true flotation by the collision of hydrophobic particles with bubbles followed by the rising of bubble–particle aggregates from the pulp phase to the froth phase and eventually the transport of bubble–particle aggregates into the concentrate is the primary mechanism for selective separation of valuable minerals from gangue minerals. In this study the true flotation of combustible matter was calculated to evaluate the net benefit of saline water and clay minerals in the coarse coal flotation. Results are

330

B. Wang, Y. Peng / Fuel 134 (2014) 326–332

shown in Fig. 5. Combustible recovery by true flotation was calculated based on the following equations [39]:

ENT ¼



xC xT

 ð1Þ

1  Rovr  ENT  Rw 1  RW

ð2Þ

RTrue flotation ¼ Rovr  RENT

ð3Þ

RENT ¼

where ENT is the degree of entrainment which uses the water as a reference to define the classification effect of the drainage of entrained particles in the froth phase, xC and xT are the mass flow rate of the tracer (a fully-liberated or non-floating mineral) in the concentrate and in tailings respectively, and Rovr, RENT and RTrue flotation are overall recovery, recovery by entrainment and recovery by true flotation, respectively. In this study, clay minerals are the predominant mineral mater. They have colloid sizes and are liberated in flotation. As a result, xC and xT were estimated by the mass flow rates of ash. Fig. 5 indicates that the combustible recovery by true flotation was only 4–6% in the flotation of HCC using either de-ionised or saline water, which suggests that the true flotation process, the collision of hydrophobic particles with bubbles followed by the rising of bubble–particle aggregates, was almost completely inhibited. When LCC was floated, true flotation was the main mechanism and combustible recovery by true flotation was increased from 46% in de-ionised water to 67% in saline water. Obviously electrolytes in saline water contributed to the increase of true flotation significantly. Compared to LCC, the flotation of MCC resulted in a more increase (from 50% to 78%) in combustible recovery by true flotation, which indicates the beneficial effect of the additional clay minerals. The synergy between saline water and the additional 5% clay minerals resulted in 7% more combustible recovery by true flotation. 3.2. Association of clay platelets in de-ionised water and saline water

Combustible recovery by true flotation (%)

The different flotation behaviour of these clayey coal samples in de-ionised water and saline water may be linked with the association of clay platelets. Higher recovery of clay minerals may be obtained if they enter the froth as aggregates. This in turn increases froth stability and combustible recovery. The association of clay platelets was investigated first by settling tests and then CryoSEM analysis. Fig. 6 shows the photograph of sediment of LCC, MCC and HCC suspensions in de-ionised water after 0.5 h of settling. It was found

100

80

LCC-DI

LCC-SW

MCC-DI

MCC-SW

HCC-DI

HCC-SW

60

40

20

0 0

2

4

6

8

10

Flotation time (min) Fig. 5. Combustible recovery by true flotation as a function of flotation time from the flotation of LCC, MCC and HCC samples using de-ionised and saline water.

that the sediment thickness decreased with time while the interface between the sediment and supernatant was vague. Nasser and James [10] indicated that when clay particles are associated, the van der Waals and other attractive colloidal forces were dominant and the aggregated particles settled together forming a sharp interface between the sediment network and supernatant, while when the particles settled in a dispersed form the gravitational force was the governing force, particles settle individually and there was no clear distinct interface between the accumulated sediment bed and overlying water. In this current study, flotation and settling tests were conducted at pH 9 and both the edges and faces of clay platelets are negatively charged in de-ionised water [40], resulting in repulsion among them. The interface in Fig. 6(A) and (B) might be caused by the settling of coal particles while clay particles were mainly dispersed. Fig. 6 also indicates that with the proportion of clay minerals, the settling rate of the suspension decreased. For example, after 0.5 h, LCC suspension settled to 150 mL from 500 mL but MCC suspension only settled to 380 mL. It is likely that the high particle concentration reduced the mobility of the clay particles and the strong repulsion between the faces of clay platelets forced the particles into adopting E–E association [9]. A higher clay particle concentration promoted more E–E association of clay platelets which may result in higher viscosity of the suspension with slower settling of coal particles. The settling rate of HCC suspension was extremely slow and the sediment barely formed in 0.5 h. This indicates strong E–E association and high viscosity resulting in poor flotation behaviour in terms of both entrainment and true flotation. Fig. 7 shows photograph of sediment of LCC, MCC and HCC suspensions after 0.5 h of setting in saline water. Again, the sediment thickness decreased with time but a clear interface occurred between the sediment and supernatant, which indicates the settling in an associated form. Compared to the settling in de-ionised water, saline water facilitated the settling of each coal suspension. It has been found that F–F aggregates have a denser structure than E–E aggregates resulting in lower viscosity and faster settling, while E–F aggregates have a lower density with more intra-aggregation water trapped compared with E–E aggregates resulting in higher viscosity and slower settling [10]. Obviously, saline water promoted the F–F association in the coal suspensions in this study. It has been known that F–F association of clay platelets is produced in salt solutions at the critical salt concentration and the increased salt concentration produces F–F association [20]. However, the faster settling of F–F aggregates in saline water should cause less entrainment in flotation than the slower settling of E–E aggregates or dispersed particles in de-ionised water. This is opposite to the observation in flotation where entrainment as indicated by ash recovery was higher in saline water than in de-ionised water. It should be noted that flotation is different from the settling in that agitation and aeration make the flotation a dynamic process which may break some aggregates forming in settling tests. This aspect was examined by Cryo-SEM measurements. Fig. 8 shows SEM BSE images and EDS analyses on the concentrate from the flotation of MCC in de-ionised water and saline water. The particles chosen for EDS analysis were indicated by the arrows. As shown in Fig. 8, when de-ionised water was used, clay minerals were recovered to the flotation concentrate in a dispersed form without clear association of clay platelets. However, when saline water was used, clay minerals were recovered in either F–F or E–E associated forms. The presence of the clay minerals in these forms was confirmed by the EDS analysis which detected the Si and Al signals clearly. In the settling tests, E–E association of clay platelets was predicted in de-ionised water. However, this association as a result of the restriction of particle mobility could not be sustained in the dynamic flotation condition.

B. Wang, Y. Peng / Fuel 134 (2014) 326–332

Fig. 6. Photograph of sediment of coal suspensions after 0.5 h of setting in de-ionised water: (A) LCC, (B) MCC and (C) HCC.

Fig. 7. Photograph of sediment of coal suspensions after 0.5 h of setting in de-ionised water: (A) LCC, (B) MCC and (C) HCC.

Fig. 8. SEM BSE images (top) and EDX analysis (bottom) on the concentrate from the flotation of the MCC sample in de-ionised water (A) and saline water (B).

331

332

B. Wang, Y. Peng / Fuel 134 (2014) 326–332

In contrast, the E–E and F–F association in saline water as a result of the attractive van der Waals forces and the compression of electrical double layers [10] was stronger and went through the flotation process. These aggregates resulting from the interaction of saline water and clay minerals not only increased the recovery of clay minerals by entrainment, but also increased froth stability and true flotation of coarse particles. However, when the pulp viscosity was sufficient high, the mobility of E–E and F–F aggregates or dispersed particles was limited resulting in poor flotation as observed in the flotation of HCC. 4. Conclusions In coarse coal flotation, a synergistic interaction between clay minerals and saline water occurred and affected the recovery by entrainment and true flotation. At pH 9, saline water facilitated the formation of clay aggregates which sustained in the dynamic flotation process and resulted in the recovery of more clay minerals. This in turn increased froth stability and promoted higher combustible recovery by true flotation. Flotation of the highclay-content coal was poor in terms of both entrainment and true flotation as a result of high pulp viscosity limiting bubble and particle mobility. Acknowledgements The authors greatly appreciate financial support from the Australian Coal Industry’s Research Program (ACARP), and New Start-up Grant awarded to Dr. Yongjun Peng by the University of Queensland as well as discussions and suggestion from Frank Mercuri and John Gartlan at Xstrata Coal and Ian Brake, Ben Cronin and Susan Watkins from BHP Billiton Mitsubishi Alliance (BMA). Thanks also to the scientific and technical assistance of the Australian Microscopy and Microanalysis Research Facility at the University of Queensland on the coal surface analysis. References [1] Arnold BJ, Aplan FF. The effect of clay slimes on coal flotation, part I: the nature of the clay. Int J Miner Process 1986;17:225–42. [2] Arnold BJ, Aplan FF. The effect of clay slimes on coal flotation, part II: the role of water quality. Int J Miner Process 1986;17:243–60. [3] Zhao S, Peng Y. The oxidation of copper sulfide minerals during grinding and their interactions with clay particles. Powder Technol 2012;230:112–7. [4] Wang B, Peng Y. The behaviour of mineral matter in fine coal flotation using saline water. Fuel 2013;109:309–15. [5] Bergaya F, Theng BKG, Lagaly G. Handbook of clay science. Burlington: Elsevier Science; 2006. [6] van Olphen H. An introduction to clay colloid chemistry: for clay technologists geologists, and soil scientists. 2nd ed. New York, London, Sydney: Wiley; 1977. [7] Luckham PF, Rossi S. The colloidal and rheological properties of bentonite suspensions. Adv Colloid Interface Sci 1999;82:43–92. [8] Lagaly G, Ziesmer S. Colloid chemistry of clay minerals: the coagulation of montmorillonite dispersions. Adv Colloid Interface Sci 2003;100–102:105–28. [9] Cruz N, Peng Y, Farrokhpay S, Bradshaw D. Interactions of clay minerals in copper–gold flotation: Part 1 – rheological properties of clay mineral suspensions in the presence of flotation reagents. Miner Eng 2013;50–51: 30–7. [10] Nasser MS, James AE. Settling and sediment bed behaviour of kaolinite in aqueous media. Sep Purif Technol 2006;51:10–7.

[11] Klassen VI, Mokrousov VA. An introduction to the theory of flotation. 2nd ed. London: Butterworths; 1963. [12] Blake TD, Kitchener JA. Stability of aqueous films on hydrophobic methylated silica. J Chem Soc Faraday Trans 1972:68. [13] Marrucci G, Nicodemo L. Coalescence of gas bubbles in aqueous solutions of inorganic electrolytes. Chem Eng Sci 1967;22:1257–65. [14] Lessard RR, Zieminski SA. Bubble coalescence and gas transfer in aqueous electrolytic solutions. Ind Eng Chem Fundam 1971;10:260–9. [15] Craig VSJ, Ninham BW, Pashley RM. Effect of electrolytes on bubble coalescence. Nature 1993;364:317–9. [16] Henry CL, Dalton CN, Scruton L, Craig VSJ. Ion-specific coalescence of bubbles in mixed electrolyte solutions. J Phys Chem C 2007;111:1015–23. [17] Henry CL, Craig VSJ. The link between ion specific bubble coalescence and hofmeister effects is the partitioning of ions within the interface. Langmuir 2010;26:6478–83. [18] Henry CL, Craig VSJ. Ion-specific influence of electrolytes on bubble coalescence in nonaqueous solvents. Langmuir 2008;24:7979–85. [19] Laskowski JS. Coal flotation in solution with a raised concentration of inorganic salts. Colliery Guardian 1965;211:361–6. [20] Lagaly G. Colloid clay science. In: Bergaya F, Theng BKG, Lagaly G, editors. Handbook of clay science. Burlington: Elsevier; 2006. p. 141–245. [21] Sposito G, Grasso D. Electrical double layer structure, force, and fields at the clay water interface. In: Hsu JP, editor. Interfacial forces and fields: theory and applications. New York: Marcel Dekker; 1999. p. 207–49. [22] Secor RB, Radke CJ. Spillover of the diffuse double layer on montmorillonite particles. J Colloid Interface Sci 1985;103:237–44. [23] Keren R, Shainberg I, Klein E. Settling and flocculation value of sodiummontmorillonite particles in aqueous media. Soil Sci Soc Am J 1988;52:76–80. [24] Heller H, Keren R. Rheology of Na-rich montmorillonite suspension as affected by electrolyte concentration and shear rate. Clays Clay Miner 2001;49:286–91. [25] Humeres E, Debacher NA. Kinetics and mechanism of coal flotation. Colloid Polym Sci 2002;280:365–71. [26] Schulze HJ, Radoev B, Geidel T, Stechemesser H, Töpfer E. Investigations of the collision process between particles and gas bubbles in flotation – a theoretical analysis. Int J Miner Process 1989;27:263–78. [27] Weber ME, Paddock D. Interceptional and gravitational collision efficiencies for single collectors at intermediate Reynolds numbers. J Colloid Interface Sci 1983;94:328–35. [28] Trahar WJ, Warren LJ. The flotability of very fine particles – a review. Int J Miner Process 1976;3:103–31. [29] Dai Z, Fornasiero D, Ralston J. Particle–bubble collision models – a review. Adv Colloid Interface Sci 2000;85:231–56. [30] Sutherland KL. Physical chemistry of flotation. XI. Kinetics of the flotation process. J Phys Colloid Chem 1948;52:394–425. [31] Van Deventer JSJ, Van Dyk WA, Lorenzen L, Feng D. The dynamic behaviour of coarse particles in flotation froths. Part III: Ore particles. Miner Eng 2002;15: 659–65. [32] Ofori P, Firth B, McNally C, Nguyen A. Working effectively with saline water in coal preparation. CSIRO Energy Technology, the University of Queensland; 2009. [33] Ozdemir O, Taran E, Hampton MA, Karakashev SI, Nguyen AV. Surface chemistry aspects of coal flotation in bore water. Int J Miner Process 2009; 92:177–83. [34] Runge K, McMaster J, Wortley M, La Rosa D, Guyot O. A correlation between visiofroth™ measurements and the performance of a flotation cell. Perth: The Australasian Institute of Mining and Metallurgy (The AusIMM); 2007. p. 79– 86. [35] Battersby BJ, Sharp JCW, Webb RI, Barnes GT. Vitrification of aqueous suspensions from a controlled environment for electron microscopy: an improved plunge-cooling device. J Microsc 1994;176:110–20. [36] Kurniawan AU, Ozdemir O, Nguyen AV, Ofori P, Firth B. Flotation of coal particles in MgCl2, NaCl, and NaClO3 solutions in the absence and presence of Dowfroth 250. Int J Miner Process 2011;98:137–44. [37] Bulatovic SM, Wyslouzil DM, Kant C. Operating practices in the beneficiation of major porphyry copper/molybdenum plants from Chile: innovated technology and opportunities, a review. Miner Eng 1998;11:313–31. [38] Rao F, Liu Q. Froth treatment in Athabasca oil sands Bitumen recovery process: a review. Energy Fuels 2013;27:7199–207. [39] Savassi ON, Alexander DJ, Franzidis JP, Manlapig EV. An empirical model for entrainment in industrial flotation plants. Miner Eng 1998;11:243–56. [40] Liu D, Peng Y. Reducing the entrainment of clay minerals in flotation using tap and saline water. Powder Technol 2014;253:216–22.