Fine coal beneficiation by column flotation

Fuel Processing Technology 126 (2014) 49–59

Contents lists available at ScienceDirect

Fuel Processing Technology journal homepage: www.elsevier.com/locate/fuproc

Fine coal beneficiation by column flotation Oh-Hyung Han a, Min-Kyu Kim a, Byoung-Gon Kim b, Nimal Subasinghe c, Chul-Hyun Park c,⁎ a b c

Department of Energy & Resources Engineering, Chosun University, Gwangju 501-759, Republic of Korea Korea Institute of Geoscience and Mineral Resources, Daejeon 305-350, Republic of Korea Department of Metallurgical & Minerals Engineering, WASM, Curtin University, Kalgoorlie 6430, Australia

a r t i c l e

i n f o

Article history: Received 25 October 2013 Received in revised form 9 April 2014 Accepted 14 April 2014 Available online xxxx Keywords: Fine coal Column flotation Combustible recovery Water recovery Separation efficiency

a b s t r a c t The amenability of beneficiating a fine hard coal from Hwa-Sun coal mine using column flotation has been studied using a CoalPro flotation column developed by Canadian Process Technologies (CPT). After initial testwork using a batch flotation cell to determine the optimal flotation conditions, tests were carried out to compare its performance with the column. The results showed that the performance of the column was far superior to that of the conventional cell owing to the large fines content present in the coal. It was established that a stable froth column of 40 cm in height and a suitable balance between froth and collection zone could be maintained at 1.0 cm/s superficial gas rate. A suitable reagent scheme and the optimal operating conditions that minimized adverse effects generally encountered in treating fine coal, such as gangue entrainment, have been identified. An increase in water recovery significantly decreased ash rejection implying entrainment of fine ash particles. Column flotation is capable of producing an acceptable clean coal concentrate of 85% combustible recovery with 81% ash rejection at a maximum separation efficiency of 62%, compared to conventional flotation which has 70% recovery with 70% ash rejection at an efficiency of 42%. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Coal is one of the main fossil fuels that supplies energy of up to 30% worldwide and has been widely perceived as a strategic energy source particularly in countries with poor energy resources, due to rising oil and gas prices. Utilization of coal for energy production requires the removal of ash as well as hazardous materials such as SOx and NOx [1]. In order to clean low rank coal physical separation techniques, such as gravity separation, electrostatic separation and froth flotation have been used extensively owing to their low cost. Of these, froth flotation is the most common technology used for fine coal beneficiation. Froth flotation exploits the differences in surface hydrophobicity of the different constituent minerals and selectively separates the valuable minerals from gangue by attaching them to air bubbles and recovering them from the mineral laden froth [2]. Conventional flotation that uses mechanical cells has been known to be ineffective in processing fine coal mainly due to entrainment of fine gangue minerals in the froth that requires complex circuit arrangements that incorporate several cleaner stages [3]. In contrast, column flotation has been effective in cleaning fine coal as it has many advantages over conventional flotation owing to its ability to effectively reduce entrainment of fine gangue minerals due to less turbulence in the pulp, having a deep froth bed and using wash water to drain back the entrained gangue. In addition, column flotation is preferred due to its simpler construction, convenience in incorporating automatic control, as well as a having ⁎ Corresponding author. E-mail address: [email protected] (C.-H. Park).

http://dx.doi.org/10.1016/j.fuproc.2014.04.014 0378-3820/© 2014 Elsevier B.V. All rights reserved.

single-stage system which embodies rougher, cleaner, and scavenger [4,5]. Several studies also have emphasized that column flotation is superior to mechanical flotation in handling both coarse and fine fractions and gives a higher recovery with lower ash content [3,6–9]. In this study, an attempt has been made to identify the most significant parameters in designing flotation columns for processing a finely-sized low rank coal including water recovery that influences fine gangue entrainment.

2. Previous work The working mechanism of a typical flotation column and its important process variables are shown schematically in Fig. 1. Several studies on column flotation have focused on the froth stability and the pulp level, entrainment of fine particles, flotation rate, product grade/recovery relationships and carrying capacity. Goodall and O'conner [10] observed that decreasing air rate and increasing frother concentration reduce entrainment and produce stable froth together with small bubbles in lab-scale column flotation. Bias rate (Jb) which is the net flow of liquid down the froth column is related to the wash water rate and is of critical importance to the performance of the column. At high values it reduces the carrying capacity and at low rates it reduces grade. Thus, Finch and Dobby have recommended that Jb should be between 0 and 0.1 cm/s to reduce carrying capacity [11]. However, other researchers have recommended values between 0.1 and 0.4 cm/s in order to reduce gangue entrainment and achieve acceptable product grades [12,13]. Entrainment is a characteristic feature

50

O.-H. Han et al. / Fuel Processing Technology 126 (2014) 49–59

and compressed) air travels at high velocity through an hour-glass shaped tube [24]. CPT flotation columns have gravity-flow wash water system, multi-point feed distribution, internal product launders and self-closing sparger systems. Especially, the advantages of CPT spargers eliminate back-flow and plugging owing to a self-closing system and also capable of being mounted outside the column to simplify inspection and replacement [24]. Hence Canadian Process Technologies (CPT) CoalPro flotation column has been installed at over 500 diverse plants and extensively released in the world [4,24]. In this work, flotation tests on a low rank anthracitic hard coal from Hwa-Sun coal mine were carried out using lab scale CPT Coalpro column. Particularly, this study was focused on investigating variables such as superficial air rate, superficial wash water rate, feed solids concentration and several chemical reagents that affect column flotation with a view in identifying important variables for plant design. 3. Experimental

Fig. 1. Schematic diagram of flotation column.

of fine particle processing and is non-selective and bears no distinction between hydrophobic or hydrophilic particles [14]. The results of Kirjavainen [15] indicated that the entrainment of hydrophilic particles to the froth product is actually a statistical phenomenon in flotation tests using −45 μm quartz. Tao et al. [16] reported that in coal flotation, the presence of raw coal fraction between 30 and 150 μm stabilizes froth but that less than 30 μm destabilizes froth. Also, contact angles in excess of 90° have a tendency to destabilize the froth [16]. At high frother concentrations, the thickness of the bubble walls increase leading to lower bubble coalescence and thus, increases bubble stability [10]. The entrainment of particles in flotation is closely related to the water recovery [14]. Smith and Warren [17] have shown that entrainment becomes significant for particles smaller than about 30 μm, and it is proportional to the recovery of water into the froth products. On the other hand, understanding and implementation of column flotation have undergone a rapid growth in relation to column design, construction, air sparging and operation since 1990s [11]. In particular, the air sparging system which generates small bubbles is an essential technology to achieve a higher flotation rate. The types of sparger designs include 1) Porous media that uses pierced plastic or fabric such as filter cloth [12], 2) Static shear contacting devices that use metal mesh packing to break up the gas stream into bubbles [18], inline static mixers [13] or cyclones with vertical baffle, 3) Jetting devices which inject gas stream from an orifice into the slurry [19,20], and their improved designs such as slamjet or cavitation. In the coal industry, porous media spargers are still being used for basic laboratory studies but are no longer commonly used in commercial plants. Their disadvantage being the fact that the column must be shut down and drained prior to assessing or replacing sparger units, whereas both static shearing and jetting devices can be accessed more easily [21]. Main features of the more advanced sparger systems such as the microcel sparger system (Microcel™ column) are non-plugging ability, externally mounted for easy maintenance and also it can produce smaller bubbles (0.1 to 0.4 mm), which provide a higher flotation rate. It can also achieve a higher degree of turbulence inside the static mixer to produce smaller bubbles [13]. It has found wide applications in the coal industry in plants such as Alpha Inc. (Middle Fork etc., USA), Horizon Ltd. (Marrowbone, USA), and BHP (Peak downs and Saraji, Australia) [22]. Other commonly used type of sparger is the jetting type devices developed by the U.S. Bureau of Mines but it has been plagued with maintenance problems due to plugging of the orifices [23]. Recently, Eriez and Canadian Process Technologies (CPT) have developed CoalPro column flotation cell, which is equipped with Slamjet sparger that generates smaller bubbles using an air-lance and also Cavitation-Tube sparger using cavitation phenomena occurring when the slurry (pulp

3.1. Materials Coal sample (anthracite) used in this study was obtained from HwaSun Coal Mine in South Korea. It is a low rank coal which was rejected as waste from the main processing plant. On analysis it was revealed that the ash in this coal could be liberated at about 150 μm size. Thus it was decided to grind the raw coal to this size and test the amenability of removing the ash by flotation. 3.2. Equipments and procedures Preliminary tests to determine the effective reagent combination for flotation of this coal were carried out using a laboratory Denver sub-A batch flotation machine with 4 L cell. The minus 150 μm fraction was fed into flotation cell at 20% solids by mass and agitated for 5 min. The reagents tested were: collectors (kerosene and a patented collector developed for coal, DMU-101), frothers (MIBC, Dowfroth 250, Aerofroth 65 and pine oil) and depressants (Sodium metaphosphate (SMP) and Sodium silicate). These preliminary tests revealed that the optimal combination was DMU-101, MIBC and SMP. The laboratory CPT CoalPro column flotation system used is shown in Fig. 2. Features of this column are: 1) height and diameter of column; 1500 mm and 55 mm, 2) Sparger; porous HDPE or optional cavitation tube and 3) automatic control for level, sparger air and wash water. The column's upper section consists of wash water distributor, froth collection launder and feeder while the lower section houses the sparger (bubble generator) and tailings outlet. The feed slurry inlet is at a point of 1/3 of column height from the top. The wash water was fed at the top of the column while the tailings were collected at the bottom of the column. The concentrate and tailings products were collected and dried in the dryer at 105 °C for 24 h. The contents of ash, fixed carbon, volatile materials and moisture were measured using proximate analyzer (TGA601, LECO Ltd., USA). The composition of samples of coal and ash was analyzed using SEM, XRD and XRF techniques. The following procedures were followed in the preparation of samples for these analyses. 3.2.1. Scanning electron microscopic analysis A field emission-scanning electron microscope (FE-SEM, S-4800, Hitachi, Japan) equipped with an energy dispersive X-ray spectrometer (EDS, Link Isis 3.0, Oxford Instruments plc, U.K.) was used for this analysis of coal samples. Generally when using this equipment, carbon coating of the sample surface is required in order to make it conductive. However since the sample contains carbon a platinum coating was applied instead. First, each coal product sample was mixed with a volatile solvent alcohol (ethanol), dispersed and cleaned in an ultrasonic bath. The sample was then placed on a stub over a thin layer of silver (Ag)


51

Table 1 Proximate analysis of raw coal sample. Proximate analysis (%), dry basis Ash

Volatile

Fixed carbon

37.56

6.31

56.12

respectively. The XRD pattern on raw sample using Cu Kα target is shown in Fig. 3. It shows that the contents in ash comprised mostly of SiO2 and Al2O3 and amounts to about 37.5% by mass and deleterious elements such as sulfur and phosphorus were not detected. Main gangue minerals were silicate minerals which consist of quartz (SiO2), Muscovite (K(OHF2)2Al3Si3O10) and Kaolinite(Al2Si2O5(OH)4) which could be separated by froth flotation techniques. The results of wet screen analysis for raw sample used in the tests are given in Table 3. Mean size (d50) was 66 μm and the amount of minus 37 μm particles was about 26% by mass. A plot of weight % retained and fixed carbon distribution (%) of raw coal versus particle size (μm) is shown in Fig. 4. As evidenced by the fact that the distribution of coal and weight at all coarse sizes are nearly identical it may be concluded that the coal is uniformly distributed in coarser particle sizes and is liberated only at very small sizes, ie below about 20 μm. Thus, the separation of coal from ash gangue requires processing at fine sizes. 4.2. Performance measures In coal preparation, the performance of a separation device is generally determined by evaluating the combustible recovery (Rc), and ash rejection (θa) in the concentrate product. These are calculated using the following formula based on the well-known two product formula in mineral processing [25]. Fig. 2. Photograph and details of laboratory CPT CoalPro column flotation system.

Rc ¼

ð f−tÞð100−cÞ 100 ðc−tÞð100−f Þ

ð1Þ

θa ¼

ð f−cÞt 100 ðt−cÞf

ð2Þ

paste. After drying the sample, it was coated using a Pt electron E-1030 sputter for 1 min which yielded a platinum layer of 6–7 μm. The coated samples were used in the SEM analysis. 3.2.2. XRD analysis Samples for XRD analysis were prepared by grinding the ash samples to minus 50 μm sizes and inserted into a holder with cover glass and inserted in sample changer. The X-ray diffraction was recorded at 40 kV and 30 mA for a Cu-target tube. The samples were analyzed using an X-ray diffractometer (Rigaku D-max-2500PC, Rigaku/MSC, Inc., TX, USA).

where, Rc is the combustible recovery and θa is the ash rejection. Also, f, c and t are the ash contents of the feed, the clean coal and tailings, respectively. The overall separation efficiency (Se) of the process is defined as the recovery of valuables minus the recovery of gangue [25] which for coal separation is given as Se ¼ Rc −ð100−θa Þ:

3.2.3. XRF analysis Ash ground to minus 100 μm was mixed with Lithium tetraborate (1 g:5.5 g) in a mixer mill, molded in a Pt/Au (95%:5%) crucible of capacity 20 ㎖ and fused in a Fusion Glass Disc Casting furnace at 1100 °C for 10 min. An X-ray fluorescence spectroscopy (MXF-2400 Multi channel Spectrometer, SHIMADZU Co., Japan) was used for the analysis. Measurement was conducted at spectrometer atmosphere of vacuum mode and measuring rates of 40 kV, 70 mA, and 2.8 kW using X-Ray tube of End-window type with Rh target. X-ray intensities are converted into concentration values of the elements through the use of calibration curves predetermined using standard samples. The results of these tests are given in Section 4. 4. Results and discussion

ð3Þ

In this work, the above performance measures have been used to quantify the effects of variables discussed below. 4.3. Effect of significant variables In column flotation, separation of combustibles from ash is significantly affected by the variables relating to: 1) gas dispersion, such as superficial gas rate (Jg), gas hold up (εg), bubble size (db) 2) feed solid concentration (Cw) 3) pulp chemistry affected by chemical reagents such as collector, frother and regulators (e.g. depressant) and 4) machine operating parameters, such as the superficial wash water rate Table 2 Chemical analysis of ash.

4.1. Characterization of test coal

Chemical Component (%)

The proximate analysis of the raw coal sample and the chemical analysis of the ash sample used in this work are given in Tables 1 and 2,

SiO2

Al2O3

Fe2O3

CaO

MgO

K2O

50.28

32.15

4.99

1.85

5.59

4.25

52


Counts Q M K

1400 1200

K : Kaolinate(JCPDS card No, 14-0164) M : Muscovite(JCPDS card No, 19-0814,10-490) Q : Quartz(JCPDS card No, 33-1161)

1000 800 600 400

M K Q

K

M M

M K

K M

Q M M Q

M M

200

K

Q

Q M

M Q

0 10

20

30

40

50

60

70

80

90

Position [ ¨¬ 2Theta] (Copper(Cu)) Fig. 3. XRD pattern on raw coal sample using Cu Kα target.

(Jw), bias water rate (Jb), pulp level, froth height, and water recovery (Rw). These variables have been studied in this work using CPT column and their influence on the processing of the test coal is reported below. It must be noted that in column flotation it is customary to report the flowrates of water and gas as superficial rates (J), which is defined as the volumetric flowrate per unit area of column cross-section. It has units of velocity and sometimes referred to as superficial velocity. 4.3.1. Gas dispersion The nature of gas dispersion within a flotation column is of vital importance in determining its performance. Dobby and Finch [26] have identified three important variables that relate to gas dispersion, namely, bubble size, gas hold up and bubble surface area flux. Measurement of these variables is difficult. However, these quantities are related to the superficial gas rate (Jg) within the collection zone and the relationships are discussed below. 4.3.1.1. Superficial gas rate. The superficial gas rate affects the formation of bubbles, froth column and concentrate carrying capacity among others. It is defined as: Jg ¼ Volumetric air rate=cross‐sectional area of column:

quantified in terms of the bubble surface area of flux (Sb). The bubble surface area flux is related to the superficial gas velocity by the relationship: Sb ¼ 6

Jg : db

ð4Þ

The bubble surface area flux is directly proportional to the flotation rate constant in the collection zone. The effect of superficial gas rate on water recovery, combustible recovery and ash rejection in CPT column flotation is shown in Fig. 5. It can be seen that the combustible coal recovery increased consistently with increasing gas rate over the range tested. However, above 1.0 cm/s the ash rejection decreased significantly possibly due to the entrainment of fine gangue particles, caused by rapidly rising bubbles. At high superficial gas rates both bubble size and gas hold up tend to increase. Smaller bubbles destabilize the froth while larger mineral laden bubbles, make the froth more stable. Thus, considering combustible recovery and ash rejection, the optimum superficial gas rate was determined at 1.0 cm/s for this coal, which seems to maintain stable froth bed and good balance between froth and pulp.

It has been observed that the size of bubbles increases as air rate is increased [26]. As the size of bubbles decreases their rise velocity also decreases and in turn the gas hold up within the collection zone increases. The surface area of bubbles which affect the probability of collision and attachment with particles increases as the size of bubbles decreases. Gorain [27] has shown that the combined effect could be

Table 3 Wet screen analysis for raw coal sample. Size (micron)

Weight % retained

Cum. % passing

Fixed carbon dist. (%)

+149 −149 + 113 −113 + 74 −74 + 53 −53 + 44 −44 + 37 −37 + 25 −25 Total

0.36 17.44 33.22 9.73 8.22 4.84 10.06 16.13 100.0

99.64 82.2 48.98 39.25 31.03 26.19 16.13 0

0.34 17.54 33.69 9.81 8.55 4.91 10.46 14.69 100.00

Fig. 4. Weight % retained and fixed carbon distribution (%) of raw coal versus particle size (μm).

100

100

80

80 0.3

Comb. Re. Ash Rej. Water Re.

40

40

Bias (Jb, cm/s)

60

60

Jb

0.2

0.1 20

20

0

0 1.0

1.3

0.0 0.0

1.6

Superficial Gas Rate (cm/s)

4.3.2. Froth cleaning One of the advantages of using a column is that it gives rise to a higher froth column than a conventional cell. This increases the residence time of particles in the froth zone which allows any entrained gangue particles to be drained back to the pulp. With the addition of wash water, the entrained gangue particles are washed back into the pulp zone thereby increasing the grade of the concentrate i.e. higher ash rejection.

100

100

80

80

60

60 Comb. Re.

Ash Rej. Water Re.

20

40

Ash Rejection (%)

4.3.2.1. Superficial wash water rate. The effect of superficial wash water rate on water recovery, combustible recovery and ash rejection in CPT column flotation is shown in Fig. 6. As shown, water recovery initially decreased rapidly with increasing wash water rate (Jw) from 0.1 cm/s to 0.3 cm/s. Combustible recovery decreased consistently with increasing wash water rate but ash rejection increased. The former is due to drainage of coal back to the pulp as a result of detachment from bubbles. The latter is due to a similar effect that washes away entrained ash particles due to the bias water rate. The superficial wash water determines the bias water

20

0

0 0.1

0.2

0.3

0.2

0.3

0.4

0.5

0.4

Fig. 7. Variation of bias water rate determined from Eq. (5) with superficial water rate.

and the bias water rate affects the recovery of particles and also the removal of entrained particles. Fig. 7 shows the variation of bias water determined from Eq. (5) with superficial water rate. It can be seen that the zero bias water occurs at superficial water rate of 0.086 cm/s. In several studies on bias rate, Finch and Dobby [11] recommended it should be between 0 and 0.1 cm/s to prevent loss of capacity. However, Luttrel et al. [13] reported a bias rate 0.2 cm/s to remove ash from coal and Yianatos et al. [12] a bias rate of 0.3–0.4 cm/s. Considering combustible recovery and ash rejection for the test coal, the optimum superficial wash water rate was determined at 0.3 cm/s. 4.3.2.2. Water recovery. Water recovery into the concentrate is an important variable in the flotation of fine particles as it has been shown to be directly proportional to the entrainment of gangue particles [17]. There are two sources of water that can contribute to water recovery. i.e. wash water and feed water. Dobby and Finch [26] have taken that feed water does not report to the concentrate at low gas rates while Tao et al. [16] have taken that the fraction of water reporting to the concentrate from feed water and wash water is equal as evidenced by their equation to

Water & Combustible Recovery (%)

Fig. 5. Effect of superficial gas rate on water recovery, combustible recovery and ash rejection in CPT column flotation (Jw: 0.3 cm/s, Cw: 7%, frother (MIBC): 0.15 ml/L, collector (DMU101): 0.2 kg/t, depressant (SMP): 3 kg/t).

40

0.1

Superficial Wash Water Rate (Jw, cm/s)

100

100

80

80

60

60 Comb.Re.

Ash Rej. Water Re.

40

40

20

Ash Rejection (%)

0.7


53

0.4

Ash Rejection (%)



20

0

0 1

4

7

10

Superficial Wash Water Flow Rate (cm/s)

Feed Solids Concentration (%)

Fig. 6. Effect of superficial wash water rate on water recovery, combustible recovery and ash rejection in CPT column flotation (Jg: 1.0 cm/s, Cw: 7%, frother (MIBC): 0.15 ml/L, collector (DMU101): 0.2 kg/t, depressant (SMP): 3 kg/t).

Fig. 8. Effect of feed solids concentration on water recovery, combustible recovery and ash rejection in CPT column flotation (Jg: 1.0 cm/s, Jw: 0.3 cm/s, frother (MIBC): 0.15 ml/L, collector (DMU101): 0.2 kg/t, depressant (SMP): 3 kg/t).

80

80

60

60 Comb. Re. Ash Rej.

40

40

20

20

100

100

80

80


100

Ash Rejection (%)

100

0.0

0.2

0.4

Comb. Re. Ash Rej. Water Re.

40

40

20

20

0

0

0

60

60

Ash Rejection (%)


Combustible Recovery (%)

54

0 0.10

0.6

0.15

0.20

0.25

Dosage of Collector (DMU-101, kg/ton)

Dosage of Frother (MIBC, ml/L)

Fig. 9. Effect of collector dosage on combustible recovery and ash rejection in CPT column flotation (Jg: 1.0 cm/s, Jw: 0.3 cm/s, Cw: 7%, Frother (MIBC): 0.15 ml/L, Depressant (SMP): 3 kg/t).

Fig. 11. Effect of frother dosage on water recovery, combustible recovery and ash rejection in CPT column flotation (Jg: 1.0 cm/s, Jw: 0.3 cm/s, Cw: 7%, collector (DMU101): 0.2 kg/t, depressant (SMP): 3 kg/t).

calculate bias water rate. Bias water has been taken as the water that enters the collection zone from the froth zone (positive bias). They defined the bias water rate as [16]: Jb ¼ Jw −Jwp ¼ ð1−Rw ÞJw −ðRw Jwf Þ

ð5Þ

where, Jb is bias water rate, Jw — wash water rate, Jwp — water rate to concentrate (the froth product), Jwf — water rate in the feed, and Rw is water recovery. Rearranging Eq. (5) yields the water recovery (Rw) as: Rw ¼

Jwp : Jw þ Jwf

ð6Þ

As shown in Fig. 5, water recovery (Rw, the fraction of water reporting to the froth product) increased with increasing the gas rate (Jg) and more sharply at above 1.0 cm/s. This is because more water was carried into the froth due to the increase in gas hold-up in collection zone resulting in shorter froth depth.

4.3.3. Feed solids concentration The effect of feed solids concentration on water recovery, combustible recovery and ash rejection in CPT column flotation is shown in Fig. 8. As shown, water recovery remained constant with increasing feed solids concentration, Cw (% solids in feed by weight). Both combustible recovery and ash rejection decreased sharply at Cw over 7%. This may mainly be due to entrainment of ash in coal at higher solids concentrations. It has also been reported that the change in solids concentration results in changes in pulp viscosity which in turn changes bubble size due to coalescence [10]. Changes in bubble size affect recovery. As the coal tested had a significant fines fraction (less than 37 μm) of about 26% that comprises silicates and clay minerals, it too would have contributed to the changes in pulp viscosity which suppresses turbulence and reduces collisions [11,28]. O'Connor et al. [29] also reported that an increase in feed solids concentration resulted in an increase in bubble size. Tao et al. [16] demonstrated that raw coal fraction between 30 and 150 μm stabilized froth

20

40

60

80

0 10

20

20

60

60 0

40

80

=

40

80

E

Comb. Re. Ash Rej.


60

60

Ash Rejection (%)

80

80

S.


0

=

100

E S.

100

100

40

40

20

20

CPT Column Conventional

0

0 0.0

1.5

3.0

4.5

Dosage of Depressant (SMP, kg/ton)

0 0

20

40

60

80

0 100

Ash Rejection (%) Fig. 10. Effect of depressant dosage on combustible recovery and ash rejection in CPT column flotation (Jg: 1.0 cm/s, Jw: 0.3 cm/s, Cw: 7%, Frother (MIBC): 0.15 ml/L, Collector (DMU101): 0.2 kg/t).

Fig. 12. Comparison of combustible recovery, ash rejection and separation efficiency for CPT column and conventional flotation.


55

Counts

1400

Raw sample 1200

Q M K


1000 800 600 M K

400

K

M K

M K

Q M M

M

200

Q M M Q

M

Q

K

Q M

M Q

1400

Concentrate 1200 M : Muscovite(JCPDS card No, 19-0814) C : Coal (Graphite) (JCPDS card No, 23-0064)

C

1000 800 600 M

400

M

M

200

M

1400

Tailings 1200


1000 M Q

800 K

600 400

K M

M M

200

M

M

K M

M Q

K

M K M

K M

K M

0 10

20

30

40

Position [

o

50

60

70

80

90

2Theta] (Copper(Cu))

Fig. 13. XRD pattern on raw coal, concentrate and tailings using Cu Kα target.

but that of less than 30 μm destabilized froth. Johansson [30] found that 26–44 μm quartz particles of greater hydrophobicity (contact angle: N74–90°) penetrated the interface to a much greater extent and ruptured the film, suppressing foam stability. However, hydrophobic coarse particles play a role as buffer between bubbles [14]. It was therefore considered that the feed solids concentration in processing this coal to be below 7% by mass. 4.3.4. Chemical reagents The appropriate flotation reagent combination for cleaning the given coal was identified as the collector (DMU-101), frother (MIBC) and depressant (Sodium metaphosphate). The effect of collector (DMU-101) dosage on combustible recovery and ash rejection in CPT column flotation is shown in Fig. 9. It can be

seen that in the absence of collector the combustible recovery was lowest with 68.7% and ash rejection highest with 88.9%. It demonstrates that both ash fractions were strongly depressed by depressant (SMP) along with some reduction in coal recovery. Combustible recovery increased gradually with increasing collector dosage and reached a maximum at about 0.2 kg/t. However ash rejection decreased gradually with increasing collector dosage. This may be due to the entrapment of ash particles within coal particle aggregates. Thus, the collector dosage was selected at 0.2 kg/t, with acceptable combustible recovery and ash rejection. Depressant SMP is generally used in coal flotation for preventing the flotation of finer ash. The effect of depressant (SMP) dosage on combustible recovery and ash rejection is shown in Fig. 10. In the absence of depressant (zero SMP), combustible recovery was highest and ash rejection lowest due to the effect of collector. However, combustible

56


(a) Raw sample

200 µm

SEM

200 µm

C

200 µm

Si

200 µm

Al

(b) Concentrate

200 µm

SEM

200 µm

C

200 µm

Si

200 µm

Al

Fig. 14. SEM and mapping by element of raw coal, concentrate and tailing produced from CPT column flotation.


57

(c) Tailing (×150)

50 µm

50 µm

SEM

C

50 µm

50 µm

Si

Al Fig. 14 (continued).

recovery decreased continuously with increasing depressant dosage while the ash rejection increased. The optimum depressant dosage was determined at 3 kg/t, considering economically acceptable combustible recovery and ash rejection. The effect of frother dosage on water recovery, combustible recovery and ash rejection is shown in Fig. 11. Overall, water and combustible recovery increased with increase in frother dosage while ash rejection decreased. It was observed that a stable froth of about 40 cm height was formed at the top of the column above 0.15 ml/L frother dosage, which is attributed to a reduction in the bubble coalescence in the froth zone. Tao et al. [16] considered that the frother concentration in the froth column may vary considerably at heights above the interface. Water recovery increased significantly at frother concentrations above 0.15 ml/L and ash rejection decreased. There seem to be several reasons for these effects. Many researchers have found that [14] the entrainment of fine particles in flotation is closely related to the water recovery. The thickness of the bubble walls and the bubble surface area flux generally increase at higher frother concentrations [10] and contain more water trapped within the bubble bed [31]. Higher froth concentrations may also result in poor washing characteristics of the froth and cause entrapment of fine ash particles between close-packed bubbles, which make the froth too stable particularly with smaller bubbles [10]. With these considerations, frother dosage was selected at 0.15 ml/L, considering acceptable combustible recovery and ash rejection. 4.4. CPT column vs. conventional flotation The performance of CPT column and conventional flotation was compared in terms of combustible recovery, ash rejection and separation efficiency measures as described in Section 4.2. The results are

shown in Fig. 12. Conventional flotation data was obtained from preliminary tests and rougher-cleaner stages which were conducted using a laboratory flotation cell. The recovery-rejection curve is useful for estimating the qualities of clean coal and ash that are expected in response to variations in the feed coal quality and the operating circumstances [32]. Separation efficiency reflects the combined effectiveness of the separation both with regard to combustible recovery and ash rejection. Based on the contours of separation efficiency as shown in Fig. 12, the CPT Coalpro column could produce a clean coal concentrate of 85% combustible recovery with about 81% ash rejection at a maximum separation efficiency of 62%, compared to conventional flotation which produces a relatively low combustible recovery of 70% with 70% ash rejection at a maximum efficiency of 42%. The results demonstrate that the CPT column is superior to conventional flotation for fine coal beneficiation.

Table 4 EDS analysis of each product. Elements

Raw coal wt. %

Concentrate wt. %

Tailings wt. %

CK OK Na K Mg K Al K Si K KK Fe K

63.96 26.06 0.18 0.20 3.66 4.23 1.01 0.71 100.0

77.49 19.59 – – 1.12 1.22 0.25 0.33 100

38.50 42.48 – 0.23 7.38 9.52 1.10 0.78 100.00

58


XRD patterns and scanning electron microscopic (SEM) images of the feed, concentrate and tailings products obtained from the CPT column are shown in Figs. 13 and 14, respectively. Scanning electron microscopy images of the feed, concentrate and tailing products as shown in Fig. 14, reveal that the concentrate predominantly contained coal with very little silicates in comparison to feed and tailings product. The SEM image of the feed shows that coal, Si and Al are distributed uniformly throughout the feed particles. This fact was confirmed in Fig. 4 in which the distribution of coal and mass was nearly identical in coarser particles. However, the SEM image of concentrate sample shows that the presence of Si and Al in the concentrate is minimal implying that most of the ash has reported to the tailings stream. It can also be seen in the XRD data that there is more graphitic coal prevalent in the concentrate compared to gangue minerals, such as muscovite, in the tailings product as expected. This fact is confirmed by the data obtained from EDS analysis of each product which is shown in Table 4. This confirms the efficiency of the CPT flotation column in cleaning low rank fine grained coal over conventional cells particularly when the coal is liberated only at very fine particle sizes. This fact is confirmed as the particle sizes in the tailings product shown in SEM image are more finer than those of the concentrate. It also implies that the mechanism of separation in the column is predominantly flotation and the effect of gangue entrainment of the gangue particles is minimal possibly due to the wash water. Regression analysis of the data for the prediction of combustible recovery (Rc) and ash rejection (θa) using Minitab statistical analysis software gave the following relationships and the corresponding t-values for the independent variables. For combustible recovery, Rc:

The r-squared values for the relationships for Rc and θa are 80 and 83% respectively, indicating their satisfactory level of accuracy. The corresponding t-values and their p-values reveal that the combustible recovery is significantly dependent on all variables tested, i.e. air rate, frother concentration, depressant concentration, collector dosage, wash water rate and feed solids concentration. The ash rejection however, is significantly affected only by air rate, wash water rate and collector dosage. The fact that ash rejection is not affected by depressant implies that the ash reports to the concentrate not by flotation but as a result of gangue entrainment. The effects of feed solids concentration and frother dosage do not affect ash rejection.

Rc ¼ 77:3684 þ 17:9522 Jg ðcm=sÞ þ 66:8321 Frotherðml=LÞ‐2:10996

Acknowledgments

Depressantðkg=tÞ‐51:4324 Jwðcm=sÞ þ 0:0267618 Collectorðkg=tÞ‐0:882758 Cw ð%Þ

On testing the amenability of using column flotation for the beneficiation of a fine low rank coal, it has been shown that column flotation was more effective than conventional flotation. The most significant variables for column flotation in relation to combustible coal recovery and ash rejection were: • For combustible coal recovery: Air rate, wash water rate, collector • For ash rejection: Air rate, collector. The recommended values of the process variables, air rate, wash water rate, feed solids concentration, frother (MIBC) dosage, collector (DMU-101) dosage and depressant (SMP) dosage are 1 cm/s, 0.3 cm/ s, 7% by mass, 0.15 ml/L, 0.2 kg/t and 3 kg/t, respectively. The maximum separation achievable for this coal using a column flotation has been shown to be combustible coal recovery of 85% at an ash rejection of 81%.

The authors would like to express special thanks to the Korea Institute of Geoscience and Mineral Resources for the financial support. References

Coefficients Term

Coef

SE coef

T

P

Constant Jg(cm/s) Frother(ml/L) Depressant(kg/t) Jw(cm/s) Collector(kg/t) Cw %

77.3684 17.9522 66.8321 −2.1100 −51.4324 0.0268 −0.8828

7.3103 3.8940 23.2775 0.7759 11.8793 0.0057 0.3880

10.5835 4.6103 2.8711 −2.7193 −4.3296 4.7266 −2.2754

0.000 0.000 0.010 0.014 0.000 0.000 0.035

Summary of model S = 2.80745

R-sq = 80.51%

For ash rejection, θa: θa ¼ 119:815‐42:9823 Jgðcm=sÞ‐60:3938 Frotherðml=LÞ þ 1:9909 Depressantðkg=tÞ þ 48:4716 Jwðcm=sÞ‐0:0251829 Collectorðkg=tÞ‐0:57677 Cw ð%Þ

Coefficients Term

Coef

SE coef

T

P

Constant Jg(cm/s) Frother(ml/L) Depressant(kg/t) Jw(cm/s) Collector(kg/t) Cw (%)

119.815 −42.982 −60.394 1.991 48.472 −0.025 −0.577

10.1138 5.3873 32.2045 1.0735 16.4350 0.0078 0.5367

11.8466 −7.9784 −1.8753 1.8546 2.9493 −3.2148 −1.0746

0.000 0.000 0.077 0.080 0.009 0.005 0.297

Summary of model S = 3.88412

5. Conclusions

R-sq = 82.79%

[1] X. Gui, G. Cheng, J. Liu, Y. Cao, S. Li, Q. He, Effects of energy consumption on the separation performance of fine coal flotation, Fuel Processing Technology 115 (2013) 192–200. [2] N.E. Altun, C. Hicyilmaz, J.Y. Hwang, A.S. Bagci, Beneficiation of Himmetoglu oil shale by flotation as a solid fuel substitute, Energy and Fuels 20 (2006) 214–221. [3] S.R.S. Sastri, P.S.R. Reddy, K.K. Bhattacharyya, S.G. Kumar, K.S. Narasimhan, Recovery of coal fines using column flotation, Minerals Engineering 1 (1988) 359–363. [4] J.K.M. Mankosa, J. Furey, A Comeback for Fine Coal Recovery, ERIEZ, Flotation Division, USA, 2001. 1–4. [5] M.T. Ityokumbul, A.I.A. Salama, A.M.A. Taweel, Estimation of bubble size in flotation columns, Minerals Engineering 8 (1995) 77–89. [6] N.E. Altun, C. Xiao, J.Y. Hwang, Separation of unburned carbon from fly ash using a concurrent flotation column, Fuel Processing Technology 90 (2009) 1464–1470. [7] H. Hacifazlioglu, H. Sutcu, Optimization of some parameters in column flotation and a comparison of conventional cell and column cell in terms of flotation performance, Journal of the Chinese Institute of Chemical Engineers 38 (2007) 287–293. [8] J. Barraza, A. Portilla, J. Piñeres, Direct liquefaction of vitrinite concentrates obtained by column flotation, Fuel Processing Technology 92 (2011) 776–779. [9] M.S. Jena, S.K. Biswal, S.P. Das, P.S.R. Reddy, Comparative study of the performance of conventional and column flotation when treating coking coal fines, Fuel Processing Technology 89 (2008) 1409–1415. [10] C.M. Goodall, C.T. O'Connor, Pulp-froth interactions in a laboratory column flotation cell, Minerals Engineering 4 (1991) 951–958. [11] J.A. Finch, G.S. Dobby, Column flotation: a selected review. Part I, International Journal of Mineral Processing 33 (1991) 343–354. [12] J.B. Yianatos, J.A. Finch, A.R. Laplante, Selectivity in column flotation froths, International Journal of Mineral Processing 23 (1988) 279–292. [13] G.H. Luttrell, R.R. Yoon, A flotation column simulator based on hydrodynamic principles, International Journal of Mineral Processing 33 (1991) 355–368. [14] T.V. Subrahmanyam, E. Forssberg, Froth stability, particle entrainment and drainage in flotation — a review, International Journal of Mineral Processing 23 (1988) 33–53. [15] V.M. Kirjavainen, Application of a probability model for the entrainment of hydrophilic particles in froth flotation, International Journal of Mineral Processing 27 (1989) 63–74. [16] D. Tao, G.H. Luttrell, R.H. Yoon, A parametric study of froth stability and its effect on column flotation of fine particles, International Journal of Mineral Processing 59 (2000) 25–43. [17] P.G. Smith, L.J. Warren, Entrainment of particles into flotation froths, Mineral Processing and Extractive Metallurgy Review 5 (1989) 123–145.

O.-H. Han et al. / Fuel Processing Technology 126 (2014) 49–59 [18] D.C. Yang, A New Packed Column Flotation System, ‘88 Column Flotation’, SME, Phoenix, USA, 1988. 257–265. [19] H. Hacifazlioglu, Recovery of coal from cyclone overflow waste coals by using a combination of jameson and column flotation, Energy Sources, Part A: Recovery, Utilization, and Environmental Effects 33 (2011) 2044–2057. [20] G.J. Jameson, A New Concept in Flotation Column Design ‘88 Column Flotation’, SME, Phoenix USA, 1988. 281. [21] J.N. Kohmuench, M.J. Mankosa, L. Christodoulou, Methods for improving ultrafine and coarse particle recovery, in: A.L. Mular, D.N. Halbe, D.J. Barrate (Eds.), Mineral Processing Plant Design, Practice, and Control Proceeding, SME Inc., Colorado, 2012, pp. 1241–1242. [22] J.N. Kohmuench, M.S. Davy, W.S. Ingram, I.R. Brake, G.H. Luttrell, Benefits of column flotation using the Eriez Microcel, Tenth Australian Coal Preparation Conference, Pokolbin, Australia, 2004, pp. 272–284. [23] G.S. Dobby, J.A. Finch, Column flotation: a selected review, part II, Minerals Engineering 4 (1991) 911–923. [24] J.N. Kohmuench, M. J Mankosa, J.T. Fury, G.H. Luttrell, Coalpro Columns Operators and Engineers Handbook, Eriez Manufacturing Co., 2010, 1–34. [25] B.A. Wills, J.J. Napier-Munn, E. Nunez, Mineral Processing Technology, seventh ed. Butterworth-Heinemann, Oxford, 2006.

59

[26] J.A. Finch, G.S. Dobby, Column Flotation, Pergamon Press, Oxford, 1990. [27] B.K. Gorain, J.P. Franzidis, E.V. Manlapig, The empirical prediction of bubble surface area flux in mechanical flotation cells from cell design and operating data, Minerals Engineering 12 (1999) 309–322. [28] A.R. Laplante, J.M. Toguri, H. Smith, The effect of air rate on the kinetics of flotation. Part 2: the transfer of material from the froth over the cell lip, International Journal of Mineral Processing 11 (1983) 221–234. [29] C.T. O'Connor, E.W. Randall, C.M. Goodall, Measurement of the effects of physical and chemical variables on bubble size, International Journal of Mineral Processing 28 (1990) 139–149. [30] G. Johansson, R.J. Pugh, The influence of particle size and hydrophobicity on the stability of mineralized froths, International Journal of Mineral Processing 34 (1992) 1–21. [31] C.E. Hemmineg, An alternative view point on flotation behaviour of ultrafine particles, Transactions of the Institution of Mining and Metallurgy 89 (1980) C113–C120. [32] G.H. Luttrell, A. Zufiqar, R. Bratton, A. Dynys, L. Watters, Optimal design of fine coal circuitry, in: C.A. Young, G.H. Luttrell (Eds.), Separation Technologies for Minerals, Coal, and Earth Resources, SME Inc., Colorado, 2012, pp. 139–148.