Powder Technology 305 (2017) 166–173

Contents lists available at ScienceDirect

Powder Technology

journal homepage: www.elsevier.com/locate/powtec

Effect of compound collector and blending frother on froth stability and

flotation performance of oxidized coal
Yaowen Xing a, Xiahui Gui b,⁎, Yijun Cao b, Yingwei Wang a, Mengdi Xu a, Dongyue Wang a, Chenwei Li a
a
School of Chemical Engineering and Technology, China University of Mining and Technology, Xuzhou 221116, Jiangsu, China
b
Chinese National Engineering Research Center of Coal Preparation and Purification, China University of Mining and Technology, Xuzhou 221116, Jiangsu, China

a r t i c l e i n f o a b s t r a c t

Article history: The beneficiation of oxidized coal is difficult to achieve by conventional flotation with oily collectors. Oxidized
Received 31 July 2016 coal samples were prepared by the peroxide solution oxidation and a new type of oxygenated polar compound
Received in revised form 10 September 2016 collector (CC) and a powerful blending frother (BF) were used to intensify oxidized coal flotation in the present
Accepted 3 October 2016
study. The effects of the CC and BF on the froth stability and flotation performance (i.e., recovery and selectivity)
Available online 06 October 2016
were investigated. The results show that only 44.51% recovery with 9.71 s dynamic froth stability index (DFS)
Keywords:
was obtained for oxidized coal flotation by using conventional diesel and octanol as the collector and frother, re-
Oxidized coal spectively. Lots of oxygen functional groups (−OH and C-O) were introduced after peroxide oxidation indicated
Flotation by Fourier Transform Infrared Spectroscopy (FTIR) and X-ray photoelectron spectroscopy (XPS) analysis, leading
Collector to poor hydrophobicity and low flotation recovery. An optimal separation result with 76.24% oxidized coal recov-
Frother ery and 19.43 s DFS was obtained using the CC combined with octanol, which was attributed to the hydrogen
Froth stability bonding between the hydrophilic sites on coal surface and polar oxygen functional groups in CC. An identical
DFS (19.43 s) was obtained using conventional diesel combined with the BF, while with only 21.38% recovery.
Diesel did not sufficiently render the coal hydrophobicity and as a result only a small amount of coal particles
was collected by attaching to bubble surface in pulp phase. The Fuerstenau upgrading curves show that both
the maximum recovery and the best flotation selectivity could be obtained using the CC combined with octanol.
© 2016 Elsevier B.V. All rights reserved.

1. Introduction is difficult to achieve by conventional flotation with oily collectors be-

Coal is the main fossil fuel used in many countries and may be the
largest contributor to industrial growth, accounting for almost 70% of
primary energy consumption, particularly in China. In recent years,
with the depletion of high-quality coal resources and the complication
of geological conditions, low-grade coal, such as oxidized coal refuse,
has been reconsidered for use and is becoming an important resource.
The demineralization upgrading of low-grade coal to obtain available
clean fuels has attracted increased attention in the field [1]. A strategy
to recover and utilize both the organic matter and mineral matter pres-
ent in oxidized coal refuse could yield an important energy source offer-
ing environmental and economic benefits [2,3].
Froth flotation is an effective separation method for fine coal
cleaning and upgrading that is based on the differences in the surface
hydrophobicitied of organic matter and mineral matter [4–6]. Common
oily collectors (diesel and kerosene) are added to the flotation pulp to
make coal particles more hydrophobic and separable from the pulp by
their attachment to rising air bubbles, which form a particle-rich froth
on the suspension surface. However, the beneficiation of oxidized coal
⁎ Corresponding author.
E-mail address: guixiahui1985@163.com (X. Gui).
cause of the presence of a large quantity of surface oxygen functional
groups and poor froth stability [7,8]. The surface physicochemical char-
acteristics, surface functional group composition, and surface roughness
are significantly changed by the oxidation process [9,10]. The results of
X-ray photoelectron spectroscopy (XPS) analyses have shown that the
content of hydrophobic C-C and C-H groups decreases, whereas the con-
tent of hydrophilic C-O and C=O groups increases. The surface rough-
ness and cracking also increase after oxidation, according to scanning
electron microscopy (SEM) results [9]. A firm hydration film is formed
at the hydrophilic sites of oxidized coal surfaces because of hydrogen
bonding [11,12]. It is therefore important to develop an enhanced flota-
tion separation process that can optimize the flotation efficiency of
oxidized coal, similar to the chemical process of multiphase reactions
[13–15].
Numerous methods have been employed to intensify oxidized or
difficult-to-float coal flotation, such as the use of new types of collec-
tors; particle surface pre-treatment using grinding, ultrasonic, micro-
wave, and thermal methods; bubble surface modification; and high-
efficiency flotation process design. The addition of surfactants or oxy-
genated functional groups to oily collectors can significantly enhance
the oxidized coal flotation yield because of hydrogen bonding between
the polar part of the coal surface and the new reagent [16]. A series of

http://dx.doi.org/10.1016/j.powtec.2016.10.003
0032-5910/© 2016 Elsevier B.V. All rights reserved.
 Y. Xing et al. / Powder Technology 305 (2017) 166–173 167

non-ionic surfactants containing oxygenated and aromatic functional
groups, i.e., tetrahydrofurfuryl esters (THF), were used as collectors to
enhance low-rank/oxidized-coal flotation by Jia et al. [17]. The members
of the THF series were found to be more effective collectors than oily
collectors for both oxidized and un-oxidized coals. Xia et al. [18] found
that the flotation recovery of oxidized coal was improved by grinding
pretreatment, which removed the oxidation layer and exposed some
fresh surfaces and hydrophobic sites. Atesok and Celik [19] used bitumi-
nous coal pitch in dry grinding, followed by flotation of a difficult-to-
float coal, and achieved a remarkable increase in the recovery rate,
from 22.1% to 89.9%, and a decrease in the ash content, from 11.8% to
10.1%, because of the higher viscosity of the pitch. The floatability of ox-
idized coal can also be restored by ultrasonication preconditioning. Feng
and Aldrich [20] found that ultrasonication pretreatment had a signifi-
cant effect on the flotation recovery rate and kinetics of oxidized coal,
especially for smaller-size fractions. XPS results indicated that a thin
clay layer and oxidized layer were removed using ultrasonication [21].
Buttermore and Slomka [22] reported that the sonication of oxidized
coal restored its flotability to a level close to or exceeding that of fresh
coal. Microwave and thermal treatments are also common methods
for improving the floatability of oxidized coal by removing the pore
water and some organic OH functional groups [8]. The characteristics
of bubbles, which act as the carriers of particles in flotation, have an im-
portant effect on the separation performance. The flotation of hard-to-
float coal can be enhanced by suitable bubble surface modification tech-
nology. Liu et al. [23] proposed a novel concept for using reactive oily
bubbles as the carriers in flotation. Oily bubbles have much higher con-
tact angles than air bubbles, generating a strong collecting power that is
conducive to the flotation of both low-rank and oxidized coal [24]. Re-
cently, researchers have found that direct-contact and zero-condition-
ing flotation process design can make collectors adsorb onto oxidized
coal surfaces directly without the prevention of hydration film and
can effectively improve the floatability of difficult-to-float oxidized
coal [25–27]. Xia and Yang [28] found that reverse flotation was partic-
ularly well suitable for the upgrading of Taixi oxidized coal. It is well ac-
cepted that oxygenated polar collectors are the most effective method
to enhance oxidized coal flotation among above mentioned methods.
However, the effect of oxygenated polar collector on oxidized coal flota-
tion performance is complex and required to be further studied.
To date, little attention has been given to frothers and their potential
effect on oxidized coal flotation. Frothers are used for froth generation
and to facilitate air dispersion into fine bubbles, which contributes to a
high flotation recovery rate and good kinetics [29]. Frothers are com-
monly characterised as either selective or powerful and chosen on the
basis of laboratory and/or pilot plant tests [30]. Low-molecular-weight
frothers are more selective but less effective than high-molecular-
weight frothers and thus result in the loss of combustible matter in tail-
ings [31]. Tan [32] found that better foaming properties could be obtain-
ed using a mixture of selective and powerful polypropylene glycols
compared to the use of single frother. Similar findings were reported
by Dey et al. [31]. A powerful frother is more effective than selective
frother in recovering combustibles in coking coal flotation and it may
be well suited to difficult-to-float oxidized coal. Therefore, it is essential
to explore the effect of powerful frother on oxidized coal flotation.
Froth structure and stability plays an important role in determining
flotation recovery and grade. It is well accepted that poor froth stability
is incapable of getting a satisfactory flotation recovery and grade. How-
ever, it doesn't mean that a high flotation recovery can be obtained def-
initely with good froth stability. Froth flotation is a complex system and
the final recovery is the result of the interaction of many factors such as
were examined in this study. Oxidized coal samples were prepared
using the peroxide solution oxidation method and the surface function-
al groups were characterised using Fourier transform infrared spectros-
copy (FTIR) and XPS before and after oxidation.
2. Experimental
2.1. Materials
Fine high-metamorphosed anthracite coal samples were collected
from the coal storage yard of the Xuehu Coal Preparation Plant in
Yongcheng City, Henan Province, China. Samples that were smaller
than 0.25 mm in size were prepared by screening. The particle size dis-
tributions are presented in Table 1. Dry coal particles with a size smaller
than 0.25 mm were immersed in 30% peroxide solutions in a magnetic
stirrer for 24 h at room temperature to obtain the final oxidized coal
samples [33]. It should be noted that large amounts of heat and gases
were released in the initial stage of oxidation, which indicated that the
H2O2 solution method was quick and effective in preparing the oxidized
coal. Flotation reagents, diesel, octanol, and the CC and BF were pur-
chased from the Zhuzhou Flotation Reagents & Chemicals Co., Ltd.
2.2. Method
2.2.1. FTIR
FTIR (Nicolet is5, Thermo Scientific) was used to identify the func-
tional groups of the coal and flotation reagents in the wave number
range of 4000–400 cm− 1. A mixture of 2 mg of coal and 300 mg of
KBr was prepared and ground to a fineness of 2 μm in an agate mortar.
A pressure of 28 MPa was applied to the powder to form thin circular
plates, and the influences of the background and vapour were
accounted for in the data analysis process. For measurement of the func-
tional groups of the liquid flotation reagents, the painting method, in
which a micro-drop of reagent was placed on a KBr pellet, was used to
create a homogeneous distribution of reagent film. All of the experi-
ments were conducted at room temperature 20 °C.
2.2.2. XPS
Flotation is a type of interface separation technique. Interfacial prop-
erties play a crucial role in bubble–particle interaction sub-processes.
Recently, XPS has emerged as an advanced interface chemistry analysis
tool and has been used extensively in flotation and energy engineering
applications. In the present study, an XPS system (ESCALAB 250Xi,
Thermo Scientific, Waltham, MA, America) with Al Ka radiation
(hv = 1486.6 eV) and a 900 μm light spot size was used to characterise
the surface chemistry changes before and after oxidation. Each XPS
spectra data set was fitted using the XPS Peakfit software. It should be
noted that binding energy calibration was performed by setting the
C1s hydrocarbon peak to 284.6 eV.
2.2.3. Flotation column kinetics experiments
Column flotation tests were carried out at an 80 g/L solid concentra-
tion using a traditional laboratory countercurrent flotation column-that
Table 1
Particle size distributions of the coal sample.
Size, mm Yield, % Ash,% Positive Negative
cumulative cumulative

properties particles, reagents, bubbles, froth stability, and solution. Yield, % Ash,% Yield, % Ash,%
Based on previously summarised the literature review, a new type of 0.25–0.125 35.95 18.15 35.95 18.15 100.00 21.18
oxygenated polar compound collector (CC), a mixture of diesel and ali- 0.125–0.074 17.44 19.45 53.40 18.57 64.05 22.88
phatic acid, and powerful blending frother (BF), a mixture of octanol 0.074–0.045 11.21 21.66 64.61 19.11 46.60 24.16
and aliphatic acid salt, were used to enhance oxidized coal flotation. −0.045 35.39 24.95 100.00 21.18 35.39 24.95
Total 100 21.18 – – – –
The effects of the CC and BF on froth stability and flotation performance
168 Y. Xing et al. / Powder Technology 305 (2017) 166–173

Air compressor In Eq. (1), Vf is the foam volume, Hmax is the maximum height of the
froth at equilibrium, A is the cross-sectional area of the flotation column,
Flotation column and Q is the gas volumetric flow rate.

2.2.5. Fuerstenau upgrading curves analysis

Rotameter
Fig. 1. Traditional countercurrent flotation column system.
was 400 mm high and 50 mm in diameter, as shown in Fig. 1. Coal was
mixed with the collector and foaming agent using normal tap water and
conditioned in a magnetic stirrer for 5 min. The dosages of the different
collectors and frothers were kept constant at 0.2 kg/t and 0.2 kg/t, re-
spectively. The resulting coal slurry was fed to the column using a peri-
staltic pump. Air was introduced through a ceramic sparger at four
different pore size ranges (80–120 μm, 40–80 μm, 15–40 μm, and 1–
15 μm) at the bottom of column, and its flow rate was kept constant
at 0.04 m3/h controlled by using an air rotameter and air compressor.
The flotation process was divided into five stages and continued for
5 min. Flotation concentrate products were collected after 30, 60, 120,
180, and 300 s. The flotation concentrates and tailings were filtered,
dried at 80 °C, and weighed. Approximately 10 g of dried sample from
each product was ground using a mortar and pestle. Then, approximate-
ly 1 g of the ground sample from each product was burned in an oven at
815 °C for 2 h for ash analysis. The left-over ash was weighed to calcu-
late the ash content. The ash analysis procedure has been described
elsewhere [4].
The Fuerstenau upgrading curves were used for the characterization,
comparison, and analysis of the whole separation selectivity. These
curves have proven to be very useful in the analysis of the de-ashing
and desulfurization of coal flotation results [36]. The kinetic equations
relating the recoveries to time for the two components of the separation
products, when combined together to eliminate the time parameter,
provided Fuerstenau upgrading curves, which relate the recoveries of
the two components in concentrate [37,38]. Commonly, the flotation ki-
netics of both the combustible and ash materials can be successfully de-
scribed using a classical first-order equation, as shown in Eqs. (2) and
(3) [4]. Therefore, the combination of the two first-order kinetic Eqs.
(2) and (3) yield Eq. (4), which is a mathematical formula for approxi-
mation of the Fuerstenau curves. A new parameter,k ¼ kk12, was defined as
a separation selectivity index, the value of which can be obtained by
fitting the flotation kinetics using Eq. (4).


ε ¼ ε ∞ 1−e−k1 t ð2Þ


ε a ¼ εa∞ 1−e−k2 t ð3Þ
32
 k1
ε −ε k2
ε ¼ ε∞ 41− 5
a∞ a
ð4Þ
εa∞
where ε and εa are the combustible matter recovery and ash recovery in
the concentrate, respectively; ε∞ and εa∞ are the maximum combustible
recovery and ash recovery in the concentrate, respectively; k1 and k2 are
the first-order flotation rate constants of the combustible and ash-
forming material, respectively; and t is the flotation time.

2.2.4. Dynamic froth stability measurement 3. Results and discussion

The froth structure and stability play important roles in determining
the flotation product grade and separation efficiency [34]. The froth sta-
bility was determined by calculating the dynamic froth stability index
(DFS) as a function of the measured maximum steady-state froth height
with continuous bubbling in the same flotation column, according to Eq.
(1) [35]. During the experiments, the froth height was recorded as a
function of time at a constant airflow rate until the froth reached equi-
librium. All of the experiments were carried out twice, and the average
values were calculated.
V f Hmax A
DFS ¼ ¼ :
Q Q
3.1. Effect of oxidation on flotation performance
Preliminary tests to determine the effective bubble size for column
flotation were carried out using the following series of ceramic sparger
diameters: 120–80 μm (#1), 80–40 μm (#2), 40–15 μm (#3), and 15–
1 μm (#4). The effect of the sparger diameter on the raw coal flotation
DFS and performance with conventional diesel and octanol is illustrated
in Fig. 2. The froth growth rate, maximum froth height, and DFS initially
increased and then nearly levelled off at 40.62 s as the bubble size
decreased. These results indicated that small bubbles were conducive
ð1Þ
to forming froth with high quality and good stability. This could be

30 100
#1 A #1 B
Cumulated combustible recovery

#2 90 #2
25
#3 80 #3
Froth height (cm)

20 #4 70 #4
60
(%)

15 50
DFS (s) 40
10
#1 28.26 30
#2 31.79 20
5 #3 40.62
#4 42.39
10
0 0
0 20 40 60 80 100 7 8 9 10 11 12
Time (s) Cumulated ash content (%)

Fig. 2. The effect of sparger diameter on raw coal flotation DFS and performance with conventional diesel and octanol.
 Y. Xing et al. / Powder Technology 305 (2017) 166–173 169

30 100
Oxidized coal with diesel and octanol

Cumulated combustible recovery(%)

(A) Oxidized coal with diesel and octanol (B)
Raw coal with diesel and octanol 90 Raw coal with diesel and octanol
25
DFS (s) 80
Oxidized coal: 9.71

Froth height (cm)

20 70
Raw coal: 40.62
60
15 50
40
10
30
5 20
10
0 0
0 20 40 60 80 100 8 9 10 11 12
Time (s) Cumulated ash content (%)

Fig. 3. The effect of oxidation on flotation DFS and performance with conventional diesel and octanol.

explained by the fact that small bubbles promoted the transportation of
water towards the froth layer, which effectively inhibited bubble coales-
cence and bursting. The same phenomenon has also been observed by
other researchers [35]. With respect to the flotation performance, the
coal recovery and flotation kinetics were observed to increase with de-
creasing bubble size. A substantial increase was observed (91.02% re-
covery with 11.45% ash content) when the #3 ceramic sparger was
used. This indicated that bubbles in the diameter range of 40–15 μm
were found to be suitable for coal flotation in the present study. There-
fore, the #3 ceramic sparger was selected for the subsequent experi-
ments. A good correlation was observed between the DFS and the
ultimate flotation recovery. The high dynamic froth stability results ob-
tained with small bubbles corresponded to the high recovery values.
These results were consistent with the findings of other researchers
[35,39,40].
Oxidized coal samples were prepared by the peroxide solution oxi-
dation method, and the effect of oxidation on the flotation DFS and per-
formance were examined. The results are shown in Fig. 3. The DFS
decreased dramatically, from 40.62 s to 9.71 s, after oxidation. The flota-
tion recovery also decreased, from 91.02% to 44.51%, using conventional
diesel and octanol as the collector and frother, respectively. This illus-
trated that particles cannot be successfully attached by bubbles for oxi-
dized coal. It is well known that the characteristics of the particles
presented in the froth have a crucial effect on the froth stability. Bubbles
armoured by intermediate-hydrophobicity particles with contact angles
of approximately 65° become more resistant to drainage coalescence
and facilitate the production of a stable froth because a steric barrier
to coalescence is introduced by the particles [41,42]. Particles with
higher or lower contact angles are detrimental to froth stability. The
contact angle of the oxidized coal in this study was only 42°. The poor
hydrophobicity leads to a thick hydration film on the oxidized coal sur-
face and hinders the attachment of coal particles to bubbles. As a result,
the froth stability deteriorated, which further decreased the flotation
recovery.
To identify the surface chemistry changes that occurred, the surface
functional groups were characterised by FTIR before and after oxidation.
The FTIR spectrums of raw coal and oxidized coal are shown in Fig. 4. Ac-
cording to the literature [43–48], the oxygen functional groups of –OH
and C-O with peaks of 3415 cm− 1 and 1091 cm− 1, respectively, in-
crease after peroxide solution oxidation. The strong and sharp peaks
of –OH and C-O suggested that oxidation improved the hydrophilicity,
which is detrimental to coal flotation. However, changes in the O-
C=O and C=O peaks at 1734 cm−1and 1652 cm−1 were not observed.
The kaolin peaks at 3693 and 3619 cm−1, Si-O-Si peak at 1031 cm−1, Si-
O peaks at 1009 and 470 cm− 1, and Fe-O peak at 537 cm− 1 also in-
creased, which indicated that the ash content of a coal sample may in-
crease after oxidation and further decreased the coal hydrophobicity.
It should be noted that the CH3 andCH2 peaks at 2850 cm−1–
2930 cm−1 and at 1430 cm−1 and the benzene rings or C=C peaks at
1600 cm− 1 did not exhibit a tendency to decrease. This behaviour
may indicate that the coal sample was not oxidized sufficiently by the
peroxide solution and that these slight changes cannot be detected
using FTIR. However, it was essential to confirm the changes in the car-
bon after oxidation by XPS.
To obtain more comprehensive information about the coal surface
before and after oxidation, XPS experiment was conducted. The XPS
wide energy spectrums of raw coal and oxidized coal are shown in
Fig. 5. Semi-quantitative results for the surface chemical composition
of the carbon, oxygen, aluminium, silicon, nitrogen, sodium, and calci-
um contents before and after oxidation, based on the XPS wide energy

Fig. 4. The FTIR spectrums of raw coal and oxidized coal.

170 Y. Xing et al. / Powder Technology 305 (2017) 166–173

5 O1s C1s
4x10

5
3x10

Counts (s)
5
2x10

5
1x10

0
1400 1200 1000 800 600 400 200 0
Binding Energy (eV)
(A) Raw coal

5
4x10
O1s
5
3x10
Counts (s)

C1s
5
2x10

5
1x10

0
1400 1200 1000 800 600 400 200 0
Binding Energy (eV)
(B) Oxidized coal

Fig. 5. The XPS wide energy spectrums of raw coal and oxidized coal.

spectra, are listed in Table 2. The carbon content of the oxidized coal was
much lower than that of the raw coal, with a 10.72% gap, which was not
consistent with the FTIR results. It illustrated that a fraction of the or- C-C/C-H
ganic carbon was indeed oxidized to inorganic carbon dioxide during
the oxidation stage. However, the contents of oxygen and inorganic
mineral elements (Al, Si, Na, and Ca) were higher in the oxidized coal
compared with the raw coal, which is consistent with the FTIR results.
The organic fractions on the coal surface were attacked by oxygen,
and fresh minerals were fully exposed to the surface.
The C1s peaks of the fresh and oxidized coals are shown in Fig. 6. The C-O
O=C-O C=O
binding energy peaks at 284.60, 285.60, 287.54 and 290.25 eV represent
the groups of C-C and C-H, C-O, C=O, and O=C-O, respectively [49,50].
The peak split result of C for the fresh and oxidized coals is shown in 298.0 296.0 294.0 292.0 290.0 288.0 286.0 284.0 282.0 280.0
Table 3. Apparently, more of the C-O group (20.95%) was detected on (A) Raw coal
the oxidized coal surface than on the raw coal surface (11.32%). The con-
C-C/C-H
tent of C-C and C-H (81.60%) on the raw coal surface was higher than
that on the oxidized coal surface (71.03%). These findings are in good
agreement with the XPS wide energy spectra and FTIR results. It was
also interesting that the content of C=O and O = C-O increased slightly,
indicating that the surface oxidation was not very strong, which is

C-O
Table 2 O=C-O C=O
Semi-quantitative results of the surface chemical composition of raw coal and oxidized
coal.
298.0 296.0 294.0 292.0 290.0 288.0 286.0 284.0 282.0 280.0
Types C1s,% O1s,% Al2p,% Si2p,% N1 s,% Na1s,% Ca2p,%
(B) Oxidized coal
Raw coal 63.90 22.97 4.95 6.02 1.53 0.31 0.32
Oxidized coal 53.18 31.82 6.38 6.95 1.67 0.00 0.00
Fig. 6. C1s peaks of fresh and oxidized coals.
 Y. Xing et al. / Powder Technology 305 (2017) 166–173 171

Table 3
Peak split result of C of fresh and oxidized coals.
Types C-C/C-H,% C-O,%
Raw coal 81.60 11.32
Oxidized coal 71.03 20.95
consistent with the FTIR results. Typically, C-C and C-H groups are first
oxidized to C-O and then to C=O or O=C-O and finally to CO2 through
intensive oxidation. In conclusion, the oxidized coal examined in this
paper was not strongly oxidized, and its poor hydrophobicity could be
attributed to the introduction of –OH and C-O oxygen functional groups.
3.2. Effect of CC and BF on oxidized coal flotation
3.2.1. Froth stability and flotation recovery
CC and BF were used to intensify the oxidized coal flotation. The ef-
fect of different reagent combinations (i.e., CC with octanol, diesel with
BF, and CC with BF) on DFS and recovery of oxidized coal flotation are
shown in Fig. 7. When CC and octanol were used, both the froth stability
and flotation recovery increased significantly, with a DFS of 19.43 s and
76.24% recovery. This illustrated that CC could make the oxidized coal
more hydrophobic than conventional diesel. The froth was well
armoured by CC-coated particles, leading to a stable froth. To identify
the reason for this, the FTIR spectrums of CC and diesel were measured.
The results are shown in Fig. 8. The hydrophobic benzene rings or C=C,
CH3, and CH2, which had peaks at 2955, 2924, 2853, 1464, and
1377 cm− 1 did not show obvious changes for the CC and diesel.
However, the polar hydrophilic oxygen peaks at 3424 cm−1(− OH)
and 1695–1654 cm−1(C = O) were sharpened and enhanced with the
C = O,% O = C-O,% CC, compared with those in diesel. This indicated that these polar oxy-
5.15 1.93
gen functional groups would bond with the oxygen groups on oxidized
5.92 2.10 coal surfaces through hydrogen bonding. The residual non-polar hydro-
phobic groups in the CC would bond with the non-polar groups on ox-
idized coal surfaces by hydrophobic function. As a result, the polar
groups on the coal surfaces were fully passivated by the coverage of
the CC and the floatability was restored. It was interesting to note that
when diesel and BF were used, the identical DFS (19.43 s) could be ob-
tained with only 21.38% recovery. A satisfactory recovery cannot be ob-
tained only by increasing the froth stability. It can be explained that the
BF may belong to a powerful frother family, which can generate stable
froth by reducing bubble surface tension and preventing bubble coales-
cence effectively. However, the low recovery is highly attributed to the
poor hydrophobicity of oxidized coal. Diesel did not sufficiently render
the coal hydrophobicity and as a result only a small amount of coal par-
ticles was collected by attaching to bubble surface in pulp phase. It
should be noted that the flotation recovery by using diesel and BF is
lower than that of by using diesel and octanol. BF may hinder the ab-
sorption of diesel on oxidized coal surface. The FTIR spectrums of the
BF and octanol are shown in Fig. 9. Obvious discrepancies can be observ-
able at peaks of 3354 cm− 1 and 1732–1711 cm−1. The peak at
3354 cm−1 was attributed to –OH, and the peak in the range of 1732–
1711 cm−1 was attributed to O=C-O. It illustrated that a more power-
ful carboxylic acid frother was added into BF. A small fraction of carbox-
ylic acid was enough to significantly reduce the interfacial tension at the
air water interface and change the foam properties from weak to pow-
erful because of the Gibbs–Marangoni effect. The extra local stresses

12 100
(A) Oxidized coal with diesel and octanol (B)
Cumulated combustible recovery(%)

90 Oxidized coal with CC and octanol

10 Oxidized coal with diesel and BF
80
Oxidized coal with CC and BF
70
8
Froth height (cm)

60
6 50
40
4
Oxidized coal with diesel and octanol 30
Oxidized coal with CC and octanol 20
2
Oxidized coal with diesel and BF 10
Oxidized coal with CC and BF
0 0
0 20 40 60 80 100 0 100 200 300
Time (s) Time (s)

Fig. 7. The effect of different reagents combinations on DFS and recovery of oxidized coal flotation.

Fig. 8. The FTIR spectrums of CC and diesel.

172 Y. Xing et al. / Powder Technology 305 (2017) 166–173
Fig. 9. The FTIR spectrums of BF and octanol.
were generated due to the opposed surface tension gradient forces [31].
When the CC was used with the BF, a moderate DFS (15.01 s) and recov-
ery (52.59%) were obtained. BF may hinder the absorption of CC on ox-
idized coal surface and weaken the particle-armour effect of the
bubbles. In addition, a certain synergistic effect between the CC and BF
may exist during the flotation process. Further study on this point is
required.
3.2.2. Flotation selectivity analysis
To evaluate the flotation selectivity of different reagents combina-
tions, the Fuerstenau upgrading curves analysis was conducted. Assum-
ing that ε∞=εa∞=100%, Eq. 4 can be simplified to Eq. 5.
"   #
100−εa k
ε ¼ 100 1−
100
The Fuerstenau upgrading curves of different reagent combinations
for oxidized coal flotation are shown in Fig. 10. High R2 values were ob-
tained, confirming the validity and robustness of the mathematical
model. The selectivity index k of oxidized coal flotation with octanol
was much higher than that of with BF. It indicated that better flotation
Fig. 10. The Fuerstenau upgrading curves of different reagents combinations for oxidized
coal flotation.
selectivity can be obtained when weak frother octanol was used.
Frother has a major impact on decreasing bubble size because of the
prevention of bubble coalescence [40]. BF is more efficient in preventing
bubble coalescence and provides more stable and viscous foams com-
pare to octanol. Melo and Laskowski found that a high water recovery
was obtained when a powerful fother was used [29,30]. It is well accept-
ed that the ultra-fine high ash particles entrainment strongly depend on
the water recovery. This is probably the reason for poor flotation selec-
tivity when BF was used. In contrast, there is little difference between
CC and diesel in flotation selectivity index when same kind of frother
was used. The optimal k value (3.13) was obtained using the new CC
and octanol, which illustrated that the CC could absorb on coal surface
with high selectivity. Both the maximum recovery and the best flotation
selectivity were obtained using the new CC combined with octanol.
ð5Þ
4. Conclusions
A new type of oxygenated polar compound collector (CC), a mixture
of diesel and aliphatic acid, and powerful blending frother (BF), a mix-
ture of octanol and aliphatic acid salt, were used to intensify oxidized
coal flotation. Lots of oxygen functional groups (−OH and C-O) on oxi-
dized coal surface were identified by FTIR and XPS, leading to poor hy-
drophobicity and low flotation recovery. Not only optimal separation
results (76.24% recovery and 19.43 s DFS) and but also the better flota-
tion selectivity can be obtained by using CC combined with weak
octanol for oxidized coal flotation. The polar groups on the coal surfaces
were fully passivated by oxygen functional groups in CC through
hrdrogen bonding. An identical DFS (19.43 s) was obtained using con-
ventional diesel combined with powerful BF, while with only 21.38% re-
covery. Diesel did not sufficiently render the coal hydrophobicity and as
a result only a small amount of coal particles was collected by attaching
to bubble surface in pulp phase. Poor flotation selectivity was obtained
when powerful BF was used. Effective intensification of oxidized coal
flotation was found to be achievable only by using CC combined with
weak octanol in the present study. However, more variety of frothers
should be selected to look for high effective one for oxidized coal flota-
tion. The synergistic effect between the CC and BF during the oxidized
coal flotation process in the present study also required to be studied
further.
Acknowledgements
This research was supported by the National Nature Science Founda-
tion of China (51574236) and the Fundamental Research Funds for Cen-
tral Universities (2015XKMS095) for which the authors express their
appreciation.
 Y. Xing et al. / Powder Technology 305 (2017) 166–173
References
[1] P. Meshram, B.K. Purohit, M.K. Sinha, S.K. Sahu, B.D. Pandey, Demineralization of low
grade coal – a review, Renew. Sust. Energ. Rev. 41 (2015) 745–761.
[2] D. Tao, B. Li, S. Johnson, B. Parekh, A flotation study of refuse pond coal slurry, Fuel
Process. Technol. 76 (2002) 201–210.
[3] J. Barraza, J. Guerrero, J. Pineres, Flotation of refuse tailing fine coal slurry, Fuel Pro-
cess. Technol. 106 (2013) 498–500.
[4] Y. Xing, X. Gui, J. Liu, Y. Cao, Y. Zhang, S. Li, Flotation behavior study of hard-to-sep-
arate and high-ash fine coal, Physicochem. Probl. Miner. Process. 52 (2) (2016)
703–718.
[5] A.S. Reis, M.A.S. Barrozo, A study on bubble formation and its relation with the per-
formance of apatite flotation, Sep. Purif. Technol. 161 (2016) 112–120.
[6] A. Nguyen, H. Schulze, Colloid Science of Flotation, Marcel Dekker, New York, 2004.
[7] R. Jia, G. Harris, D. Fuerstenau, An improved class of universal collectors for the flo-
tation of oxidized and/or low-rank coal, Int. J. Miner. Process. 58 (2000) 99–118.
[8] W. Xia, J. Yang, C. Liang, A short review of improvement in flotation of low rank/ox-
idized coals by pretreatments, Powder Technol. 237 (2013) 1–8.
[9] W. Xia, J. Yang, Changes in surface properties of anthracite coal before and after in-
side/outside weathering processes, Appl. Surf. Sci. 313 (2014) 320–324.
[10] W. Xia, Y. Li, Role of roughness change on wettability of taixi anthracite coal surface
before and after the heating process, Energy Fuel 30 (2016) 281–284.
[11] D.W. Fuerstenau, M.C. Williams, K.S. Narayanan, Assessing the wettability and de-
gree of oxidation of coal by film flotation, Energy Fuel 2 (3) (1988) 237–241.
[12] D.W. Fuerstenau, J.M. Rosenbaum, J.S. Laskowski, Effect of surface functional groups
in the floatability of coal, Colloids Surf. 8 (2) (1983) 153–173.
[13] C.S. Li, X.P. Zhang, S.J. Zhang, K. Suzuki, Environmentally conscious design of chem-
ical processes and products: multi-optimization method, Chem. Eng. Res. Des. 87
(2) (2009) 233–243.
[14] C.S. Li, G. Wozny, K. Suzuki, Design and synthesis of separation process based on a
hybrid method, Asia Pac. J. Chem. Eng. 4 (6) (2009) 905–915.
[15] C.S. Li, X.P. Zhang, S.J. Zhang, X.Z. He, Design of separation process of azeotropic mix-
tures based on the green chemical principles, J. Clean. Prod. 15 (2007) 690–698.
[16] S. Dey, Enhancement in hydrophobicity of low rank coal by surfactants-a critical
overview, Fuel Process. Technol. 94 (2012) 151–158.
[17] R. Jia, G.H. Harris, D.W. Fuerstenau, Chemical reagents for enhanced coal flotation,
Int. J. Coal Prep. Util. 22 (2002) 123–149.
[18] W. Xia, J. Yang, B. Zhu, Flotation of oxidized coal dry-ground with collector, Powder
Technol. 228 (2012) 324–326.
[19] G. Atesok, M.S. Celik, A new flotation scheme for a difficult-to-float coal using pitch
additive in dry grinding, Fuel 79 (2000) 1509–1513.
[20] D. Feng, C. Aldrich, Effect of preconditioning on the flotation of coal, Chem. Eng.
Commun. 192 (2005) 972–983.
[21] S.G. Ozkan, Effects of simultaneous ultrasonic treatment on flotation of hard coal
slimes, Fuel 93 (2012) 576–580.
[22] W.H. Buttermore, B.J. Slomka, The effect of sonic treatment on the flotability of ox-
idized coal, Int. J. Miner. Process. 32 (3–4) (1991) 251–257.
[23] J. Liu, T. Mak, Z. Zhou, Z. Xu, Fundamental study of reactive oily-bubble flotation,
Miner. Eng. 15 (2002) 667–676.
[24] Y. Wang, J. Liu, W. Yu, One System and Method of Coal Deep Purification. China,
2011 20111095118.0.
[25] H. Ahmed, J. Drzymala, Effect of flotation procedure and composition of reagents on
yield of a difficult-to-float coal, Physicochem. Probl. Miner. Process. 38 (2004)
53–63.
[26] H. Ahmed, J. Drzymala, Upgrading difficult-to-float coal using microemulsion,
Miner. Metall. Process. 29 (2) (2012) 88–96.
[27] D.P. Patil, J.S. Laskowski, Development of zero conditioning procedure for coal re-
verse flotation, Miner. Eng. 21 (2008) 373–379.
173
[28] W. Xia, J. Yang, Reverse flotation of taixi oxidized coal, Energy Fuel 27 (2013)
7324–7329.
[29] Y. Liao, Y. Cao, S. Huang, J. He, X. Zhang, S. Li, S. Shao, Water-carrying properties of
flotation frothers and its effect on fine coal flotation, Int. J. Coal Prep. Util. 35 (2)
(2015) 88–98.
[30] F. Melo, J.S. Laskowski, Fundamental properties of flotation frothers and their effect
on flotation, Miner. Eng. 19 (2006) 766–773.
[31] S. Dey, S. Pani, R. Singh, Study of interactions of frother blends and its effect on coal
flotation, Powder Technol. 260 (2014) 78–83.
[32] S.N. Tan, R.J. Pugh, D. Fornasiero, S. Sedev, J. Ralston, Foaming of polypropylene gly-
cols and glycol/MIBC mixtures, Miner. Eng. 18 (2) (2005) 179–188.
[33] Z. Liu, Z. Liu, Z. Zong, X. Wei, J. Wang, C. Lee, GC/MS analysis of water-soluble prod-
ucts from the mild oxidation of Longkou brown coal with H2O2, Energy Fuel 17 (2)
(2003) 424–426.
[34] X. Qu, L.G. Wang, A.V. Nguyen, Correlation of air recovery with froth stability and
separation efficiency in coal flotation, Miner. Eng. 41 (2013) 25–30.
[35] A.U. Kurniawan, O. Ozdemir, A.V. Nguyen, P. Ofori, B. Firth, Flotation of coal particles
in MgCl2, NaCl, and NaClO3 solutions in the absence and presence of Dowfroth 250,
Int. J. Miner. Process. 98 (2011) 137–144.
[36] A. Bakalarz, J. Drzymala, Interrelation of the Fuerstenau upgrading curves with ki-
netics of separation, Physicochem. Probl. Miner. Process. 49 (2) (2013) 443–451.
[37] J. Drzymala, Evaluation and comparison of separation performance for varying feed
composition and scattered separation results, Int. J. Miner. Process. 75 (2005)
189–196.
[38] J. Drzymala, H. Ahmed, Mathematical equations for approximation of separation re-
sults using the Fuerstenau upgrading curves, Int. J. Miner. Process. 76 (2005) 55–65.
[39] B. McFadzean, T. Marozva, J. Wiese, Flotation frother mixtures: decoupling the sub-
processes of froth stability, froth recovery and entrainment, Miner. Eng. 85 (2016)
72–79.
[40] A.K. Gupta, P.K. Banerjee, A. Mishra, P. Satish, Pradip, Effect of alcohol and polyglycol
ether frothers on foam stability, bubble size and coal flotation, Int. J. Miner. Process.
82 (2007) 126–137.
[41] G. Johansson, R.J. Pugh, The influence of particle size and hydrophobicity on the sta-
bility of mineralized froths, Int. J. Miner. Process. 34 (1992) 1–21.
[42] E.C. Cilek, S. Karaca, Effect of nanoparticles on froth stability and bubble size distri-
bution in flotation, Int. J. Miner. Process. 138 (2015) 6–14.
[43] R. Pietrzak, H. Wachowska, Low temperature oxidation of coals of different rank and
different sulphur content, Fuel 82 (6) (2003) 705–713.
[44] R. Pietrzak, H. Wachowska, Thermal analysis of oxidized coals, Thermochim. Acta
419 (1) (2004) 247–251.
[45] T. Grzybek, R. Pietrzak, H. Wachowska, The influence of oxidation with air in com-
parison to oxygen in sodium carbonate solution on the surface composition of
coals of different ranks, Fuel 85 (7) (2006) 1016–1023.
[46] J.M. Sokolovic, R.D. Stanojlovic, Z.S. Markovic, Activation of oxidized surface of an-
thracite waste coal by attrition, Physicochem. Probl. Miner. Process. 48 (2012) 5–18.
[47] J.A. Orrego-Ruiz, R. Cabanzo, E. Mejía-Ospino, Study of Colombian coals using photo-
acoustic Fourier transform infrared spectroscopy, Int. J. Coal Geol. 85 (3) (2011)
307–310.
[48] M.S. Biswal, S.K. Jena, M.V. Rudramuniyappa, Study on flotation characteristics of
oxidized Indian high ash sub-bituminous coal, Int. J. Miner. Process. 87 (2008)
42–50.
[49] B. Wang, Y. Peng, S. Vink, Diagnosis of the surface chemistry effects on fine coal flo-
tation using saline water, Energy Fuel 27 (2013) 4869–4874.
[50] W.J. Zou, Selective flocculation-flotation separation of coking coal's middllings, Xu-
zhou: China University of Mining & Technology, School of Chemical Engineering
and Technology (2014) 36–38.