A coal desulfurization process via sodium metaborate electroreduction with pulse voltage using a boron-doped diamond thin film electrode

Article abstract of DOI:10.1039/c2ra21604c

A preliminary study was conducted on the coal desulfurization process via NaBO₂ electroreduction with pulse voltage using a boron-doped diamond (BDD) thin film electrode. It has been proved that NaBO₂ was converted into NaBH₄ by ¹¹B NMR. The factors that influence the conversion rate of NaBO₂ into NaBH₄ and coal desulfurization efficiency were investigated. Under the conditions of -1.5 V forward pulse voltage, +0.5 V reverse pulse voltage, 0.2 mol L^-1 NaBO₂ concentration, 0.5 mol L^-1 NaOH concentration, 2 s forward pulse duration, 1 s reverse pulse duration, 50 g L^-1 coal concentration, 0.8 mmol L^-1 NiCl₂ concentration and 2.5 h electrolytic time, a higher desulfurization efficiency (64%) was obtained compared with the common electrochemical desulfurization (ECDS) process using a Pt electrode. By analyzing and comparing the coal samples and electrolytes before and after desulfurization it was indicated that the removed S from the coal sample in the form of gaseous H₂S was mainly converted into Na₂S and Na₂S_x and boron (B) recycling was realized during coal desulfurization. Particularly, the combustion characteristics of coal were improved after desulfurization. Finally, the desulfurization mechanism was proposed. All these results indicated that the desulfurization process via NaBO₂ electroreduction with pulse voltage using a BDD thin film electrode is effective and highly promising for coal desulfurization.

Full text of DOI:10.1039/c2ra21604c

RSC Advances
View Article Online
View Journal | View Issue
PAPER
A coal desulfurization process via sodium metaborate
electroreduction with pulse voltage using a boron-
doped diamond thin film electrode3
Cite this: RSC Advances, 2013, 3,
1476
Chenhua Shu, Tonghua Sun,* Jinping Jia, Ziyang Lou and Yafei Shen
A preliminary study was conducted on the coal desulfurization process via NaBO2 electroreduction with
pulse voltage using a boron-doped diamond (BDD) thin film electrode. It has been proved that NaBO2 was
converted into NaBH4 by 11B NMR. The factors that influence the conversion rate of NaBO2 into NaBH4 and
coal desulfurization efficiency were investigated. Under the conditions of 21.5 V forward pulse voltage,
+0.5 V reverse pulse voltage, 0.2 mol L21 NaBO2 concentration, 0.5 mol L21 NaOH concentration, 2 s
forward pulse duration, 1 s reverse pulse duration, 50 g L21 coal concentration, 0.8 mmol L21 NiCl2
concentration and 2.5 h electrolytic time, a higher desulfurization efficiency (64%) was obtained compared
with the common electrochemical desulfurization (ECDS) process using a Pt electrode. By analyzing and
comparing the coal samples and electrolytes before and after desulfurization it was indicated that the
removed S from the coal sample in the form of gaseous H2S was mainly converted into Na2S and Na2Sx and
boron (B) recycling was realized during coal desulfurization. Particularly, the combustion characteristics of
coal were improved after desulfurization. Finally, the desulfurization mechanism was proposed. All these
results indicated that the desulfurization process via NaBO2 electroreduction with pulse voltage using a
BDD thin film electrode is effective and highly promising for coal desulfurization.
Received 29th July 2012,
Accepted 14th November 2012
DOI: 10.1039/c2ra21604c
www.rsc.org/advances
NaBO2, the hydrolytic by-product of NaBH4, is much cheaper
and more stable. It is of great importance to recycle NaBO2
1 Introduction
Sulfur dioxide (SO2) emissions from coal combustion have
been implicated as a cause of acid rain and other air pollution
related problems which can cause enormous economic
into NaBH4. Because electroreduction methods can realize the
conversion of NaBO2 into NaBH4, it is a good conception to
combine the NaBO2 electroreduction (electrochemical process)
with the NaBH4 reduction (chemical process) for reductive
desulfurization of coal to realize boron (B) recycling. Namely,
NaBO2 was converted into NaBH4 by electroreduction and
subsequently the generated NaBH4 was converted into NaBO2
again while it was used for the reductive desulfurization of
coal. Then, the above two processes were repeated. B recycling
can be realized at the same time as coal desulfurization. This
process is an integration of electrochemical and chemical
processes for reductive desulfurization (ECCRD). Based on this
conception, some research has been done.20,21 However, the
shortcomings of lower desulfurization efficiency due to the
lower conversion rate of NaBO2 into NaBH4 and the use of
inappropriate electrodes remain. Thus, it is necessary to
explore an effective method to improve the conversion rate of
NaBO2 into NaBH4. Many studies on the electroreduction of
NaBO2 into NaBH4 have been reported.22–27 The electroreduc-
tion mechanism is as follows:
losses.1
A number of physical, microbial and chemical
(including chemical oxidative and chemical reductive) coal
desulfurization methods have been reported. The physical
processes can only remove a portion of pyritic sulfur but
cannot remove organic sulfur.2,3 The microbial desulfurization
methods were usually not very effective.4–6 Chemical oxidative
desulfurization methods can remove most sulfurs, including
inorganic and organic forms, but the methods destroy the
macromolecular structure and properties of coal.7–14
Compared with oxidative desulfurization (OD), reductive
desulfurization (RD) has more incomparable advantages such
as mild conditions and short processing time. Sodium
borohydride is a good reducing agent in many industrial
processes, which was used in sulfur removal from gasoline and
coal recently.15–19 But the price of sodium borohydride is very
expensive due to the limited deposition of boron minerals.
School of Environmental Science and Engineering, Shanghai Jiaotong University,
Dongchuan Road 800, Shanghai, 200240, People’s Republic of China.
E-mail: sunth@sjtu.edu.cn
2
2
Cathode: BO2 + 6H2O + 8e2 A BH4 + 8OH2,
E1/2 = 21.24 V (vs. SHE)
(1)
Electronic supplementary information (ESI) available. See DOI: 10.1039/
3
c2ra21604c
Anode: 4OH2 24e2 A2H2O + O2q,E1/2 =1.229 V (vs. SHE) (2)
1476 | RSC Adv., 2013, 3, 1476–1485
This journal is ß The Royal Society of Chemistry 2013

View Article Online
RSC Advances
Paper
The above equations show that BO22 is reduced to BH42 at
the cathode, but due to charge repulsion it is very difficult for
the negatively charged BO22 to keep close to the surface of the
cathode. Thus, the conversion rate of NaBO2 into NaBH4 is
very low when constant voltage is applied. The pulse voltage is
a forward pulse (cathodic pulse) followed by a reverse pulse
(anodic pulse), which means that the working electrode will
turn into the anode after working for a period of time as the
cathode. When the working electrode works as the anode, the
BDD film electrode was used as the working electrode for
several reasons. Above all, the hydrogen evolution reaction
(2H2O + 2e2 A H2q + 2OH2) might take place at the same
time when the electroreduction of NaBO2 proceeded. The BDD
film electrode was capable of inhibiting the hydrogen
evolution reaction due to its low potential of hydrogen
evolution. Likewise, its high potential of oxygen evolution
can avoid the oxygen evolution reaction (4OH2 2 4e2 A 2H2O
+ O2q) when the reverse pulse was supplied (Fig. S1,
ESI ).31,34–36 Besides, its corrosion resistance makes it stable
2
3
negative charged BO2 will be attracted to the surface of the
in strongly alkaline solution and its low background current is
beneficial to improving current efficiency.
working electrode, and then the working electrode works as
2
the cathode, the BO2 gathered on the surface of the working
electrode will be reduced immediately into BH42 before being
excluded. Afterwards, the directions of pulse voltage continue
alternating, which means that the attraction and electroreduc-
In this paper, in order to improve the desulfurization
efficiency, the electroreduction of NaBO2 into NaBH4 was fulfilled
with pulse voltage using a BDD thin film electrode. The factors
that influence the conversion rate of NaBO2 into NaBH4 and the
desulfurization efficiency were investigated. Furthermore, the
results were compared with the common electrochemical
desulfurization (ECDS) using a Pt electrode. Finally, the desulfur-
ization was proposed by analysing and comparing the coal
samples and electrolytes before and after desulfurization.
2
tion of BO2 alternates on the working electrode. Thus, the
conversion rate of NaBO2 into NaBH4 will be improved greatly
by supplying pulse voltage.
In previous studies,20,21 Pb electrodes were mainly used as
the cathode in ECCRD processes. However, Pb was not a
perfect cathode material due to its narrow potential window,
high background current etc. Furthermore, Pb electrodes were
prone to corrosion in strongly alkaline solution. The voltages
for ECCRD processes were above 2 V and the Pb electrodes
wore down after working for a period of time in previous
studies. In recent years, the boron-doped diamond (BDD) film
has been considered a very promising electrode material. It
possesses several advantageous properties such as excellent
chemical stability, high electrical conductivity, a wide poten-
tial window and low background current.28–33 In this study the
2 Experimental
2.1 Materials
A coal sample was collected from Zunyi, Guizhou province, China,
which was ground and passed through a 140 mesh sieve. The
proximate and ultimate analyses of coal sample are given in
Table 1. A BDD thin film electrode was prepared by a hot filament
Table 1 Characterization of the coal samples
Treated coal
ECDSb
Parameters
Original coal
ECCRDa
ECODS
ECRDS
Ultimate analysis (wt%, db)
C
H
71.27
6.04
69.29(22.8%)
7.13
65.24(28.4%)
5.21
69.37(22.7%)
6.94
N
S
Odiff.
1.02
6.16
15.51
0.84
2.21(264.1%)
20.53
0.95
2.69(256.3%)
25.91
0.81
3.98(235.4%)
18.9
Sulfur content of different forms
(wt%, db)
Pyritic sulfur (PS)
Sulfate sulfur (SS)
Organic sulfur (OS)diff.
Combustion characteristic
1.93
0.57
3.66
0.35(281.9%)
0.12(278.9%)
1.74(252.5%)
0.39(283.4%)
0.11(280.7%)
2.19(240.1%)
1.07(244.6%)
0.34(240.3%)
2.57(229.8%)
Calorific value (J g21
)
24 267
483
24 703(+1.8%)
473(210)
20 187(216.8%)
476(27)
24 089(20.7%)
478(25)
Ignition (T/uC)
Proximate analysis (wt%, db)
Ash
20.1
8.2
1.6
13.8(231.3%)
6.5
1.4
25.3(+25.8%)
8.5
1.5
18.6(27.5%)
7.5
1.4
Volatile matter
Moisturec
Fixed carbondiff.
70.1
78.3(+11.7%)
64.7(27.7%)
72.5(+3.4%)
a
Process conditions: 21.5 V forward pulse voltage, +0.5 V reverse pulse voltage, 0.2 mol L21 NaBO2 concentration, 0.5 mol L21 NaOH
concentration, 2 s forward pulse duration, 1 s reverse pulse duration, 50 g L21 coal concentration, 0.8 mmol L21 NiCl2 concentration, 2.5 h
electrolytic time. Process conditions: 1 mol L21 NaOH concentration, 50 g L21 coal concentration, 2.5 V vs. SCE constant voltage.
b
c
Atmospheric conditions.
This journal is ß The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 1476–1485 | 1477

View Article Online
Paper
RSC Advances
chemical vapor deposition (HFCVD) technique on tantalum
substrate from acetone and hydrogen mixtures. Trimethyl borate
served as the boron source. NaBO2?4H2O (.99%, AR), NaOH
(.96%, AR) and NiCl2?6H2O (.98%, AR) were purchased from
Sinopharm Chemical Reagent Co. Ltd. (Shanghai China).
(40 6 10 mm) was used as the working electrode (WE). The
counter electrode (CE) was a graphite electrode (40 6 10 mm).
The reference electrode (RE) was a saturated calomel electrode
and all the experimental potentials reported were normalized to
this reference electrode. Pulse voltages were obtained from an
electrochemical workstation (Autolab PGSTAT30) with PC soft-
ware control (GPES 4.9). The generated NaBH4 was analyzed by
iodometric titration and 11B nuclear magnetic resonance. The
conversion rate of NaBO2 into NaBH4 was calculated by the
following eqn (3), where C1 (mol L21) is the NaBO2 concentra-
tion in original electrolytes and C2 (mol L21) is the NaBH4
concentration in treated electrolytes.
2.2 Electroreduction of NaBO2 into NaBH4 experiment
As shown in Fig. 1(a), the electroreduction was carried out in an
electrolytic cell with a cation exchange membrane separating
the working electrode and the counter electrode compartments.
An aqueous solution containing NaOH and a mixed aqueous
solution containing NaOH and NaBO2 were added into the
counter electrode and working electrode compartments,
respectively. The cation exchange membrane does not allow
BO22 or BH42 to transfer to the counter electrode compartment
Conversation rate (%) = (C2/C1) 6 100%
(3)
2
so as to maintain a high concentration of BO2 near the
2.3 Coal desulfurization experiment by ECCRD process
2
working electrode, which prevents the electrooxidation of BH4
As shown in Fig. 1(b), a magnetic stirrer and a gas pipeline for
H2S were added into the experimental setup of coal desulfur-
and BO22 on the working electrode. A BDD thin film electrode
Fig. 1 Schematic diagram of electroreduction experiment (a) and coal desulfurization experiment by ECCRD process (b).
1478 | RSC Adv., 2013, 3, 1476–1485
This journal is ß The Royal Society of Chemistry 2013

View Article Online
RSC Advances
Paper
ization, compared with the experimental setup of electro-
reduction (Fig. 1(a)). In order to collect the generated H2S, the
WE compartment was sealed. First, an aqueous solution
containing NaOH and a mixed aqueous solution containing
NaOH and NaBO2 were added into the CE and WE compart-
ments, respectively. Meanwhile, the prepared coal sample and
NiCl2 catalyst were added into the WE compartment. Then, the
magnetic stirrer and power were turned on in sequence and
the reaction began immediately. After the end of the reaction
the clean coal was obtained by filtering and leaching the coal
slurry in the working compartment. At last, the clean coal
sample and filtrate were used for s-content analysis and ICP
analysis, respectively. Simultaneously the electrolytes in the
counter compartment were also analyzed by ICP. The
desulfurization efficiency was calculated by the following eqn
(4), where TS1 is the sulfur content in original coal samples
and TS2 is the sulfur content in treated coal samples.
metric titration.38 The elemental analysis of electrolytes and
filtrates was determined using an inductively coupled plasma
(ICP, Iris Advantage 1000, Thermo King-CordCo., USA) instru-
ment. The data was presented as the average of two replicates
(data error ,5%) in each treatment.
3 Results and discussion
3.1 Electroreduction of NaBO2 into NaBH4 process
3.1.1 Effect of forward pulse voltage, reverse pulse voltage,
NaBO2 concentration, NaOH concentration on conversion
rate. Forward pulse voltage. The role of forward pulse voltage
2
was to reduce BO2 into BH42. According to the cyclic
voltammograms of BDD thin film electrode (Fig. S1, ESI ),
3
the reduction peak of NaBO2 to NaBH4 was observed at around
21.5 V. Although reducing the forward pulse voltage probably
improved the conversion rate, too low a voltage would lead to
hydrogen evolution reaction. Fig. 2(a) showed the effect of
forward pulse voltage on conversion rate, compared with
constant voltage. There was no NaBH4 to generate when
constant voltage was higher than 21.2 V; below that, the
maximum value of conversion rate reached 6.28% at 21.8 V.
However, when the pulse voltage was supplied, the maximum
value was up to 15.33% at the forward pulse voltage of 21.5 V.
The results suggest that the performance of pulse voltage
electrolysis was significantly better than that of constant
voltage electrolysis. When the forward pulse voltage was lower
Desulfurization efficiency (wt%)=1/TS1 (TS1 2TS2)6100% (4)
2.4 Coal desulfurization experiment by ECDS process
In order to compare with the ECCRD processes using BDD thin
film electrodes, the common electrochemical desulfurization
(ECDS) experiments were carried out using Pt electrodes.
According to the desulfurization mechanism, ECDS prcesses
can be divided into two categories: electrochemical oxidation
desulfurization (ECODS) and electrochemical reduction desul-
furization (ECRDS).37 The ECDS experiments were carried out
in an electrolytic cell with a cation exchange membrane
separating the anodic and cathodic compartments. For the
ECODS process, a Pt electrode (10 6 10 mm) was used as the
anode and a graphite electrode (10 6 10 mm) was used as the
cathode. For the ECRDS process, on the contrary, a Pt
electrode (10 6 10 mm) was used as the cathode and a
graphite electrode (10 6 10 mm) was used as the anode. Two
aqueous solutions containing NaOH were added into the the
anodic and cathodic compartments, respectively. Meanwhile,
the prepared coal sample was added into the Pt electrode
compartment. Then, the magnetic stirrer and power were
turned on in sequence and the reaction began immediately.
After the end of the reaction, the clean coal was obtained by
filtering and leaching the coal slurry. At last, the clean coal
samples were used for s-content analysis.
than 21.5
V the conversion rate decreased instead of
increasing due to the hydrogen evolution reaction. Thus,
21.5 V was considered as the optimum value of forward pulse
voltage.
Reverse pulse voltage. The role of reverse pulse voltage was to
2
attract BO2 to the surface of BDD thin film electrode. A high
2
reverse pulse voltage was beneficial for BO2 attraction to the
BDD thin film electrode. However, the oxygen evolution
2
reaction (Fig. S1, ESI ) and conversion of BO
to borax39
3
2
might take place when the reverse pulse voltage was too high.
On the contrary, a low reverse pulse voltage was not enough to
2
attract BO2 to the surface of the BDD thin film electrode. As
shown in Fig. 2(b), the conversion rate initially increased with
the increasing of reverse pulse voltage, and then decreased.
The maximum value of conversion rate was obtained when the
reverse pulse voltage was +0.5 V.
2.5 Analysis methods
Total sulfur (TS) was determined using an elemental analyzer
(Elementar Vario EL III, Germany); sulfate sulfur (SS) was
analyzed gravimetrically, using the Eschka method; pyritic
sulfur (PS) was analyzed by measuring the amount of iron
combined in the pyritic state (GB/T 215-2003); and organic
sulfur (OS) was estimated by the difference. The proximate
analysis of the coal sample followed the Chinese standard
method (GB/T 212-2001). The analysis of calorific value and
ignition temperature followed the Chinese standard methods
(GB/T 213-2003) and (GB/T 18511-2001), respectively. The
qualitative and quantitative detection of NaBH4 was fulfilled
by 11B nuclear magnetic resonance (Bruker Avance III 400
NMR spectrometer, room temperature, 128 MHz) and iodo-
NaBO2 concentration. NaBO2 is the source of NaBH4, so the
concentration of NaBO2 is a key factor of producing NaBH4. As
shown in Fig. 2(c), the conversion rate increased with the
increasing of NaBO2 concentration when the NaBO2 concen-
tration was less than 0.2 mol L21, but decreased when the
NaBO2 concentration was more than 0.2 mol L21. The reason
was that the oxidation and hydrolysis of NaBH4 intensified
with the increase of concentration of NaBH4. Namely, the total
NaBH4 concentration would remain constant while increasing
to a certain value. According to eqn (3), the conversion rate
would decrease with the increase of NaBO2 concentration
under constant NaBH4 concentration. However, the maximum
This journal is ß The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 1476–1485 | 1479

View Article Online
Paper
RSC Advances
Fig. 2 Effect of forward pulse voltage (a), reverse pulse voltage (b), NaBO2 concentration (c) and NaOH concentration (d) on conversion rate.
value of total NaBH4 concentration can be observed at the
NaBO2 concentration of 0.3 mol L21. The 0.2 mol L21 of NaBO2
concentration was used in the following experiments.
NaOH concentration. The maximum value of conversion rate
was obtained when the NaOH concentration was 0.5 mol L21
.
3.1.2 Effect of forward pulse duration, reverse pulse
duration and electrolytic time on conversion rate. The role of
forward pulse voltage was to reduce NaBO2 to NaBH4.
Increasing the forward pulse duration can improve the
conversion rate within a certain range, but if the forward
pulse duration is too long, the conversion rate would also
decrease because pulse numbers would reduce under constant
total electrolytic time. The role of reverse pulse voltage was to
NaOH concentration. There are two contradictory effects of
increasing NaOH concentration. On the one hand, it is well
known that NaBH4 is unstable and hydrolyzes spontaneously
when stored in aqueous solution. This hydrolysis reaction can
be greatly inhibited by the addition of NaOH because NaBH4
can be stable in the strong alkaline solutions. For example,
when the pH is 8, more than 50% of NaBH4 can hydrolyze
within 30 s. However, the half-life of NaBH4 is 430 d when the
pH is 14.40 On the other hand, increasing NaOH concentration
can reduce the potential of oxygen evolution reaction, which
makes it easier for oxygen evolution to take place when a
reverse pulse voltage is supplied. Furthermore, increasing
NaOH concentration can improve the viscosity of electrolyte,
2
attract BO2 to the surface of the BDD thin film electrode. If
the reverse pulse duration was too short, it wouldn’t be
2
enough to transfer BO2 to the surface of the BDD thin film
electrode. Conversely, the conversion rate would also decrease
2
2
because the electrooxidation of BO2 and BH4 would
increase. Tables 2a and 2b presented the experimental and
analytical results of the orthogonal array for the three factors.
Obviously, the most important influence factor was the reverse
pulse duration. However, the forward pulse duration had a
lesser effect on the conversion rate. The optimum values of the
forward and reverse pulse durations are 2 s and 1 s,
respectively. Increasing the electrolytic time was evidently
beneficial for improving the conversion rate, ranking only
2
which makes it difficult for BO2 to transfer to the surface of
the BDD thin film electrode. Thus, it was necessary to
investigate a reasonable NaOH concentration for the electro-
chemical reduction. As shown in Fig. 2(d), the conversion rate
increased initially, and then decreased with the increase of
1480 | RSC Adv., 2013, 3, 1476–1485
This journal is ß The Royal Society of Chemistry 2013

View Article Online
RSC Advances
Paper
Table 2 (a) Experimental design and results for electroreduction processes according to L9 (33) orthogonal arraya. (b) Data handling of the orthogonal array
Table 2a
Run no.
Forward pulse duration (s)
Reverse pulse duration (s)
Electrolytic time (h)
Conversion rate (%)
1
2
3
4
5
6
7
8
9
1
1
1
2
2
2
3
3
3
0.5
1
1.5
0.5
1
1.5
0.5
1
0.5
1
2
1
2
0.5
2
0.5
1
5.78
12.45
5.56
7.55
15.33
4.34
9.23
6.45
3.67
1.5
a
Other reaction conditions: 21.5 V forward pulse voltage, +0.5 V reverse pulse voltage, 0.2 mol L21 NaBO2 concentration, 0.5 mol L21 NaOH
concentration.
Table 2b
Vib (%)
Parameters
Forward pulse duration(s)
Reverse pulse duration(s)
Electrolytic time(h)
V1
V2
V3
Range
7.930
9.073
6.45
7.52
5.523
7.89
10.04
4.517
11.41
4.523
6.887
2.623
b
Vi: average of the conversion rate at the same level; range = (Vi)max 2 (Vi)min.
second to the reverse pulse duration. To further validate the
effect of electrolytic time, it was necessary to carry out some
supplementary experiments (Fig. 3(b)).
using a boron-doped diamond thin film electrode with pulse
voltage.
The electrolytic time is also of great importance to the
conversion rate of NaBO2 into NaBH4. As shown in Fig. 3(b), the
conversion rate increased before 2 h, after which the growth
trend slowed down rapidly. Considering economic efficiency, 2
h was selected as the optimum value of electrolytic time.
3.1.3 Confirmation of electroreduction of NaBO2 into NaBH4
by 11B NMR. The electrolytes before and after electroreduction
were analyzed by 11B NMR. Fig. 3(a) showed the results of 11B
NMR analysis. The resonance lines near 1.6 ppm can be
assigned to NaBO239 and those near 42 ppm can be assigned to
3.2 Coal desulfurization process
41
NaBH4
which appeared only in the electrolytes after
3.2.1 Effect of coal concentration on desulfurization
efficiency and total sulfur removal in the ECCRD process.
Coal concentration has an effect on the mass transfer, and the
electrochemical treatment. The result was in good agreement
with the results of iodometric titration which suggested that
NaBH4 was obtained through the electroreduction of NaBO2
Fig. 3 11B NMR spectrogram of electrolytes before and after electroreduction, Process conditions: 21.5 V forward pulse voltage, +0.5 V reverse pulse voltage, 0.2 mol
21 NaBO2 concentration, 0.5 mol L21 NaOH concentration , 2 s forward pulse duration, 1 s reverse pulse duration, 2 h electrolytic time (a). Effect of electrolytic time
L
on conversion rate (b).
This journal is ß The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 1476–1485 | 1481

View Article Online
Paper
RSC Advances
probability of contact with the active hydrogen atom (H*). Low
coal concentration benefited mass transfer, but it decreased
the contact frequency of H* and coal particles to some extent,
resulting in the reduction of desulfurization efficiency.
However, a high coal concentration inhibited mass transfer,
which also decreases desulfurization efficiency. As shown in
Fig. 4(a), the maximum value of desulfurization efficiency
(64.1%) was observed at a coal concentration of 50 g L21, while
the maximum value of total sulfur removal reached 2.08 g
low after electrolyzing for 5 h without NiCl2. However, with the
increasing of NiCl2 concentration, the desulfurization effi-
ciency increased and the treatment time decreased. When the
NiCl2 concentration increased to 0.8 mmol L21, the desulfur-
ization efficiency and treatment time remained almost stable.
The desulfurization efficiency was above 60% for 2.5 h
electrolysis. The possible reason (Fig. 1) is that nickel borides
were prepared by the reaction of NiCl2 with NaBH4 in protic
solvents (4NaBH4 + 2NiCl2 + 9H2O A Ni2B + 3H3BO3 + 4NaCl +
12.5H2q).47,48 Under the catalysis of nickel borides, NaBH4
can immediately react with H2O, producing amounts of H2 (or
active hydrogen). Then, the activated hydrogen was adsorbed
on the surface of nickel borides (Ni–B), forming into a kind of
Ni–MH intermediate product. On the other hand, the S atom
in coal contains lone-pair electrons, and Ni2+ had an empty
d-orbital, so S can react easily with Ni2+. Finally, the highly
activated hydrogen adsorbed on the surface of nickel boride
(Ni–B) attacked S bonded with a-C, resulting in C–S bond
cleavage.49,50
when the coal concentration was 60 g L21
.
3.2.2 Effect of NiCl2 concentration on desulfurization
efficiency in the ECCRD process. It is well known that metal
borides (e.g. NiB, CoB) are highly active catalysts which can be
generated readily from metal halides (e.g. NiCl2, CoCl2) and
boron hydrides (e.g. NaBH4) in protic conditions.42
Particularly, nickel borides have been employed as efficient
and convenient reagents for reductive desulfurization.43–46 In
this work, in order to improve the desulfurization efficiency,
NiCl2 was added to enhance the reducibility of NaBH4. As
shown in Fig. 4(b), the desulfurization efficiency was still very
Fig. 4 Effect of coal concentration on desulfurization efficiency and total sulfur removal in the ECCRD process; other reaction conditions: 21.5 V forward pulse
voltage, +0.5 V reverse pulse voltage, 0.2 mol L21 NaBO2 concentration, 0.5 mol L21 NaOH concentration, 2 s forward pulse duration, 1 s reverse pulse duration, 0.8
mmol L21 NiCl2 concentration, 2.5 h electrolytic time (a). Effect of NiCl2 concentration on desulfurization efficiency in the ECCRD process; other reaction conditions:
21.5 V forward pulse voltage, +0.5 V reverse pulse voltage, 0.2 mol L21 NaBO2 concentration, 0.5 mol L21 NaOH concentration, 2 s forward pulse duration, 1 s reverse
pulse duration, 50 g L21 coal concentration (b). Effect of electrolytic time on desulfurization efficiency in the ECDS process; reaction conditions: 1 mol L21 NaOH
concentration, 50 g L21 coal concentration, 2.5 V vs. SCE constant voltage (c).
1482 | RSC Adv., 2013, 3, 1476–1485
This journal is ß The Royal Society of Chemistry 2013

View Article Online
RSC Advances
Paper
deashing were 2.8 and 31.3%, respectively. The calorific value
of the coal sample increased from 24 267 to 24 703 J g21 (or by
1.8%). For the ECODS process, the efficiencies of decarboniza-
tion was 8.4%, but the ash content increased by 25.8%, so the
calorific value decreased from 24 267 to 20 187 J g21 (or by
16.8%). For the ECRDS process, the calorific value of the coal
sample remained almost unchanged. After desulfurization, all
three processes slightly lowered the ignition temperature of
the coal samples. These results indicate that the desulfuriza-
tion process by ECCRD is a relatively mild desulfurization
method, because not only can it remove S but can also improve
the combustion characteristics of coal.
3.2.3 Effect of electrolytic time on desulfurization efficiency.
Fig. 4(b) also showed the effect of electrolytic time on
desulfurization efficiency in the ECCRD process. The desulfur-
ization efficiency increased rapidly with the increasing of
electrolytic time when electrolytic time was less than 2 h, after
which, the growth trend slowed down. The desulfurization
efficiency reached more than 64% at 2.5 h of electrolytic time.
Fig. 4(c) showed the effect of electrolytic time on desulfuriza-
tion efficiency in the ECDS process. The maximum values of
desulfurization efficiency (56.3% for the ECODS process and
35.4% for the ECRDS process) were obtained at 4 h of
electrolytic time. In the same electrolytic time, the desulfur-
ization efficiency by the ECODS process was obviously higher
than that by the ECRDS process. In contrast to the ECRDS and
ECODS processes, the ECCRD process improved the desulfur-
ization efficiency in a shorter electrolytic time.
3.3 Mechanism of coal desulfurization in the ECCRD process
According to previous reports in the literature16–21 and the
analysis of experimental results, the possible desulfurization
mechanism was proposed as follows (eqn (1), (2) and (5)–(11)
and Fig. 1):
3.2.4 Mass balance in the ECCRD process. Table 3 presents
the ICP analytical results of the filtrates from the working
electrode compartment and the electrolytes from the counter
electrode compartment. The mass of TS removal: experimental
value (g L21) = desulfurization efficiency (wt%) 6 total sulfur
content (wt%) 6 coal concentration (g L21) = 64.1% 6 6.16%
6 50 g L21 = 2.12 g L21; estimated value (g L21) = S1 (sulfur in
the working electrode compartment, g L21) + S2 (sulfur in the
counter electrode compartment, g L21) = (0.95 + 1.11) = 2.06 g
WE: BO2 + 6H2O + 8e2 A BH4 + 8OH2
(1)
(5)
2
2
2
2
BH4 + 2H2O A BO2 + 4H2 (H*)
FeS2 + 2H* A Fe + S + H2Sq
2H* + R–SH A R–H + H2Sq
4H*+ R–S–R9 A R–H + R9H + H2Sq
H2S + 2NaOH A Na2S + 2H2O
CE: 4OH2 2 4e2 A 2H2O + O2q
H2S + 2NaOH A Na2S + 2H2O
2H2O + 2Na2S + O2 A 2S + 4NaOH
(x 2 1)S + Na2S A Na2Sx
(6)
(7)
L
21. Obviously, the estimated value of TS reduction is almost
in accordance with its corresponding experimental value,
which demonstrated that the removed is practically
(8)
S
converted into H2S. After treatment, Ni and B were reduced
(9)
from 0.047 to 0.043 g L21 (by 8.5%), and from 2.16 to 2.12 g
L
21 (by 1.8%), respectively. Thus, a large proportion of Ni and
(2)
B could be recycled and reused.
3.2.5 Coal characterization. Calorific value and ignition
temperature are two important combustion characteristics of
coal. Oxidative desulfurization usually tended to decrease the
calorific value due to carbon loss. However, the calorific value
of coal can also increase after desulfurization because the
desulfurization process can lead to its simultaneous deminer-
alization.51 Thus, how the calorific value changes after
desulfurization may depend on the ratio of decarbonitation
and demineralization.52 Table 1 presents the changes of coal
characteristics before and after desulfurization. For the
ECCRD process, the efficiencies of decarbonization and
(9)
(10)
(11)
As shown in the above eqn (1), (2) and (5)–(11) and Fig. 1, the
removed S in the form of gaseous H2S from the coal sample
was mainly converted into Na2S and Na2Sx. B recycling was
realized at the same time of coal desulfurization. Thus, the
ECCRD process reduced the additional pollution and realized
Table 3 ICP analytical results of filtrates and electrolytesa
Compartment
Element
Before reactionb (g L21
)
After reaction(g L21
)
Working electrode compartment
S
0
0
0.95
0.35
15.9
0.028
2.12
1.11
0.28
10.8
0.015
Fe
Na
Ni
B
16.1
0.047
2.16
0
0
11.5
0
Counter electrode compartment
S
Fe
Na
Ni
a
Process conditions: 21.5 V forward pulse voltage, +0.5 V reverse pulse voltage, 0.2 mol L21 NaBO2 concentration, 0.5 mol L21 NaOH
concentration, 2 s forward pulse duration, 1 s reverse pulse duration, 0.8 mmol L21 NiCl2 concentration, 50 g L21 coal concentration, 2.5 h
electrolytic time. Theoretical value.
b
This journal is ß The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 1476–1485 | 1483

View Article Online
Paper
RSC Advances
7 K. C. Chaung, R. Markuszewesky and T. D. Wheelock, Fuel
Process. Technol., 1983, 7, 43–57.
8 R. A. Rodriguez, C. C. Jul and D. G. Limon, Fuel, 1996, 75,
606–612.
the reasonable application and sustainable development of
resources which only used electricity and green solutions (B
recycling by electrolysis).
9 S. Mukherjee, S. Mahiuddin and P. C. Borthakur, Energy
Fuels, 2001, 15, 1418–1424.
4 Conclusions
10 R. Alvarez, C. Clemente and D. Gomez-Limon, Fuel, 2003,
82, 2007–2015.
Electrochemical synthesis of NaBH4 with pulse voltage using a
BDD thin film electrode and its application to coal desulfur-
ization was presented in this work. The results of iodometric
titration and 11B NMR suggested that NaBH4 was obtained
through the electrochemical reduction of NaBO2. The factors
that influenced the conversion rate of NaBO2 into NaBH4 and
coal desulfurization efficiency were investigated. Under the
conditions of 21.5 V forward pulse voltage, +0.5 V reverse
pulse voltage, 0.2 mol L21 NaBO2 concentration, 0.5 mol L21
NaOH concentration, 2 s forward pulse duration, 1 s reverse
pulse duration, 50 g L21 coal concentration, 0.8 mmol L21
NiCl2 concentration and 2.5 h electrolytic time, the desulfur-
ization efficiency reached more than 64%. In contrast with the
ECRDS and ECODS processes, the ECCRD process improved
the desulfurization efficiency with a lower voltage and in a
shorter electrolytic time. By analyzing and comparing the coal
samples and electrolytes before and after desulfurization it
was indicated that the removed S from coal sample in the form
of gaseous H2S was mainly converted into Na2S and Na2Sx and
that B recycling was realized simultaneous to coal desulfuriza-
tion. Additionally, after desulfurization, the combustion
characteristics of coal were improved and the BDD thin film
electrode remained nearly intact. Finally, the desulfurization
mechanism was proposed. In summary, the coal desulfuriza-
tion process via sodium metaborate electroreduction with
pulse voltage using a BDD thin film electrode is effective and
highly promising, which will open up new possibilities for coal
desulfurization.
11 K. Charutawai, S. Ngamprasertsisith and P. Prasassarakich,
Fuel Process. Technol., 2003, 84, 207–216.
12 W. D. Li and E. H. Cho, Energy Fuels, 2005, 19, 499–507.
13 B. P. Baruah, B. K. Saikia, P. Kotoky and P. G. Rao, Energy
Fuels, 2006, 20, 1550–1555.
14 H. G. Alam, A. Z. Mohaddam and M. R. Omidkhah, Fuel
Process. Technol., 2009, 90, 1–7.
15 Z. X. Wang, S. J. Wang, G. Y. Liu and W. B. Wang, Pet.
Process. Petrochem., 2007, 38, 6–9.
16 Z. L. Li, T. H. Sun and J. P. Jia, Fuel Process. Technol., 2010,
91, 1162–1167.
17 Y. F. Shen, T. H. Sun and J. P. Jia, Energy Fuels, 2011, 25,
2963–2967.
18 Y. F. Shen, T. H. Sun and J. P. Jia, RSC Adv., 2012, 2,
3123–3132.
19 Y. F. Shen, T. H. Sun and J. P. Jia, RSC Adv., 2012, 2,
4189–4197.
20 Y. F. Shen, X. L. Yang, T. H. Sun and J. P. Jia, Energy Fuels,
2011, 25, 5007–5014.
21 Y. F. Shen, T. H. Sun and J. P. Jia, Fuel, 2012, 96, 250–256.
22 G. F. Hu, F. Chapel, A. D. Mceiroy, E. City and R. M. Adams,
US Pat., 2 855 353, 1958.
23 H. B. H. Cooper, US Pat., 3 734 842, 1973.
24 H. Sharifian and J. S. Dutcher, US Pat., 4 904 357, 1990.
25 C. H. Hale and H. Sharifian, US Pat., 4 931 154, 1990.
26 S. Amendola, US Pat., 6 497 973 B1, 2002.
˙
27 A. E. Sanli, I. Kayacan, B. Z. Uysal and M. L. Aksu, J. Power
Sources, 2010, 195, 2604–2606.
28 M. Panizza and G. Cerisola, Electrochim. Acta, 2005, 51,
191–199.
29 C. Saez, M. Panizza, M. A. Rodrigo and G. Cerisola, J. Chem.
Technol. Biotechnol., 2007, 82, 575–581.
30 M. Panizza and A. Kapalka, Electrochim. Acta, 2008, 53,
2289–2295.
Acknowledgements
31 M. Panizza and G. Cerisola, Chemosphere, 2009, 77,
1060–1064.
32 S. C. Elaoud, M. Panizza, G. Cerisola and T. Mhiri, J. Chem.
Technol. Biotechnol., 2012, 87, 381–386.
This work is financially supported by the National Key
Technology R&D Program (No. 2010BAK69B24) and the
National Natural Science Foundation of China (No. 41173108).
33 G. M. Swain and R. Ramesham, Anal. Chem., 1993, 65,
345–351.
References
34 H. B. Martin, A. Argoitia, U. Landau, A. B. Anderson and J.
C. Angus, J. Electrochem. Soc., 1996, 143, L133–L136.
35 M. C. Granger, M. Witek, J. S. Xu, J. Wang, M. Hupert,
A. Hanks, M. D. Koppang, J. E. Butler, G. Lucazeau,
M. Mermoux, J. W. Strojek and G. M. Swain, Anal. Chem.,
2000, 72, 3793–3804.
36 D. A. Tryk, K. Tsunozaki, T. N. Rao and A. Fujishima,
Diamond Relat. Mater., 2001, 10, 1804–1809.
37 V. Lam, G. C. Li, C. J. Song, J. W. Chen, C. Fairbridge,
R. Hui and J. J. Zhang, Fuel Process. Technol., 2012, 98,
30–38.
1 C. F. You and X. C. Xu, Energy, 2010, 35, 4467–4472.
2 P. K. Naik, P. S. R. Reddy and V. N. Misra, Fuel Process.
Technol., 2004, 85, 1473–1485.
3 Z. F. Luo, Y. M. Zhao, Q. R. Chen, M. M. Fan and X. X. Tao,
Fuel Process. Technol., 2002, 79, 63–69.
4 C. Aacharya, R. N. Kar and L. B. Sukla, Fuel, 2001, 80,
2207–2216.
5 C. Aacharya, L. B. Sukla and V. N. Misra, Fuel, 2005, 84,
1597–1600.
6 H. He, F. F. Hong, X. X. Tao, L. Li, C. Y. Ma and Y. D. Zhao,
Fuel Process. Technol., 2012, 101, 73–77.
38 D. A. Lyttle, E. H. Jensen and W. A. Struck, Anal. Chem.,
1952, 24, 1843–1844.
1484 | RSC Adv., 2013, 3, 1476–1485
This journal is ß The Royal Society of Chemistry 2013

View Article Online
RSC Advances
Paper
39 E. H. Park, S. U. Jeong, U. H. Jung and S. H. Kim, Int. J.
Hydrogen Energy, 2007, 32, 2982–2987.
46 J. M. Khurana and D. Magoo, Synth. Commun., 2010, 40,
2908–2913.
40 W. Rong, J. Fudan Univ (Nat Sci), 1998, 37, 276–278.
41 S. Garroni, C. Milanese, D. Pottmaier, G. Mulas, P. Nolis,
A. Girella, R. Caputo, D. Olid, F. Teixdor, M. Baricco,
A. Marini, S. Surinach and M. Dolors, J. Phys. Chem. C,
2011, 115, 16664–16671.
42 B. Ganem and J. O. Osby, Chem. Rev., 1988, 86, 763–780.
43 J. M. Khurana, A. Ray and S. Singh, Tetrahedron Lett., 1998,
39, 3829–3832.
47 T. G. Back, J. K. Yang and H. R. Krouse, J. Org. Chem., 1992,
57, 1986–1990.
48 T. G. Back, D. L. Baron and K. Yang, J. Org. Chem., 1993, 58,
2407–2413.
49 J. C. Walter, A. Zurawski, D. Montgomery, M. Thornburg
and S. Revankar, J. Power Sources, 2008, 179, 335–339.
´
´
50 S. Schouten, D. Pavlovic, J. S. Sinninghe Damste and J.
W. de Leeuw, Org. Geochem., 1993, 19, 901–909.
51 B. P. Baruah and P. Khare, Energy Fuels, 2007, 21,
2156–2164.
44 J. M. Khurana, G. Kukreja and G. Bansal, J. Chem. Soc.,
Perkin Trans. 1, 2002, 2520–2524.
45 J. M. Khurana and G. Kukreja, J. Heterocycl. Chem., 2003,
40, 677–679.
52 K. C. Liu, J. Yang, J. P. Jia and Y. L. Wang, Chemosphere,
2008, 71, 183–188.
This journal is ß The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 1476–1485 | 1485