Optimization of Microporous Carbon Structures for Lithium–Sulfur Battery Applications in Carbonate-Based Electrolyte

Article abstract of DOI:10.1002/smll.201603533

Developing appropriate sulfur cathode materials in carbonate-based electrolyte is an important research subject for lithium-sulfur batteries. Although several microporous carbon materials as host for sulfur reveal the effect, methods for producing microporous carbon are neither easy nor well controllable. Moreover, due to the complexity and limitation of microporous carbon in their fabrication process, there has been rare investigation of influence on electrochemical behavior in the carbonate-based electrolyte for lithium-sulfur batteries by tuning different micropore size(0–2 nm) of carbon host. Here, we demonstrate an immediate carbonization process, self-activation strategy, which can produce microporous carbon for a sulfur host from alkali-complexes. Besides, by changing different alkali-ion in the previous complex, the obtained microporous carbon exhibits a major portion of ultramicropore (<0.7 nm, from 54.9% to 25.8%) and it is demonstrated that the micropore structure of the host material plays a vital role in confining sulfur molecule. When evaluated as cathode materials in a carbonate-based electrolyte for Li-S batteries, such microporous carbon/sulfur composite can provide high reversible capacity, cycling stability and good rate capability.

Full text of DOI:10.1002/smll.201603533

www.advancedsciencenews.com
Lithium–Sulfur Batteries
Optimization of Microporous Carbon Structures
for Lithium–Sulfur Battery Applications in
Carbonate-Based Electrolyte
Lei Hu, Yue Lu, Xiaona Li, Jianwen Liang, Tao Huang, Yongchun Zhu,*
and Yitai Qian*
Developing appropriate sulfur cathode materials in carbonate-based electrolyte
is an important research subject for lithium-sulfur batteries. Although several
microporous carbon materials as host for sulfur reveal the effect, methods for
producing microporous carbon are neither easy nor well controllable. Moreover, due
to the complexity and limitation of microporous carbon in their fabrication process,
there has been rare investigation of influence on electrochemical behavior in the
carbonate-based electrolyte for lithium-sulfur batteries by tuning different micropore
size(0–2 nm) of carbon host. Here, we demonstrate an immediate carbonization
process, self-activation strategy, which can produce microporous carbon for a sulfur
host from alkali-complexes. Besides, by changing different alkali-ion in the previous
complex, the obtained microporous carbon exhibits a major portion of ultramicropore
(<0.7 nm, from 54.9% to 25.8%) and it is demonstrated that the micropore structure
of the host material plays a vital role in confining sulfur molecule. When evaluated
as cathode materials in a carbonate-based electrolyte for Li-S batteries, such
microporous carbon/sulfur composite can provide high reversible capacity, cycling
stability and good rate capability.
studies of the lithium–sulfur batteries based on ether-based
1. Introduction
electrolyte.[5–10] Since the polysulfide shuttle effect in ether-
Lithium–sulfur batteries are considered to be a promising based electrolyte hampers lithium–sulfur battery applica-
electrochemical system owing to its high theoretical energy tions, efforts have been made to suppress solubility of lithium
density (2567 Wh Kg−1).[1–4] Until now, there are extensive polysulfide products. For instance, metal oxides,[11] metal
sulfides,[12] and special methods[13] have been introduced
to improve electrochemical performance. Compared with
ether-based electrolyte, the carbonate-based electrolyte as
traditional electrolyte in Li-ion batteries, with lower melting
point, lower cost, and higher oxidation potential, has raised
more expectations.[14,15] However, the sulfur cathode is not
Dr. L. Hu, Dr. Y. Lu, Dr. X. N. Li, Dr. J. W. Liang,
Dr. T. Huang, Prof. Y. C. Zhu, Prof. Y. T. Qian
Hefei National Laboratory for Physical
Science at Micro-Scale
Department of Chemistry
compatible with the conventional carbonate electrolytes,
since polysulfide anions react with carbonate solvents due to
the strong nucleophilic reactions.[16,17]
In order to avoid the nucleophilic reactions, recent studies
suggest that microporous carbon as host can effectively sup-
press the direct contact between the solvent molecule in
carbonate-based electrolytes and sulfur (or polysulfide inter-
mediates) molecule due to the pore size limitation.[18,19]
University of Science and Technology of China
Hefei, Anhui 230026, P. R. China
E-mail: ychzhu@ustc.edu.cn; ytqian@ustc.edu.cn
Dr. J. W. Liang, Prof. Y. T. Qian
School of Chemistry and Chemical Engineering
Shandong University
Jinan, Shandong 250100, P. R. China
DOI: 10.1002/smll.201603533
(1 of 10) 1603533
© 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
small 2017, 1603533
www.small-journal.com

full papers
www.advancedsciencenews.com
Thus, when sulfur is confined in micropores, a “quasi-solid- the restriction of sulfur and polysulfide intermediates during
state” reaction can only occur when Li ions migrate into the charge/discharge process in carbonate-based electrolyte.
the pores.[20,21] For the first time, Guo and co-workers dem- It is reported that an ultra-microporous material (d < 0.7 nm)
onstrated that a narrow pore size distribution of ≈0.5 nm has three times effective polysulfide intermediates absorption
microporous carbon as host material for sulfur cathode can as much as a supermicropore (0.7 < d < 2 nm) one.[29] Thus,
obtain a reversible capacity of 1142 mA h g−1 after 200 cycles optimization of microporous carbon structures by using a
at 0.1 C in carbonate-based electrolyte.[22] Lately, Huang simple and efficient approach is an important route in devel-
et al. adopted microporous carbon with pore size distribution opment of Li–S cells in carbonate-based electrolyte.
about 0.46 nm as host to avoid the reversible reaction during
In this work, we designed direct conversion of potas-
the cycling process. The specific capacity of obtained product sium or sodium complex precursors to different microporous
exhibits 608 mA h g−1 after 170 cycles at 0.5 A g−1 in car- carbon through carbonization process. Benefit from a high
bonate-based electrolyte.[20] Guo and co-workers exhibited pack density of precursor crystals, the fabrication process of
that porous carbon fibers with average pore size vary from microporous structure is efficient and uniform. By changing
0.67 to 1.2 nm as host materials for sulfur cathodes, which different alkali-ion in the previous complex, the obtained
exhibited reversible capacity of 900 mA h g−1 after 100 cycles microporous carbon exhibited a major portion of ultra-
at 0.16 A g−1.[23] Meanwhile, Du et al. designed 2D space micropore (<0.7 nm, from 54.9% to 25.8%). When evaluated
carbon from graphite and the distance between the carbon as host for sulfur cathode materials in carbonate-based elec-
layers is 0.4 nm, sulfur can be confined in sub-nanometer- trolyte, such a microporous carbon/sulfur composite provided
sized 2D space carbon interlayers. Thus, the electrochemical high reversible capacity of 968.4 mA h g−1 at 0.1 C after
performance shows a reversible capacity stabilizes around 50 cycles, cycling stability with more than 616.3 mA h g−1 at
600 mA h g−1.[24]
1 C after 600 cycles, and good rate capability (≈775.7 mA h g−1
Current methods of preparing microporous carbon typi- at 1.675 A g−1). Moreover, with the increase from 25.8% to
cally employ acid[19,25] or alkali[26] activated strategy, both 54.9% of the ultra-micropore in the resulting microporous
of which require using a relatively denser precursor carbon carbon materials, the Coulombic efficiency of their relevant
first. For acid activated strategy, Gao and co-workers used cells can be increased from 22.8% to 77.8%. These results
carbon microspheres as precursor by H2SO4 refluxing treat- demonstrated the potential of using microporous carbon or
ment, followed by high temperature activated process at even ultra-microporous carbon as a sulfur host in carbonate-
1000 °C for 2 h, to achieve microporous carbon. When further based electrolyte.
combined with sulfur (sulfur content of 40%), it exhibited
a stable discharge capacity of 650 mA h g−1 at a current
density of 400 mA g−1 after 500 cycles in carbonate-based
electrolyte.[19] In the meantime, KOH as alkali activator
2. Results and Discussion
was also employed to create microporous carbon due to the Scheme 1 illustrates the fabrication process for micropo-
formation of intercalated metallic K in carbon at high tem- rous carbon/sulfur composites. Using microporous carbon
perature. After the removal of the by-product by washing, derived from potassium tartrate as the example, the potas-
the expanded micropore carbon matrix formed during the sium tartrate was first carbonized at 800 °C under an inert
activation.[27] For instance, using KOH as activator, densest atmosphere of Ar for 1 h to obtain the carbonized sample
carbon microspheres that are synthesized from hydrothermal (Scheme 1a,b). X-ray diffraction (XRD) pattern of the
process can be changed to the microporous carbon with an resulting carbonized sample is shown in Figure S1a (Sup-
average pore size of 0.5 nm.[26] When using the resulting porting Information). In the figure, several main peaks that
microporous carbon as host to confine sulfur (sulfur content connect to the peaks of K2CO3⋅1.5H2O (JCPDF card number
of 40%), a stable capacity of 600 mA h g−1 after 800 cycles #01-073-0470) have been observed. After washing with
at 1675 mA g−1 in carbonate-based electrolyte was obtained.
hydrochloric acid and deionized water (Scheme 1b,c), the
Recently, Guo and co-workers have been focused on the by-product of K2CO3⋅1.5H2O can be removed completely.
use of one-step hydrothermal pyrolysis of glucose to fabri- Then, an amorphous material of carbon can be achieved (The
cate microporous carbon layer (pore size ≈0.5 nm) on the KC denoted in the following text indicates the microporous
surface of multiwalled carbon nanotubes.[22] Similarly, Cheng carbon derived from potassium tartrate, and NC and KNC
and co-workers reported a microporous–mesoporous carbon were similarly obtained from sodium tartrate and sodium
by direct carbonization of phenolic resol. When further com- potassium tartrate tetrahydrate, respectively.)). By using
posited with sulfur (sulfur content of 16%), a stable capacity this resulting carbon as host, a microporous carbon/sulfur
of 200 mA h g−1 after 800 cycles at 200 mA g−1 in carbonate- composite material has been produced by using melting
based electrolyte has been achieved.[28] Thus, the facile impregnation process (Scheme 1d,e). XRD patterns of the
control of microporous structure of carbon, with large pore microporous carbon/sulfur composite reveal the existent
volume, high surface area, and narrow pore range, by using of monoclinic crystal structure sulfur (γ-sulfur, JCPDF card
low-cost raw material, simple and controllable one-step syn- number #01-079-1517) and amorphous carbon (broad peak
thesis method, can improve the use of sulfur as cathode in the at about 23°) (Figure S2, Supporting Information). Finally,
application of Li–S batteries in carbonate-based electrolyte.
this composite was further heated at 250 °C in flowing
Moreover, we have discovered that different microporous argon to remove excessive external sulfur (Scheme 1e,f).
size (0–2 nm) of carbon host provided very different effect for The final products retain the amorphous nature (Figure S1d,
1603533 (2 of 10)
© 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
small 2017, 1603533
www.small-journal.com

www.advancedsciencenews.com
Scheme 1. Schematic illustration of the fabrication process for sulfur/microporous carbon composites.
Supporting Information), indicating the impregnation of mesopores (d < 2 nm). The surface area and pore volume dis-
sulfur into the pores of porous carbon by capillary force (the tribution analyses show in Figure 1b that the KNC sample
finial composite is denoted as KC/S).[30,31] The relating prod- exhibits less distinct features of the mesopores compared
ucts and XRD analysis for the fabrication process of NC/S to the structure of KC and NC. In the meantime, the KC
and KNC/S are also shown in Figures S1 and S2 (Supporting sample demonstrates higher ratio of ultra-microporous
Information).
volume among these samples. The pore size distribution of
The morphology of KC material which with lots of the resulting samples (KC, NC, and KNC) is displayed in
macropores exists in the interconnected 3D carbon has been Figure 1c. Also, an enlargement image for clearly revealing
revealed (Figure S3a, Supporting Information). Higher mag- the pore size distributions is shown in Figure 1d. From the
nification scanning electron microscopy (SEM) image in the figure, it can be seen that the micropores of KC are relatively
inset of Figure S3b (Supporting Information) further reveals narrow pore diameter at about 0.5 nm, which is consistent
the information on the surface of 3D porous carbon, whose with HK-plot method (Table 1). In contrast, the sample KNC
surface is rough. Transmission electron microscopy (TEM) shows pore sizes of 0.52, 0.58, 0.61, and 0.63 nm, respec-
image from Figure S4 (Supporting Information) further tively, which suggest a vast majority of the pores fall within
reveals that KC material is composed of some sheet-like the range of ultra-micropores (<0.7 nm) for the KC sample.
structure. Similarly, the TEM and SEM images of NC and In the meantime, Figure 1b displays pore volume of 0.3131,
KNC materials are provided in Figures S3 and S4 in the Sup- 0.4452, and 0.7037 cm3 g−1, respectively. Especially, KC and
porting Information. From the figure, it can be seen that NC KNC possess micropore volume as high as 67% and 91.4%,
and KNC materials have a similar morphology with KC.
respectively. While, the micropore volume of NC is only
To further examine the surface and characterize the about 44.5%. At the same time, the ratios of ultra-micropores
pore size of KC, NC, and KNC, N2 adsorption–desorption volume of KC, NC, and KNC are 54.9%, 25.8%, and 52%,
isotherms are measured. Figure 1a shows nitrogen adsorp- respectively. All of the detailed data of these porous struc-
tion–desorption isotherms of KC, NC, and KNC, indicating tures and pore volume are shown in Table 1 and Table 2
a typical type I isotherms, with high Brunauer–Emmett– for summary. After removing excessive external sulfur, the
Teller (BET) surface area of 732.61, 341.9, and 1579.8 m2 g−1, BET surface area of obtained composites (KC/S, NC/S,
respectively. According to IUPAC classification and previous and KNC/S) decreases from 732.61, 341.9, 1579.8 m2 g−1 to
reports,[29] the micropores are divided into ultra-micropores 21.11, 125.1, 25.52 m2 g−1, respectively. The corresponding
(d < 0.7 nm), super-micropores (0.7 < d < 2 nm), and pore volume decreases from 0.4452, 0.3131, 0.7037 cm3 g−1 to
(3 of 10) 1603533
© 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
small 2017, 1603533
www.small-journal.com

full papers
www.advancedsciencenews.com
Figure 1. a) Nitrogen adsorption–desorption isotherms of KC, NC, and KNC. b) Influence of various carbon sources on the pore volumes of the
activated carbons prepared from alkali-complexes. c) Pore size distributions of the three different micropore carbon. d) Enlargement image of
pore size distributions (0.3–1.0 nm).
0.0403, 0.1406, 0.0312 cm3 g−1 simultaneously. Data (Table S4 to amorphous) with few graphite layers. Elemental mapping
and Figure S10, Supporting Information) for KC/S, NC/S, and of KC/S composite reveals a uniform distribution of sulfur
KNC/S also demonstrate well dispersion of sulfur into the (Figure 2c−e). The energy-dispersive spectroscope (EDS)
micropores of the carbon materials.
mapping images taken from the region show that sulfur is
When using these resulting microporous carbons as host, well dispersed in the microporous carbon. Furthermore, the
the relating microporous carbon/sulfur composite materials KC/S composite is selected to test the distribution of sulfur
have been produced through melting impregnation and exces- by line-scanning (Figure 2f). The result demonstrates a sim-
sive external sulfur-removing process. Also, using the KC/S as ilar tendency at both sulfur and carbon, which is consistent
an example, TEM images of the KC/S composite exhibit the with the EDS mapping images.
similar morphology structure of the KC materials with some
X-ray photoelectron spectroscopy (XPS) is utilized to
sheet-like structure (TEM image in Figure 2a). From SEM analyze the extent of the binding between microspore carbon
images in Figure S5a,c (Supporting Information), it can be and sulfur. The XPS spectra of KC/S composite are shown
obviously seen that a certain amount of sulfur is adsorbed in Figure S6a (Supporting Information), disclosing the pres-
on the surface of the microporous carbon before removing ence of O 1s, C 1s, S 2s, and S 2p. The decomposition of the C
superficial sulfur. After removing superficial sulfur, the com- 1s leads to two peaks with their binding energy of 284.8 eV
[32]
[33]



posite almost has the same surface as the pure carbon and (C C/C C),
285.6eV (C O)
(Figure S6c, Supporting
no extra bulk sulfur is observed (Figure S5b,d, Supporting Information). For XPS spectrum of S 2p, the two peaks
Information). High resolution transmission electron micros- at 165.1 and 164.0 eV are assigned to S 2p1/2 and S 2p3/2,
copy (HRTEM) micrograph (Figure 2b) confirms that the respectively. Figure S6b (Supporting Information) also shows
KC/S composite indicates highly disordered structure (close the additional peak at 168.5 eV, revealing the existence of
Table 1. The specific surface area, pore size, total pore volume, micropore volume, mesopore volume, and pore volume ratios of microspore
carbon hosts.
a)
b)
c)
d)
Samples
Sbet
[m2 g−1
Pore size
[nm]
VT
Vmicro
Vmes
Vmicro
/
Vmicro(<0.7 nm)
/
Vmicro(0.7–2 nm)/
]
[cm3 g−1
0.3131
0.4452
0.7037
]
[cm3 g−1
0.1395
0.2985
0.6436
]
[cm3 g−1
0.1736
0.1467
0.0601
]
VT[%]
44.5
67.0
91.4
VT [%]
VT[%]
NC
341.95
732.61
1579.8
0.54
0.49
0.51
25.8
54.9
52.0
18.7
12.1
39.4
KC
KNC
a)The specific area (SBET) was calculated by the BET method; b)VT represented the total pore volume calculated by the BET method; c)The volume of micropore (Vmicro) calculated by HK-plot method;
d)The volume of mesopore (Vmeso) was obtained by subtracting the microporous volume from the total pore volume.
1603533 (4 of 10)
© 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
small 2017, 1603533
www.small-journal.com

www.advancedsciencenews.com
Table 2. The relationship between pore volume and electrochemical performance.
Sample
Vultramicro [d < 0.7 nm]
Vsupermicro [0.7 < d < 2 nm]
Vmicro [d < 2 nm]
CE
[%]
Stability
[%]
C rate [1 C]
[%]
[%]
[%]
[mAh g−1
]
KC/S
54.9
25.8
52.0
12.1
18.7
39.4
67.0
44.5
91.4
77.8
22.8
66.0
0.038 Per cycle
0.041 Per cycle
0.084 Per cycle
775
NC/S
KNC/S
70 mAh g−1
592 mAh g−1
sulfur molecules confined in the microporous carbon, which structure. Thermogravimetric (TG) analysis (Figure S7b,
is consistent with previous studies.[21,34] Similarly, NC/S and Supporting Information) suggests that the sulfur contents
KNC/S composite materials show the similar morphology of KC/S, NC/S, and KNC/S sample are 42.5%, 25.66%, and
and XPS spectra with KC/S.
49.73%, respectively. A single state in the TG curves indicates
Moreover, the Raman spectra (Figure S7a, Supporting superficial sulfur has removed, which is consistent with the
Information) of KC/S, NC/S, and KNC/S exhibit two prom- previous study.[21]
inent shifts, the peak located at ≈1596 cm−1 ascribes to the
The cyclic voltammetry (CV) tests of Li–S batteries using
E2g phonon of C sp2 atom (G-band), while the peak at KC/S (Figure 3a), NC/S (Figure 3b), and KNC/S (Figure 3c)
≈1356 cm−1 corresponds to the sp3-hybridized defect site as cathodes are performed within a potential window of
(D-band).[35,36] The intensity ratio (R) of D-band to G-band 1–3 V versus Li+/Li at 0.1 mV s−1. In the figure, all of the CV
(R = ID/IG) is an index of the disorder degree of the carbon. curves also display a couple of oxidation/reduction peaks.
The R values for the KC/S, NC/S, and KNC/S are 2.78, The CVs of KC, NC, and KNC show a single pair of redox
2.11, and 2.54, respectively, indicating a highly disordered peaks. Only one discharge plateau in CV curve implies the
Figure 2. a) TEM micrograph of KC/S, b) HRTEM micrograph of KC/S, c) high angle annular dark field scanning transmission electron
microscope(HAADF-STEM) image of KC/S. d,e) Elemental mapping of KC/S composite. f) Corresponding line-scanning of sulfur and carbon
elements across a selected area (white line) of the sulfur–carbon composite. g) The distribution of carbon and sulfur profiles.
(5 of 10) 1603533
© 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
small 2017, 1603533
www.small-journal.com

full papers
www.advancedsciencenews.com
Figure 3. The first five cycling voltammograms of a) KC/S, b)NC/S, and c) KNC/S in carbonate-based electrolyte at 0.1 mV s−1
.
reduction from sulfur molecules to Li2S, which is consistent exhibits good cycling stability. After 50 cycles, the KC/S elec-
with “quasi-solid-state” reaction mechanism of Li–S batteries trode still maintains a charge capacity of 968.4 mA h g−1, which
in sub-nanoconfinement reported by other groups.[21,34] For is ≈95% of the capacity in the second cycle (1023.5 mA h g−1).
better comparison of the CV curves of KC/S, NC/S, and In contrast, after 50 cycles at 0.1 C, the NC/S and KNC/S elec-
KNC/S, the locations of oxidation and reduction peaks in trodes exhibit a charge capacity of 232 and 803.4 mA h g−1,
each scan are listed in Table S1 (Supporting Information). respectively. The capacities of NC/S and KNC/S electrodes
The Eo and Er are corresponding to the oxidation potential are much lower than that of KC/S electrode. The detailed
and reduction potential, respectively. The voltage polariza- galvanostatic discharge/charge profiles of KC/S, NC/S, and
tion of KC/S sample is about 0.6 V for a redox couple, which KNC/S electrodes are shown in Figure 4b–d. From the figure,
is lower than the other samples, indicating the good electro- it can be calculated that the initial Coulombic efficiency (CE)
chemical kinetics and low electrode polarizations compared of KC/S, NC/S, and KNC/S are 77.8%, 22.8%, and 66.0%,
with NC/S and KNC/S composites. The sharp rise of reduc- respectively. It is found that the CE of KC/S in the first cycle
tion peaks at the end of first scan should be responsible for is much higher than those obtained by other groups.[22,25,26]
the formation of solid electrolyte interphase (SEI) layer.[37] Moreover, the potential of the KC/S electrode in the first
After the first cycle, the subsequent CV curves of KC/S are galvanostatic cycle is slightly higher than that of the other
well overlapped, which indicates the good reversibility for two samples, which match well with CV tests. After the first
the electrochemical reaction.
cycle, the average CE of the electrode from the 2nd to the
To understand the electrochemical performance of the 50th cycle is nearly 100% (Figure 4a), which suggests that the
as-prepared microporous carbon/sulfur composite electrodes, microporous carbon/sulfur composite electrode with a stable
the cycling performance of the as-prepared electrodes is and reversible cycling performance is under the protection of
measured at 0.1 C. As shown in Figure 4a, the KC/S electrode SEI layer and microporous carbon.
Figure 4. a) Cycling performance of KC/S, NC/S, and KNC/S electrodes at 0.1 C. b) Galvanostatic discharge/charge profiles of KC/S in the first
and second cycles. c) Galvanostatic discharge/charge profiles of NC/S in the first and second cycles. d) Galvanostatic discharge/charge profiles
of KNC/S in the first and second cycles.
1603533 (6 of 10)
© 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
small 2017, 1603533
www.small-journal.com

www.advancedsciencenews.com
Moreover, we also evaluate the rate performance of these cells of microporous carbon/sulfur cathodes are also
electrodes. Figure 5a–c shows galvanostatic discharge/ performed in a long cycling charge/discharge process at the
charge profiles of KC/S, NC/S, and KNC/S at different cur- current density of 1 C (Figure 5e). As a result, cyclic perfor-
rent densities, respectively. For KC/S, when the current rate mance of KC/S cathode can maintain 616.3 mAh g−1 after
gradually increased from 0.05 to 0.1 C, the corresponding 600 cycles, with CE stabilized at 99%, indicating an excellent
discharge capacities are measured to be 984.9, 903.2, 828.8, stability of the electrode upon cycling, which shows the best
and 775.7 mA h g−1, respectively. After undergoing various electrochemical performance. The capacity decay rate is as
charge–discharge rates, the KC/S swiftly resumes its capacity low as 0.038%. Whereas, the NC/S cathode also shows a high
of about 915 mA h g−1 when the rate is brought back to value (0.041% per cycle) and KNC/S suffers fastest capacity
0.05 C and the CE is nearly 100%. By contrast, NC/S and fading (0.084% per cycle). Long-term cycle stability fur-
KNC/S get lower rate capacities and deteriorated cycling sta- ther demonstrates ultra-micropore (<0.7 nm) of carbon can
bility, KNC/S composite electrode exhibits almost the same more effectively trap elemental sulfur and polysulfide due to
capacity of 967.6 mA h g−1 at 0.05 C compared with KC/S. stronger absorption. To get further insight into the excellent
Remarkably, when the current rate is switched back to 0.05 C, electrochemical performance of KC/S composite cathode,
the capacity can only be restored to ≈826.7 mA h g−1. In addi- the morphologies of the KC/S composite cathode after 50
tion, the NC/S composite shows worst rate performance, with cycles at 0.1 C are characterized. As shown in Figure S9
specific capacities of 278.7, 230.6, 122.4, and 70 mA h g−1 at (Supporting Information), after 50 cycles, the KC/S com-
0.05, 0.1, 0.5, and 1 C, respectively (Figure 5d).
posite still maintains its integrity and shows no large cracks.
In application, long-term cycling performance is also In addition, the list of microporous carbon/sulfur com-
another point for the evaluation of the electrode. Thus, posites reported for lithium–sulfur battery applications in
Figure 5. a) Galvanostatic discharge/charge profiles of KC/S with different current densities. b) Galvanostatic discharge/charge profiles of NC/S
with different current densities. c) Galvanostatic discharge/charge profiles of KNC/S with different current density. d) Rate capability at different
rates (increased from 0.05 to 1 C). e) Cycling performance of KC/S, NC/S, and KNC/S at a current density of 1 C.
(7 of 10) 1603533
© 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
small 2017, 1603533
www.small-journal.com

full papers
www.advancedsciencenews.com
are occupied with sulfur molecules (Figure S10, Supporting
Information), sulfur molecules in supermicropores may react
with carbonate-based electrolyte during cycling. For KC/S
and KNC/S, there are no significant differences in the ini-
tial capacity. However, during the long-term cycle process,
capacity decay rate has a similar tendency with ratios of
super-micropore volume. Therefore, KNC cannot host sulfur
molecules well in carbonate-based electrolyte, although
KNC provides the highest specific area. Among these sam-
ples, KC/S composite demonstrates excellent reversibility in
conventional carbon-based electrolyte due to the existence
of ultra-micropores. Furthermore, ultra-micropore provides
a stronger absorption of sulfur and polysulfide intermediates
during the charge/discharge process.[29] Hence, we simply
draw a conclusion that ultra-microporous carbon struc-
ture is an ideal sulfur host in carbonate-based electrolyte.
To increase sulfur loading in composites and improve the
cycle performance in carbonate-based electrolyte, specific
efforts need to focus on (1) increasing the ultra-micropores
volume ratio in carbon, and (2) avoiding producing larger
Figure 6. Electrochemical impedance spectra of as-prepared batteries.
Inset: The equivalent-circuit model of C/S composites for as-prepared
batteries.
carbonate-based electrolyte is provided in Table S2 (Sup- sub-nanometer space (0.7–2 nm). This research hopes to pro-
porting Information). vide a considerable choice of using ultra-micropore structure
Figure 6 displays the impedance spectra of as-prepared carbon as a host for lithium–sulfur batteries and contributing
batteries at the fresh state. For all three electrodes, the semi- to further enhance battery life and energy density in car-
circle in the high- and medium-frequency regions and an bonate-based electrolyte.
inclined line in the low-frequency zone are observed. The
depressed semicircle in the high- and medium-frequency
regions can be related to the combination of a solid/electro-
lyte interface film resistance and charge-transfer resistance
3. Conclusion
of electrochemical reaction. The line in the low-frequency In summary, a series of microporous carbons were prepared
region is designated to the diffusion-controlled Warburg through a novel and facile strategy of immediate carboni-
impedance.[38,39] Based on this, we use the R(CR)W equiva- zation of alkali-complexes. It is found that the micropore
lent circuit to fitting the data.The fitted results and equivalent structure of the microporous carbon host material has a key
circuit model were built in Figure 6. Besides, the fitting results influence on confining sulfur molecule, which determines the
of R1 and R2 were displayed in Table S3 (Supporting Informa- basic electrochemical behavior of the sulfur cathodes in car-
tion), respectively. In the equivalent circuit, the R1 repre- bonate-based electrolyte. As a demonstration, KC/S delivers
sents the resistance of the electrolyte. The semicircle noted a high specific capacity of 968.4 mA h g−1 after 50 cycles at
by R2//CPE is attributed to the resistance and capacitance 0.1 C or 616.3 mAh g−1 after 600 cycles at 1 C and manifests
in the high- and medium-frequency regions. In the mean- an outstanding rate performance among these composites.
time, the inclined line was modeled by Warburg impedance Furthermore, this work would open an avenue for developing
Zw, which refers to the diffusion of ions.[38,39] The results in microporous or even ultra-microporous carbon for future
Table S3 (Supporting Information) and Figure 6 show that sulfur host in carbonate-based electrolyte, which contributes
the impedance of R2 in KC/S electrode is the smallest, while to the further enhancement of battery performances.
the impedance of NC/S and KNC/S composites increased.
The charge-transfer resistance (R2) of KC/S is smaller than
the other S cathodes, indicating the KC/S sample with good
electrochemical kinetics.
4. Experimental Section
Table 2 shows the relationship between pore structure of
Materials: Potassium tartrate (C4H4K2O6⋅0.5H2O, 99 wt%),
the microporous carbon and electrochemical performance sodium tartrate (C4H4Na2O6⋅2H2O, 99 wt%), sodium potassium
of the microporous carbon/sulfur composite. Obviously, tartrate tetrahydrate (C4H4KNaO6⋅4H2O, 99 wt%), and hydrochloric
the ratio of pore volume about super-micropore or ultra- acid (HCl, 36.0–38.0 wt%) were purchased from Shanghai Chem-
micropore of carbon host influences the electrochemical ical Ltd. All chemicals were of analytic grade and used without
performance of sulfur cathodes deeply. The initial Coulombic further purification.
efficiency of the cell turns higher as the ratio of ultra-
Preparation of Microporous Carbon: 5 g of potassium tartrate
micropore volume became higher, owing to the fact that was loaded in a corundum crucible, and then directly heated to
irreversible electrochemical reactions happen only at slightly a target temperature (800 °C) in a horizontal tube furnace at a
larger space(>0.7 nm).[40] In the meantime, the irreversible heating rate of 5 °C min−1. The sample was kept at this tempera-
reaction in the initial cycle is controlled well in ultra-micro- ture for 1 h under flowing Ar atmosphere. After cooling to room
porous matrix. Considering almost all of the micropores temperature, the obtained black powder was washed with diluted
1603533 (8 of 10)
© 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
small 2017, 1603533
www.small-journal.com

www.advancedsciencenews.com
hydrochloric acid, subsequent deionized water, and ethanol until
pH came to 7. Finally, the black carbon (KC) was dried at 80 °C at
oven. NC and KNC were similarly obtained from sodium tartrate
and sodium potassium tartrate tetrahydrate, respectively.
Acknowledgements
This work was financially supported by the National Natural Sci-
ence Fund of China (Grant Nos. 21471142, 21521001, 21671181,
21671183, and 51602303) and the China Postdoctoral Science
Foundation (Grant Nos. 2015M580548, 2015M582083, and
2016T90581).
Synthesis of Microporous Carbon/Sulfur Composites: 0.2 g
of above microporous carbon(KC, NC, or KNC) and sulfur pow-
ders (1 g) were ballmilled with ethanol media in a planetary
ball mill (QM-3SP04) at 300 rpm for 0.5 h, then the mixture was
sealed in a quartz tube under vacuum and heated at 300 °C for
5 h (heating rate 5 °C min−1). After heating treatment of the mix-
ture, the tube was cooled to room temperature naturally. The
products were heated to 250 °C for 2 h under Ar atmosphere to
remove superficial sulfur from outside the microporous carbon.
The finial composites were denoted as NC/S, KC/S, and KNC/S,
respectively.
Material Characterization: The intermediate products and
microporous carbon/sulfur composites were characterized by XRD
(Philips X’ Pert Super diffractometer, Cu Kα, λ = 1.54178 Å). The
morphology of the microporous carbon was characterized using
SEM (JSM-6700F) and TEM (H7650). EDS mapping was digitally
acquired using HRTEM (Talos F200X). TEM with the as-prepared
samples disperses on 200 mesh lacey-carbon copper TEM grids.
The Raman spectrum was performed at the wavelength of
514.5 nm. XPS data were recorded on an ESCALAB 250 spec-
trometer (Perkin-Elmer) at room temperature with a Kratos Ana-
lytical spectrometer and a monochromatic Al Kα (1486.6 eV) X-ray
source. The thermogravimetric analysis (TGA) was performed using
TGA Q5000IR system under Ar atmosphere. The BET surface area
and HK-plot pore distribution were measured by BEL SORP-max
machine, BEL, Japan. All the pore size tests were carried out before
the vacuum degree reached 3 × 10−4 Pa.
[1] J. Liang, Z.-H. Sun, F. Li, H.-M. Cheng, Energy Storage Mater.
2016, 2, 76.
[2] A. Manthiram, S. H. Chung, C. Zu, Adv. Mater. 2015, 27, 1980.
[3] Z. W. Seh, Y. Sun, Q. Zhang, Y. Cui, Chem. Soc. Rev. 2016, 45,
5605,
[4] A. Rosenman, E. Markevich, G. Salitra, D. Aurbach, A. Garsuch,
F. F. Chesneau, Adv. Energy Mater. 2015, 5, 1500212.
[5] B. Li, S. Li, J. Xu, S. Yang, Energy Environ. Sci. 2016, 9, 2025.
[6] G. Li, J. Sun, W. Hou, S. Jiang, Y. Huang, J. Geng, Nat. Commun.
2016, 7, 10601.
[7] Z. Yuan, H. J. Peng, T. Z. Hou, J. Q. Huang, C. M. Chen, D. W. Wang,
X. B. Cheng, F. Wei, Q. Zhang, Nano Lett. 2016, 16, 519.
[8] L. Sun, D. Wang, Y. Luo, K. Wang, W. Kong, Y. Wu, L. Zhang,
K. Jiang, Q. Li, Y. Zhang, J. Wang, S. Fan, ACS Nano 2016, 10,
1300.
[9] X. Liang, L. F. Nazar, ACS Nano 2016, 10, 4192.
[10] L. Ma, H. L. Zhuang, S. Wei, K. E. Hendrickson, M. S. Kim, G. Cohn,
R. G. Hennig, L. A. Archer, ACS Nano 2016, 10, 1050.
[11] X. Liang, C. Hart, Q. Pang, A. Garsuch, T. Weiss, L. F. Nazar, Nat.
Commun. 2015, 6, 5682.
[12] Z. Ma, Z. Li, K. Hu, D. Liu, J. Huo, S. Wang, J. Power Sources 2016,
325, 71.
[13] Z. Ma, X. Huang, Q. Jiang, J. Huo, S. Wang, Electrochim. Acta
2015, 182, 884.
Electrochemical Measurement: The active material slurries
were prepared by ballmilling mixture of 80% as-prepared S com-
posites (KC/S, NC/S, and KNC/S), 10% super P carbon black, 10%
poly(vinylidene fluoride) binder in N-methyl pyrrolidone. The slur-
ries were pasted onto an Al foil and dried at 60 °C under vacuum
for 15 h. After drying, the Al foil was pressed by rolling and cut
for a wafer (Φ = 12 mm) to fabricate the sulfur electrode. The
[14] Z. Xu, J. Wang, J. Yang, X. Miao, R. Chen, J. Qian, R. Miao, Angew.
Chem. Inter. Ed. 2016, 128, 10528.
[15] W. Weng, V. G. Pol, K. Amine, Adv. Mater. 2013, 25, 1608.
[16] J. Gao, M. A. Lowe, Y. Kiya, H. D. Abruña, J. Phys. Chem. C 2011,
115, 25132.
[17] T. Yim, M.-S. Park, J.-S. Yu, K. J. Kim, K. Y. Im, J.-H. Kim, G. Jeong,
Y. N. Jo, S.-G. Woo, K. S. Kang, I. Lee, Y.-J. Kim, Electrochim. Acta
2013, 107, 454.
loading mass on the Al foil was determined to be about 3 mg cm−2 [18] J. Wang, Y. S. He, J. Yang, Adv. Mater. 2015, 27, 569.
(sulfur loading about 1 mg cm−2). The coin-type half cells (2016
R-type) were carefully packed under an argon-filled glove box
[19] B. Zhang, X. Qin, G. R. Li, X. P. Gao, Energy Environ. Sci. 2010, 3,
1531.
[20] Y. Xu, Y. Wen, Y. Zhu, K. Gaskell, K. A. Cychosz, B. Eichhorn,
K. Xu, C. Wang, Adv. Funct. Mater. 2015, 25, 4312.
[21] Z. Li, L. Yuan, Z. Yi, Y. Sun, Y. Liu, Y. Jiang, Y. Shen, Y. Xin,
(H2O, O2 < 1 ppm). Lithium foils were used as the counter elec-
trode, and 1.0 mol L−1 LiPF6 dissolved in mixed solvent of ethylene
carbonate (EC) and dimethyl carbonate (DMC) (v/v = 1/1) was
used as electrolyte. CV was carried out using an electrochemical
workstation (AUT302N, Metrohm, Switzerland) at a scan rate of
0.1 mV s−1. The alternating-current (AC) impedance spectra were
performed on electrochemical workstation (AUT302N, Metrohm,
Switzerland) with AC amplitude of 5 mV in the frequency range of
100 kHz to 0.01 Hz. The obtained spectra were analyzed by using
Nova software.
Z. Zhang, Y. Huang, Adv. Energy Mater. 2014, 4, 1301473.
[22] S. Xin, L. Gu, N. H. Zhao, Y. X. Yin, L. J. Zhou, Y. G. Guo, L. J. Wan,
J. Am. Chem. Soc. 2012, 134, 18510.
[23] C. Fu, B. M. Wong, K. N. Bozhilov, J. Guo, Chem. Sci. 2016, 7, 1224.
[24] W. C. Du,J. Zhang, Y. X. Yin, Y. G. Guo, L. J. Wan, Chem.-Asian
J. 2016, 11, 2690.
[25] W. Zhang, D. Qiao, J. Pan, Y. Cao, H. Yang, X. Ai, Electrochim. Acta
2013, 87, 497.
[26] H. Ye, Y.-X. Yin, S. Xin, Y.-G. Guo, J. Mater. Chem. A 2013, 1, 6602.
[27] J. Wang, S. Kaskel, J. Mater. Chem. 2012, 22, 23710.
[28] D. W. Wang, G. Zhou, F. Li, K. H. Wu, G. Q. Lu, H. M. Cheng,
I. R. Gentle, Phys. Chem. Chem. Phys. 2012, 14, 8703.
[29] F. Hippauf, W. Nickel, G.-P. Hao, K. Schwedtmann, L. Giebeler,
S. Oswald, L. Borchardt, S. Doerfler, J. J. Weigand, S. Kaskel, Adv.
Mater. Interfaces 2016, 3, 1600508.
Supporting Information
[30] X. Li, Y. Cao, W. Qi, L. V. Saraf, J. Xiao, Z. Nie, J. Mietek,
J.-G. Zhang, B. Schwenzer, J. Liu, J. Mater. Chem. 2011, 21,
16603.
Supporting Information is available from the Wiley Online Library
or from the author.
(9 of 10) 1603533
© 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
small 2017, 1603533
www.small-journal.com

full papers
www.advancedsciencenews.com
[31] X. Li, J. Liang, K. Zhang, Z. Hou, W. Zhang, Y. Zhu, Y. Qian, Energy [37] C. Luo, Y. Zhu, O. Borodin, T. Gao, X. Fan, Y. Xu, K. Xu, C. Wang,
Environ. Sci. 2015, 8, 3181. Adv. Funct. Mater. 2016, 26, 745.
[32] L. Fei, X. Li, W. Bi, Z. Zhuo, W. Wei, L. Sun, W. Lu, X. Wu, [38] W. Cai, G. Li, F. He, L. Jin, B. Liu, Z. Li, J. Power Sources 2015, 283,
K. Xie, C. Wu, H. L. Chan, Y. Wang, Adv. Mater. 2015, 27,
5936.
[33] S. Niu, W. Lv, G. Zhou, Y. He, B. Li, Q. H. Yang, F. Kang, Chem.
Commun. 2015, 51, 17720.
[34] Z. Li, Y. Jiang, L. Yuan, Z. Yi, C. Wu, Y. Liu, P. Strasser, Y. Huang,
ACS Nano 2014, 8, 9295.
524.
[39] Z. Deng, Z. Zhang, Y. Lai, J. Liu, J. Li, Y. Liu, J. Electrochem. Soc.
2013, 160, A553–A558-.
[40] E. Markevich, G. Salitra, A. Rosenman, Y. Talyosef, F. Chesneau,
D. Aurbach, J. Mater. Chem. A 2015, 3, 19873.
[35] A. C. Ferrari, D. M. Basko, Nat. Nanotechnol. 2013, 8, 235.
[36] Z. Ma, S. Dou, A. Shen, L. Tao, L. Dai, S. Wang, Angew. Chem. Int.
Ed. 2015, 54, 1888.
Received: October 21, 2016
Revised: November 22, 2016
Published online:
1603533 (10 of 10)
© 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
small 2017, 1603533
www.small-journal.com