Disordered mesoporous carbon as polysulfide reservoir for improved cyclic performance of lithium–sulfur batteries

Disordered mesoporous carbon as polysulfide reservoir for improved cyclic performance of lithium–sulfur batteries

CARBON 6 8 ( 2 0 1 4 ) 2 6 5 –2 7 2 Available at www.sciencedirect.com ScienceDirect journal homepage: www.elsevier.com/locate/carbon Disordered m...

2MB Sizes 4 Downloads 67 Views

CARBON

6 8 ( 2 0 1 4 ) 2 6 5 –2 7 2

Available at www.sciencedirect.com

ScienceDirect journal homepage: www.elsevier.com/locate/carbon

Disordered mesoporous carbon as polysulfide reservoir for improved cyclic performance of lithium–sulfur batteries Min-Sik Park a, Bo Ock Jeong b, Tae Jeong Kim b, Seok Kim c, Ki Jae Kim a, Ji-Sang Yu a, Yongju Jung b,*, Young-Jun Kim a,* a

Advanced Batteries Research Center, Korea Electronics Technology Institute, 68 Yatap-dong, Bundang-gu, Seongnam 463-816, Republic of Korea b Department of Chemical Engineering, Korea University of Technology and Education, Cheonan 330-708, Republic of Korea c Department of Chemical Engineering, Pusan National University, San 30, Jangjeon-dong, Geumjeong-gu, Pusan 609-735, Republic of Korea

A R T I C L E I N F O

A B S T R A C T

Article history:

Mesoporous carbon, which was templated by colloidal silica, was added to a sulfur cathode

Received 15 September 2013

as a functional material to confine polysulfides to improve the cyclic performance of

Accepted 1 November 2013

lithium–sulfur batteries. To investigate the effect of the pore size and the pore volume of

Available online 9 November 2013

mesoporous carbon on the absorption characteristics of the Li-polysulfides, mesoporous carbons with various pore sizes and total pore volumes were prepared by varying the size and the amount of colloidal silica templates. The results show that mesoporous carboncontaining sulfur cathode enhanced the cyclic performance of the batteries significantly. Comparable performances were observed regardless of pore size, suggesting that the pore size is not a critical factor affecting the absorption characteristics of the Li-polysulfides. However, the cyclic performance was affected by the total pore volume, suggesting that a certain pore volume is necessary to confine the majority of the soluble Li-polysulfides generated during cycling and to enhance sulfur utilization. The novel results obtained in this study will contribute to the consolidation of S electrochemistry and further development of high-energy lithium–sulfur batteries. Ó 2013 Elsevier Ltd. All rights reserved.

1.

Introduction

Lithium–sulfur (Li–S) batteries hold great potential as an energy storage system that can overcome the limited energy density of currently marketed lithium-ion batteries [1–4]. Over the past decade, the feasibility of Li–S batteries has been extensively demonstrated from various perspectives. One of the main challenges facing the development of a self-sustaining Li–S battery is improving the battery’s poor cyclic performance and rate capability to meet the rigorous requirements for commercial use [5–8]. In this regard, one of the most important

tasks is to confine soluble high-order Li-polysulfides (Li2Sx, 4 6 x 6 8) in the cathode during cycling because the spontaneously dissolved Li-polysulfides promote the chemical oxidation of Li2S2 or Li2S on the surface of the metallic Li anode and shuttle reactions of the intermediate phases between S and Li. This process also results in a gradual loss of the active S in the cathode, leading to a significant decrease in capacity. To address the issue, much attention has been devoted to the rational design of the cathode architecture [9–15]. In particular, remarkable progress has recently been made using an ordered mesoporous silica (OMS) additive as a

* Corresponding author: Fax: +82 41 560 1224. E-mail address: [email protected] (Y. Jung). 0008-6223/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.carbon.2013.11.001

266

CARBON

6 8 ( 2 0 1 4 ) 2 6 5 –2 7 2

reservoir for highly soluble Li-polysulfides (Li2Sx, 4 6 x 6 8) in the cathode [16,17]. This additive has resulted in outstanding cyclic performance and has provided promising insight into Li–S battery technology. The role of the OMS additive is believed to be to confine the dissolved Li-polysulfides to preclude them from migrating spontaneously to the anode and thus preventing further chemical reductions or shuttle reactions [17]. This approach is quite effective in minimizing irreversible reactions due to side reactions associated with the dissolved Li-polysulfides and to ensure stable cyclic performance. This premise warrants a more comprehensive understanding of the effects of OMS or other additives with a similar functionality from both structural and electrochemical viewpoints. In this study, we investigated the use of colloidal silica templated mesoporous carbon (CSC), one of disordered mesoporous carbons, as a promising functional additive for improving the cyclic performance of Li–S batteries. Compared to that of insulating OMS, the use of conductive CSC would have distinctive advantages in minimizing the impedance while achieving the goal of confining the Li-polysulfides in the mesopores [18–20]. This characteristic of CSC should also result in an enhancement in the cyclic performance as well as the electrochemical kinetics of Li–S batteries. The positive effects of CSC addition were demonstrated by the results of various structural and electrochemical analyses performed in this study. To determine the key factors governing the interaction between CSC and Li-polysulfides (e.g., absorption), we also systematically investigated the effects of the structural parameters of the CSC, including pore size and pore volume, on the electrochemical performance of Li–S batteries. The results of this study and discussion provided herein will contribute to a better understanding of S electrochemistry and therefore to the improvement of Li–S batteries.

2.

Experimental

To synthesize CSC, 6.5 g of sucrose (99%, Junsei) as a carbon source was dissolved in de-ionized water and mixed with commercial colloidal silica solutions containing particles of different sizes: 4 nm (silicon (IV) oxide, 15% in H2O, Alfa Aesar), 10 nm (silicon (IV) oxide, 30% in H2O, Alfa Aesar), and 20 nm (silicon (IV) oxide, 40% in H2O, Alfa Aesar). To make sure the pore volumes of the prepared CSC samples were the same, the amount of each colloidal silica template was fixed at 5.5 g. For comparison, 2.75 g of a colloidal silica template (20 nm) was also used for the preparation of CSC to reduce the pore volume by half. After solvent evaporation by drying at 100 °C for 6 h and 160 °C for another 6 h, a series of CSC powders were yielded by carbonization at 900 °C for 3 h under an Ar atmosphere [20]. The collected powders were chemically etched using 10% hydrofluoric acid (HF) solution to remove the silica templates, as illustrated in Fig. 1. The final products were then repeatedly washed with de-ionized water and dried at 100 °C before use. The morphology and microstructure of the CSC powders were examined using field-emission scanning electron microscopy (FESEM, JEOL JSM-7000F) combined with energydispersive X-ray spectroscopy (EDS) and high-resolution

Fig. 1 – A schematic of the preparation of CSC as a functional additive for use in Li–S batteries: (i) Mixing of sucrose and colloidal silica template in water, (ii) carbonization at 900 °C under Ar atmosphere, (iii) removal of silica templates by chemical etching with 10% HF solution. (A colour version of this figure can be viewed online.)

transmission electron microscopy (HRTEM, JEOL JEM3010). The surface area and pore volume of the mesoporous carbons were measured via the Brunauer–Emmett–Teller (BET) and Barrett–Joyner–Halenda (BJH) methods, respectively, using a surface area and porosimetry analyzer (Tristar II 3020, Micromeritics). The volume conductivities of the composites were estimated as a function of pressure using a powderresistivity measurement system (Loresta). Further structural information regarding the mesoporous carbon was obtained using Raman spectroscopy (Bruker Optik GmbH). The amount of residual silica template in the CSC powder was confirmed using a thermogravimetric analyzer (TGA, PerkinElmer TG/ DTA 6300). To evaluate the electrochemical properties of the powders, cathodes were prepared by coating Al foil with a slurry containing S powder (60 wt%), a conducting agent (Ketjenblack, 10 wt%), CSC (10 wt%), and a binder (polyethylene oxide (PEO), 20 wt%) dissolved in acetonitrile. For comparison, cathodes composed of S powder (60 wt%), a conducting agent (10 wt% or 20 wt%), and a binder (30 wt% or 20 wt%) were also prepared in the same way. The loading and density of the cathodes were fixed at 3.9 mg/cm2 and 1.0 g/cm3, respectively. The prepared cathodes were carefully dried in a vacuum for 12 h and pressed. The thickness of the cathodes was measured to be approximately 40 lm after pressing. A total of 2016 coin-type test cells were assembled using the prepared cathodes (2.01 cm2) and a lithium metal anode (2.54 cm2) in an Ar-filled glove box. A polyethylene separator and 1 M lithium bis(trifluoromethanesulfonyl)imide (LiTFSI) dissolved in a mixture of dimethyl ether (DME), diglyme (DG), and 1,3dioxolane (DOL) in a volumetric ratio of 6:2:2, with 0.3 M LiNO3 as an additive, were used as the separator and the electrolyte, respectively. The cells were galvanostatically discharged to 1.8 V vs. Li/Li+ and charged to 2.7 V vs. Li/Li+, repeatedly, at discharging currents of 0.05, 0.1, 0.25, 0.5, 1.0, 1.5, and 2.5 C (1 C = 1674 mA/g) and a fixed charging current of 0.1 C to examine the rate capability of the cells. The cyclic performance of the cells was also evaluated in the voltage range of 1.8–2.7 V vs. Li/Li+ at a constant current of 0.25 C after the formation step of the first two cycles at a constant current of 0.1 C. To confirm whether CSC particles work as a polysulfide reservoir, ex-situ EDS analyses of the cathodes with

CARBON

6 8 (2 0 1 4) 2 6 5–27 2

CSC-20a, prepared by discharge to different cut-off voltages such as OCV, 2.06, and 1.8 V vs. Li/Li+, were carried out.

3.

Results and discussion

Although many kinds of mesoporous carbon materials with different pore structures have been proposed as conducting frameworks for Li–S batteries [20–22], CSC is one of the most suitable candidate for demonstrating its pore size and pore volume effect on the electrochemical performance of Li–S batteries because other effects induced by different pore structures of mesoporous carbon can be excluded. The CSC exhibits a highly porous structure, which allows for the absorption or adsorption of soluble Li-polysulfides during cycling. The microstructure of CSC can be easily controlled by adjusting the size and amount of inorganic templates (i.e., colloidal silica particles) used for CSC synthesis (Fig. 1) [23]. To determine the effects of the pore size and the pore volume of the CSC powders on the absorption characteristics of the Li-polysulfides, we prepared a series of CSC samples with different pore sizes and pore volumes and examined their effects on the electrochemical performance of the corresponding Li–S batteries. The CSC samples were prepared using commercial colloidal silica templates with different particle sizes, e.g., 4 (CSC-4a), 10 (CSC-10a), and 20 nm (CSC20a). To ensure that the same total pore volume was achieved regardless of the pore size, the amount of each template was carefully controlled depending on the template size. For comparison, we also prepared a sample designated CSC-20b using half the normal amount of template (20 nm). The morphologies of the series of CSC samples prepared by using templates

267

of different sizes and in different amounts are compared in Fig. 2. According to the FESEM images, CSC-20a has a highly porous structure, in which mesopores measuring approximately 20 nm are randomly distributed in the amorphous carbon matrix (Fig. 2c). As expected, CSC-4a (Fig. 2a) and CSC-10a (Fig. 2b) possess similar morphologies, but much smaller mesopores than those in CSC-20a can be clearly observed. In the case of CSC-20b (Fig. 2d), mesopores measuring approximately 20 nm were effectively created, but it is clear that a portion of the non-porous carbon matrix still remained because a smaller amount of silica template was added to synthesize this sample, leading to a relatively lower pore density in the carbon matrix. For further inspection, we examined the microstructures of a series of CSC samples using HRTEM, as shown in Fig. 3. All of the CSC samples showed a typical amorphous carbon structure without the formation of any impurities at the synthesis temperature used, and the silica templates were successfully removed by HF treatment, which has been confirmed by TGA analyses. The results show that all CSC samples contained well-defined spherical mesopores directly transferred from the silica templates, which were randomly distributed in an amorphous carbon matrix. As expected, the pore sizes of the CSC samples increased with the template size from 4 to 20 nm. It should be noted that some of the mesopores were inter-connected by the merging of mesopores, but most of them were isolated in CSC-20a (Fig. 3c). On the other hand, CSC-20b possessed thicker carbon walls and a non-porous carbon matrix, as shown in Fig. 3d, which exhibited a relatively lower pore density by reducing the amount of the template used by half. These results confirm that the

Fig. 2 – FESEM images of the series of CSC additives prepared using colloidal silica templates with different pore sizes and volumes: (a) CSC-4a, (b) CSC-10a, (c) CSC-20a, and (d) CSC-20b. (A colour version of this figure can be viewed online.)

268

CARBON

6 8 ( 2 0 1 4 ) 2 6 5 –2 7 2

Fig. 3 – HRTEM images of the series of CSC additives prepared using colloidal silica templates with different pore sizes and volumes: (a) CSC-4a, (b) CSC-10a, (c) CSC-20a, and (d) CSC-20b.

microstructure (e.g., pore size and pore volume) of the CSC samples could be easily controlled by altering the size and amount of silica template used. Fig. 4a shows the nitrogen adsorption/desorption isotherms of CSC samples with different pore sizes and pore volumes. Note that the distinctive hysteresis loop shapes were obtained, depending on the pore sizes. The CSC-20a exhibits steep condensation steps in the range of P/P0 = 0.75–0.98, which is relatively wider than the ranges of mesoporous materials with cylindrical pore structures. For CSC-10a and CSC-4a, the steps appeared over broader ranges of P/P0, forming wider hysteresis curves, which represents a general characteristic of CSC materials with disordered pore structures [24,25]. It appears that the mesopores could be easily merged or inter-connected during synthesis when a larger template (20 nm) was employed, whereas more isolated mesopores could be created by reducing the template size [26]. On the other hand, there was no significant difference in the areas of the hysteresis loops for CSC samples with different pore sizes. We only observed, by comparing samples CSC-20a and CSC-20b, that a notable reduction in surface area occurred by decreasing the amount of the template used. Fig. 4b shows the pore size distribution curves of a series of CSC samples obtained using the BJH method based on the adsorption branch. Mesopores measuring 19.9 and 19.4 nm dominated the structures of samples CSC-20a and CSC-20b, respectively. As expected, CSC-20b showed a smaller amount

of mesopores in the carbon matrix. By reducing the template size, CSC-10a and CSC-4a developed smaller mesopores measuring 11.8 and 5.5 nm, respectively. The overall results reveal that CSCs with different pore sizes and pore volumes were successfully synthesized by controlling the template size and content. The surface areas and total pore volumes of the series of CSCs estimated using the BET and BJH methods, respectively, are summarized in Table 1. CSC-20a showed a surface area of 992 m2/g and a pore volume of 1.74 cm3/g, which is slightly lower than the theoretically calculated volume (1.80 cm3/g) based on the added silica template content. This discrepancy can be reasonably explained by the formation of merged or inter-connected mesopores in the carbon matrix using a large silica template, leading to a decrease in the total pore volume [27]. The surface area of CSC-10a was not very different from that of CSC-20a, but the estimated pore volume (1.79 cm3/g) is similar to the calculated volume because most of the mesopores seemed to have been isolated from one another in the carbon matrix. After further decreasing the template size to 4 nm, a relatively high surface area of 1054 m2/g was attained for CSC-4a owing to the large surface area of the template (4 nm). The CSC-4a maintained the same pore volume of 1.79 cm3/g. On the other hand, we note that the surface area and pore volume of CSC-20b were estimated to be 626 m2/g and 0.94 cm3/g, respectively. These results demonstrate that the total pore volume of CSC-20b was successfully reduced.

CARBON

269

6 8 (2 0 1 4) 2 6 5–27 2

Fig. 4 – (a) N2 sorption isotherms and (b) pore size distribution curves of CSC-4a (h), CSC-10a (s), CSC-20a (m), and CSC-20b (). (A colour version of this figure can be viewed online.)

To investigate the effect of CSC as a functional additive on the electrochemical performance of Li–S batteries, a fixed amount (10 wt%) of each CSC sample was incorporated into a battery cathode during electrode fabrication, and the resulting Li–S batteries were characterized. Fig. 5 shows the galvanostatic charge and discharge profiles of the cells with and without CSC recorded at currents of 0.05, 0.1, 0.25, 0.5, 1.0, 1.5, and 2.5 C (1 C = 1674 mA/g). To avoid side reaction arising from reduction of LiNO3, the cells were tested in the voltage range of 1.8–2.7 V vs. Li/Li+ [28]. As shown in Fig. 5a and b, the cell containing CSC-20a exhibited an initial discharge capacity of 1214.1 mAh/g at a current density of 0.05 C, which

is slightly higher than that of the cell without CSC (1180.8 mAh/g). The other cells containing CSC-10a and CSC-4a additives also showed similar increases in their initial discharge capacities (data not shown). This improvement induced by the addition of CSC was more pronounced at a higher current density of 2.5 C. The cells containing CSC20a, CSC-10a, and CSC-4a still exhibited higher discharge capacities of 592.6, 458.8, and 488.3 mAh/g, respectively, than the discharge capacity of the cell without CSC (307.2 mAh/g). These results demonstrate that the initial discharge capacity of Li–S batteries can be increased by the addition of CSC. Furthermore, it can be concluded that undesirable side reactions associated with the dissolved Li-polysulfides were effectively minimized by trapping them in the mesopores of the CSCs. It should be noted that there was no notable variation in the initial discharge capacities as a function of the mesopore sizes of the CSCs, which suggests that Li-polysulfide absorption or the adsorption characteristics of the CSC do not greatly depend on the size of mesopores. In contrast, the addition of CSC-20b was not as effective as the addition CSC-20a, even though the two CSCs showed the same pore size. The cell with CSC-20b presented an initial discharge capacity of 1184.9 mAh/g at a low current density of 0.05 C. Compared to the other CSCs, it would be difficult to induce any enhancement in battery performance by the addition of CSC-20b. We only observed a small increase in the discharge capacity at 2.5 C. These results support the notion that the pore volume is a predominant factor in determining the extent of Li-polysulfide absorption or the adsorption characteristics of the CSCs, which is responsible for trapping the dissolved Li-polysulfides during cycling. In addition, we also confirmed that the effect of the pore size of the CSCs on the electrochemical performance of the Li–S batteries is negligible. Fig. 6 shows the cyclic performance of Li–S batteries featuring cathodes with CSC additives examined at a constant current of 0.25 C. For comparison, cathodes with mass ratios of 6:1:3 or 6:2:2 (S:KB:PEO) were also prepared without the addition of CSC. We observed a notable improvement in the cyclic performance of the Li–S batteries with the addition of CSC (Fig. 6a). In particular, the capacity observed in the early cycles was notably reduced after the addition of CSC, which supports the notion that CSC can effectively trap the dissolved Li-polysulfides during the initial discharge and promote further electrochemical reactions. These processes would also suppress undesired side reactions associated with the Li-polysulfides. As a result, the cells with CSC additives exhibited 1.5 times higher capacities over 50 cycles, regardless of the mesopore size, relative to the capacities of the cells with a 6:1:3 mass ratio. The fact that the addition of CSC-4a, CSC-10a, or CSC-20a led to a comparable enhancement in

Table 1 – Physical properties of the series of CSC additives prepared using colloidal silica templates of different sizes.

Template size (nm) Pore size (nm) Surface area (m2/g) Pore volume (cm3/g)

CSC-4a

CSC-10a

CSC-20a

CSC-20b

4 5.5 1054 1.79

10 11.8 936 1.79

20 19.9 974 1.96

20 19.4 626 0.94

270

CARBON

6 8 ( 2 0 1 4 ) 2 6 5 –2 7 2

Fig. 6 – (a) Cyclic performance of Li–S battries employing CSC additives with different pore sizes and the same pore volume: CSC-4a (j), CSC-10a (d), and CSC-20a (m). (b) Cyclic performance of Li–S battries employing CSC additives with different pore volumes and the same pore size: CSC-20a (m) and CSC-20b (). (A colour version of this figure can be viewed online.)

Fig. 5 – Galvanostatic charge and discharge profiles of Li–S batteries employing 10 wt% CSC additives recorded over the voltage range of 1.8–2.7 V vs. Li/Li+ at different discharging currents (0.05, 0.1, 0.25, 0.5, 1.0, 1.5, and 2.5 C, 1 C = 1674 mA/g) and a fixed charging current of 0.1 C: (a) No additive, (b) CSC-20a, and (c) CSC-20b. (A colour version of this figure can be viewed online.)

capacity suggests that the electrochemical performance of the resulting cells was not highly sensitive to the pore size, at least in the range of pore sizes investigated in this study. On the other hand, a comparison of the cyclic properties of the cells containing CSC-20a and CSC-20b shows that the

addition of CSC-20b was less effective in improving the cyclic performance of the corresponding cell (Fig. 6b). After 50 cycles, the cell containing CSC-20a exhibited a discharge capacity of 835 mAh/g, whereas a capacity of 770 mAh/g was attained for the cell containing CSC-20b. In addition, the cell containing CSC-20b still showed a considerable initial decrease in capacity indicating that side reactions occurred due to dissolved Li-polysulfides. This result suggests that it is not the pore size but rather the pore volume of CSC that has a critical effect on the cyclic performance of Li–S batteries. It appears that the amount of dissolved Li-polysulfides that can be absorbed by CSC-20b is limited due to the sample’s low pore volume. Li-polysulfide absorption or the adsorbing characteristics of the CSCs were further investigated using ex-situ EDS analyses [17]. To elucidate the nature of Li-polysulfide absorption or adsorption, cathodes containing CSC-20a were prepared by discharging to different cut-off voltages: the open-circuit voltage (OCV), 2.06, and 1.8 V vs. Li/Li+. For reliable measurements, the cells were assembled using an electrolyte containing 1 M LiPF6 salt for the exclusion of the S signal

CARBON

6 8 (2 0 1 4) 2 6 5–27 2

271

Fig. 7 – (a) The initial discharge profiles of Li–S battery containing CSC-20a additive toward different cut-off voltages: OCV (point A), 2.06 V vs. Li/Li+ (point B), and 1.8 V vs. Li/Li+ (ponit C). (b) FESEM images of CSC-20a additive collected at different cutoff voltages and corresponding S/F ratios estimated from the areal EDS results. (A colour version of this figure can be viewed online.)

arising from LiTFSI salt. As shown in Fig. 7a, the cells were aged for 24 h at the OCV (point A) and then carefully disassembled. It was concluded that the active S did not dissolve by forming soluble high-order Li-polysulfides at a given voltage. When the cell was discharged to 2.06 V vs. Li/Li+ (point B), the active S was fully dissolved in the cell. Then, further reactions promoting the formation of low-order Li-polysulfides occurred until the cell voltage reached 1.8 V vs. Li/Li+ (point C). At each point, we carried out ex-situ EDS analyses for the CSC additives in the disassembled cathodes to determine how much Li-polysulfide could be absorbed in or adsorbed onto the CSCs during the initial discharge cycle. Based on the EDS results (Fig. 7b), we estimated the ratio of the S to F signals (S/F ratio). The S/F ratio was estimated to be 2.6 at the OCV after aging for 24 h. The small S signal detected at the OCV may have been caused by the formation of Li-polysulfides that originated from the chemical reaction between chemically dissolved S and Li metal during aging after cell assembly. It should be noted that the S/F ratio was significantly increased to 8.2 when the cell voltage reached 2.06 V vs. Li/Li+, which was mainly caused by the absorption or adsorption of Li-polysulfides because active S had fully transformed into soluble high-order Li-polysulfides at that voltage. The notable increase in the S/F ratio indicates that

the CSC absorbed the Li-polysulfides in its mesopores or adsorbed the Li-polysulfides onto its surface. After the first discharge to 1.8 V vs. Li/Li+, the S/F ratio further increased to 29.6, may have been chiefly responsible for the formation of discharge products such as Li2S2 or Li2S in the mesopores and on the surface of the CSC. The results clearly demonstrate that the enhanced cyclic performance and capacity retention of Li–S batteries observed when a CSC additive was incorporated into the cathode were due to the effective confinement of the soluble Li-polysulfides in the CSC mesopores, which led to an increase in S utilization by reducing active S loss and minimizing the side reactions associated with the Li-polysulfides during cycling.

4.

Conclusion

The possibility of utilizing CSC as a functional additive in the cathodes of Li–S batteries was investigated. The goal was to utilize CSC with conductive and highly porous features for the effective absorption or adsorption of dissolved Li-polysulfides generated during cycling. With CSC incorporated into the cathode, the cyclic performance of Li–S batteries could be notably enhanced, suggesting that the CSC reduced the loss of active S and suppressed undesirable side reactions

272

CARBON

6 8 ( 2 0 1 4 ) 2 6 5 –2 7 2

associated with Li-polysulfides. The effects of the pore size and pore volume of CSC additives on the electrochemical performance of Li–S batteries were investigated by systematically varying the pore size and pore volume of the CSCs incorporated into S cathodes. The results showed that although battery performance was insensitive to the variation in pore size, it was notably affected by the variation in pore volume. This finding suggests that the pore volume is a critical factor in determining the amount of soluble Li-polysulfides that are confined within CSC mesopores and therefore the degree of S utilization. The Li-polysulfide absorption or the adsorption characteristics of the CSCs were further investigated using ex-situ EDS analyses, and the results are in good agreement with those obtained from electrochemical characterizations. The highly improved cyclic performance achieved by utilizing CSC as a functional cathode additive should help accelerate the development of self-sustaining Li–S batteries.

[10]

[11]

[12]

[13]

[14]

[15]

Acknowledgements [16]

The authors thank the Technology Innovation Center, SK Holdings, for their financial support in the development of LixSy solid-state batteries.

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/ j.carbon.2013.11.001.

[17]

[18]

[19]

[20] R E F E R E N C E S

[21] [1] Tarascon JM, Amand M. Issues and challenges facing rechargeable lithium batteries. Nature 2001;414:359–67. [2] Goodenough JB, Kim Y. Challenges for rechargeable Li batteries. Chem Mater 2010;22:587–603. [3] Whittingham MS. Lithium batteries and cathode materials. Chem Rev 2004;104:4271–302. [4] Bruce PG, Scrosati B, Tarascon JM. Nanomaterials for rechargeable Lithium batteries. Angew Chem In Ed 2008;47:2930–46. [5] Shim J, Striebel KA, Cairns EJ. The lithium/sulfur rechargeable cell: effects of electrode composition and solvent on cell performance. J Electrochem Soc 2002;149:A1321–5. [6] Ji X, Lee KT, Nazar LF. A highly ordered nanostructured carbon–sulphur cathode for lithium–sulphur batteries. Nat Mater 2009;8:500–6. [7] Evers S, Nazar LF. Graphene-enveloped sulfur in a one pot reaction: a cathode with good coulombic efficiency and high practical sulfur content. Chem Commun 2012;48:1233–5. [8] Yang Y, McDowell MT, Jackson A, Cha JJ, Hong SS, Cui Y. New nanostructured Li2S/silicon rechargeable battery with high specific energy. Nano Lett 2010;10:1486–97. [9] Park MS, Yu JS, Kim KJ, Jeong G, Kim JH, Jo YN, et al. One-step synthesis of sulphur-impregnated graphene cathode for

[22]

[23]

[24]

[25]

[26] [27]

[28]

lithium–sulphur batteries. Phys Chem Chem Phys 2012;14:6796–804. Lin T, Tang Y, Wang Y, Bi H, Liu Z, Huang F, et al. Scotch-tapelike exfoliation of graphite assisted with elemental sulfur and graphene–sulfur composites for high-performance lithium–sulfur batteries. Energy Environ Sci 2013;6:1283–90. Jayaprakash N, Shen J, Moganty SS, Corona A, Archer LA. Porous hollow [email protected] composites for high-power lithium–sulfur batteries. Angew Chem In Ed 2011;50:5904–8. Elazari R, Salitra G, Garsuch A, Panchenko A, Aurbach D. Sulfur-impregnated activated carbon fiber cloth as a binderfree cathode for rechargeable Li–s batteries. Adv Mater 2011;23:5641–4. Lee KT, Black R, Yim T, Ji X, Nazar LF. Surface-initiated growth of thin oxide coatings for Li–sulfur battery cathodes. Adv Energy Mater 2012;2:1490–6. Park MS, Yu JS, Kim KJ, Jeong G, Kim JH, Yim T, et al. Porous carbon sphere as a functional conducting framework for use in lithium–sulfur batteries. RSC Adv 2013;3:11774–81. Sun XG, Wang XQ, Mayes RT, Dai S. Lithium–sulfur batteries based on nitrogen-doped carbon and an ionic-liquid electrolyte. Chem Sus Chem 2012;5:2079–85. Evers S, Nazar LF. New approaches for high energy density lithium–sulfur battery cathodes. Acc Chem Res 2013;46:1135–43. Ji X, Evers S, Black R, Nazar LF. Stabilizing lithium–sulphur cathodes using polysulphide reservoirs. Nat Commun 2011;2:325–31. Shi QH, Liang HJ, Feng D, Wang JF, Stucky GD. Porous carbon and carbon/metal oxide microfibers with well-controlled pore structure and interface. J Am Chem Soc 2008;130:5034–5. Zakhidov AA, Baughman RH, Iqbal Z, Ciu C, Khayrullin I, Dantas SO, et al. Carbon structures with three-dimensional periodicity at optical wavelengths. Science 1998;282:897–901. Lai C, Gao XP, Zhang B, Yan TY, Zhou Z. Synthesis and electrochemical performance of sulfur/highly porous carbon composites. J Phys Chem C 2009;113:4712–6. Wang DW, Zhou G, Li F, Wu KH, Lu GQ, Cheng HM, et al. A microporous-mesoporous carbon with graphitic structure for a high-rate stable sulfur cathode in carbonate solvent-based Li–S batteries. Phys Chem Chem Phys 2012;14:8703–10. Zhang B, Qin X, Li GR, Gao XP. Enhancement of long stability of sulfur cathode by encapsulating sulfur into micropores of carbon spheres. Energy Environ Sci 2010;3:1531–7. Lee HI, Stucky GD, Kim JH, Pak CH, Chang H, Kim JM. Spontaneous phase separation mediated synthesis of 3D mesoporous carbon with controllable cageand window size. Adv Mater 2011;23:2357–61. Kruk M, Jaroniec M, Sayari A. Application of large-pore Mcm41 molecular-sieves to improve pore-size analysis using nitrogen adsorption measurements. Langmuir 1997;13:6267–73. Kruk M, Jaroniec M. Determination of mesopore size distributions from argon adsorption data at 77 K. J Phys Chem B 2002;106:4732–9. Lee J, Kim J, Hyeon T. Recent progress in the synthesis of porous carbon materials. Adv Mater 2006;18:2073–94. Lee HI, Kim JH, You DJ, Lee JE, Kim JM, Ahn WS, et al. Rational synthesis pathway for ordered mesoporous carbon with controllable 30- to 100-angstrom pores. Adv Mater 2008;20:757–62. Zhang SS. Roles of LiNO3 in rechargeable lithium/sulfur battery. Electrochim Acta 2012;70:344–8.