- Email: [email protected]
A high-performance slurry-coated polysulfide cathode for lithium–sulfur battery
Journal Pre-proof A high-performance slurry-coated polysulfide cathode for lithium–sulfur battery Jian Liang Cheong, Ayman A. AbdelHamid, Jackie Y. Ying PII:
S2211-2855(19)30821-3
DOI:
https://doi.org/10.1016/j.nanoen.2019.104114
Reference:
NANOEN 104114
To appear in:
Nano Energy
Received Date: 2 August 2019 Revised Date:
4 September 2019
Accepted Date: 11 September 2019
Please cite this article as: J.L. Cheong, A.A. AbdelHamid, J.Y. Ying, A high-performance slurrycoated polysulfide cathode for lithium–sulfur battery, Nano Energy (2019), doi: https://doi.org/10.1016/ j.nanoen.2019.104114. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier Ltd.
A High-Performance Slurry-Coated Polysulfide Cathode for Lithium-Sulfur Battery Jian Liang Cheonga, Ayman A. AbdelHamida, and Jackie Y. Yinga* a
NanoBio Lab, 31 Biopolis Way, The Nanos 138669, Singapore
*E-mail address: [email protected] (J. Y. Ying)
Abstract There has been tremendous progress in enhancing the electrochemical performance of lithium-sulfur batteries. For example, the use of dissolved polysulfide (PS) as cathode improves sulfur (S) utilization and redox kinetics. Notably, free-standing PS cathodes have exceptional performance, but their preparation is not scalable as compared to the slurrycoating process. However, the electrochemical performance of slurry-coated PS electrodes such as Pt/graphene, Super P, metal nitrides and hierarchical porous carbon, is not satisfactory.
Herein,
a high-performance
lithium-PS
battery using inexpensive
commercially available materials has been developed. Specific capacities between 1220 and 1007 mAh g-1 were achieved at charge rates of 0.2–2.0 C, having capacity fade of lower than 0.14% per cycle over 200 cycles. At higher S loading, a practical areal capacity of > 4 mAh g-1 was achieved. The difference between PS and melt-diffusion S cathodes was also examined. Remarkably, the PS cathode offered 48% higher specific capacity and 26% lower capacity fade than the melt-diffused S cathode due to differences in morphology, surface area and ohmic resistance of the cathodes. This work provides a strong case towards a paradigm shift away from conventional cathode preparation approaches to improve the electrochemical performance of lithium-S batteries.
Keywords: Lithium-sulfur batteries, preformed cathode scaffold, polysulfide, slurry coating, reduced graphene oxide
1
1. Introduction The high theoretical energy density, low material cost and high abundance of S makes lithium sulfur (Li-S) battery system very attractive for energy storage. Although already in the market for niche applications, full-scale commercialization of Li-S batteries will not be realized until certain challenges are overcome [1]. These problems are largely related to the cathode: low active material utilization due to low electronic conductivity of S and lithium sulfide (Li2S), collapse of cathode structure from volume changes during cycling, and polysulfide shuttling effect leading to poor cycling stability [2-4]. Although various strategies have been implemented to overcome the above-mentioned problems, they require complicated procedures to prepare the cathodes [3, 5-7]. Unlike lithium ion battery (LIB), LiS battery has an intricate electrochemistry. It begins with the dissolution of elemental S into polysulfides (PS), which can be long chain S82- and S62- and/or shorter chain S42- and S22-, depending on the state of discharge. The dissolution process results in the loss of contact between the binder and cathode materials, which ultimately leads to structural collapse [8 ,9]. Since S dissolution to PS is inevitable, why not use PS as the S source for Li-S batteries? In fact, it has been shown that the use of PS over solid S as cathode material offers several advantages, namely, improved S utilization and enhanced redox kinetics [10-23]. Earlier work on Li-PS batteries established that PS cathodes have reduced polarization, high ionic conductivity and high capacity retention, as compared to solid S cathodes [14, 15]. Subsequent research efforts were shifted towards the use of PS on free-standing cathode structures [11-13, 16-18, 20-22, 24-26]. Although these cathodes were reported to have excellent electrochemical performance, it might be costly to prepare free-standing structures at an industrial scale, as compared to the slurry coating process for current collectors for Liion batteries [27]. On the other hand, there have been very few reports on slurry-coated dissolved PS cathodes and their electrochemical performances were poor [10, 28-33]. In addition, it remains unclear how and why PS cathodes, whereby the S source (i.e. PS) is introduced on a preformed cathode structure prepared via slurry coating, are better than S cathodes prepared by conventional methods (e.g. melt-diffusion, whereby S and its host was introduced into the slurry before coating) (Scheme 1). As such, it is paramount to investigate the differences between these two distinctly different methods of cathode preparation, and develop a slurry-coated PS cathode with excellent electrochemical performance. Herein, we report a high-performance PS cathode (PS/rGO) prepared via scalable slurry-coating process. The major component of the PS/rGO cathode was selected to be N-
2
doped reduced graphene oxide (rGO), which is highly conductive and have N sites that have high affinity for PS absorption [34]. The high conductivity of rGO-based cathodes is known to impart good electrochemical performance in Li-S batteries [35, 36]. Vapor grown carbon fiber (VGCF), which imparts mechanical strength to the cathode structure, and water-soluble LA-132 binder, which is non-toxic as compared to the conventional PVDF/NMP binder/solvent system were also selected [8, 16, 37]. This is different from the reported slurry-coated PS cathodes, whereby Super P carbon, hierarchical silica-etch carbon, metal nitrides nanoparticles and Pt/graphene, were employed as cathode materials [10, 29-33]. These materials have either lower electronic conductivity (Super P carbon and silica-etch carbon) or surface area (Super P carbon and metal nitride nanoparticles), as compared to rGO. Although the Pt/graphene material was expected to work well, its cathode microstructure appeared less porous and much denser as compared to the PS/rGO cathode [31]. Thus, PS interaction with active Pt/graphene surface would not be optimal for electrochemical performance. Using the PS/rGO cathode, initial specific capacities of 1220, 1112, 1087 and 1007 -1
mAh g were obtained at charge rates of 0.2 C, 0.5 C, 1.0 C and 2.0 C, respectively, with more than 82% capacity being retained over 200 cycles. To the best of our knowledge, this performance is the best among the reported slurry-coated PS cathodes (Table S1) [10, 29-33]. In addition, we have studied the difference in battery performance between melt-diffused S cathode (S/rGO) and PS/rGO cathode using electron microscopy, physisorption and electrochemical methods. The performance difference was found to be substantial; PS/rGO cathode gave a 48% higher specific capacity and 26% lower capacity fade as compared to S/rGO cathode. This difference could be attributed to the difference in morphology, surface area and Ohmic resistance, factors which were strongly influenced by how the cathodes were prepared. For optimal electrochemical performance, future Li-S systems should decouple S incorporation from the typical cathode preparation process.
2. Materials and Methods N-doped rGO was purchased from Nanjing JCNANO Technology Co. Ltd. VGCF was purchased from Zhongke Leiming (Beijing) Science And Technology Co Ltd. LA-132 binder was purchased from Chengdu Indigo Power Sources Co. Ltd. Sublimed S, Li2S, dimethoxyethane (DME) and CS2 were purchased from Sigma Aldrich.
3
2.1. Cathode preparation For the preformed rGO host structure, 80 wt% rGO, 10 wt% VGCF and 10 wt% LA132 binder were stirred overnight before coating on carbon-coated Al current collector via the doctor’s blade method. Cathode host was then left to dry in a 60°C oven for a few hours. Cathode mass used was between 3.30 to 3.70 mg. The electrolyte was prepared by adding 1 M LiTFSI and 2 wt% LiNO3 to a 1:1 volume mixture of 1,3-dioxolane (DOL) and 1,2dimethoxyethane (DME). Li2S6 (PS) solution was prepared by stirring a mixture of S (160.5 mg) and Li2S (46.0 mg) at 50°C overnight in an Ar-filled glovebox. 2.85 M sulfur was prepared by adding 2 mL electrolyte. 5.42 M sulfur was prepared by adding 1 mL electrolyte. For S loading of 1.50 mg cm-2, 21 µL of 2.85 M sulfur, which is equivalent to 10 µL electrolyte/mg S, was added to the preformed rGO host structure in a glovebox. The wt% of S, rGO, VGCF, binder in the cathode was 35, 52, 6.5 and 6.5, respectively. For S loading of 5.05 mg cm-2, 37 µL of 5.42 M sulfur was added, corresponding to 64 wt% of S in the cathode. Electrolyte volume per S weight was fixed at 8 µL/mg for a loading density of 5.05 mg cm-2. To prepare the conventional melt-diffused S-host cathode structure, 87 wt% S-rGO composite, prepared by melt diffusion at 160°C in a hydrothermal vessel overnight, 6.5 wt% VGCF and 6.5 wt% LA-132 were stirred overnight and coated on carbon-coated Al current collector via the doctor’s blade method. By controlling the wet film thickness, a S loading density of ~1.50 mg cm-2 was obtained. The S content in the S-rGO composite was ~40 wt% based on elemental analysis. 20 µL of electrolyte (~10 µL/mg S) was added to the meltdiffused S cathode.
2.2. Coin cell preparation and electrochemical testing Standard 2032-type coin cells were used for cell cycling and rate capability tests. Assembly was done in an Ar-filled glovebox, with the 12.7-mm cathodes and lithium foil as the anode/reference electrode. A glass fiber membrane (GF/A, GE Healthcare) and a Celgard membrane, soaked with electrolyte, were used as separator. Both membranes were soaked with electrolyte. Galvanostatic charge-discharge cycling was done using a LAND CT2001 battery tester (Wuhan LAND electronics) between 1.6 V and 2.8 V vs. Li/Li+ for the rate capability test, and between 1.6 V and 2.8 V at high loading (S = 5.05 mg cm-2). For a S loading of 1.50 mg cm-2, fixed rate cycling at 0.2 C, 0.5 C and 1.0 C was conducted between 1.8 V and 2.8 V, while cycling at 2.0 C was carried out between 1.6 and 2.8 V. Cyclic voltammograms were obtained at a scan rate of 0.05 mV s−1. Electrochemical impedance spectra were collected with a 10-mV amplitude at open circuit potential and a frequency 4
range of 1 MHz and 0.01 Hz on an M204 Autolab potentiostat (Metrohm) fitted with a frequency response analyzer module.
2.3. Materials Characterization Field emission scanning electron microscopy (SEM) was performed on a JSM-7400F (JEOL) with energy-dispersive X-ray (EDX) spectroscopy (Oxford Instruments) at an accelerating voltage of 6 kV. Fresh and spent cathode were washed with DME several times and dried under vacuum before SEM imaging was conducted. Nitrogen adsorption-desorption isotherms at -196 °C were collected using Micromeritics ASAP 2420 physiorption analyzer. Samples (~40-60 mg) were degassed at 60°C for 12 h before measurement. Specific surface areas were calculated using the BET (Brunauer–Emmet–Teller) method. Pore size and pore size distribution (PSD) were obtained by the BJH method using the cylindrical pore model. Data obtained were analyzed using the BJH model, and pore volume was taken at P/P0 = 0.988. Samples for physisorption were prepared by removing cathodes coated on an Al current collector. Prior to physisorption measurements, the melt-diffused S host cathode was washed several times with CS2 before drying under vacuum overnight. Elemental analysis of S content was performed on a Flashsmart elemental analyzer (Thermo Scientific).
3. Results and Discussion Scanning electron microscopy (SEM) of the preformed rGO cathode before PS addition revealed a highly porous, 3D structure of interconnected VGCF tubes and crumpled rGO sheets that were well separated (Figure 1). Elemental mapping showed that carbon, oxygen and nitrogen were homogeneously distributed throughout the material, indicating that the cathode components were well-dispersed during cathode preparation (Figure S1). Mapping of the PS/rGO cathode further illustrated high distribution of S from PS (Figure S2). In comparison, other slurry-coated PS cathodes appeared less porous and more densely-packed [30-33]. High porosity is important to accommodate the volumetric changes during interconversion of S and Li2S, and to provide the structure interconnectivity that is essential for long-range and rapid electron transfer [3, 11, 19, 20]. These features are important to achieve high electrochemical performance for Li-S batteries. To determine the robustness and stability of the PS/rGO electrode, rate capability study was conducted. The study involved increasing the charge rate from 0.1 C to 2.0 C, followed by lowering the charge rate to 0.2 C (Figure 2a). The average discharge capacities
5
of the PS/rGO electrode at 0.1 C, 0.2 C, 0.5 C, 1.0 C and 2.0 C were 1499, 1265, 1102, 999 and 879 mAh g-1, respectively (Table S3). When the charge rate was abruptly reduced to 0.2 C, the capacity recovered to 1191 mAh g-1, indicating high structural stability of the PS/rGO electrode. The charge-discharge curves for the 5th cycle at each 0.1 C, 0.2 C, 0.5 C, 1.0 C and 2.0 C rate were also examined (Figure 2b). The charge curves, which showed a continuous plateau, indicated the successive oxidation reaction from Li2S to LiPS to S. The discharge curves consisted of a short upper plateau, corresponding to the reduction of S to LiPS, and a longer lower plateau, corresponding to the reduction of LiPS to Li2S. The increase in C rate led to a decrease in specific capacity, which could be attributed to the sluggish reaction kinetics that resulted from the inherent poor electrical conductivity of S [3]. Long-term cycling at fixed C rates was also performed (Figure 2c). The initial discharge capacities of the PS/rGO electrode were 1220, 1112, 1087 and 1007 mAh g-1 at 0.2, 0.5, 1.0 and 2.0 C, respectively. After 200 cycles, high discharge capacities of 999, 948, 906 and 866 mAh g-1 were retained at 0.2, 0.5, 1.0 and 2.0 C, respectively. The above performance surpassed other slurry-coated PS reported in the literature [10, 29-33]. At 0.2 C, PS/rGO cathode gave a higher initial (1220 vs. 1000 mAh g-1) and retained discharge capacity (999 vs. 780 mAh g-1), at a higher S loading (1.50 vs. 1.21 mg cm-2) and larger number of cycles (200 vs. 100), as compared to the Pt/graphene PS electrode, which showed the best performance amongst the previously reported slurry-coated PS cathodes (Table S1) [31]. Representative reports on free-standing cathodes based on rGO are shown in Table S2. Although these cathodes have excellent electrochemical performance, they are difficult to scale up and often involve a low S concentration (i.e. require more electrolyte) [38]. The advantage of PS/rGO cathode lies in its high scalability, while maintaining excellent electrochemical performance. To determine if the PS/rGO electrode could reach a practical areal capacity as LIB (4 mAh cm-2), S loading density was increased [4, 39]. Low S utilization was expected at high S loadings due to a thicker layer of insulating S on the cathode surface. At 0.1 C, a S loading of 5.05 mg cm-2 and a high S concentration of 5.43 M, the PS/rGO electrode gave an initial specific capacity of 858 mAh g-1, corresponding to an areal capacity of 4.33 mAh cm-2 (Figure 2d). After 50 cycles, 798 mAh g-1 was retained, equivalent to an areal capacity of 4.03 mAh cm-2. The electrochemical performance results illustrate that the PS/rGO cathode is capable of achieving a practical areal capacity comparable to current LIB technology, and a superior electrochemical performance to other slurry-coated PS electrode.
6
To address the difference between the distinctly different method of preparing PS and melt-diffused cathodes, the electrochemical performance of the S/rGO was also evaluated. The average discharge capacities of S/rGO cathode were 1186, 920, 810, 739 and 650 mAh g1
at 0.1, 0.2, 0.5, 1.0 and 2.0 C, respectively, and recovered to 872 mAh g-1 at 0.2 C (Figure
3a, Table S4). The charge-discharge curves at all C rates clearly revealed longer charge and discharge plateaus for PS/rGO, indicating a greater S utilization and hence, specific capacity, as compared to S/rGO. In addition, the voltage hysteresis, determined by the voltage difference between the middle of the charge and the discharge curves, was found to be smaller for the PS/rGO than S/rGO, indicating enhanced redox kinetics in PS/rGO as compared to S/rGO (Figures 3b–f). For long-term cycling of the S/rGO cathode, initial discharge capacities were 867, 821, 747 and 659 mAh g-1 at 0.2, 0.5, 1.0 and 2.0 C rate, respectively (Figure 4). After 200 cycles, the discharge capacities became 653, 650, 605 and 533 mAh g-1 at 0.2, 0.5, 1.0 and 2.0 C rate, respectively. Similar to the trend observed in the rate studies, the first and last (200th) charge-discharge curves of the long-term cycling for PS/rGO featured longer charge and discharge plateaus, illustrating a higher S utilization and a smaller voltage hysteresis as compared to S/rGO (Figures S3 and S4). Capacity fade, known to be positively correlated with PS shuttling effect, was also determined for the two electrodes. Capacity fade values for PS/rGO electrode were 0.100%, 0.080%, 0.091%, 0.075% per cycle at 0.2, 0.5, 1.0 and 2.0 C rate, respectively. For the S/rGO electrode, the capacity fade values were 0.142%, 0.117%, 0.105% and 0.106% per cycle at 0.2, 0.5, 1.0 and 2.0 C rate, respectively. The lower capacity fade could be attributed to the enhanced redox kinetics due to the smaller voltage hysteresis, allowing faster conversion of PS to insoluble sulfides, decreasing the PS concentration and thereby limiting the PS shuttle [40]. Based on the above analysis, PS/rGO electrode showed 48% higher specific capacity and 26% lower capacity fade per cycle, on average, as compared to S/rGO electrode. The difference in electrochemical performance was found to be correlated to the difference in cathode structure. The PS/rGO cathode was highly porous and interconnected before (Figure 5a, b) and after cycling (Figure 5c, d). The pores were uniform in size and evenly distributed, suggesting that the reversible reaction of Li-S system during battery cycling did not affect the porous structure of the PS/rGO cathode. This could be attributed to the preparation method of the PS/rGO cathode, whereby the S source (in the form of Li2S6) was added after the rGO scaffold was constructed (Scheme 1). As such, the chemical reactions only occurred on the surface of the robust rGO scaffold, rendering the structure intact even after cycling. On the other hand, fresh S/rGO cathode structure, although 7
interconnected, consisted of a mixture of large and small pores (Figure 6a, b). After the rate capability test, the cathode underwent a major structural transformation, forming a highly dense and closely packed structure (Figure 6c, d). Unlike the PS/rGO cathode, S, that was melt-diffused with the crumpled rGO sheets, was already incorporated within the structure of S/rGO (Scheme 1). During discharge, the dissolution of insoluble S to soluble PS resulted in disconnections within the S/rGO cathode structure, leading to structural changes. Structural changes were known to occur for composite S cathodes due to the dissolution of S to PS species during battery discharge [8, 41, 42]. The difference in structure of the two electrodes was quantified by nitrogen adsorption, which was performed on both cathodes in the absence of S or LiPS. S removal was necessary to simulate the effect of structural changes observed in SEM. For the PS/rGO cathode, analysis was conducted on the preformed rGO cathode, whereas the S/rGO cathode was washed with CS2 to remove the S. The nitrogen adsorption-desorption isotherm for both cathodes corresponded to a type II isotherm with H3 hysteresis loop (Figure 7a) [43, 44]. The surface area and pore volume of the preformed rGO cathode were 264 m2/g and 0.31 cm3/g, respectively. These values were higher than that of the washed S/rGO cathode (181 m2/g and 0.25 cm3/g). In addition, pore size distribution analysis revealed the presence of mesopores (3–4 nm) and macropores (60–90 nm) for both cathodes (Figure 7b). The rGO cathode had a comparable mesopore size (3.5 nm vs. 3.6 nm) and slightly larger macropore size (82 nm vs. 65 nm) than the washed S/rGO cathode. Since both electrodes were essentially identical in terms of composition, electrolyte volume and S loading density, the higher specific capacities and lower capacity fade values could be attributed to the higher surface area of the PS/rGO cathode, as compared to the S/rGO cathode. The higher surface area of the PS/rGO cathode led to an increased availability of electrochemically active sites for S species, such as S, Li2S and PS, allowing both nucleation and binding to occur on the cathode surface, which led to higher specific capacities [16]; this agreed well with the electrochemical performance discussed earlier (Figure 3 and 4). This resulted in a decrease in the concentration of dissolved PS in bulk, reducing the undesired PS shuttling effect, consistent with the lower capacity fading values of PS/rGO electrode, as compared to the S/rGO electrode [40]. In addition to surface area difference, the structural change, or the lack thereof, was found to have a pronounced effect on the ohmic resistance of the electrodes. Both S/rGO and PS/rGO electrodes were further examined by cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) after rate capability studies. CV curves of both electrodes revealed features typical of a Li-S battery system [45]: two sharp reduction peaks (I and II) 8
and a broader oxidation peak (III) that consisted of several overlapping peaks (Figure 8). Peaks I (~2.3 V) and II (~2.0 V) corresponded to the reduction of elemental S to soluble higher order PS (Li2Sx, 4 ≤ x ≤8), and its further reduction to insoluble Li2S2 and Li2S, respectively, while Peak III (~2.3–2.4V) was attributed to the reverse process. Peak shape of both cathodes resembled one another as these electrodes consisted of essentially identical constituents, ratios and S loading (which was fixed at 1.50 mg cm-2). This implied that the structural difference between these cathodes did not affect the potential at which redox reactions occurred. The area under the curve (typically correlated with capacity) was found to be larger for the PS/rGO cathode as compared to the S/rGO cathode, which was consistent with the galvanostatic cycling results discussed earlier. The EIS data collected were mathematically transformed into Nyquist plots (Figure 9). In general, these plots appeared to be an overlap of several semicircles ending with a steep upward slope. The semicircle at the high frequency region was fitted with an equivalent circuit (Figure 9, inset). The first intercept at the high frequency region of the real (Z’) axis gives the value of electrolyte resistance, Re [16, 20, 46]. The difference between the Z’ axis intercepts of the fitted semicircles gives the charge transfer resistance (RCT) value, which is associated with the charge transfer process between S and the electrode. The Re values of PS/rGO and S/rGO cells were found to be 4.7 and 7.4 Ω, respectively. The RCT values of PS/rGO and S/rGO cells were 3.0 and 7.8 Ω, respectively. As discussed earlier, the PS/rGO cathode has a higher surface area than the S/rGO cathode, implying that the amount of electrochemically active sites would be greater in PS/rGO than S/rGO. With the electrolyte to S ratio (10 µL/mgS) being identical for both cathodes, the effects of surface area versus ohmic resistance could be established. The higher surface area of the PS/rGO cathode allowed more S species in the electrolyte to be captured as compared to S/rGO cathode, reducing the viscosity and hence, lowering Re. At the same time, the insulating S layer would be thinner in PS/rGO than S/rGO, resulting in a lower RCT for PS/rGO cathode as compared to the S/rGO cathode [16]. In addition, structural changes that occurred in the S/rGO cathode could lead to disconnectivity between conductive elements within the cathode [41], contributing to a higher resistance as compared to the PS/rGO cathode. In the structurally intact PS/rGO cathode, the conductive elements within the structure remained interconnected, ensuring continuous and unimpeded electron conduction pathways from the current collector throughout the entire 3D cathode structure. The lower ohmic resistance of the PS/rGO electrode, as compared to the S/rGO cathode, suggested enhanced redox kinetics, which agreed well with the charge-
9
discharge curve analysis discussed earlier, resulting in the improved rate, specific capacity and cycling performance of the Li-S batteries (Figures 3 and 4).
4. Conclusion In conclusion, using inexpensive and non-toxic commercially available materials, we have developed the PS/rGO cathode via slurry coating for Li-S batteries. At a S loading of 1.50 mg cm-2, the initial discharge capacities of the PS/rGO electrode were 1220, 1112, 1087 and 1007 mAh g-1 at 0.2, 0.5, 1.0 and 2.0 C rate, while retaining capacities of 999, 948, 906 and 866 mAh g-1 over 200 cycles, respectively. These values corresponded to an average capacity fade of only 0.086% per cycle. At a higher S loading of 5.05 mg cm-2, the PS/rGO cathode attained a practical areal capacity of > 4 mAh cm-2 over 50 cycles. The electrochemical performance of the PS/rGO cathode is superior to the other slurry-coated PS cathodes reported in the literature. We have also evaluated the difference between a PS cathode (PS/rGO) and cathode prepared via melt-diffusion (S/rGO). The structural change that occurred in the S/rGO cathode led to a highly densified and compact structure, resulting in a lower surface area with smaller pores, greater voltage hysteresis and a higher ohmic resistance, as compared to the structurally intact PS/rGO cathode. The PS/rGO cathode showed an average of 48% increase in specific capacity and 26% lower in capacity fade as compared to S/rGO cathode at various C rates (0.2 C to 2.0 C). These results clearly demonstrated that the preparation of S cathode strongly influenced the electrochemical performance. Excellent electrochemical performance could be achieved by using commercially available materials without any processing. To maximize the electrochemical performance, future design of practical Li-S batteries should consider different cathode preparation approaches, such as incorporating S source into a preformed cathode host structure, and focus on methods that are industrially scalable.
Appendix: Supplementary Data Supplementary data associated with this article can be found in the online version at http://www.journals.elsevier.com/nano-energy.
Acknowledgements We gratefully acknowledge Dr. Su Seong Lee and Dr. Jinhua Yang for their useful discussion, and Carina Lim Yi Jing for assistance in some of the experimental work. This 10
work was funded by NanoBio Lab (Biomedical Research Council, Agency for Science, Technology and Research, Singapore).
References [1]
T. Cleaver, P. Kovacik, M. Marinescu, T. Zhang, G. Offer, J. Electrochem. Soc. 165 (2017) A6029.
[2]
S.S. Zhang, J. Power Sources 231 (2013) 153.
[3]
A. Manthiram, Y. Fu, S.-H. Chung, C. Zu, Y.-S. Su, Chem. Rev. 114 (2014) 11751.
[4]
R. Fang, S. Zhao, Z. Sun, D.W. Wang, H.M. Cheng, F. Li, Adv. Mater. 29 (2017) 1606823.
[5]
H. Hu, B. Zhao, H. Cheng, S. Dai, N. Kane, Y. Yu, M. Liu, Nano Energy 57 (2019) 635.
[6]
H. Hu, H. Cheng, Z. Liu, G. Li, Q. Zhu, Y. Yu, Nano Lett. 15 (2015) 5116.
[7]
C. Dong, W. Gao, B. Jin, Q. Jiang, iScience, 6 (2018) 151.
[8]
R. Elazari, G. Salitra, Y. Talyosef, J. Grinblat, C. Scordilis-Kelley, A. Xiao, J. Affinito, D. Aurbach, J. Electrochem. Soc. 157 (2010) A1131.
[9]
G. Li, S. Wang, Y. Zhang, M. Li, Z. Chen, J. Lu, Adv. Mater. 30 (2018) 1705590.
[10] S.S. Zhang, J.A. Read, J. Power Sources 200 (2012) 77. [11] Y. Fu, Y.S. Su, A. Manthiram, Angew. Chem. Int. Ed. 52 (2013) 6930. [12] C. Zu, Y. Fu, A. Manthiram, J. Mater. Chem. A 1 (2013) 10362. [13] Y. Yang, G. Zheng, Y. Cui, Energy Environ. Sci. 6 (2013) 1552. [14] R. Xu, I. Belharouak, J.C.M. Li, X. Zhang, I. Bloom, J. Bareño, Adv. Energy Mater. 3 (2013) 833. [15] R. Demir-Cakan, M. Morcrette, Gangulibabu, A. Guéguen, R. Dedryvère, J.-M. Tarascon, Energy Environ. Sci. 6 (2013) 176. [16] K. Han, J. Shen, S. Hao, H. Ye, C. Wolverton, M.C. Kung, H.H. Kung, ChemSusChem 7 (2014) 2545. [17] C. Zu, A. Manthiram, Adv. Energy Mater. 4 (2014) 1400897. [18] X. Pu, G. Yang, C. Yu, Adv. Mater. 26 (2014) 7456. [19] G. Zhou, Y. Zhao, C. Zu, A. Manthiram, Nano Energy 12 (2015) 240. [20] G. Zhou, E. Paek, G.S. Hwang, A. Manthiram, Nat. Commun., 6 (2015) 7760. [21] R. Fang, S. Zhao, S. Pei, Y. Cheng, P. Hou, M. Liu, H.-M. Cheng, C. Liu, F. Li, Carbon, 109 (2016) 719. [22] X. Li, X. Pu, S. Han, M. Liu, C. Du, C. Jiang, X. Huang, T. Liu, W. Hu, Nano Energy 30 (2016) 193. 11
[23] G. Babu, N. Masurkar, H. Al Salem, L.M. Arava, J. Am. Chem. Soc. 139 (2017) 171. [24] S. Luo, M. Yao, S. Lei, P. Yan, X. Wei, X. Wang, L. Liu, Z. Niu, Nanoscale 9 (2017) 4646. [25] S.H. Chung, A. Manthiram, Adv. Mater. 30 (2018) 1705951. [26] R. Yu, S.-H. Chung, C.-H. Chen, A. Manthiram, Energy Storage Mater. 18 (2019) 491. [27] A. Kraytsberg, Y. Ein-Eli, Adv. Energy Mater. 6 (2016) 1600655. [28] R.D. Rauh, K.M. Abraham, G.F. Pearson, J.K. Surprenant, S.B. Brummer, J. Electrochem. Soc. 126 (1978) 523. [29] M. Agostini, D.-J. Lee, B. Scrosati, Y.K. Sun, J. Hassoun, J. Power Sources 265 (2014) 14. [30] K.E. Hendrickson, L. Ma, G. Cohn, Y. Lu, L.A. Archer, Adv. Sci. 2 (2015) 1500068. [31] H. Al Salem, G. Babu, C. V. Rao, L.M.R. Arava, J. Am. Chem. Soc. 137 (2015) 11542. [32] N. Mosavati, V.R. Chitturi, S.O. Salley, K.Y.S. Ng, J. Power Sources 321 (2016) 87. [33] N. Mosavati, S.O. Salley, K.Y.S. Ng, J. Power Sources 340 (2017) 210. [34] K. Chen, Z. Sun, R. Fang, Y. Shi, H.-M. Cheng, F. Li, Adv. Funct. Mater. 28 (2018) 1707592. [35] Y. Song, W. Zhao, N. Wei, L. Zhang, F. Ding, Z. Liu, J. Sun, Nano Energy 53 (2018) 432. [36] M.Q. Zhao, Q. Zhang, J.Q. Huang, G.L. Tian, J.Q. Nie, H.J. Peng, F. Wei, Nat. Commun. 5 (2014) 3410. [37] J. Pan, G. Xu, B. Ding, J. Han, H. Dou, X. Zhang, RSC Adv. 5 (2015) 13709. [38] Y. Gogotsi, P. Simon, Science 334 (2011) 917. [39] H.-J. Peng, J.-Q. Huang, X.-B. Cheng, Q. Zhang, Adv. Energy Mater. 7 (2017) 1700260. [40] A.N. Mistry, P.P. Mukherjee, J. Phys. Chem. C 122 (2018) 23845. [41] S.-E. Cheon, K.-S. Ko, J.-H. Cho, S.-W. Kim, E.-Y. Chin, H.-T. Kim, J. Electrochem. Soc. 150 (2003) A796. [42] P. Barai, A. Mistry, P.P. Mukherjee, Extreme Mech. Lett. 9 (2016) 359. [43] H.C. Youn, S.M. Bak, M.S. Kim, C. Jaye, D.A. Fischer, C.W. Lee, X.Q. Yang, K.C. Roh, K.B. Kim, ChemSusChem 8 (2015) 1875. [44] B.D. Ossonon, D. Bélanger, RSC Adv. 7 (2017) 27224. [45] M. Wild, L. O’Neill, T. Zhang, R. Purkayastha, G. Minton, M. Marinescu, G.J. Offer, Energy Environ. Sci. 8 (2015) 3477. [46] Z. Deng, Z. Zhang, Y. Lai, J. Liu, J. Li, Y. Liu, J. Electrochem. Soc. 160 (2013) A553.
12
Jian Liang Cheong is a Senior Lab Officer at the NanoBio Lab, Singapore. He received his B. Sc. (Honours) in Chemistry from the National University of Singapore (NUS), Singapore in 2010. Currently, he is doing his Ph.D. under the supervision of Prof. Jackie Ying. He is interested in developing practical and scalable systems for energy storage.
Ayman A. AbdelHamid is a Research Scientist at the NanoBio Lab, Singapore. He received his M.Sc. in Nanoscience and Technology from Nile University, Egypt in 2012, and his Ph.D. in Materials Science and Engineering from Nanyang Technological University, Singapore in 2017. His research interests include development of nanomaterials and nanocomposites for energy storage applications.
13
Jackie Y. Ying received her B.E. and Ph.D. from The Cooper Union and Princeton University, respectively. She joined the faculty at Massachusetts Institute of Technology in 1992, where she was Professor of Chemical Engineering until 2005. She was the Founding Executive Director of the Institute of Bioengineering and Nanotechnology in Singapore from 2003 to 2018. Prof. Ying currently leads the NanoBio Lab as A*STAR Senior Fellow. For her research on nanostructured materials, Prof. Ying has been recognized with the American Ceramic Society Ross C. Purdy Award, David and Lucile Packard Fellowship, Office of Naval Research Young Investigator Award, National Science Foundation Young Investigator Award, Camille Dreyfus Teacher-Scholar Award, American Chemical Society Faculty Fellowship Award in Solid-State Chemistry, Technology Review’s Inaugural TR100 Young Innovator Award, American Institute of Chemical Engineers (AIChE) Allan P. Colburn Award, International Union of Biochemistry and Molecular Biology Jubilee Medal, Materials Research Society Fellowship, Royal Society of Chemistry Fellowship, Crown Prince Grand Prize in the Brunei Creative, Innovative Product and Technological Advancement (CIPTA) Award, American Institute for Medical and Biological Engineering Fellowship, Academy of Sciences of Iran Medal of Honor, American Association for the Advancement of Science Fellowship, Singapore National Academy of Science Fellowship, and Turkish Academy of Sciences Academy Prize in Science and Engineering Sciences. Prof. Ying was elected a World Economic Forum Young Global Leader, and inducted to the German National Academy of Sciences, Leopoldina and the U.S. National Academy of Inventors. She was named one of the “One Hundred Engineers of the Modern Era” by AIChE in its Centennial Celebration, and a Highly Cited Researcher (Cross-Field Category) by Clarivate Analytics. She was the inaugural winner of the Mustafa Prize “Top Scientific Achievement Award” in 2015 for her research in bio-nanotechnology. She is the Editor-inChief of Nano Today.
14
Scheme 1. Preparation of PS/rGO cathode (top) and S/rGO cathode (bottom).
b)
a)
2 µm
Figure 1. SEM images of rGO cathodes: (a) top view and (b) cross-sectional view.
15
2 µm
a)
b)
c)
d)
Figure 2. Electrochemical performance of PS/rGO cathode. (a) Rate capability at S = 1.50 mg cm-2. (b) 5th cycle charge-discharge curves from rate capability study at different C rates of ( ) 0.1 C, ( ) 0.2 C, ( ) 0.5 C, ( ) 1.0 C and ( ) 2.0 C, and S = 1.50 mg cm-2. (c) Longterm cycling at S = 1.50 mg cm-2 and various C rates of ( ) 0.2 C, ( ) 0.5 C, ( ) 1.0 C and ( ) 2.0 C. (d) Long-term cycling at 0.1 C and S = 5.05 mg cm-2.
16
b)
a)
c)
d)
e)
f)
Figure 3. (a) Rate capability studies of ( ) PS/rGO and ( ) S/rGO cathodes, and (b–f) the corresponding 5th cycle charge-discharge curves of ( ) PS/rGO and ( ) S/rGO cathodes at (b) 0.1 C, (c) 0.2 C, (d) 0.5 C, (e) 1.0 C and (f) 2.0 C, with a S loading of 1.50 mg cm-2.
17
a)
b)
c)
d)
Figure 4. Long-term cycling performance of ( ) PS/rGO and ( ) S/rGO cathodes at (a) 0.2 C, (b) 0.5 C, (c) 1.0 C and (d) 2.0 C, with a S loading of 1.50 mg cm-2.
18
a)
b)
10 µm
c)
2 µm
d)
10 µm
2 µm
Figure 5. SEM images of PS/rGO cathodes: (a, b) fresh cathode, (c, d) after rate capability studies.
19
a)
b)
10 µm
c)
2 µm
d)
10 µm
2 µm
Figure 6. SEM images of S/rGO cathodes: (a, b) fresh cathode, (c, d) after rate capability studies.
20
a)
b)
Figure 7. Nitrogen adsorption-desorption of ( ) PS/rGO and ( ) S/rGO. (a) Isotherm and (b) BJH desorption pore size distribution.
21
II
I
I
Figure 8. Cyclic voltammogram of ( ) PS/rGO and ( ) S/rGO cathodes.
Re
RCT
CPE
Figure 9. Nyquist plots with Re and RCT values of cycled ( ) PS/rGO and ( ) S/rGO cells after rate capability studies. Inset: high frequency region with electrochemical fitted circuit.
22
TOC Graphics
23
Research Highlights •
A high-performance lithium-polysulfide battery using inexpensive commercially available materials has been developed
•
Specific capacities between 1220 and 1007 mAh g-1 were achieved at charge rates of 0.2–2.0 C, having capacity fade of lower than 0.14% per cycle over 200 cycles
•
At higher sulfur loading, a practical areal capacity of > 4 mAh g-1 was achieved
•
The difference between sulfur cathodes prepared via polysulfide addition and meltdiffusion was examined at identical sulfur loadings
•
The polysulfide cathode offered higher specific capacity and lower capacity fade than the melt-diffused sulfur cathode due to differences in morphology, surface area and ohmic resistance
1
S2211-2855(19)30821-3
DOI:
https://doi.org/10.1016/j.nanoen.2019.104114
Reference:
NANOEN 104114
To appear in:
Nano Energy
Received Date: 2 August 2019 Revised Date:
4 September 2019
Accepted Date: 11 September 2019
Please cite this article as: J.L. Cheong, A.A. AbdelHamid, J.Y. Ying, A high-performance slurrycoated polysulfide cathode for lithium–sulfur battery, Nano Energy (2019), doi: https://doi.org/10.1016/ j.nanoen.2019.104114. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier Ltd.
A High-Performance Slurry-Coated Polysulfide Cathode for Lithium-Sulfur Battery Jian Liang Cheonga, Ayman A. AbdelHamida, and Jackie Y. Yinga* a
NanoBio Lab, 31 Biopolis Way, The Nanos 138669, Singapore
*E-mail address: [email protected] (J. Y. Ying)
Abstract There has been tremendous progress in enhancing the electrochemical performance of lithium-sulfur batteries. For example, the use of dissolved polysulfide (PS) as cathode improves sulfur (S) utilization and redox kinetics. Notably, free-standing PS cathodes have exceptional performance, but their preparation is not scalable as compared to the slurrycoating process. However, the electrochemical performance of slurry-coated PS electrodes such as Pt/graphene, Super P, metal nitrides and hierarchical porous carbon, is not satisfactory.
Herein,
a high-performance
lithium-PS
battery using inexpensive
commercially available materials has been developed. Specific capacities between 1220 and 1007 mAh g-1 were achieved at charge rates of 0.2–2.0 C, having capacity fade of lower than 0.14% per cycle over 200 cycles. At higher S loading, a practical areal capacity of > 4 mAh g-1 was achieved. The difference between PS and melt-diffusion S cathodes was also examined. Remarkably, the PS cathode offered 48% higher specific capacity and 26% lower capacity fade than the melt-diffused S cathode due to differences in morphology, surface area and ohmic resistance of the cathodes. This work provides a strong case towards a paradigm shift away from conventional cathode preparation approaches to improve the electrochemical performance of lithium-S batteries.
Keywords: Lithium-sulfur batteries, preformed cathode scaffold, polysulfide, slurry coating, reduced graphene oxide
1
1. Introduction The high theoretical energy density, low material cost and high abundance of S makes lithium sulfur (Li-S) battery system very attractive for energy storage. Although already in the market for niche applications, full-scale commercialization of Li-S batteries will not be realized until certain challenges are overcome [1]. These problems are largely related to the cathode: low active material utilization due to low electronic conductivity of S and lithium sulfide (Li2S), collapse of cathode structure from volume changes during cycling, and polysulfide shuttling effect leading to poor cycling stability [2-4]. Although various strategies have been implemented to overcome the above-mentioned problems, they require complicated procedures to prepare the cathodes [3, 5-7]. Unlike lithium ion battery (LIB), LiS battery has an intricate electrochemistry. It begins with the dissolution of elemental S into polysulfides (PS), which can be long chain S82- and S62- and/or shorter chain S42- and S22-, depending on the state of discharge. The dissolution process results in the loss of contact between the binder and cathode materials, which ultimately leads to structural collapse [8 ,9]. Since S dissolution to PS is inevitable, why not use PS as the S source for Li-S batteries? In fact, it has been shown that the use of PS over solid S as cathode material offers several advantages, namely, improved S utilization and enhanced redox kinetics [10-23]. Earlier work on Li-PS batteries established that PS cathodes have reduced polarization, high ionic conductivity and high capacity retention, as compared to solid S cathodes [14, 15]. Subsequent research efforts were shifted towards the use of PS on free-standing cathode structures [11-13, 16-18, 20-22, 24-26]. Although these cathodes were reported to have excellent electrochemical performance, it might be costly to prepare free-standing structures at an industrial scale, as compared to the slurry coating process for current collectors for Liion batteries [27]. On the other hand, there have been very few reports on slurry-coated dissolved PS cathodes and their electrochemical performances were poor [10, 28-33]. In addition, it remains unclear how and why PS cathodes, whereby the S source (i.e. PS) is introduced on a preformed cathode structure prepared via slurry coating, are better than S cathodes prepared by conventional methods (e.g. melt-diffusion, whereby S and its host was introduced into the slurry before coating) (Scheme 1). As such, it is paramount to investigate the differences between these two distinctly different methods of cathode preparation, and develop a slurry-coated PS cathode with excellent electrochemical performance. Herein, we report a high-performance PS cathode (PS/rGO) prepared via scalable slurry-coating process. The major component of the PS/rGO cathode was selected to be N-
2
doped reduced graphene oxide (rGO), which is highly conductive and have N sites that have high affinity for PS absorption [34]. The high conductivity of rGO-based cathodes is known to impart good electrochemical performance in Li-S batteries [35, 36]. Vapor grown carbon fiber (VGCF), which imparts mechanical strength to the cathode structure, and water-soluble LA-132 binder, which is non-toxic as compared to the conventional PVDF/NMP binder/solvent system were also selected [8, 16, 37]. This is different from the reported slurry-coated PS cathodes, whereby Super P carbon, hierarchical silica-etch carbon, metal nitrides nanoparticles and Pt/graphene, were employed as cathode materials [10, 29-33]. These materials have either lower electronic conductivity (Super P carbon and silica-etch carbon) or surface area (Super P carbon and metal nitride nanoparticles), as compared to rGO. Although the Pt/graphene material was expected to work well, its cathode microstructure appeared less porous and much denser as compared to the PS/rGO cathode [31]. Thus, PS interaction with active Pt/graphene surface would not be optimal for electrochemical performance. Using the PS/rGO cathode, initial specific capacities of 1220, 1112, 1087 and 1007 -1
mAh g were obtained at charge rates of 0.2 C, 0.5 C, 1.0 C and 2.0 C, respectively, with more than 82% capacity being retained over 200 cycles. To the best of our knowledge, this performance is the best among the reported slurry-coated PS cathodes (Table S1) [10, 29-33]. In addition, we have studied the difference in battery performance between melt-diffused S cathode (S/rGO) and PS/rGO cathode using electron microscopy, physisorption and electrochemical methods. The performance difference was found to be substantial; PS/rGO cathode gave a 48% higher specific capacity and 26% lower capacity fade as compared to S/rGO cathode. This difference could be attributed to the difference in morphology, surface area and Ohmic resistance, factors which were strongly influenced by how the cathodes were prepared. For optimal electrochemical performance, future Li-S systems should decouple S incorporation from the typical cathode preparation process.
2. Materials and Methods N-doped rGO was purchased from Nanjing JCNANO Technology Co. Ltd. VGCF was purchased from Zhongke Leiming (Beijing) Science And Technology Co Ltd. LA-132 binder was purchased from Chengdu Indigo Power Sources Co. Ltd. Sublimed S, Li2S, dimethoxyethane (DME) and CS2 were purchased from Sigma Aldrich.
3
2.1. Cathode preparation For the preformed rGO host structure, 80 wt% rGO, 10 wt% VGCF and 10 wt% LA132 binder were stirred overnight before coating on carbon-coated Al current collector via the doctor’s blade method. Cathode host was then left to dry in a 60°C oven for a few hours. Cathode mass used was between 3.30 to 3.70 mg. The electrolyte was prepared by adding 1 M LiTFSI and 2 wt% LiNO3 to a 1:1 volume mixture of 1,3-dioxolane (DOL) and 1,2dimethoxyethane (DME). Li2S6 (PS) solution was prepared by stirring a mixture of S (160.5 mg) and Li2S (46.0 mg) at 50°C overnight in an Ar-filled glovebox. 2.85 M sulfur was prepared by adding 2 mL electrolyte. 5.42 M sulfur was prepared by adding 1 mL electrolyte. For S loading of 1.50 mg cm-2, 21 µL of 2.85 M sulfur, which is equivalent to 10 µL electrolyte/mg S, was added to the preformed rGO host structure in a glovebox. The wt% of S, rGO, VGCF, binder in the cathode was 35, 52, 6.5 and 6.5, respectively. For S loading of 5.05 mg cm-2, 37 µL of 5.42 M sulfur was added, corresponding to 64 wt% of S in the cathode. Electrolyte volume per S weight was fixed at 8 µL/mg for a loading density of 5.05 mg cm-2. To prepare the conventional melt-diffused S-host cathode structure, 87 wt% S-rGO composite, prepared by melt diffusion at 160°C in a hydrothermal vessel overnight, 6.5 wt% VGCF and 6.5 wt% LA-132 were stirred overnight and coated on carbon-coated Al current collector via the doctor’s blade method. By controlling the wet film thickness, a S loading density of ~1.50 mg cm-2 was obtained. The S content in the S-rGO composite was ~40 wt% based on elemental analysis. 20 µL of electrolyte (~10 µL/mg S) was added to the meltdiffused S cathode.
2.2. Coin cell preparation and electrochemical testing Standard 2032-type coin cells were used for cell cycling and rate capability tests. Assembly was done in an Ar-filled glovebox, with the 12.7-mm cathodes and lithium foil as the anode/reference electrode. A glass fiber membrane (GF/A, GE Healthcare) and a Celgard membrane, soaked with electrolyte, were used as separator. Both membranes were soaked with electrolyte. Galvanostatic charge-discharge cycling was done using a LAND CT2001 battery tester (Wuhan LAND electronics) between 1.6 V and 2.8 V vs. Li/Li+ for the rate capability test, and between 1.6 V and 2.8 V at high loading (S = 5.05 mg cm-2). For a S loading of 1.50 mg cm-2, fixed rate cycling at 0.2 C, 0.5 C and 1.0 C was conducted between 1.8 V and 2.8 V, while cycling at 2.0 C was carried out between 1.6 and 2.8 V. Cyclic voltammograms were obtained at a scan rate of 0.05 mV s−1. Electrochemical impedance spectra were collected with a 10-mV amplitude at open circuit potential and a frequency 4
range of 1 MHz and 0.01 Hz on an M204 Autolab potentiostat (Metrohm) fitted with a frequency response analyzer module.
2.3. Materials Characterization Field emission scanning electron microscopy (SEM) was performed on a JSM-7400F (JEOL) with energy-dispersive X-ray (EDX) spectroscopy (Oxford Instruments) at an accelerating voltage of 6 kV. Fresh and spent cathode were washed with DME several times and dried under vacuum before SEM imaging was conducted. Nitrogen adsorption-desorption isotherms at -196 °C were collected using Micromeritics ASAP 2420 physiorption analyzer. Samples (~40-60 mg) were degassed at 60°C for 12 h before measurement. Specific surface areas were calculated using the BET (Brunauer–Emmet–Teller) method. Pore size and pore size distribution (PSD) were obtained by the BJH method using the cylindrical pore model. Data obtained were analyzed using the BJH model, and pore volume was taken at P/P0 = 0.988. Samples for physisorption were prepared by removing cathodes coated on an Al current collector. Prior to physisorption measurements, the melt-diffused S host cathode was washed several times with CS2 before drying under vacuum overnight. Elemental analysis of S content was performed on a Flashsmart elemental analyzer (Thermo Scientific).
3. Results and Discussion Scanning electron microscopy (SEM) of the preformed rGO cathode before PS addition revealed a highly porous, 3D structure of interconnected VGCF tubes and crumpled rGO sheets that were well separated (Figure 1). Elemental mapping showed that carbon, oxygen and nitrogen were homogeneously distributed throughout the material, indicating that the cathode components were well-dispersed during cathode preparation (Figure S1). Mapping of the PS/rGO cathode further illustrated high distribution of S from PS (Figure S2). In comparison, other slurry-coated PS cathodes appeared less porous and more densely-packed [30-33]. High porosity is important to accommodate the volumetric changes during interconversion of S and Li2S, and to provide the structure interconnectivity that is essential for long-range and rapid electron transfer [3, 11, 19, 20]. These features are important to achieve high electrochemical performance for Li-S batteries. To determine the robustness and stability of the PS/rGO electrode, rate capability study was conducted. The study involved increasing the charge rate from 0.1 C to 2.0 C, followed by lowering the charge rate to 0.2 C (Figure 2a). The average discharge capacities
5
of the PS/rGO electrode at 0.1 C, 0.2 C, 0.5 C, 1.0 C and 2.0 C were 1499, 1265, 1102, 999 and 879 mAh g-1, respectively (Table S3). When the charge rate was abruptly reduced to 0.2 C, the capacity recovered to 1191 mAh g-1, indicating high structural stability of the PS/rGO electrode. The charge-discharge curves for the 5th cycle at each 0.1 C, 0.2 C, 0.5 C, 1.0 C and 2.0 C rate were also examined (Figure 2b). The charge curves, which showed a continuous plateau, indicated the successive oxidation reaction from Li2S to LiPS to S. The discharge curves consisted of a short upper plateau, corresponding to the reduction of S to LiPS, and a longer lower plateau, corresponding to the reduction of LiPS to Li2S. The increase in C rate led to a decrease in specific capacity, which could be attributed to the sluggish reaction kinetics that resulted from the inherent poor electrical conductivity of S [3]. Long-term cycling at fixed C rates was also performed (Figure 2c). The initial discharge capacities of the PS/rGO electrode were 1220, 1112, 1087 and 1007 mAh g-1 at 0.2, 0.5, 1.0 and 2.0 C, respectively. After 200 cycles, high discharge capacities of 999, 948, 906 and 866 mAh g-1 were retained at 0.2, 0.5, 1.0 and 2.0 C, respectively. The above performance surpassed other slurry-coated PS reported in the literature [10, 29-33]. At 0.2 C, PS/rGO cathode gave a higher initial (1220 vs. 1000 mAh g-1) and retained discharge capacity (999 vs. 780 mAh g-1), at a higher S loading (1.50 vs. 1.21 mg cm-2) and larger number of cycles (200 vs. 100), as compared to the Pt/graphene PS electrode, which showed the best performance amongst the previously reported slurry-coated PS cathodes (Table S1) [31]. Representative reports on free-standing cathodes based on rGO are shown in Table S2. Although these cathodes have excellent electrochemical performance, they are difficult to scale up and often involve a low S concentration (i.e. require more electrolyte) [38]. The advantage of PS/rGO cathode lies in its high scalability, while maintaining excellent electrochemical performance. To determine if the PS/rGO electrode could reach a practical areal capacity as LIB (4 mAh cm-2), S loading density was increased [4, 39]. Low S utilization was expected at high S loadings due to a thicker layer of insulating S on the cathode surface. At 0.1 C, a S loading of 5.05 mg cm-2 and a high S concentration of 5.43 M, the PS/rGO electrode gave an initial specific capacity of 858 mAh g-1, corresponding to an areal capacity of 4.33 mAh cm-2 (Figure 2d). After 50 cycles, 798 mAh g-1 was retained, equivalent to an areal capacity of 4.03 mAh cm-2. The electrochemical performance results illustrate that the PS/rGO cathode is capable of achieving a practical areal capacity comparable to current LIB technology, and a superior electrochemical performance to other slurry-coated PS electrode.
6
To address the difference between the distinctly different method of preparing PS and melt-diffused cathodes, the electrochemical performance of the S/rGO was also evaluated. The average discharge capacities of S/rGO cathode were 1186, 920, 810, 739 and 650 mAh g1
at 0.1, 0.2, 0.5, 1.0 and 2.0 C, respectively, and recovered to 872 mAh g-1 at 0.2 C (Figure
3a, Table S4). The charge-discharge curves at all C rates clearly revealed longer charge and discharge plateaus for PS/rGO, indicating a greater S utilization and hence, specific capacity, as compared to S/rGO. In addition, the voltage hysteresis, determined by the voltage difference between the middle of the charge and the discharge curves, was found to be smaller for the PS/rGO than S/rGO, indicating enhanced redox kinetics in PS/rGO as compared to S/rGO (Figures 3b–f). For long-term cycling of the S/rGO cathode, initial discharge capacities were 867, 821, 747 and 659 mAh g-1 at 0.2, 0.5, 1.0 and 2.0 C rate, respectively (Figure 4). After 200 cycles, the discharge capacities became 653, 650, 605 and 533 mAh g-1 at 0.2, 0.5, 1.0 and 2.0 C rate, respectively. Similar to the trend observed in the rate studies, the first and last (200th) charge-discharge curves of the long-term cycling for PS/rGO featured longer charge and discharge plateaus, illustrating a higher S utilization and a smaller voltage hysteresis as compared to S/rGO (Figures S3 and S4). Capacity fade, known to be positively correlated with PS shuttling effect, was also determined for the two electrodes. Capacity fade values for PS/rGO electrode were 0.100%, 0.080%, 0.091%, 0.075% per cycle at 0.2, 0.5, 1.0 and 2.0 C rate, respectively. For the S/rGO electrode, the capacity fade values were 0.142%, 0.117%, 0.105% and 0.106% per cycle at 0.2, 0.5, 1.0 and 2.0 C rate, respectively. The lower capacity fade could be attributed to the enhanced redox kinetics due to the smaller voltage hysteresis, allowing faster conversion of PS to insoluble sulfides, decreasing the PS concentration and thereby limiting the PS shuttle [40]. Based on the above analysis, PS/rGO electrode showed 48% higher specific capacity and 26% lower capacity fade per cycle, on average, as compared to S/rGO electrode. The difference in electrochemical performance was found to be correlated to the difference in cathode structure. The PS/rGO cathode was highly porous and interconnected before (Figure 5a, b) and after cycling (Figure 5c, d). The pores were uniform in size and evenly distributed, suggesting that the reversible reaction of Li-S system during battery cycling did not affect the porous structure of the PS/rGO cathode. This could be attributed to the preparation method of the PS/rGO cathode, whereby the S source (in the form of Li2S6) was added after the rGO scaffold was constructed (Scheme 1). As such, the chemical reactions only occurred on the surface of the robust rGO scaffold, rendering the structure intact even after cycling. On the other hand, fresh S/rGO cathode structure, although 7
interconnected, consisted of a mixture of large and small pores (Figure 6a, b). After the rate capability test, the cathode underwent a major structural transformation, forming a highly dense and closely packed structure (Figure 6c, d). Unlike the PS/rGO cathode, S, that was melt-diffused with the crumpled rGO sheets, was already incorporated within the structure of S/rGO (Scheme 1). During discharge, the dissolution of insoluble S to soluble PS resulted in disconnections within the S/rGO cathode structure, leading to structural changes. Structural changes were known to occur for composite S cathodes due to the dissolution of S to PS species during battery discharge [8, 41, 42]. The difference in structure of the two electrodes was quantified by nitrogen adsorption, which was performed on both cathodes in the absence of S or LiPS. S removal was necessary to simulate the effect of structural changes observed in SEM. For the PS/rGO cathode, analysis was conducted on the preformed rGO cathode, whereas the S/rGO cathode was washed with CS2 to remove the S. The nitrogen adsorption-desorption isotherm for both cathodes corresponded to a type II isotherm with H3 hysteresis loop (Figure 7a) [43, 44]. The surface area and pore volume of the preformed rGO cathode were 264 m2/g and 0.31 cm3/g, respectively. These values were higher than that of the washed S/rGO cathode (181 m2/g and 0.25 cm3/g). In addition, pore size distribution analysis revealed the presence of mesopores (3–4 nm) and macropores (60–90 nm) for both cathodes (Figure 7b). The rGO cathode had a comparable mesopore size (3.5 nm vs. 3.6 nm) and slightly larger macropore size (82 nm vs. 65 nm) than the washed S/rGO cathode. Since both electrodes were essentially identical in terms of composition, electrolyte volume and S loading density, the higher specific capacities and lower capacity fade values could be attributed to the higher surface area of the PS/rGO cathode, as compared to the S/rGO cathode. The higher surface area of the PS/rGO cathode led to an increased availability of electrochemically active sites for S species, such as S, Li2S and PS, allowing both nucleation and binding to occur on the cathode surface, which led to higher specific capacities [16]; this agreed well with the electrochemical performance discussed earlier (Figure 3 and 4). This resulted in a decrease in the concentration of dissolved PS in bulk, reducing the undesired PS shuttling effect, consistent with the lower capacity fading values of PS/rGO electrode, as compared to the S/rGO electrode [40]. In addition to surface area difference, the structural change, or the lack thereof, was found to have a pronounced effect on the ohmic resistance of the electrodes. Both S/rGO and PS/rGO electrodes were further examined by cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) after rate capability studies. CV curves of both electrodes revealed features typical of a Li-S battery system [45]: two sharp reduction peaks (I and II) 8
and a broader oxidation peak (III) that consisted of several overlapping peaks (Figure 8). Peaks I (~2.3 V) and II (~2.0 V) corresponded to the reduction of elemental S to soluble higher order PS (Li2Sx, 4 ≤ x ≤8), and its further reduction to insoluble Li2S2 and Li2S, respectively, while Peak III (~2.3–2.4V) was attributed to the reverse process. Peak shape of both cathodes resembled one another as these electrodes consisted of essentially identical constituents, ratios and S loading (which was fixed at 1.50 mg cm-2). This implied that the structural difference between these cathodes did not affect the potential at which redox reactions occurred. The area under the curve (typically correlated with capacity) was found to be larger for the PS/rGO cathode as compared to the S/rGO cathode, which was consistent with the galvanostatic cycling results discussed earlier. The EIS data collected were mathematically transformed into Nyquist plots (Figure 9). In general, these plots appeared to be an overlap of several semicircles ending with a steep upward slope. The semicircle at the high frequency region was fitted with an equivalent circuit (Figure 9, inset). The first intercept at the high frequency region of the real (Z’) axis gives the value of electrolyte resistance, Re [16, 20, 46]. The difference between the Z’ axis intercepts of the fitted semicircles gives the charge transfer resistance (RCT) value, which is associated with the charge transfer process between S and the electrode. The Re values of PS/rGO and S/rGO cells were found to be 4.7 and 7.4 Ω, respectively. The RCT values of PS/rGO and S/rGO cells were 3.0 and 7.8 Ω, respectively. As discussed earlier, the PS/rGO cathode has a higher surface area than the S/rGO cathode, implying that the amount of electrochemically active sites would be greater in PS/rGO than S/rGO. With the electrolyte to S ratio (10 µL/mgS) being identical for both cathodes, the effects of surface area versus ohmic resistance could be established. The higher surface area of the PS/rGO cathode allowed more S species in the electrolyte to be captured as compared to S/rGO cathode, reducing the viscosity and hence, lowering Re. At the same time, the insulating S layer would be thinner in PS/rGO than S/rGO, resulting in a lower RCT for PS/rGO cathode as compared to the S/rGO cathode [16]. In addition, structural changes that occurred in the S/rGO cathode could lead to disconnectivity between conductive elements within the cathode [41], contributing to a higher resistance as compared to the PS/rGO cathode. In the structurally intact PS/rGO cathode, the conductive elements within the structure remained interconnected, ensuring continuous and unimpeded electron conduction pathways from the current collector throughout the entire 3D cathode structure. The lower ohmic resistance of the PS/rGO electrode, as compared to the S/rGO cathode, suggested enhanced redox kinetics, which agreed well with the charge-
9
discharge curve analysis discussed earlier, resulting in the improved rate, specific capacity and cycling performance of the Li-S batteries (Figures 3 and 4).
4. Conclusion In conclusion, using inexpensive and non-toxic commercially available materials, we have developed the PS/rGO cathode via slurry coating for Li-S batteries. At a S loading of 1.50 mg cm-2, the initial discharge capacities of the PS/rGO electrode were 1220, 1112, 1087 and 1007 mAh g-1 at 0.2, 0.5, 1.0 and 2.0 C rate, while retaining capacities of 999, 948, 906 and 866 mAh g-1 over 200 cycles, respectively. These values corresponded to an average capacity fade of only 0.086% per cycle. At a higher S loading of 5.05 mg cm-2, the PS/rGO cathode attained a practical areal capacity of > 4 mAh cm-2 over 50 cycles. The electrochemical performance of the PS/rGO cathode is superior to the other slurry-coated PS cathodes reported in the literature. We have also evaluated the difference between a PS cathode (PS/rGO) and cathode prepared via melt-diffusion (S/rGO). The structural change that occurred in the S/rGO cathode led to a highly densified and compact structure, resulting in a lower surface area with smaller pores, greater voltage hysteresis and a higher ohmic resistance, as compared to the structurally intact PS/rGO cathode. The PS/rGO cathode showed an average of 48% increase in specific capacity and 26% lower in capacity fade as compared to S/rGO cathode at various C rates (0.2 C to 2.0 C). These results clearly demonstrated that the preparation of S cathode strongly influenced the electrochemical performance. Excellent electrochemical performance could be achieved by using commercially available materials without any processing. To maximize the electrochemical performance, future design of practical Li-S batteries should consider different cathode preparation approaches, such as incorporating S source into a preformed cathode host structure, and focus on methods that are industrially scalable.
Appendix: Supplementary Data Supplementary data associated with this article can be found in the online version at http://www.journals.elsevier.com/nano-energy.
Acknowledgements We gratefully acknowledge Dr. Su Seong Lee and Dr. Jinhua Yang for their useful discussion, and Carina Lim Yi Jing for assistance in some of the experimental work. This 10
work was funded by NanoBio Lab (Biomedical Research Council, Agency for Science, Technology and Research, Singapore).
References [1]
T. Cleaver, P. Kovacik, M. Marinescu, T. Zhang, G. Offer, J. Electrochem. Soc. 165 (2017) A6029.
[2]
S.S. Zhang, J. Power Sources 231 (2013) 153.
[3]
A. Manthiram, Y. Fu, S.-H. Chung, C. Zu, Y.-S. Su, Chem. Rev. 114 (2014) 11751.
[4]
R. Fang, S. Zhao, Z. Sun, D.W. Wang, H.M. Cheng, F. Li, Adv. Mater. 29 (2017) 1606823.
[5]
H. Hu, B. Zhao, H. Cheng, S. Dai, N. Kane, Y. Yu, M. Liu, Nano Energy 57 (2019) 635.
[6]
H. Hu, H. Cheng, Z. Liu, G. Li, Q. Zhu, Y. Yu, Nano Lett. 15 (2015) 5116.
[7]
C. Dong, W. Gao, B. Jin, Q. Jiang, iScience, 6 (2018) 151.
[8]
R. Elazari, G. Salitra, Y. Talyosef, J. Grinblat, C. Scordilis-Kelley, A. Xiao, J. Affinito, D. Aurbach, J. Electrochem. Soc. 157 (2010) A1131.
[9]
G. Li, S. Wang, Y. Zhang, M. Li, Z. Chen, J. Lu, Adv. Mater. 30 (2018) 1705590.
[10] S.S. Zhang, J.A. Read, J. Power Sources 200 (2012) 77. [11] Y. Fu, Y.S. Su, A. Manthiram, Angew. Chem. Int. Ed. 52 (2013) 6930. [12] C. Zu, Y. Fu, A. Manthiram, J. Mater. Chem. A 1 (2013) 10362. [13] Y. Yang, G. Zheng, Y. Cui, Energy Environ. Sci. 6 (2013) 1552. [14] R. Xu, I. Belharouak, J.C.M. Li, X. Zhang, I. Bloom, J. Bareño, Adv. Energy Mater. 3 (2013) 833. [15] R. Demir-Cakan, M. Morcrette, Gangulibabu, A. Guéguen, R. Dedryvère, J.-M. Tarascon, Energy Environ. Sci. 6 (2013) 176. [16] K. Han, J. Shen, S. Hao, H. Ye, C. Wolverton, M.C. Kung, H.H. Kung, ChemSusChem 7 (2014) 2545. [17] C. Zu, A. Manthiram, Adv. Energy Mater. 4 (2014) 1400897. [18] X. Pu, G. Yang, C. Yu, Adv. Mater. 26 (2014) 7456. [19] G. Zhou, Y. Zhao, C. Zu, A. Manthiram, Nano Energy 12 (2015) 240. [20] G. Zhou, E. Paek, G.S. Hwang, A. Manthiram, Nat. Commun., 6 (2015) 7760. [21] R. Fang, S. Zhao, S. Pei, Y. Cheng, P. Hou, M. Liu, H.-M. Cheng, C. Liu, F. Li, Carbon, 109 (2016) 719. [22] X. Li, X. Pu, S. Han, M. Liu, C. Du, C. Jiang, X. Huang, T. Liu, W. Hu, Nano Energy 30 (2016) 193. 11
[23] G. Babu, N. Masurkar, H. Al Salem, L.M. Arava, J. Am. Chem. Soc. 139 (2017) 171. [24] S. Luo, M. Yao, S. Lei, P. Yan, X. Wei, X. Wang, L. Liu, Z. Niu, Nanoscale 9 (2017) 4646. [25] S.H. Chung, A. Manthiram, Adv. Mater. 30 (2018) 1705951. [26] R. Yu, S.-H. Chung, C.-H. Chen, A. Manthiram, Energy Storage Mater. 18 (2019) 491. [27] A. Kraytsberg, Y. Ein-Eli, Adv. Energy Mater. 6 (2016) 1600655. [28] R.D. Rauh, K.M. Abraham, G.F. Pearson, J.K. Surprenant, S.B. Brummer, J. Electrochem. Soc. 126 (1978) 523. [29] M. Agostini, D.-J. Lee, B. Scrosati, Y.K. Sun, J. Hassoun, J. Power Sources 265 (2014) 14. [30] K.E. Hendrickson, L. Ma, G. Cohn, Y. Lu, L.A. Archer, Adv. Sci. 2 (2015) 1500068. [31] H. Al Salem, G. Babu, C. V. Rao, L.M.R. Arava, J. Am. Chem. Soc. 137 (2015) 11542. [32] N. Mosavati, V.R. Chitturi, S.O. Salley, K.Y.S. Ng, J. Power Sources 321 (2016) 87. [33] N. Mosavati, S.O. Salley, K.Y.S. Ng, J. Power Sources 340 (2017) 210. [34] K. Chen, Z. Sun, R. Fang, Y. Shi, H.-M. Cheng, F. Li, Adv. Funct. Mater. 28 (2018) 1707592. [35] Y. Song, W. Zhao, N. Wei, L. Zhang, F. Ding, Z. Liu, J. Sun, Nano Energy 53 (2018) 432. [36] M.Q. Zhao, Q. Zhang, J.Q. Huang, G.L. Tian, J.Q. Nie, H.J. Peng, F. Wei, Nat. Commun. 5 (2014) 3410. [37] J. Pan, G. Xu, B. Ding, J. Han, H. Dou, X. Zhang, RSC Adv. 5 (2015) 13709. [38] Y. Gogotsi, P. Simon, Science 334 (2011) 917. [39] H.-J. Peng, J.-Q. Huang, X.-B. Cheng, Q. Zhang, Adv. Energy Mater. 7 (2017) 1700260. [40] A.N. Mistry, P.P. Mukherjee, J. Phys. Chem. C 122 (2018) 23845. [41] S.-E. Cheon, K.-S. Ko, J.-H. Cho, S.-W. Kim, E.-Y. Chin, H.-T. Kim, J. Electrochem. Soc. 150 (2003) A796. [42] P. Barai, A. Mistry, P.P. Mukherjee, Extreme Mech. Lett. 9 (2016) 359. [43] H.C. Youn, S.M. Bak, M.S. Kim, C. Jaye, D.A. Fischer, C.W. Lee, X.Q. Yang, K.C. Roh, K.B. Kim, ChemSusChem 8 (2015) 1875. [44] B.D. Ossonon, D. Bélanger, RSC Adv. 7 (2017) 27224. [45] M. Wild, L. O’Neill, T. Zhang, R. Purkayastha, G. Minton, M. Marinescu, G.J. Offer, Energy Environ. Sci. 8 (2015) 3477. [46] Z. Deng, Z. Zhang, Y. Lai, J. Liu, J. Li, Y. Liu, J. Electrochem. Soc. 160 (2013) A553.
12
Jian Liang Cheong is a Senior Lab Officer at the NanoBio Lab, Singapore. He received his B. Sc. (Honours) in Chemistry from the National University of Singapore (NUS), Singapore in 2010. Currently, he is doing his Ph.D. under the supervision of Prof. Jackie Ying. He is interested in developing practical and scalable systems for energy storage.
Ayman A. AbdelHamid is a Research Scientist at the NanoBio Lab, Singapore. He received his M.Sc. in Nanoscience and Technology from Nile University, Egypt in 2012, and his Ph.D. in Materials Science and Engineering from Nanyang Technological University, Singapore in 2017. His research interests include development of nanomaterials and nanocomposites for energy storage applications.
13
Jackie Y. Ying received her B.E. and Ph.D. from The Cooper Union and Princeton University, respectively. She joined the faculty at Massachusetts Institute of Technology in 1992, where she was Professor of Chemical Engineering until 2005. She was the Founding Executive Director of the Institute of Bioengineering and Nanotechnology in Singapore from 2003 to 2018. Prof. Ying currently leads the NanoBio Lab as A*STAR Senior Fellow. For her research on nanostructured materials, Prof. Ying has been recognized with the American Ceramic Society Ross C. Purdy Award, David and Lucile Packard Fellowship, Office of Naval Research Young Investigator Award, National Science Foundation Young Investigator Award, Camille Dreyfus Teacher-Scholar Award, American Chemical Society Faculty Fellowship Award in Solid-State Chemistry, Technology Review’s Inaugural TR100 Young Innovator Award, American Institute of Chemical Engineers (AIChE) Allan P. Colburn Award, International Union of Biochemistry and Molecular Biology Jubilee Medal, Materials Research Society Fellowship, Royal Society of Chemistry Fellowship, Crown Prince Grand Prize in the Brunei Creative, Innovative Product and Technological Advancement (CIPTA) Award, American Institute for Medical and Biological Engineering Fellowship, Academy of Sciences of Iran Medal of Honor, American Association for the Advancement of Science Fellowship, Singapore National Academy of Science Fellowship, and Turkish Academy of Sciences Academy Prize in Science and Engineering Sciences. Prof. Ying was elected a World Economic Forum Young Global Leader, and inducted to the German National Academy of Sciences, Leopoldina and the U.S. National Academy of Inventors. She was named one of the “One Hundred Engineers of the Modern Era” by AIChE in its Centennial Celebration, and a Highly Cited Researcher (Cross-Field Category) by Clarivate Analytics. She was the inaugural winner of the Mustafa Prize “Top Scientific Achievement Award” in 2015 for her research in bio-nanotechnology. She is the Editor-inChief of Nano Today.
14
Scheme 1. Preparation of PS/rGO cathode (top) and S/rGO cathode (bottom).
b)
a)
2 µm
Figure 1. SEM images of rGO cathodes: (a) top view and (b) cross-sectional view.
15
2 µm
a)
b)
c)
d)
Figure 2. Electrochemical performance of PS/rGO cathode. (a) Rate capability at S = 1.50 mg cm-2. (b) 5th cycle charge-discharge curves from rate capability study at different C rates of ( ) 0.1 C, ( ) 0.2 C, ( ) 0.5 C, ( ) 1.0 C and ( ) 2.0 C, and S = 1.50 mg cm-2. (c) Longterm cycling at S = 1.50 mg cm-2 and various C rates of ( ) 0.2 C, ( ) 0.5 C, ( ) 1.0 C and ( ) 2.0 C. (d) Long-term cycling at 0.1 C and S = 5.05 mg cm-2.
16
b)
a)
c)
d)
e)
f)
Figure 3. (a) Rate capability studies of ( ) PS/rGO and ( ) S/rGO cathodes, and (b–f) the corresponding 5th cycle charge-discharge curves of ( ) PS/rGO and ( ) S/rGO cathodes at (b) 0.1 C, (c) 0.2 C, (d) 0.5 C, (e) 1.0 C and (f) 2.0 C, with a S loading of 1.50 mg cm-2.
17
a)
b)
c)
d)
Figure 4. Long-term cycling performance of ( ) PS/rGO and ( ) S/rGO cathodes at (a) 0.2 C, (b) 0.5 C, (c) 1.0 C and (d) 2.0 C, with a S loading of 1.50 mg cm-2.
18
a)
b)
10 µm
c)
2 µm
d)
10 µm
2 µm
Figure 5. SEM images of PS/rGO cathodes: (a, b) fresh cathode, (c, d) after rate capability studies.
19
a)
b)
10 µm
c)
2 µm
d)
10 µm
2 µm
Figure 6. SEM images of S/rGO cathodes: (a, b) fresh cathode, (c, d) after rate capability studies.
20
a)
b)
Figure 7. Nitrogen adsorption-desorption of ( ) PS/rGO and ( ) S/rGO. (a) Isotherm and (b) BJH desorption pore size distribution.
21
II
I
I
Figure 8. Cyclic voltammogram of ( ) PS/rGO and ( ) S/rGO cathodes.
Re
RCT
CPE
Figure 9. Nyquist plots with Re and RCT values of cycled ( ) PS/rGO and ( ) S/rGO cells after rate capability studies. Inset: high frequency region with electrochemical fitted circuit.
22
TOC Graphics
23
Research Highlights •
A high-performance lithium-polysulfide battery using inexpensive commercially available materials has been developed
•
Specific capacities between 1220 and 1007 mAh g-1 were achieved at charge rates of 0.2–2.0 C, having capacity fade of lower than 0.14% per cycle over 200 cycles
•
At higher sulfur loading, a practical areal capacity of > 4 mAh g-1 was achieved
•
The difference between sulfur cathodes prepared via polysulfide addition and meltdiffusion was examined at identical sulfur loadings
•
The polysulfide cathode offered higher specific capacity and lower capacity fade than the melt-diffused sulfur cathode due to differences in morphology, surface area and ohmic resistance
1