Tunnel Structure Enhanced Polysulfide Conversion for Inhibiting “Shuttle Effect” in Lithium-Sulfur Battery

The Lithium sulfur (Li-S) battery has a great potential to replace lithium-ion batteries due to its high-energy density. However, the “shuttle effect” of polysulfide intermediates (Li2S8, Li2S6, Li2S4, etc.) from the cathode can lead to rapid capacity decay and low coulombic efficiency, thus limiting its further development. Anchoring polysulfide and inhibiting polysulfide migration in electrolytes is one of the focuses in Li-S battery. It is well known that polar metal oxides-manganese oxides (MnO2) are normally used as an effective inhibitor for its polysulfide inhibiting properties. Considering the natural 1D tunnel structure, MnO2 with three kinds of typical tunnel-type were screened to study the effects of the tunnel size on the adsorption capacity of polysulfide. We found that MnO2 with larger tunnel sizes has stronger chemisorption capacity of polysulfide. It promotes the conversion of polysulfide, and corresponding cathode exhibits better cycle reliability and rate performance in the cell comparison tests. This work should point out a new strategy for the cathode design of advanced Li-S battery by controlling the tunnel size.


Introduction
With increasing demand of chemical energy storage, the lithium sulfur (Li-S) battery with a high theoretical capacity of 1675 mAh g −1 , is considered to be a promising candidate to replace the state-of-the-art lithium-ion batteries [1][2][3]. Meanwhile, sulfur is abundant, environmentally friendly and cost effective, which fully meets the requirements of secondary energy storage power [4][5][6]. However, the practical application of the Li-S battery is still hindered by the rapid capacity decay, low coulombic efficiency and poor rate performance, which are mostly due to the "shuttle effect" [7][8][9][10][11]. Recently, meso/microporous carbon [12][13][14], carbon nanotubes [15][16][17] and other porous carbon-based materials which have good conductivity and chemical stability, have been applied to suppress the shuttle of polysulfide, keep good conductivity and improve sulfur utilization. However, their ability for polysulfide adsorption is very limited due to their nonpolar nature.
As for enhancing the polysulfide adsorption ability and suppressing the "shuttle effect", metal oxides are noticed because of their strong polar-polar chemical interaction with polysulfide [18][19][20][21][22][23][24][25]. More recently, lots of literature has reported that anchoring polysulfide by chemical reaction is much more promising and metal oxides such as SiO 2 , V 2 O 5 , Al 2 O 3 , TiO 2 , and MnO 2 could significantly improve cycling performance and realize high-loading sulfur in Li-S battery [26,27]. Among known metal oxides which are used as cathode materials, MnO 2 has the most potential in the Li-S battery due to its abundant resources, low cost, and nontoxicity. The previous research indicates MnO 2 has a particularly strong adsorption capacity for polysulfide compared to carbon materials and other metal oxides [28]. Moreover, MnO 2 has a large variety of crystal structures including α, β, δ, λ and so on, depending on the edge-sharing and angle-sharing MnO 6 octahedron.
Thus, 1D tunnel, 2D layer, and 3D mesh structure can be formed [29][30][31][32][33]. δ-MnO 2 as a layered structure can be used as an effective sulfur host material. Nazar's group firstly reported that δ-MnO 2 can be considered as a remarkable chemical inhibitor for polysulfide based on mediating polysulfides redox [34]. For the tunnel-type MnO 2 , Ni and coworkers prepared γ-MnO 2 covered with sulfur nanospheres, forming a core-shell structure to trap polysulfide through physical and chemical effects [35]. Zhang and researchers fabricated Mo-doping γ-MnO 2 to accelerate the transformation of polysulfide [36]. Wang and coworkers have synthesized highly ordered mesoporous β-MnO 2 to encapsulate sulfur and found that the thin mesoporous walls could provide short diffusion distances for Li ions [37]. These hybrid structures can physically and chemically encapsulate polysulfide and enhance the electrochemical performance of sulfur cathode. The studies above have reported synthesizing tunnel-type MnO 2 by in-situ recombination or ion doping and its application as sulfur host, but little work has focused on the effects of tunnel sizes on the polysulfide conversion abilities.
Herein, MnO 2 with three different tunnel sizes have been screened and prepared by hydrothermal reaction to investigate the effects of tunnel sizes on the polysulfide conversion. It is found that MnO 2 with a larger tunnel size has stronger chemisorption capacity of polysulfide, and therefore is more favorable to inhibiting the polysulfide shuttle. In addition, the larger tunnel size possesses faster reaction kinetics in the redox reaction in Li-S battery, hence larger-tunnel-size-MnO 2 based cathode exhibits better cycle reliability and rate performance in the cell tests. This work should provide a new perspective for the cathode design of advanced Li-S battery. Figure 1 illustrates the theoretical structures of the three tunnel-types of MnO 2 and how each tunnel phase is conventionally named as M × N, where M and N stand for the number of the MnO 6 octahedra constituting the height and width of the tunnel, respectively [38,39]. Depending on the number of MnO 6 units in the MnO 2 , the tunnel size increases from β-MnO 2 (1 × 1 tunnel) to α-MnO 2 (2 × 2 tunnel), and to todorokite MnO 2 (3 × 3 tunnel) [40][41][42][43]. The β-MnO 2 was obtained as follows: First, 1.0 g of MnSO 4 ·H 2 O was dissolved in 60 mL of distilled deionized water, and then 0.37 g KMnO 4 was added into the solution with stirring. Then, the slurry was transferred into a Teflon-lined stainless-steel autoclave and heated at 120 • C for 12 h. The resulting product was washed and dried at 60 • C overnight.

Synthesis of Manganese Oxides
The α-MnO 2 was prepared as follows: Firstly, 3.67 g of MnAC 2 ·4H 2 O and 2.5 mL of CH 3 COOH were dissolved in 35 mL deionized water with stirring. Then, 40 mL of 2.17 g KMnO 4 was added to the above mixture. Finally, the mixed solution was maintained at 100 • C for 24 h in a Teflon-lined autoclave. The resulting product was collected by centrifugation, washed with deionized water and dried at 100 • C overnight.
The todorokite MnO 2 (t-MnO 2 ) was synthesized as follows: First, 30 mL 6.0 M of NaOH aqueous solution was dropped into the solution (20 mL) containing 1.7 g of MnSO 4 ·H 2 O with stirring. Then, 0.35 g of MgSO 4 ·7H 2 O and 1.90 g of K 2 S 2 O 8 were added to the above mixture. After stirring for 3 h at room temperature and being washed with distilled water, the wet sample was dispersed in 300 mL of 1.0 M of MgCl 2 ·6H 2 O and stirred for 24 h for ion exchange. The obtained mixture was transferred into a Teflon-lined autoclave and heated at 160 • C for 24 h. The resulting mixture was filtered and washed with distilled water. Lastly, the product was dried at 100 • C for 12 h.

Preparation of the Sulfur Composites
Sulfur and manganese oxides were mixed with a ratio of 70:30. The mixture was then transferred into a sealed stainless-steel vial and heated at 155 • C for 12 h in an oven.

Cell Assembling and Testing
The cathodes were prepared with 60 wt% active material, 30 wt% Super P carbon, and 10 wt% polyvinylidene fluoride (PVDF) binder and the slurry was casted onto Al foil current collector. The electrodes were dried at 60 • C in a vacuum for 24 h. The CR2032 coin cells with lithium metal (counter electrode) were fabricated. Sulfur cathode size is d = 12 mm, pure sulfur loading is~1 mg cm −2 , and overall sulfur content is 42 wt%. The electrolyte was 1.0 M LiTFSI in dioxolane/dimethoxyethane solvent (DOL/DME volume ratio1:1) with 2 wt% lithium nitrate (LiNO 3 ) as an additive and PE was used as the separator. Galvanostatic measurements were carried out between 1.8 and 2.8 V (vs. Li/Li + ) on a Land CT2001A system (LANHE, Wuhan, China). The cyclic voltammetry (CV) experiments were performed with a CHI600E electrochemical workstation (CH, Shanghai, China) at a scanning rate of 0.1 mV s −1 between 1.7 and 2.8 V.

Polysulfide Adsorption Test
Typically, 5 mM Li 2 S 6 solution was prepared by the reaction Li 2 S with S in DOL and DME (v/v = 1:1). Then three samples of the same weight (5 mg) were added into three glass vials, respectively, and dispersed in 2 mL Li 2 S 6 solution. Optical images were taken to compare the adsorption ability.

Structure Characterization
The morphology was examined by using scanning electron microscopy (SEM, JSM-2100F, JEOL, Tokyo, Japan). XRD patterns were collected using a D/max-TTR III (Rigaku Corporation, Shibuya-ku, Japan) with Cu Ka radiation, 40 kV, 200 mA). Nitrogen adsorption/desorption isotherms were performed on a Quantachrome Autosorb-IQ system (Quantachrome Instruments, Boynton Beach, FL, USA). The nanostructures of manganese oxides were characterized by high-resolution transmission electron microscopy (HRTEM, JEOL, Tokyo, Japan, 2010). For XPS, the samples were sealed in a vial before being quickly transferred to the chamber of an ultra-high vacuum Imaging XPS Microprobe system for analysis (Thermo Scientific ESCALAB 250Xi, Waltham, MA, USA).

Results
As illustrated in Figure 2, the morphology of prepared three tunnel-type MnO 2 was carefully characterized. In the Figure 2a,c, the SEM images clearly showed the nanofiber morphology of the β-MnO 2 and α-MnO 2 , with a length ranging from tens of nanometers to hundreds of nanometers. No extra phase was found on the surface of nanofiber. Because of the different reaction mechanism, the synthesized t-MnO 2 is nano-flake and with some long nanoribbons (Figure 2e). HTEM was used to measure the tunnel sizes of β-MnO 2 , α-MnO 2 and t-MnO 2 . In the Figure 2b, the HRTEM image shows the interlayer distance is 0.31 nm, which agrees with the (110) plane in the crystal structure of β-MnO 2 , and hence the existence of 1 × 1 tunnel is confirmed. The clear lattice fringes in Figure 2d show that the crystal plane spacing of α-MnO 2 is 0.70 nm, which corresponds to the (110) plane. For the t-MnO 2 structure, the crystallinity can be observed from the distinct lattice fringes in HRTEM image (Figure 2f). The interlayer distance of 0.97 nm corresponds to the (100) plane at an angle of 9.2 • , thereby confirming the existence of 3 × 3 tunnel. X-ray diffraction (XRD) measurement was used to confirm the crystal structure of synthetic tunnel-type MnO 2 ( Figure 3). As shown in Figure 3a, the diffraction pattern for the β-MnO 2 has five sharp peaks at 28.  To compare the polysulfide adsorption capacity of three tunnel-type MnO 2 , polysulfide adsorption test was performed. Typically, three samples of the same weight (5 mg) were added into three glass vials, respectively, and dispersed in 2 mL Li 2 S 6 solution. After 1 h adsorption, the solution changed from dark-yellow to light yellow (Figure 4d). Obviously, the solution containing t-MnO 2 was noticeably lighter in color than other solutions, which indicates t-MnO 2 has greater ability to adsorb polysulfide faster. It is usually believed that larger surface area means stronger adsorption capacity, but the BET surface area of t-MnO 2 is 35.47 m 2 g, smaller than half of that of α-MnO 2 (103.82 m 2 g). As a result, adsorption capacity of these atomic tunnels of MnO 2 were not well reflected by normal surface area tests. Similar adsorption independent of specific surface area by MnO 2 on methylene blue has also been reported by ref. [44], which apparently exhibits the special adsorption properties. Therefore, the stronger adsorption of t-MnO 2 of polysulfide mentioned above should be partially attributed to the special adsorption properties of the larger tunnel size of the polysulfide. X-ray photoelectron spectroscopy (XPS) analysis was used to find out the absorption type between polysulfide and tunnel-type MnO 2 . By examining the specimens taken out from the Li 2 S 6 solution after the adsorption test, it is found that all the S2p XPS spectra of β-MnO 2 , α-MnO 2 and t-MnO 2 with polysulfides reveal four types of sulfur environment, as shown in Figure 4a-c. In the lower-binding-energy region from 163 eV to 165 eV, two pairs of S2p peaks refer to the same terminal (ST) and bridging sulfur (SB) peak, coming from the sulfides and polysulfides. The peaks in the higher-binding-energy region between 171 and 166 eV correspond with the binding energy of thiosulfate and polythionate species, which arise from the redox reaction between Li 2 S 6 and MnO 2 [34]. Therefore, the chemisorption process occurs in such a way: the tunnel-type MnO 2 reacts with polysulfide and converts polysulfide to thiosulfate and polythionate, thereby anchoring polysulfide on the surface of MnO 2 .
Considering the excellent inhibiting effects on the "shuttle effect" by the tunnels of MnO 2 , then, electrochemical experiments with the cathodes of tunnel-type MnO 2 were performed. Cyclic voltammograms (CV) of cells in the first cycle were shown in Figure 5a. Two typical cathodic peaks of the three cathodes were all observed at~2.3 V and~2.0 V, which could be ascribed to the reduction process of sulfur. These two peaks are, respectively, assigned to the transformation of S 8 to the long-chain polysulfide and then the reduction of long-chain sulfide species to solid products (Li 2 S 2 /Li 2 S). When sweeping back, two sharp peaks of t-MnO 2 /S cathode were found at 2.35 V and 2.4 V, which is mainly due to the oxidation of short-chain sulfide to polysulfide and S 8 . Similarly, the β-MnO 2 /S cathode exhibits two peaks located at 2.44 V and 2.5 V. However, for the α-MnO 2 /S cathode, only one peak was observed at 2.4 V, indicating the complete conversion of Li 2 S and polysulfides into element sulfur. A large number of studies reported that it was a common phenomenon in the first CV scan of Li-S battery [45,46]. Notably, the peak current of t-MnO 2 /S electrode is greater than that of the α-MnO 2 /S and β-MnO 2 /S electrodes, which represents higher kinetics and reversibility of redox reaction. It makes sense that ions diffuse more easily in larger channels or more open structures. Moreover, the larger the cavity of tunnel-type MnO 2 , the higher the ionic conductivity [47]. Relevant studies on kinetics of MnO 2 have already been widely reported [37,48]. Thus, the t-MnO 2 with 3 × 3 tunnel could provide wider pathways and more spacious cavities for the incorporation of Li + ions into the material bulk than other smaller tunnel sizes.  Figure 5b shows the typical galvanostatic discharge/charge profiles of the Li-S battery at 1 C rate. For the initial cycle, the t-MnO 2 /S based cell delivers an excellent discharge capacity of 1431.9 mAh g −1 at 1 C, which is significantly larger than the 935.9 mAh g −1 of β-MnO 2 /S and 956.7 mAh g −1 of α-MnO 2 /S cathodes. The maximum discharge capacity reflects the utilization degree of sulfur when the battery carries on galvanostatic discharge. In order to enhance sulfur utilization, the sulfur in the electrode needs to be fully exposed to lithium ions and electrons, so as to effectively reduce to the lithium polysulfide species. The high initial discharge of t-MnO 2 /S cathode is due to the larger tunnel structure, which enhances sulfur utilization. In addition, the first discharge curves exhibited two plateaus at 2.0 V and~2.3 V, which coincides with the CV results. Upon charge process, the platform voltage is also consistent with the corresponding CV results.
In order to directly demonstrate the effects of strong chemisorption on polysulfide by tunnel-type MnO 2 , we further studied the cycle performance and rate capacity of Li-S cells assembled with β-MnO 2 /S, α-MnO 2 /S and t-MnO 2 /S as cathodes. As shown in the Figure 5c, the long-term cycle stability of Li-S battery in terms of discharge capacity and coulombic efficiency was also studied at 1 C. The cycle performance of the three electrodes was compared, and the cell with t-MnO 2 /S was found to have the best cycle stability in the long cycle. The residual capacity maintains in 583.7 mAh g −1 even over 500 cycles, which is still superior to 408 mAh g −1 of α-MnO 2 /S and 345 mAh g −1 of β-MnO 2 /S. Moreover, impressively, the coulombic efficiency of the t-MnO 2 /S based battery remain over 99% during the long-term cycling, which indicates its excellent intrinsic reversibility. In addition, SEM images of the discharged MnO 2 /S cathode after 100 cycles under 1 C, which further proves the stable structure of t-MnO 2 /S cathode to buffer the volume changes during repeated lithiation processes. Therefore, the enhanced chemisorption of t-MnO 2 to polysulfide effectively promotes anchoring polysulfide and inhibiting the shuttle of polysulfide, and thus greatly improving the cyclic stability and coulomb efficiency of the Li-S battery.
The rate performance of the cell was obtained in Figure 5d. It is noteworthy that the cell with t-MnO 2 /S shows a superior rate capability of 447 mAh g −1 at 2 C, which is 82 mAh g −1 higher than the specific discharge capacity of α-MnO 2 /S cathodes and 291 mAh g −1 higher than that of β-MnO 2 /S cathodes. In addition, t-MnO 2 /S cathode shows 1371, 676, 581, 498 mAh g −1 at 0.1, 0.2, 0.5 and 1 C, respectively. The preferable rate capability of t-MnO 2 /S based cell should mainly be attributed to the larger 3 × 3 tunnel structure, which accelerates the conversion of polysulfide and improves the reaction kinetics at higher current density.

Conclusions
In summary, MnO 2 with three different types of tunnel sizes were designed to investigate their chemisorption capacity for polysulfide. MnO 2 with larger tunnel size shows stronger chemisorption capacity through the redox reaction. The larger tunnel size of t-MnO 2 not only anchors and accelerates the conversion of polysulfide but also possesses faster reaction kinetics in Li-S battery. Thus, the t-MnO 2 based battery, with an ultrahigh initial capacity of 1431.9 mAh g −1 and a capacity of 583.7 mAh g −1 after 500 cycles, shows better cell performance and higher coulombic efficiency than the other two electrodes. Apparently, this work will bring some new concepts and strategies to the material selection and structure design for advanced Li-S battery.