Application of Carbon Nanotube-Based Materials as Interlayers in High-Performance Lithium-Sulfur Batteries: A Review

With the ever-increasing demands of electrochemical energy storage, lithium–sulfur (Li–S) batteries have drawn more attention because of their superior theoretical energy density and high specific capacity. However, practical applications of Li–S batteries suffer from problems such as low conductivity of sulfur and discharged products, severe polysulfide shuttling effect, and large volume change of sulfur during cycling, resulting in sluggish rate performance, and unsatisfactory cycle life. Various nanostructured carbon materials have been served as barrier layers to overcome these problems. In particular, carbon nanotubes (CNTs) with unique 1D nanostructure, have been introduced to Li–S batteries as the intermediate layers because of its superior flexibility, excellent electrical conductivity, and good chemical stability. Moreover, CNTs and CNTs-based barrier layers could also curb lithium polysulfides shuttling. In the minireview, we summarize recent works of CNTs-based materials as modifying interlayers for Li–S batteries. In addition, the strategies to enhance electrochemical performances of the batteries are summarized and discussed. Finally, the challenges and prospects for future research of CNTs-based materials as interlayer are proposed. We hope this review will be useful for designing and fabricating high-performance Li–S batteries and boost their practical applications.

However, there are some challenges for Li-S batteries to overcome, including the electrical insulating characteristic of sulfur and the discharge product (Li 2 S 2 /Li 2 S), the large volumetric change of sulfur during the charge/discharge process (approximately 80%), as well as the notorious shuttle effect caused by dissolved lithium polysulfides (LiPSs) (Huang et al., 2015;Cao et al., 2019;Chen Y.T. et al., 2019).
To address these problems, researchers have made considerable effort with regard to cathode and separator modification Wei et al., 2019;Li et al., 2020a), optimization of electrolytes (Ding et al., 2016), and lithium metal anode protection and stabilization (Paolella et al., 2019;Pei et al., 2019). Among them, the interlayers could selectively control the shuttle effect of polysulfides, while not disturbing the Li + transfer. As a superior type of carbon material, carbon nanotubes (CNTs) have been intensively investigated as intermediate layers due to their excellent mechanical durability and high electrical conductivity as well as stable chemical properties. Together with other materials, CNTs-based interlayers are receiving considerable attentions for Li-S batteries in recent years Jiang et al., 2018;Son et al., 2019). For example, Yao M. et al. (2018) reported the multifunctional layer consisting of MnO 2 and multi-walled CNTs modified polypropylene (PP) separator could tightly attract polysulfides and promote the transfer of electrons and ions, which could enhance the electrochemical performance of Li-S batteries.  designed and constructed a MoS 2 /CNTs interlayer to insert between separator and cathode, exhibited an optimized cycle performance and excellent rate capability. Some reviews also discussed recent developments on the interlayer materials for Li-S batteries. For instance, Chen et al. reviewed on interlayer design of Li-S batteries by coating and inserting different types of carbon-based materials . Furthermore, Jiao et al. reviewed recent articles regarding to the interlayers for Li-S batteries including carbon-based interlayers ; Pang and co-workers reviewed on recent research work on CNT-based materials for Li-S batteries (Zheng et al., 2019). However, there is still a lack of review to exclusively cover the state-of-the-art developments of CNTs-based interlayer materials for Li-S batteries.
Herein, we provide an overview on carbon nanotubesbased materials as inter-layers of Li-S batteries. Their nano/microstructure and electrochemical performances are summarized. In addition, some reasonable suggestion on their design and development are proposed to facilitate breakthroughs. We hope this review could attract more attentions to CNTsbased interlayer materials for Li-S batteries and boost their practical applications.

MWCNTs
CNTs can be regarded as the curled graphene sheets, and they could be classified into two catagories, including multi-walled CNTs (MWCNTs) and single-walled CNTs (SWCNTs) Ali et al., 2019). MWCNTs could be seen as the collection of multiple curled graphene sheets. Compared to SWCNTs, they are cheaper and easier to fabricate. Moreover, they remains the merits of high conductivity, low thermal expansion coefficient and stable structure (Zheng et al., 2019). For interlayers for Li-S batteries, MWCNTs are used mainly in two ways, (1) as the coating layer on separator (Cheng et al., 2016;; (2) as the self-supporting film (Kim H.M. et al., 2016;. The electrochemical performances of Li-S batteries with MWCNTs-based interlayers are summarized in Table 1.

Separator-Based Interlayer
As an essential part of Li-S batteries, separators could separate the cathode and anode and prevent the internal short circuit. However, the pore size (∼100 nm) of commercial separators is much larger than the average size of long-chained polysulfides, which may cause the soluble polysulfides to easily migrate to the anodes, leading to notorious shuttle effect (Li et al., 2020d). To prevent the shuttling of dissolved polysulfides, various MWCNTs-based materials have been developed to modify separators to provide physical entrapment and chemical adsorption for LiPSs, enhancing the electrochemical performances of Li-S batteries (Chang et al., 2015;Tan et al., 2018;Li et al., 2019e;Xiang et al., 2019).

Pure MWCNTs
As the most common CNT materials, MWCNTs have been intensively applied as interlayer coated on separator for Li-S batteries. For example, Chung and Manthiram (2014) developed a MWCNT-coated separator as a barrier to anchor dissolved LiPSs, as shown in Figure 1A. The MWCNTs layer not only act as an upper current collector for rapidly electron transport and high sulfur utilization but also serve as a filter to trap and adsorb LiPSs. Meanwhile, the porous network of MWCNTs layer is conducive to electrolyte infiltration and electron/ion diffusion. Hence, the cell with MWCNTs-coated separator exhibited a superior long-term cycling performance. After 150 cycles, the cell with MWCNT layer delivered a specific capacity of 881, 809, and 798 mAh g −1 at 0.2, 0.5, and 1 C, respectively. After that, researchers have paid more attention on carbon nanotubes modified separator to improve the performance of Li-S batteries. For instance, Ponraj et al. (2017) synthesized a hydroxyl-functionalized carbon nanotube (CNTOH) coated on the separator, which could address the poor electronic conductivity of active materials and mitigate the diffusion and migration of LiPSs in Li-S batteries owing to the good conductivity and hydroxyl groups of CNTOH.

MWCNTs-Based Composites
Nano-composite materials normally have a better performance than single-component materials due to the synergistic effect of different components, which have attracted extensive attention from researchers in recent years Guan B. et al., 2019;Kim et al., 2020;Ma et al., 2020a;. Many types of MWCNTs-based composites,     Figure 1B). The NCQDs with large surface area and rich oxygenated functional groups could be regarded as an efficient adsorbent for LiPSs, meanwhile, MWCNTs are beneficial to enhance the conductivity of composites. Therefore, the cell with MWCNT/NCQDs-coated separators displayed an excellent electrochemical performance, as well as an average self-discharge value of ∼11% after resting for 48 h, which is lower than that with the pristine MWCNTs-modified separator ( Figure 1B). MWCNTs/Polymer composites have also been fabricated to trap LiPSs through physical obstructing and chemical bonding in Li-S batteries (Chang et al., 2015;Luo et al., 2016;Lee Y.-H. et al., 2018). For Lee Y.-H. et al. (2018) instance, developed a spiderweb separator consisting of three functional nanomats with good mechanical properties and excellent electrical conductivity. In this sandwich-type separator, MWCNT-wrapped polyetherimide (PEI/MWCNT) nanomats were used as the top and bottom layer, and the PVIm[TFSI]/poly(vinylidene fluorideco-hexafluoropropylene) (PVDF-HFP) with a polyethylene (PE) nanomat was the middle layer ( Figure 1C). When tested in Li-S batteries, the cell with this composites separator exhibited higher discharge capacity and more stable cycling performance than that with pristine PE separator (Lee Y.-H. et al., 2018).
Metal-based compounds are mainly including metal oxides (Ding et al., 2018;Luo L. et al., 2018), metal sulfides (Jeong et al., 2017;Xiang et al., 2019), metal phosphides , metal borides (Guan B. et al., 2019), metal hydroxides  and metal-organic frameworks (MOFs) (Lee D.H. et al., 2018;Wu et al., 2018), which could offer a polar chemical interaction with LiPSs and curb polysulfides shuttling in Li-S batteries. For instance, Hong et al. (2019) fabricated cerium-MOFs/MWCNTs composites as coating material on separators for Li-S batteries at a high sulfur loading of 6 mg cm −2 , the cell with the modified separators could deliver an initial specific capacity of 993.5 mAh g −1 at 0.1 C, and the specific capacity of 886.4 mAh g −1 was retained after 200 cycles ( Figure 1D). The superior electrochemical performance is ascribed that Ce-MOFs/MWCNTs can mitigate the transmission and diffusion of polysulfides and boost the efficient catalytic conversion of LiPSs, owing to their uniformly dispersed catalytic active sites and large specific surface area as well as good electronic conductivity. Yao S. et al. (2018) successfully exfoliated a 2D antimony sulfides (Sb 2 S 3 ) sheets and combined them with MWCNTs (Sb 2 S 3 /MWCNTs) to modify the Celgard PP separator ( Figure 1E). With Sb 2 S 3 /MWCNTs modified separator, the Li-S cell exhibited a low average capacity decay rate (∼0.049% per cycle after cycling 1000 cycles at 1 C), indicated that the Sb 2 S 3 /MWCNTs has a strong interaction with polysulfides which was confirmed by the first-principle calculation.

Free-Standing Interlayer
In addition to coating MWCNTs-based materials onto the separator to curb the shuttling of lithium polysulfides, another alternative strategy is to insert a self-supporting interlayer between the cathode and separator Sun et al., 2018). The self-supported interlayers are usually much thicker than the modified-separators interlayers, and they are supposed to have the characteristics of strong chemical and/or physical adsorption toward polysulfides, good container for polysulfides and good electronic conductor. Also, the interlayer should allow smooth Li-ion diffusion. Therefore, with the insertion of interlayers, the LiPSs could be blocked at the cathode side, and the Li + diffusion and electron tranfer could also be promoted (Huang et al., 2016).

Pure MWCNTs
Because of intrinsic interweaving properties of CNTs, it is easy to prepare freestanding CNT interlayers with good flexibility and high porosity. Manthiram (Su and Manthiram, 2012) firstly fabricated a freestanding MWCNTs membrane by ultrasonic dispersion of synthesized MWCNTs, followed by a facile vacuum filtration (Supplementary Figure S1A). With the insertion of this freestanding interlayer, interfacial resistance in cathode was greatly reduced, and the intermediate polysulfides was efficiently localized at cathode side. The enhanced cycling performance of Li-S battery with MWCNTs interlayers could be ascribed that MWCNTs can prevent the undesirable migration and diffusion of polysulfides (Su and Manthiram, 2012). After that, many other interlayers based on MWCNTs have been explored for Li-S batteries (Lee and Kim, 2015;Sun et al., 2016). For instance, Kim H.M. et al. (2016) prepared a self-assembled MWCNTs membrane by mixing MWCNTs and electrolyte, followed by being applied to the top of sulfur electrode (Supplementary Figure S1B). The self-assembled MWCNTs membrane has more tightly interwoven structure and denser than bare MWCNTs interlayer. When tested in Li-S batteries, after 100 cycles at 0.5 C, the surface of the self-assembled MWCNTs interlayer only has tiny pores and more densely packed shape than that of bare MWCNTs interlayer. The enhanced electrochemical performance of the cell with self-assembled MWCNTs interlayer could be attributed to efficiently hinder polysulfide diffusion and enhance sulfur utilization through supplying polysulfide capturing sites (Kim H.M. et al., 2016).

MWCNTs-Based Composites
Owing to the abundance of functional groups, large specific surface area, and excellent electrical conductivity, graphene has attracted more interest in energy storage materials fields (Wu et al., 2019b;Ma et al., 2020c;Sui et al., 2020). In Li-S batteries, many researchers combined graphene with MWCNTs as barrier materials to inhibit polysulfides migration Shi et al., 2019). For instance, Huang et al. synthesized a freestanding graphene oxide/MWCNTs (GO/MWCNTs) film through combining MWCNTs with GO sheets (Huang et al., 2016), which exhibited high electronic conductivity and strong ability to immobilize LiPSs in Li-S batteries, resulting in improved electrochemical performance. The cell with GO/MWCNTs depicted an outstanding initial specific capacity of 1370 mAh g −1 at 0.1 C. After 300 cycles at 0.2 C, it still maintained a discharge capacity of 671 mAh g −1 , and the corresponding capacity decay rate was about 0.043% per cycle. More recently, Shi et al. (2019) reported a three-dimensional graphene/multi-walled carbon nanotube aerogels (G/MWCNTs) as self-supporting interlayer to improve the performance of Li-S cells. As displayed in Supplementary Figure S1D, this 3D G/MWCNTs aerogels with a 3D interconnected porous network were fabricated by the rapid reduction of graphene oxide/MWCNTs aerogel via self-propagating combustion. Cross-linked MWCNTs with graphene can give the membrane good mechanical flexibility and abundant pores, so that it can prevent the damage of interlayer during battery assembly and has more active sites to anchor polysulfides by chemical interaction and physical confinement.
In addition to combining with graphene, some researchers also combined MWCNTs with polar materials such as metal oxides Kong et al., 2018;Li et al., 2019b), metal sulfides Wang L. et al., 2018), and metal catalyst Ding et al., 2019), which could have produced strong chemical interactions with polysulfides. For instance, Wang L. et al. (2018) integrated a multicomponent sandwich structure interlayer as freestanding interlayer inserting between the sulfur cathode and separator, and the Li-S cell has been displayed outstanding electrochemical performance. In the structure, CNF is used as a scaffold to support CNT and vanadium disulfide compounds (VS 2 /CNT), and covered a graphene layer onto it in the end. As shown in Supplementary Figure S1C, the CNT containing VS 2 had stronger adsorption ability toward polysulfides than pure CNT, suggesting that the polar V-S bonds in VS 2 promote the adsorption toward polysulfide. With the introduction of VS 2 /CNT, the interlayer exhibited excellent synergetic effects to trap LiPSs and suppress the self-discharge effect .

SWCNTs
SWCNTs can be regarded as an ideal scaffold to anchor polysulfides in Li-S batteries due to their relatively uniform diameters and excellent mechanical and electrical properties, which could allow fast electron transfer (Zhao et al., 2012;Fang et al., 2017). Many efforts have been devoted to SWCNTs-based materials as interlayer for Li-S batteries (Kaiser et al., 2015;Chang et al., 2016;. For example, Kaiser et al. (2015) synthesized a refined and interwoven structured SWCNTs freestanding film as the barrier layer, and inserted it between sulfur cathode and separator, which could effectively restrain polysulfides, enhancing the electrochemical performance of Li-S batteries.
Later, Kim J.H. et al. (2016) fabricated a polymer-coated single-wall carbon nanotube membrane (PAA-SWCNTs) as functional interlayer, which could efficiently restrain the diffusion of polysulfides via physical blocking of SWCNTs scaffold and hydrogen-bonding interaction of PAA. Moreover, SWCNTs films can act as second current collector to smooth the transfer of electron and Li + . As a result, due to the synergic effect of PAA-SWCNT interlayer, the assembled cell exhibited an initial capacity of 770 mAh g −1 , it still retained a capacity of 573 mAh g −1 after 200 cycles at 1 C.
Apart from combined polymer with SWCNT, there are also researchers combined rGO with SWCNT as intermediate layer to anchor polysulfides . The composite membrane not only improve the internal electronic conductivity but also efficiently anchor polysulfides, resulting in an improved performance in Li-S batteries.

CONCLUSION AND OUTLOOK
In summary, we briefly review the recent development of CNT-based materials, including heteroatom-doped CNTs, carbonaceous materials/CNT composites, and metal-based materials/ CNT composites, as interlayers for Li-S batteries. Their nano/microstructure and electrochemical performances are intensively discussed. As we can see, even though the electrochemical properties of Li-S batteries with CNT-based interlayer have been significantly improved, several problems still exist and need to be addressed before practical application of Li-S batteries.
1. As mentioned above, CNTs-based materials are beneficial to immobilize the LiPSs when it is designed as an intercept layer, however, it should be noted that these materials could just relieve the shuttle effect to a certain degree physically or chemically, and the active materials still be possible to shuttle to the anode during the cycling. Moreover, the sulfur loading is normally about 1-2 mg cm −2 in most of above-mentioned work, and the highest loading does not exceed 10 mg cm −2 . With gradually increased sulfur loading, despite the existence of intercept layers, the issues of active material loss and polysulfide shuttling are severe. Hence, it is imperative to develop new materials which have strong binding energy and interactions with polysulfides. 2. The mechanism of CNTs-based materials for immobilization and catalytic conversion of LiPSs and their phase transformation during cycling are not clear enough, and still need to be explored. In-operando techniques, such as in situ electron microscopy, and in situ X-ray technique, as well as synchron X-ray based techniques, are very helpful to obtain time dependent information. So, more research using in-operando techniques need to conduct to get insight understanding of the CNTs-based materials as interlayers for Li-S batteries. 3. Another issue to be tackled for practical application is lithium metal anode for lithium-sulfur batteries. There is always a problem in battery systems in which lithium metal exists: the formation and growth of lithium dendrites and the resulting low Coulombic efficiency and volume changes Zhao Q. et al., 2019). Therefore, in order to obtain a stable and safe lithium anode, a suitable electrolyte additive can be used. At present, the research on electrolyte additives has been highly advanced. Further, it is also possible to insert an artificial layer on the side of the lithium anode . However, there is little research on application of interlayers of CNTs-based composites in anodes, and this area requires further effort. 4. The reported literatures on CNTs-based materials coating layer/freestanding interlayers have developed many highloading sulfur cathodes with excellent performances. However, the introduction of a coating/freestanding interlayer will increase the inactive weight ratio of the whole battery. Therefore, the direction to balance the relationship between the introduced CNTs-based interlayers and the total energy densities of batteries will be very important in the development of high-energy Li-S batteries.

AUTHOR CONTRIBUTIONS
YL and FR conceived the idea. HW and YL wrote the draft. All authors contributed to the writing, discussion, and revision of the final version of the manuscript.