Efficient Synthesis of Graphene Nanoscrolls for Fabricating Sulfur

Letter www.acsami.org

Efficient Synthesis of Graphene Nanoscrolls for Fabricating SulfurLoaded Cathode and Flexible Hybrid Interlayer toward HighPerformance Li−S Batteries Yi Guo,† Gang Zhao,† Naiteng Wu,† Yun Zhang,*,† Mingwu Xiang,† Bo Wang,‡ Heng Liu,† and Hao Wu*,† †

College of Materials Science and Engineering, Sichuan University, Chengdu 610064, P. R. China MIIT Key Laboratory of Critical Materials Technology for New Energy Conversion and Storage, School of Chemistry and Chemical Engineering, Harbin Institute of Technology, Harbin 15001, P.R. China

‡

S Supporting Information *

ABSTRACT: A modified lyophilization approach is developed and used for highly efficient transformation of 2D graphene oxide sheet into 1D graphene nanoscroll (GNS) with high topological transforming efficiency (∼94%). Because of the unique open tubular structure and large specific surface area (545 m2 g−1), GNS is utilized for the first time as a porous cathode scaffold for encapsulating sulfur with a high loading (81 wt %), and also as a conductive skeleton for assembling MnO2 nanowires into a flexible free-standing hybrid interlayer, both enabling high-rate and long-life Li−S battery.

KEYWORDS: graphene nanoscroll, large surface area, high sulfur loading, free-standing hybrid interlayer, lithium sulfur batteries concentration as low as 0.05 g L−1, whereas the yield of GNS greatly reduced to only 26 and 6% once the feed concentration increased to 0.5 and 1.0 g L−1, respectively. Obviously, there is still a conflict between quality and output for scalable production of GNS, which could be a problem for its subsequent applications. Theoretical calculations have demonstrated the possibility of permitting foreign molecules to be diffused into the interlayer galleries of GNS owing to its open topology.8 This feature would promise extended applications of GNS in various energy storage systems, such as hydrogen storage,9 supercapacitor4 and lithium-ion batteries.10 Currently, lithium−sulfur (Li−S) batteries have drawn growing interests for building next generation high-energy storage devices because of their high theoretical specific capacity (1675 mAh g−1) and energy density (2600 Wh kg−1).11 However, Li−S batteries still undergoes a series of cathode-related challenges that impede their practical application, such as insulating nature of sulfur, serious expansion of sulfur during charge−discharge process, and dissolution of polysulfide intermediates (Li2Sn, 4 ≤ n ≤ 8) in the electrolyte. The most popular methods aiming to address these issues is impregnating sulfur particles into various

G

raphene nanoscroll (GNS), an emerging important member of the graphitic carbon family, is normally shaped by scrolling two-dimensional (2D) graphene sheet into a 1D tubular structure that is distinct from the seamless concentric structure of 1D multiwalled carbon nanotubes (MWCNTs). Specifically, GNS has topological open structure at both ends and interlayer galleries that can be easily adjusted,1 and it inherits many favorable properties of constituent graphene, such as high mechanical strength, outstanding electrical conductivities, and good carrier mobility. Many efforts have been devoted to the theoretical predictions and calculations of GNS,2 but comprehensive experimental investigations on its properties and applications are still restricted from the technical difficulties in reliable synthesis. A few attempts have been made on GNS synthesis, including template methods,3 intercalation/exfoliation of graphite,4 Langmuir−Blodgett technique5 and ultrasonic-assisted assembly6 et.al. Despite these potentials, most of the methods have inevitably experienced harsh reaction condition, high energy consumption, and impurity contamination, thus facing a hurdle for simultaneously realizing high-quality and high-purity GNS in high yield. More recently, Gao’s group reported a simple approach to prepare neat GNS from graphene oxide (GO) sheets using well-controlled lyophilization.7 By optimizing the GO concentration, the transforming efficiency of GO sheets to GNS can be achieved as high as 92%. Nevertheless, such high yield of GNS is obtained merely at the feeding GO © XXXX American Chemical Society

Received: October 21, 2016 Accepted: December 7, 2016 Published: December 7, 2016 A

DOI: 10.1021/acsami.6b13455 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Letter

ACS Applied Materials & Interfaces

Figure 1. Schematic diagram of preparation processes for GNS and S@GNS.

Figure 2. Digital pictures of (a) GNS, (b) the mixture of GNS and sulfur, and (c) S@GNS; FESEM images of (d, g) GNS, (e, h) the mixture of GNS and sulfur, and (f, i, and j) S@GNS with elemental mapping results of (k) sulfur and (l) carbon.

activating agents (e.g., KOH,17 NaOH,18 molten salts,19 etc.) are commonly required to create adequate micropores and/or mesopores in the carbon hosts for encapsulating sulfur species. In view of the intrinsic nature of graphene such as large surface area and high electrical conductivity, GNS is supposed to serve as an alternative carbonaceous cathode scaffold for Li− S batteries. The inherent open topology of GNS not only allows sulfur particles to be easily diffused and impregnated into the interlayer galleries so that any extra templates or chemical

nanostructured, porous carbonaceous matrixes, including amorphous carbons,12 2D graphene sheets13 and 3D graphene foams.14 Nevertheless, it is worth noting that most of the sulfur/carbon cathodes reported so far still fail to deal with the dissolution of polysulfides, especially under conditions with high sulfur loading (≥70 wt %), and in addition, tedious and high cost procedures are often involved in the synthesis of porous carbon materials, in which extra organic/inorganic templates (e.g., SiO2,15 ZnO,16 etc.) or highly corrosive B

DOI: 10.1021/acsami.6b13455 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Letter

ACS Applied Materials & Interfaces

Figure 3. TEM images of (a, c) GNS and (b, d) S@GNS.

Waals interaction of the reduced GO (rGO) sheets to overcome the energy barrier of scrolling, but also keep them still dispersed in water as single layers due to the low reduction extent. The partial reduced GO suspension is then transferred into plastic tubes to continue heating up to 80 °C, before putting them into liquid nitrogen. Theoretical analysis has predicted that the topological transformation from planar graphene to tubular GNS is dominated by the competition between elastic bending energy and free energy. Hence, once the hot rGO suspension suffers the liquid nitrogen instantaneously, a huge temperature difference is provided so as to yield more energy for bending of graphene sheet, resulting in that the rolling level for the hot rGO suspension is much better than that for the room temperature one. After freezing, the completely frozen solid is subjected to lyophilization for dehydration and further reduced by annealing treatment at 1000 °C for 2 h in 5% H2/Ar atmosphere to obtain a 3D macroscopic, cottonlike flexible GNS product (Photograph as displayed in Figure 2a). Following the four-step process that consists of chemical reduction, fast quenching, lyophilization and thermal annealing, the efficiency of the topological transformation from planar to scrolling topology can be achieved as high as 94% by quantitatively assessing the proportion of scrolling graphene (Figure S1a). Moreover, it can still reach about 79% in transforming efficiency even if the feed GO concentration increases to 1.0 g L−1 (Figure S1b), demonstrating the high efficiency in our synthetic approach for mass production of high-quality GNS. Successively, the asprepared GNS is physically mixed with commercial sulfur powder into a gray mixture (Figure 2b), followed by a classical melt-diffusion treatment (155 °C, 15 h) to facilitate the

activations are not needed, but also enables a more efficient utilization of the surface area of graphene for in situ immobilization of sulfur and geometric confinement of polysulfides in tube interior. More importantly, the highly elastic tubular wall of GNS is believed to facilitate radial expansion to adapt the increasing volume by ion insertion, which would promise the function of accommodating the volume variation of enclosed sulfur particles during lithiation process. Here, we reported a modified strategy based on Gao’s method for facile and efficient synthesis of high-quality 1D GNS in high yield from GO sheets through fast quenchingcombined lyophilization. Compared with previous Gao’s method,7 the GNS output in our work can be substantially elevated even if the amount of the feeding GO concentration increases to 0.5 and 1.0 g L−1 while maintaining a high transforming efficiency of 94 and 79%, respectively. The obtained GNS has an ordered 1D CNT-like tubular nanostructure with an excellent electronic conductivity and especially a large specific surface area (SSA) of 545 m2 g−1, which is almost the highest value among the reported GNS materials. Due to the large SSA, the as-obtained GNS can be served as an ideal carrier to incorporate with sulfur for cathode assembly, by which a very high sulfur loading of up to 81 wt % is achieved in the resulted composite cathode (denoted as S@ GNS). The synthetic route is illustrated in Figure 1. In detail, GO sheets are first ultrasonically dispersed in water to get a singlelayer GO suspension, followed by adding hydrazine hydrate for reduction at 60 °C for 20 min. According to Gao’s research, this mild chemical reduction can not only help enhance the vander C

DOI: 10.1021/acsami.6b13455 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Letter

ACS Applied Materials & Interfaces

Figure 4. (a) XRD patterns of GO, GNS, and S@GNS; (b) TGA curves of GNS, S@GNS, and Gs-S; (c) N2 adsorption/desorption isotherms of GNS and S@GNS; (d) pore size distribution curves of GNS and S@GNS.

impregnation of sulfur into the interior of GNS hosts, by which the dark S@GNS composite is finally obtained (Figure 2c). Field-emission SEM (FESEM) images of GNS, the mixture of GNS and sulfur, and S@GNS composite at different magnification are presented in Figure 2d−i. As shown in Figure 2d, g, GNS exhibits the typical scrolling morphology with the diameter of 300−700 nm and the length of exceeding 10 μm. Due to the large aspect ratio, the GNS is a kind of 1D tubular nanostructure that is significantly distinct from the 2D planar topology of raw graphene sheet. The 1D GNS as filaments interlace with each other to form a continuously cross-linked network (Figure 2d), which is favorable to enhance the electronic conductivity and shorten the migration distance of Li+. In the mixture of GNS and sulfur (Figure 2e, h), it is obvious that most of the sulfur particles are separately surrounded by numerous GNS fibers. After thermal treatment at 155 °C, the resulted S@GNS composite appears compacted (Figure 2f), and the nanoscrolls are entangled closely with each other but still have the scrolling morphology (Figure 2i). Notably, sulfur particles disappear after thermal treatment while leaving behind thickened scrolls. This is due to the impregnation of molten sulfur into the internal voids of the GNS driven by a capillary force, thus enlarging the interlayer spacing within the scroll framework. EDS mapping results also reveal the homogeneous distribution of sulfur and carbon elements in S@GNS composite, as shown in Figure 2k, l. For comparison, commercial graphene is also used as sulfur host matrix to prepare a graphene sheet-supported sulfur composite (denoted as Gs-S, see Experimental Section). FESEM analysis (Figure S2a, b) indicates that the commercial graphene have the stacked multilayer structure, and there exists obvious bulk sulfur agglomerates located on the surface of the graphene sheets after sulfur loading (Figure S2c, d).

The microstructure of the GNS and S@GNS are further observed by TEM. As shown in Figure 3 and Figure S3, both of the GNS and S@GNS maintain the scrolling structure, and their average diameters are estimated to be about 500 and 1000 nm, respectively. The increased diameter of the nanoscroll in S@GNS proves that the introduction of sulfur into the interior of GNS can expand the interspaces, coinciding with above SEM observation. Note that there clearly exists wrinkled and overlapped graphene in the nanoscroll, meaning that a multiinterlayer structure is formed by layer-by-layer rolling of flat graphene sheets. The interlayer gallery between the overlapped graphene can provide enough chambers available for sulfur loading and accommodating the large volumetric expansion upon lithiation. In addition, no any aggregation of sulfur is observed from the TEM image of S@GNS, suggesting that the sulfur can be homogeneously dispersed mainly as relatively smaller nanoparticles inside the GNS nanostructure. X-ray diffraction patterns of GO, GNS, and S@GNS are presented in Figure 4a. The original GO shows a sharp peak at 2θ = 11°, whereas the GNS has a characteristic (002) peak at 2θ = 26° corresponding to the d-spacing of 0.34 nm. This indicates that most of the functional groups of GO are eliminated by chemical reduction together with thermal reduction. The XRD pattern of S@GNS proves that the sulfur in the composite exists in the same orthorhombic structure as elemental sulfur (JCPDS no. 08−0247). TGA curves of the GNS and S@GNS are shown in Figure 4b, in which the actual sulfur loading is determined as high as 81 wt %. This sulfur loading is higher than the value in the Gs-S counterpart (76 wt %). N2 adsorption/desorption isotherm analyses are further applied to investigate the pore structures of GNS and S@ GNS.As shown in Figure 4c, GNS displays a type IV isotherm with a high nitrogen uptake, suggesting an interconnected pore system. Surprisingly, without using any extra templates or D

DOI: 10.1021/acsami.6b13455 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Letter

ACS Applied Materials & Interfaces

Figure 5. (a) CV plots of S@GNS cathode at a scanning speed of 0.1 mV s−1 in 1.5−2.8 V vs Li+/Li; (b) initial charge−discharge curves of S@GNS cathode at different rate; (c) rate capability of S@GNS and Gs-S; (d) cycling performances of S@GNS and Gs-S at 1C; (e) long-term cycling performances of S@GNS at 2C.

patterns. And the ID/IG drops to 1.31, meaning that sulfur in the composite decrease the disorder degree of GNS. The XPS spectrum of S@GNS has been presented in Figure S5. In the wide-survey XPS spectrum (Figure S5a), there are three obvious peaks at 165, 285, and 533 eV that are attributed to S 2p, C 1s and O 1s, respectively. Peak at 285 eV of the highresolution C 1s corresponds to C−O/C−S, implying that S is doped into the carbon lattice of GNS. The high-resolution of S 2p has three peaks, consisting of S−S/S−C bonds at 164 and 165 eV, and sulfate species at 168 eV, respectively.21 Figure 5a depicts the first, second, and third cyclic voltammogram (CV) curves of the Li−S cell with the S@ GNS as cathode. In the first cathodic scan, a pair of strong reduction peaks located at 2.38 and 2.03 V are related to the reduction of elemental sulfur (S8) to soluble long-chain lithium polysulfides (Li2Sn, 4 ≤ n ≤ 8) and the further reduction of those polysulfides to insoluble short-chain Li2S2 and eventually to Li2S, respectively.22 In the anodic scan, there is an oxidation peak around 2.42 V related to the reverse reactions in the charging stage. In the second and third anodic scan, the anodic peak slightly shifts to lower potential. The variation in the anodic peak is attributed to the fact that the active sulfur can be rearranged from its original positions to more energetically stable sites. There are no significant changes for both the

chemical activations, the SSA of GNS is calculated as high as 545 m2 g−1 using the Brunauer−Emmett−Teller (BET) method, along with a total pore volume of 0.544 cm3 g−1, both of which are the highest values of GNS ever reported. In addition, the pore size distribution calculated based on the Barrett−Joyner−Halenda (BJH) method, as depicted in Figure 4d, reveals that there is a coexistence of micro- and mesopore structures in GNS with the average pore size of about 3.2 nm. After sulfur loading, the SSA and pore volume of the S@GNS dramatically drop to 15 m2 g−1 and 0.045 cm3 g−1, respectively, indicating that in addition to a part of sulfur spread over the external surface of GNS, most of sulfur are permeably impregnated and entrapped into the open internal voids among the interlayers of GNS, as illustrated in Figure 1. Figure S4 presents the Raman spectrum of pristine GO, as-prepared GNS and S@GNS. Two typical Raman bands of graphene can be observed, a D band at 1342 cm−1 attributed to defects and disordered carbon, a G band at 1582 cm−1 contributing from the vibration of sp2− bonded of carbon atoms.20 While pristine GO have D band and G band at 1352 and 1585 cm−1 respectively. GNS has a higher ID/IG (1.35) than GO (1.26), indicating that the former has a more disorder structure.14 For S@GNS, the peak at 473 cm−1 is the characteristic peak of the symmetric S8 vibrations,20 which is consistent with XRD E

DOI: 10.1021/acsami.6b13455 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Letter

ACS Applied Materials & Interfaces

Figure 6. FESEM images of S@GNS cathode (a) before and (b) after 100 cycles at 1C; (c) typical colors of the cycled S@GNS and Gs-S cathodes after soaking in a mixture of DOL/DME for 24 h; (d) schematic illustration on the advantages in entrapping polysulfides of the S@GNS over the Gs-S.

comfier the wonderful charge-transfer capability of S@GNS compared to the Gs-S. It can be attributed to the large surface area and surface roughness of GNS, increasing the contact surface area and adhesion between sulfur and the host materials, leading to a decrease in the contact resistance.24 SEM analysis reveals that the S@GNS cathode still maintains the characteristic scrolling morphology after cycling, as shown in Figure 6a and b. This indicates the large volume change of sulfur species can be buffered through accommodation by the flexible GNS tubular walls. The long-term cycling performance and the corresponding CE of S@GNS at a high current density of 2C are depicted in Figure 5e. Impressively, the capacity retention can achieve about 85% after 400 cycles, and the CE is always maintained at approximately 100% even under the high current density, indicating the GNS could effectively mitigate the shuttle effects of polysulfides. To view the polysulfide dissolution behavior with the different cathode material, the S@GNS and Gs-S electrode slices are separately removed from the cells after 100 cycles at 1C (charge state), and soaked in a mixture of 1,3-dioxolane/ 1,2-dimethoxyethane (DOL/DME, 1:1, vol) for 24 h. As shown in Figure 6c, the color of the solvent for the Gs-S cathode turns into much yellow, which is caused by dissolving the yellow polysulfide out from the cathode. Comparatively, the solvent for the S@GNS cathode remains a relatively lighter color, indicating that the interior pore structure of GNS could better immobilize sulfur and geometrically restrain the polysulfide within more closed interlayer space over the loosely stacked planar sheets of Gs, as illustrated in Figure 6d. Obviously, the superior electrochemical performances of S@GNS originate from the specific electrode configuration, which is associated with the excellent electrical conductivity of GNS scaffold together with its unique 1D tubular nanostructure, as well as homogeneous distribution of sulfur confined in GNS architecture. These merits facilitate the fast electronic/ionic

anodic and cathodic peaks in continuous three cycles, suggesting good electrochemical reversibility. The rate performance is tested in the voltage range of 1.5 to 2.8 V at discharge−charge rates from 0.05 to 4C (1C = 1675 mA g−1 by sulfur weight). The charge−discharge curves at different rates and the rate capability are shown in Figure 5b, c. These profiles at different rates maintain similar shapes with relatively low over potentials (Figure 5b), while the curves, even at high rates, still exhibit the typical reaction plateaus and deliver high capacities. Compared with Gs-S, the S@GNS cathode shows a superior rate capability (Figure 5c), and its discharge specific capacity can be stabilized at about 1377, 958, 766, 686, 556, and 445 mAh g−1 when cycled at 0.05, 0.1, 0.5, 1, 2, and 3C rates, respectively. Because of the excellent electronic conductivity and 1D ionic migration pathways, the S@GNS could still deliver a specific capacity of 388 mAh g−1 even at a high current rate of 4 C. Figure 5d depicts the cycling performance of S@GNS tested at a high rate of 1 C after being activated at 0.1C for initial three cycles. As shown in Figure 5d, the current rate switches from 0.1 to 1C at the fourth cycle, and the discharge specific capacity of S@GNS is about 723 mAh g−1. After 100 continuous cycles, the discharge specific capacity is still stabilized as high as 744 mAh g−1, while the discharge/ charge Coulombic efficiency (CE) remainedat about 100% during cycling. In contrast, the Gs-S cathode suffers from rapid capacity decay from 579 to 199 mAh g−1 after 100 cycles together with delivering unstable CE. Nyquist impedance plots of S@GNS and Gs-S after the first and 100th cycles are shown in Figure S6, which has two semicircles in the high-to-medium frequency region together with an inclined line at low frequency. The first semicircle reflects the resistance that Li+ diffuse through contacting interface. The second tiny one represents the charge-transfer resistance. The slope line is resulted from Warburg impedance, corresponding to diffusion of lithium ions into the electrode material.23 The results F

DOI: 10.1021/acsami.6b13455 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Letter

ACS Applied Materials & Interfaces

Figure 7. (a) Photo of the free-standing and flexible GNSM interlayer; (b) schematic model of the cell; SEM images of GNSM interlayer (c) before and (d, e) after cycling; (f) EDS mappings of interlayer after cycling; (g) rate capability and (h) long-term cycling performance of the cell with S@ GNS cathode and GNSM interlayer; the inset in h is the typical Nyquist plots of cells with and without GNSM interlayer.

highly conductive carbon. Inspired by the excellent electrical conductivity and unique swelling behavior of GNS, we show here another application of GNS in fabrication of a freestanding hybrid interlayer that is composed of interlaced 1D GNS and MnO2 nanowires (NWs). MnO2 is used to prompt the surface adsorption of polysulfides because it owns a strong electrostatic attraction (S−Mn−O) to sulfur species.30 The well-designed hybrid interlayer not only can realize advantages from the integration of GNS and MnO2 building blocks with enhanced electronic and ionic conductivities, but also manifests strong physical/chemical interactions to control the shuttling of polysulfides and ensure their continuous reutilization. As a result, a remarkably improved electrochemical property can be achieved by assembling S@GNS cathode with hybrid interlayer in Li−S batteries. In the synthetic route (see experimental section in the Supporting Information), ultralong 1D MnO2 NWs (aspect ratio >1000, Figure S7) prepared by a hydrothermal method is first well-mixed with GNS via dispersion in ethanol. Due to the spontaneous swelling behavior of GNS in organic solvents that correlates its characteristic open topology, GNS can be easily dispersed in ethanol so as to ensure a homogeneous mixing with the MnO2 NWs together in liquid phase. After that, the mixed GNS and MnO2 NWs suspension is further assembled into a flexible and free-standing film (denoted as GNSM) by vacuum filtration, as shown in Figure 7a. In the resulted GNSM film, the weight ratio of GNS to MnO2 NWs is 4:1, and the electrical conductivity of GNSM film is measured to be 1.21 S cm−1 using a four-point probe method, almost 2 orders of magnitude greater than 2.01 × 10−2 S cm−1 of pristine MnO2

transport and efficient utilization of sulfur, which enhances the Li+ reaction kinetics and leads to superior rate capability. Besides, the interconnected open porous structure of GNS is also favorable for rapid electrolyte penetration and ion access. The design of such 1D S@GNS composite nanostructure not only effectively buffers the large volume expansion of the enclosed sulfur, but also greatly mitigates the dissolution of polysulfides during cycling, thus improving the cycling stability of Li−S batteries. In addition to employing sulfur/carbon composite cathode, an interlayer material placed between the cathode and the separator can function as an inhibitor to intercept the migrating polysulfides for improving the electrochemical performance of Li−S batteries. Several free-standing carbon interlayers such as carbonized paper25 and nanofiber,26 and acetylene black mesh27 have been reported. Nevertheless, the nonpolar nature of these carbon interlayers produces a relatively weak interaction with polar polysulfide anions, thus giving rise to low efficiency in binding and/or confining these species during long-term cycling. Recently, metal oxides, such as TiO2,28 La2O3,29 and MnO2,30 have been considered as a kind of polar materials to offer relatively strong chemical bonding with polysulfide anions, thus efficiently trapping them within the cathodes. However, use of metal oxides for interlayer still remains a challenge, because the introduced oxides with intrinsic poor conductivity would bring extra electronic and ionic resistances, thus making it difficult to reactivate the trapped polysulfides within the interlayer, and reversely deteriorating the Li−S battery performance. Hence, it seems to be more efficient and practical to make hybrid interlayer by compounding metal oxide with G

DOI: 10.1021/acsami.6b13455 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Letter

ACS Applied Materials & Interfaces film (inset in Figure S7a). This implies that the hybrid GNSM film with enhanced electronic conductivity could also act as an additional collector for sulfur to alleviate the overaggregation of insulated sulfur on the cathode surface. The XPS spectrum of GNSM has been presented in Figure S8. In the wide-survey XPS spectrum (Figure S8a), there are three obvious peaks at 643, 532, and 285 eV that are attributed to Mn 2p, O 1s and C 1s, respectively. The high-resolution of C 1s spectrum has three obvious peaks at 284, 285, and 289 eV, corresponding to C−C/ CC, C−O, and O−CO, respectively.21 The highresolution of Mn 2p spectrum has two peaks, contributing from Mn 2p1/2 at 654 eV and Mn 2p3/2 at 642 eV, respectively. SEM image (Figure 7c) shows the micromorphology of the GNSM film. It can be discerned that the film has an interconnected 1D nanostructure, and relatively thin MnO2 NWs are well-dispersed into the GNS network as uniformly interlaced and intertwined with each other. Such direct attachment of the nanowires to the conductive GNS skeletons would assist in offering fast charge transport channels for continuous reactivation and reutilization of trapped sulfur active materials within the interlayer during the electrochemical process. As illustrated in Figure 7b, the obtained GNSM hybrid film is subjected to punching and serving as a free-standing interlayer that is then placed between the S@GNS cathode and the separator in Li−S cell for electrochemical tests. The rate capability of the cell is depicted in Figure 7f. The cell with the GNSM interlayer shows an amazingly high initial discharge capacity of 1544 mAh g−1 at a current rate of 0.1C, which is about 1.7 times larger than that of the cell without inserting the interlayer (Figure 5b). Even when the current rate increased higher up to 1, 2, 3, and 4C, the cell still delivers a high discharge capacity of 944, 845, 773, and 704 mAh g−1, respectively. After the high-rate charge/discharge at 4C, the discharge specific capacity can be well recovered to 1262 mAh g−1 when scanned again at 0.1C, highlighting the excellent reversible stability. Moreover, during the following cycle period from 0.1 to 4C, the cell is still able to retain excellent rate capability, showing the reversible capacity of 1093, 925, 894, 809, 765, and 694 mAh g−1 at 0.2, 0.5, 1, 2, 3, and 4C, respectively. A long-term cycling stability of the cell is further investigated at the high 4C rate, as shown in Figure 7h. Although there is a gradual decline in capacity when cycled at such a high current rate, a reversible discharge capacity of 545 mAh g−1 is still remained even after 400 deep cycles, corresponding to an ultralow capacity decay of 0.08% per cycle. This excellent rate capability and cycling stability in our work is competitive to most of previously reported interlayerengaged Li−S batteries including carbon fiber, carbon paper, MCNTs, and GO/rGO, as well as various carbon/metal oxides composites (see details in Table S1 for comparison). To further analyze the function of the GNSM interlayer on improving the electrochemical performance, we performed the EIS measurements (inset in Figure 7h). As shown, the cell with inserting the GNSM interlayer shows a relatively smaller semicircle, manifesting its lower resistance compared with the cell without GNSM. This suggests that the insertion of the highly conductive GNSM interlayer is able to further boost the electronic accessibility and the transport kinetics of the sulfur cathode. The EIS spectrum of cell with GNSM after cycling consists of two sequential semicircles at high to medium frequency region and a slope line in the low frequency region, as shown in Figure S9. Compared to the resistance obtained before cycling, it has lower resistance, which proves that the

interfacial wettability and lithium ion diffusion are improved after cycling.23 SEM analysis is further carried out on the cell after 400 cycles (charge state). Figure 7d shows that the GNSM interlayer is uniformly covered onto the S@GNS cathode, whereas the surface of the cycled interlayer is pretty rough and dense, manifesting that there might exists a deposition of sulfur species after cycling. A closer observation on the cycled interlayer toward the cathode side (Figure 7e) clearly shows a substantial amount of sulfur compounds deposited both on the MnO2 NWs (red rectangle section) and GNS (yellow circle section). EDS images in Figure 7e confirm the inhibition action of GNSM interlayer. It shows that sulfur and fluorine elements are detected obviously after cycling. That indicates a significant entrapment of polysulfides during the cycling process within the GNSM interlayer through strong chemical30 and physical interactions26 between them, thereby providing the Li−S cell with improved electrochemical performance. Because the conditions for the GNSM interlayer making are still less optimized, further works to elaborately regulate the component ratio and thickness of the interlayer with optimization, as well as design and fabricate other metal oxides-assembled hybrid interlayer are currently ongoing in our laboratory. In summary, the highly efficient synthesis method of GNS has been reformed by a fast quenching-combined lyophilization approach, and the topological transforming efficiency of GNS can be achieved as high as ∼94% even at a high feeding GO concentration, which holds a great promise for mass production of high-quality GNS to meet different demands of applications. In view of the unique 1D tubular topology and large specific surface area (545 m2 g−1), we have demonstrated for the first time the successful applications of GNS to fabricating the cathode and the interlayer materials in Li−S batteries. When applied for sulfur cathode preparation, the open tubular structure of GNS enables an efficient sulfur encapsulation with a high loading of 81 wt %, and also functions as reservoirs to geometrically confine the volume expansion for sulfur species and restrain the migration of polysulfides. Moreover, excellent electrical conductivity and unique swelling behavior of GNS make it very effective in making a flexible free-standing Li−S battery interlayer when assembled with MnO2 NWs. The insertion of the constructed hybrid interlayer between the sulfur cathode and the separator is able to reduce the overall electrode resistance with faster charge transport channels, and also offer strong physical/chemical interactions for efficiently mitigating the shuttling of polysulfides, ensuring continuous reactivation and reutilization of the trapped sulfur active materials. The work described here shows a potential of GNS as a promising functional carbon material to realize high sulfur loading and electrochemically stable electrode configuration aimed toward developing high-performance Li−S batteries.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.6b13455. Experimental details and additional figures (SEM, TEM, RM, XPS, EIS) (PDF)

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. H

DOI: 10.1021/acsami.6b13455 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Letter

ACS Applied Materials & Interfaces ORCID

(15) Chen, S.; Huang, X.; Sun, B.; Zhang, J.; Liu, H.; Wang, G. MultiShelled Hollow Carbon Nanospheres for Lithium−Sulfur Batteries with Superior Performances. J. Mater. Chem. A 2014, 2, 16199−16207. (16) Strubel, P.; Thieme, S.; Biemelt, T.; Helmer, A.; Oschatz, M.; Brückner, J.; Althues, H.; Kaskel, S. ZnO Hard Templating for Synthesis of Hierarchical Porous Carbons with Tailored Porosity and High Performance in Lithium-Sulfur Battery. Adv. Funct. Mater. 2015, 25, 287−297. (17) Wei, S.; Zhang, H.; Huang, Y.; Wang, W.; Xia, Y.; Yu, Z. Pig Bone Derived Hierarchical Porous Carbon and Its Enhanced Cycling Performance of Lithium−Sulfur Batteries. Energy Environ. Sci. 2011, 4, 736−740. (18) Guo, J.; Xu, Y.; Wang, C. Sulfur-Impregnated Disordered Carbon Nanotubes Cathode for Lithium-Sulfur Batteries. Nano Lett. 2011, 11, 4288−4294. (19) Hou, J.; Cao, C.; Idrees, F.; Ma, X. Hierarchical Porous Nitrogen-Doped Carbon Nanosheets Derived from Silk for UltrahighCapacity Battery Anodes and Supercapacitors. ACS Nano 2015, 9, 2556−2564. (20) Xu, H.; Deng, Y.; Shi, Z.; Qian, Y.; Meng, Y.; Chen, G. Graphene-Encapsulated Sulfur (GES) Composites With A Core−Shell Structure As Superior Cathode Materials For Lithium−Sulfur Batteries. J. Mater. Chem. A 2013, 1, 15142−15149. (21) Yang, J.; Chen, F.; Li, C.; Bai, T.; Long, B.; Zhou, X. Y. A Freestanding Sulfur-doped Microporous Carbon Interlayer Derived From Luffa Sponge for High Performance Lithium−Sulfur Batteries. J. Mater. Chem. A 2016, 4, 14324−14333. (22) Peng, H.; Huang, J.; Zhao, M.; Zhang, Q.; Cheng, X.; Liu, X.; Qian, W.-Z.; Wei, F. Nanoarchitectured Graphene/CNT@Porous Carbon with Extraordinary Electrical Conductivity and Interconnected Micro/Mesopores for Lithium-Sulfur Batteries. Adv. Funct. Mater. 2014, 24, 2772−2781. (23) Zhang, K.; Qin, F. R.; Fang, J.; Li, Q.; Jia, M.; Lai, Y. Q.; Zhang, Z. A.; Li, J. Nickel Foam as Interlayer to Improve the Performance of Lithium−Sulfur Battery. J. Solid State Electrochem. 2014, 18, 1025− 1029. (24) Zhou, G.; Pei, S.; Li, L.; Wang, D. W.; Wang, S.; Huang, K.; Yin, L.-C.; Li, F.; Cheng, H.-M. A Graphene-Pure-Sulfur Sandwich Structure for Ultrafast, Long-Life Lithium-Sulfur Batteries. Adv. Mater. 2014, 26, 625−631. (25) Su, Y.; Manthiram, A. Lithium-Sulphur Batteries with a Microporous Carbon Paper as a Bifunctional Interlayer. Nat. Commun. 2012, 3, 1166. (26) Chung, S.; Manthiram, A. Carbonized Eggshell Membrane as a Natural Polysulfide Reservoir for Highly Reversible Li-S Batteries. Adv. Mater. 2014, 26, 1360−1365. (27) Jeong, T. G.; Moon, Y. H.; Chun, H.; Kim, H.; Cho, B.; Kim, Y. Free Standing Acetylene Black Mesh to Capture Dissolved Polysulfide in Lithium Sulfur Batteries. Chem. Commun. 2013, 49, 11107−11109. (28) Zhang, Z.; Li, Q.; Jiang, S.; Zhang, K.; Lai, Y.; Li, J. Sulfur Encapsulated In a TiO2-Anchored Hollow Carbon Nanofiber Hybrid Nanostructure for Lithium-Sulfur Batteries. Chem. - Eur. J. 2015, 21, 1343−1349. (29) Sun, F.; Wang, J.; Long, D.; Qiao, W.; Ling, L.; Lv, C.; Cai, R. A High-Rate Lithium−Sulfur Battery Assisted by Nitrogen-Enriched Mesoporous Carbons Decorated with Ultrafine La2O3 Nanoparticles. J. Mater. Chem. A 2013, 1, 13283−13289. (30) Liang, X.; Hart, C.; Pang, Q.; Garsuch, A.; Weiss, T.; Nazar, L. F. A Highly Efficient Polysulfide Mediator for Lithium-Sulfur Batteries. Nat. Commun. 2015, 6, 5682−5690.

Yun Zhang: 0000-0001-7505-1097 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS

■

REFERENCES

The authors acknowledge the financial supports from National Basic Research Program of China (973 Program 2013CB934700), the National Natural Science Foundation of China (51502180), Foundation for the Author of an Excellent National Doctoral Dissertation of China (FANEDD201435), and the Fundamental Research Funds for the Central Universities (2016SCU04A18).

(1) Braga, S. F.; Coluci, V. R.; Legoas, S. B.; Giro, R.; Galvao, D. S.; Baughman, R. H. Structure and Dynamics of Carbon Nanoscrolls. Nano Lett. 2004, 4, 881−884. (2) Chen, Y.; Lu, J.; Gao, Z. Structural and Electronic Study of Nanoscrolls Rolled up by a Single Graphene Sheet. J. Phys. Chem. C 2007, 111, 1625−1630. (3) Kim, Y.; Min, D. Preparation of Scrolled Graphene Oxides with Multi-Walled Carbon Nanotube Templates. Carbon 2010, 48, 4283− 4288. (4) Zeng, F.; Kuang, Y.; Wang, Y.; Huang, Z.; Fu, C.; Zhou, H. Facile Preparation of High-Quality Graphene Scrolls From Graphite Oxide by a Microexplosion Method. Adv. Mater. 2011, 23, 4929−4932. (5) Gao, Y.; Chen, X.; Xu, H.; Zou, Y.; Gu, R.; Xu, M.; Jen, A. K.-Y.; Chen, H. Highly-efficient Fabrication of Nanoscrolls from Functionalized Graphene Oxide by Langmuir−Blodgett Method. Carbon 2010, 48, 4475−4482. (6) Quintana, M.; Grzelczak, M.; Spyrou, K.; Calvaresi, M.; Bals, S.; Kooi, B.; Van Tendeloo, G.; Rudolf, P.; Zerbetto, F.; Prato, M. A Simple Road for the Transformation of Few-Layer Graphene into MWNTs. J. Am. Chem. Soc. 2012, 134, 13310−13315. (7) Xu, Z.; Zheng, B.; Chen, J.; Gao, C. Highly Efficient Synthesis of Neat Graphene Nanoscrolls from Graphene Oxide by Well-Controlled Lyophilization. Chem. Mater. 2014, 26, 6811−6818. (8) Jin, Y.; Xue, Q.; Zhu, L.; Li, X.; Pan, X.; Zhang, J.; Xing, W.; Wu, T.; Liu, Z. Self-Assembly of Hydrofluorinated Janus Graphene Monolayer: A Versatile Route for Designing Novel Janus Nanoscrolls. Sci. Rep. 2016, 6, 26914−26922. (9) Mpourmpakis, G.; Tylianakis, E.; Froudakis, G. E. Carbon Nanoscrolls: A Promising Material for Hydrogen Storage. Nano Lett. 2007, 7, 1893−1897. (10) Zhang, Y.; Chen, P. H.; Gao, X.; Wang, B.; Liu, H.; Wu, H.; Liu, H. K.; Dou, S. X. Nitrogen-Doped Graphene Ribbon Assembled CoreSheath MnO@Graphene Scrolls as Hierachically Ordered 3D Porous Electrodes for Fast and Durable Lithium Storage. Adv. Funct. Mater. 2016, 26, 7754−7765. (11) Wang, L.; Liu, J.; Haller, S.; Wang, Y.; Xia, Y. A Scalable Hybrid Separator for A High Performance Lithium-Sulfur Battery. Chem. Commun. 2015, 51, 6996−6999. (12) Zheng, G.; Yang, Y.; Cha, J. J.; Hong, S. S.; Cui, Y. Hollow Carbon Nanofiber-Encapsulated Sulfur Cathodes for High Specific Capacity Rechargeable Lithium Batteries. Nano Lett. 2011, 11, 4462− 4467. (13) You, Y.; Zeng, W.; Yin, Y.; Zhang, J.; Yang, C.; Zhu, Y.; Guo, Y.G. Hierarchically Micro/Mesoporous Activated Graphene With a Large Surface Area for High Sulfur Loading In Li−S Batteries. J. Mater. Chem. A 2015, 3, 4799−4802. (14) He, J.; Chen, Y.; Li, P.; Fu, F.; Wang, Z.; Zhang, W. Threedimensional CNT/Graphene−Sulfur Hybrid Sponges with High Sulfur Loading as Superior-Capacity Cathodes For Lithium−Sulfur Batteries. J. Mater. Chem. A 2015, 3, 18605−18610. I

DOI: 10.1021/acsami.6b13455 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX