C conductive nanofibers as interlayer for lithium-sulfur batteries with ultra long cycle life and high-rate capability

C conductive nanofibers as interlayer for lithium-sulfur batteries with ultra long cycle life and high-rate capability

Chemical Engineering Journal 355 (2019) 390–398 Contents lists available at ScienceDirect Chemical Engineering Journal journal homepage: www.elsevie...

3MB Sizes 0 Downloads 3 Views

Chemical Engineering Journal 355 (2019) 390–398

Contents lists available at ScienceDirect

Chemical Engineering Journal journal homepage: www.elsevier.com/locate/cej

Electrospun Ti4O7/C conductive nanofibers as interlayer for lithium-sulfur batteries with ultra long cycle life and high-rate capability Ya Guo1, Jing Li1, Rosaiah Pitcheri, Jinghui Zhu, Peng Wen, Yejun Qiu

T



Shenzhen Engineering Lab of Flexible Transparent Conductive Films, Department of Materials Science and Engineering, Harbin Institute of Technology, Shenzhen 518055, China

H I GH L IG H T S :

were obtained easily-via elec• TCNFs trospinning followed by in-situ carbonreduction.

interlayer has both physical • The blocking and chemical adsorption ef-

G R A P H I C A L A B S T R A C T

The Ti4O7/carbon nanofibers (TCNFs) were successfully prepared by one step electrospinning and subsequent carbothermal reduction. When served as an interlayer, the CMK3-S cathode with TCNFs showed excellent cycling and high-rate performances.

fects for LiPSs.

O chemical binding of LiPSs by Ti• TiS bonds greatly reduces LiPSs shuttle. 3D TCNFs make for lower inter• The facial resistance and faster redox re4

7

action.

battery exhibits excellent electro• Li-S chemical performances, even at high rates.

A R T I C LE I N FO

A B S T R A C T

Keywords: Carbon nanofibers Ti4O7 Interlayer Electrochemical performance Lithium-sulfur battery

In this work, a free-standing and flexible Ti4O7/C nanofibers (TCNFs) interlayer is prepared via one-step electrospinning. The TCNFs interlayer acts as both physical barrier and chemical adsorption agent for lithium polysulfides (LiPSs), in which the 3D CNFs network with large volume and high conductivity contributes to LiPSs conversion and electron transfer, while the ‘sulphiphilic’ Ti4O7 with strong chemical bonding to the soluble longchain LiPSs contributes to less LiPSs dissolution and higher coulombic efficiency. In a lithium-sulfur (Li-S) cell, CMK3/S cathode with TCNFs interlayer exhibits greatly enhanced electrochemical capability than CMK3/S with CNFs interlayer and without interlayer, and a capacity of 560 mAh g−1 after 1000 cycles is obtained under a high current density of 1 C. Even at 3 C, the capacity decay is merely 0.030% per cycle over 2500 cycles. Notably, the simple, affordable and scalable fabrication method to form a controllable free-standing interlayer that brings a novel horizon on material structure design and performance improvement in practical application of Li-S batteries.

1. Introduction The rapid developments of a sustainable society have put forward

higher requirements for the electrochemical energy storage devices, but the traditional commercial batteries with limited energy densities cannot meet the demands, so it is urgent to find the next generation of



Corresponding author. E-mail address: [email protected] (Y. Qiu). 1 These authors contributed equally to this work. https://doi.org/10.1016/j.cej.2018.08.143 Received 15 June 2018; Received in revised form 16 August 2018; Accepted 21 August 2018 Available online 23 August 2018 1385-8947/ © 2018 Elsevier B.V. All rights reserved.

Chemical Engineering Journal 355 (2019) 390–398

Y. Guo et al.

current density (3 C rate) over 2500 cycles.

secondary batteries with high energy density and long cycling life [1,2]. Lithium-sulfur (Li-S) batteries are considered as one of the most promising alternatives to the commercial lithium-ion batteries (LiBs) because of high theoretical specific capacity (S cathode 1675 mAh g−1 vs LiCoO2 272 mAh g−1), high theoretical specific energy (2600 Wh kg−1), abundant natural resources, low-cost and eco-friendly nature [3,4]. However, lots of intrinsic and fatal hindrances such as low sulfur utilization and poor long-term cycling still limit the practical application of Li-S battery. Moreover, the low conductivity of sulfur/its discharge products (Li2S and Li2S2), large volume change during cycling, and the dissolution in organic electrolyte of polysulfide intermediates (Li2Sn, with n = 4–8) which cause the “shuttle effect”, severely limiting the performance of Li-S battery [5,6]. In recent years, various carbons, conductive polymers, and metal oxides have been adopted to composite with sulfur for enhancing conductivity, suppressing volume expansion and trapping LiPSs [7–9]. Nevertheless, in most cases, the non-polar carbons with weak physical adsorption of LiPSs are not efficient enough to trap LiPSs and polymers wrapping cannot avoid the loss of sulfur on the surface formed during cycling, thus leading to capacity fading and poor coulombic efficiency. As for metal oxides concern, they occupy lots of weight which further leads to lower energy density. Therefore, it is necessary to develop novel and high competent methodologies to restrain LiPSs shuttle. Of late, an interlayer is introduced between cathode and separator to block LiPSs. The employ of carbon interlayers such as carbon paper, carbon nanotube, carbon nanofiber, graphene and so on, can enhance the conductivity and suppress the insoluble LiPS migration to Li anode thus improving the cycling performance of batteries [10–13]. However, the physical barrier with weak LiPSs adsorption ability cannot prevent the migration of dissolved LiPSs. With this concern, carbon materials are composited with the metal oxides decoration (TiO2, MoO3, and V2O5 etc.) or elemental doping (nitrogen, sulfur, etc.) with stronger physical adsorption or chemical binding to achieve great effects [14–18]. Among these, TiO2 with strong adsorption of LiPSs is most commonly used material due to its strong adsorption ability and relatively low cost [19]. However, its low conductivity restricts the electrochemical reactions during redox conversions, which leads to the aggregation of short-chain polysulfides (Li2S2/Li2S) on the surface, and prevents further redox [20]. Recently, a twinborn TiO2-TiN heterostructure loaded onto graphene is served as an interlayer to block LiPSs and exhibited good performance, in which TiN was introduced to improve the conductivity of TiO2. Nevertheless, the conductivity is still unsatisfied, so the relative high-rate performances may not be enough outstanding [20]. Moreover, the high-rate performances are concerned as a key factor of the advanced electrical devices/energy storage systems. Most of the interlayers can alleviate the shuttle effect, but compromises with rate performance, mainly because of their non-ideal electrical conductivity and sulfiphilicity [21,22]. Therefore, a highlyefficient interlayer with good conductivity and strong polysulfides adsorption is still in desperate necessity in the realistic application of Li-S batteries. Herein, we employed a highly conductive and strongly LiPSs-trapping bifunctional titanic oxide-(Ti4O7) to achieve high performances. A simple and facile method was developed to fabricate the freestanding and flexible Ti4O7/carbon nanofibers (TCNFs) interlayer via one-step electrospun polyacrylonitrile (PAN) and TiO2 mixture followed by insitu carbothermal reduction of TiO2 by PAN. The newly designed Ti4O7 possesses high electric conductivity (3–30 S cm−1) and good affinity with LiPSs due to the polar O-Ti-O units, and high dispersion in carbon nanofibers without further gathering during use. In our free-standing interlayer, the carbon nanofiber matrix with Ti4O7 can physically block the short chain Li2S and Li2S2, while the polar Ti4O7 particles can chemically bind with the dissolved LiPSs thus restricting the “shuttle effects” comprehensively. The large-scale and cost-effective TCNFs interlayer with CMK3/S cathode exhibited a high coulombic efficiency, with a particularly low capacity fading of 0.030% per cycle even at high

2. Experimental section 2.1. Materials Titanium (IV) Oxide (TiO2) (∼25 nm particle size, 99.8% metals basis, Aldrich), polyacrylonitrile (PAN) (average Mw = 150000, Aldrich), polyvinyl pyrrolidone (PVP) (average Mw = 1300000, Aldrich), N,N-dimethylformamide (DMF) (anhydrous, 99.8%, Aldrich), sublimed sulfur powder (99.9%, Aldrich), and CMK3 (AR, Nanking nanometer). All chemicals were directly employed without further treatment. 2.2. Synthesis of TCNFs interlayers TCNFs were synthesized by the traditional electrospinning method. Initially, 0.6 g of TiO2 nanopowder was dispersed in 5 mL DMF and ultrasonicated for 20 min to form a homogeneous solution. Then, 0.3 g of PAN and 0.2 g of PVP were added to the resultant solution and stirred overnight to obtain the spinnable precursor solution. The process was carried out for 10 h at room temperature by a conventional electrospinning system, during which the parameters such as applied voltage, needle-to-collector distance, and flow rate are 15 kV, 15 cm, and 0.5 mL h−1, respectively. After electrospinning, NFs were collected and subsequently stabilized in a tube furnace at 280 °C for 2 h in air, and followed by the further calcination in nitrogen (N2) atmosphere at 1000 °C for 2 h (heating rate, 5° min−1). The TCNFs were obtained during carbothermal reduction process. Finally, the TCNFs were punched into a membrane with a diameter of 16 mm for utilization. The weight and areal loading of each interlayer (diameter of 16 mm) was about 0.55 mg and 0.272 mg cm−2, respectively. 2.3. Characterization The morphology and structure of the samples were observed by transmission electron microscope (TEM, PEI TECNAIG2 F30) and scanning electron microscopy (SEM: Hitachi S-4700). In order to determine the elemental composition and distribution of the samples, energy dispersive X-ray spectroscopy (EDX) and X-ray diffraction (XRD) (Diffractometer D500/501, Siemens) were employed. The electrical conductivity was estimated by a four-point probe resistivity tester (RT 300). The thermogravimetric analysis (TGA) was carried out by heating from 25 to 800 °C at a rate of 10° min−1 in air atmosphere. Raman spectroscopy studies were carried out by Raman (HORIBA Labram HR Evolution). The specific surface area and pore size distribution of the samples were analyzed by a Brunauer-Emmett-Teller surface area analyzer (BET, Micromeritics ASAP2020). X-ray photoelectron spectra (XPS) were recorded on a VG ESCALAB220I-XL spectrometer with a monochromatic Al Kα X-ray source to analyze the surface species and their chemical states. 2.4. Li-S battery assembly and electrochemical measurements The cathode, interlayer, separator, and anode of the battery were assembled into 2032 type coin cells in a glove box. The cathode was prepared with CMK3-S, conductive acetylene black and polyvinylidene fluoride (PVDF) at a ratio of 8:1:1, respectively. The anode was a Li metal wafer with a diameter of 12 mm. The electrolyte was 1 M LiTFSI salt in a solvent mixture of 1,2-dimethoxyethane (DME) and 1,3-dioxolane (DOL) (1:1 v/v) with 1% LiNO3 additive. The cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) studies were carried out using an electrochemical workstation (CHI650E). The galvanostatic charge-discharge measurements and rate-performance tests were carried out at various current densities within the cutoff voltage of 1.5–2.8 V using a Land CT-2001A battery tester. 391

Chemical Engineering Journal 355 (2019) 390–398

Y. Guo et al.

Fig. 1. (a) Fabrication processes of TCNFs, and (b) mechanism of physical barrier and chemical adsorption for LiPSs by the interlayer.

3. Result and discussion

TEM) image (Fig. 2d) reveals the lattice fringes with the fringe width of 0.242 nm and 0.359 nm, which correspond to the (0 2 0) and (1 1 4) planes of magnéli Ti4O7, respectively [23,24]. The TEM elemental mapping of carbon, oxygen, and titanium (Fig. S1) demonstrate the consistent distribution of Ti4O7 throughout the CNFs. To further verify the structure of TiOx, XRD and Raman were carried out as shown in Fig. 3a and b. From Fig. 3a, a broad peak at approximately 25° and low intense peak at 44° indicates the existence of CNF with non-graphic carbon [25]. Besides, the peaks of the TCNFs are entirely different from the spectrum of anatase TiO2 (JCPDS No. 211272) but in accordance with the standard spectrum of magnéli Ti4O7 (JCPDS No. 71-1428) [14,26,27]. Meanwhile, Raman spectra are depicted in Fig. 3b, in which two typical peaks at approximately 1346 and 1575 cm−1, corresponding to the D and G bands of carbon species, respectively. Notably, the peaks of TCNFs at 142, 258, 416 and 600 cm−1 are assigned to the typical characteristic peaks of Ti4O7, which are quite different from TiO2 with the characteristic peaks at 144, 396, 518 and 640 cm−1 [24,26]. Therefore, the XRD and Raman results are consistent with the TEM results, indicating the formation of Ti4O7 by the reduction of TiO2 at a high temperature to construct of Ti4O7 nanoparticles in/on CNFs. The pore size distribution and specific surface area of TCNFs interlayer are presented in Fig. 3c, demonstrating that the TCNFs interlayer mainly contains mesopores with pore size between 2 and 16 nm, and the inset of Fig. 3c shows the adsorption and desorption curves of the TCNFs interlayer and corresponding BET specific surface area is 40 m2 g−1. To be mentioned, the CNFs with mesopores will benefit polysulfides storage and electron transfer, thus enhancing electrochemical conversion of polysulfides. To estimate the content of Ti4O7 in the TCNFs interlayer, TGA measurements were conducted under oxygen atmosphere. In Fig. 3d, the mass of pure CNFs continues declining slowly from 80 to 500 °C, and drops sharply to zero during 500–650 °C due to the decomposition of CNFs. But, weight gain of TCNFs between 200 °C and 500 °C is ascribed to the oxidation of Ti4O7 to form TiO2 [7], and the remaining 54 wt% of mass in the curve is regarded as TiO2. According to the conservation of Ti element, the content of Ti4O7 in the TCNFs is about 51 wt%. The high percentage of

Fig. 1 illustrates the delicate fabrication processes of the Ti4O7/C NFs and the mechanism of physical barrier and chemical adsorption for LiPSs by the TCNF interlayer. The production steps (Fig. 1a) are pretty simple: first, TiO2 and PAN mixture was electrospun into TiO2/PAN NFs (white), followed by pre-oxidation (brown) and carbonization (black). During carbonization, PAN was converted into carbon and some carbon species would react with TiO2 thus forming Ti4O7. As a result, a thin conductive membrane was generated, maintaining its original thin membrane morphology without fracture and fold. Thus the freestanding TCNFs interlayer with good conductivity, processability and flexibility is suitable for practical utilization. Fig. 1b explains the role of TCNFs interlayer as an effective medium to alleviate LiPSs diffusion to the Li metal anode during cycling. Particularly, the interlayer can not only physically confine the insoluble Li2S and Li2S2 to just the cathode side, but also chemically bonding with the dissolved LiPSs due to the polar Ti4O7. During the discharge process, the soluble long chain LiPSs were anchored by Ti4O7 particles to form Ti-S bonds, and reversely converted to ring sulfur due to the high conductivity of TCNFs (∼3 S cm−1 vs pure CNFs ∼6 S cm−1, measured by four-point probe). The two synergetic effects of the TCNFs as a functional interlayer effectively contributed to the improved electrochemical performances of Li-S batteries. The structure and morphology of the CNF and TCNFs interlayers were analyzed by SEM and TEM. The smooth CNFs matrix with the diameter of 100–250 nm was closely stacked up to form a porous network (Fig. 2a), which acted as a buffer layer for the volume expansion of the sulfur cathode during discharge process and a barrier layer for blocking the insoluble Li2S2 and Li2S. The composite fibers with nanosized Ti4O7 particles decoration are described in Fig. 2b and c, in which the primary CNFs matrix structure was still remained, and the additional Ti4O7 nanoparticles were evenly distributed throughout the CNFs, forming a uniform metallic oxide network for LiPSs adsorption with chemical interaction. The insert of Fig. 2b demonstrates the flexible interlayer with a thickness of 50 μm. High-resolution TEM (HR392

Chemical Engineering Journal 355 (2019) 390–398

Y. Guo et al.

Fig. 2. SEM images of CNFs (a) and TCNFs (b), TEM image of TCNFs (c) and HR- TEM of TCNFs (d).

Fig. 3. (a) XRD patterns of CNFs, TiO2 and TCNFs; (b) Raman spectra of CNFs, TiO2 and TCNFs; (c) pore size distribution and N2 adsorption–desorption isotherms of TCNFs; (d) TGA curve of TCNFs. 393

Chemical Engineering Journal 355 (2019) 390–398

Y. Guo et al.

Fig. 4. (a) CV profiles of the Li-S battery with the TCNFs interlayer; (b) different CV performance of CMK3-S, CMK3-S/CNFs and CMK3-S/TCNFs; (c) cycling stability of CMK3-S, CMK3-S/CNFs and CMK3-S/TCNFs interlayer at 0.2 C for 100 cycles; (d) Nyquist plots of different Li-S battery.

facile electrochemical redox reaction sites for CMK3/S cathode but also greatly inhibits the LiPSs shuttle to Li anode, contributing to enhanced comprehensive electrochemical performance. But, for CMK3/S without interlayer or with CNFs interlayer, the LiPSs diffused across the separator while the Li2S2 and Li2S deposited on the cathode or separator, leading to an increased polarization and reduced Li2S2 and/or Li2S conversion efficiency. Fig. 4c shows the cycling performance of three kinds of Li-S cells (bare CMK3/S, with CNFs and TCNFs interlayers) measured at a relatively low current density of 0.2 C. (1 C = 1675 mA g−1). Generally, the shuttle effect is more obvious at low current density, because there is sufficient time for LiPSs to dissolve in the electrolyte and migrate to Li anode. Hence, low current density (0.2 C) could be suitable to evaluate the LiPSs inhibiting effects of the interlayers. The CMK3/S cathode without interlayer, with CNFs and TCNFs interlayer displays an initial discharge capacity of 1036, 1140 and 1304 mAh g−1 with sulfur utilization of 62%, 68% and 78%, respectively. After 100 cycles, the corresponding discharge capacities are 415, 454 and 945 mAh g−1, respectively. Obviously, TCNFs shows the discharge capacity more than twice that of the other two. The excellent electrochemical stability reflects from the TCNFs interlayer, in which the 3D CNFs conductive network provides speedily electron transfer paths and Ti4O7 produce strong adsorption for the LiPSs. Electrochemical impedance spectroscopy (EIS) of the cells (bare CMK3/S, with CNF and TCNF interlayers) is shown in Fig. 4d. The EIS consists of a semicircle at high frequencies, which represents the charge-transfer resistance (Rct) and an inclined line at the low frequencies [32]. The intersection of the point on the real axis of the semicircle gives the Ohmic resistance between electrode and electrolyte. The Ohmic resistances of the three electrodes displayed almost similar values of 9, 3 and 5 Ω, respectively. On the other hand, the electrode with TCNF interlayers exhibit lower Rct of 40 Ω, whereas the bare CMK3/S, CMK3/S with CNF showed the higher Rct of 138 and 100 Ω, respectively, which

Ti4O7 provides large chemical adsorption and electrochemical conversion sites for LiPSs that contributes to high-capacity retention. To evaluate the polysulfide physical blocking and chemical adsorption efficiency of the TCNFs interlayer, the coin cells were assembled using a CMK3/S composite cathode material with the sulfur content of 60 wt%, its morphology and TGA curves are shown in Fig. S2. It is worth mentioning that the CMK3 with a limited porosity (1.2–1.5 cm3/g) can merely reach a maximum theoretical sulfur loading of 63%. Fig. 4a displays the CV profiles (first four cycles) of CMK3/S cathode with the TCNFs interlayer. A broad and weak cathodic peak is observed at 1.76 V in the first cycle, which further disappears in the following cycles, owing to the irreversible reduction of LiNO3 [28]. And there are no obvious changes occurred in either of the oxidation or reduction peaks during the subsequent three cycles, indicating a high electrochemical stability [19]. In contrast, the CV and discharge-charge profiles of CMK3/S without interlayer or with CNFs interlayer are presented in Fig. 4b and Fig. S3. Interestingly, CMK3/S with the TCNFs interlayer shows higher reduction peaks and a lower oxidation peak when compared to other two cells (with CNFs interlayer and without interlayer), showing a polarization voltage (ΔE) of 0.15 V at 1 C, which is smaller than that of other two cells. The minimum ΔE suggests that the existence of the TCNFs interlayer chemically influences the electrochemical process because of its high electrical conductivity and strong adsorption for LiPSs [23,29]. Besides, in the anodic scan of CV, the broad oxidation peak at about 2.4 V is attributed to the conversion of short-chain to long-chain LiPSs and S8. Nevertheless, the sharp increased oxidation peak in CMK3-S/ TCNFs at the potential of 2.34 V should be associated with the transition of low-concentration Li2S2 and/or Li2S to high-order polysulfides [30,31]. It means that the TCNFs interlayer can enhance the conversion of short chain lithium sulfides to high-order polysulfides due to polar Ti4O7. Emphatically, the highly conductive and adsorbent interlayer with Ti4O7/CNFs not only provides 394

Chemical Engineering Journal 355 (2019) 390–398

Y. Guo et al.

Fig. 5. (a) Cycling performance of different interlayers at 1 C for 1000 long cycles; (b) the ultralong cycling stability of CMK3-S/TCNFs at high current density 3 C; (c) Rate performance of CMK3-S, CMK3-S/CNFs and CMK3-S/TCNFs interlayer at different current densities from 0.2 C to 5 C; (d) discharge and charger profiles of CMK3-S/TCNFs at different current density; (e) Cycling performance of TCNFs interlayer at 3 C for about 1500 cycles with different sulfur loadings of 1.2 mg cm−2, 1.5 mg cm−2, and 2.5 mg cm−2.

electrons. It is worth mentioning that CNFs with TiO2 exhibits a higher Rct than CNFs due to the low conductive TiO2 phase. This proves that Ti4O7 is superior to TiO2 in improving the whole conductivity of the coin cells, thus effectively enhancing the electrochemical performance

reflects that the 3D CNFs network can enhance the electrical conductivity between the cathode and separator. What’s more, CNFs with Ti4O7 can increase the conductivity more due to the additional high conductive Ti4O7 phase which provides an express pathway for the 395

Chemical Engineering Journal 355 (2019) 390–398

Y. Guo et al.

Fig. 6. Adsorption mechanism: (a) optical photos of lithium polysulfides adsorbed on TCNFs, (1) Li2S8/DOL/DME solution, (2) the mixture of CNFs and Li2S8/DOL/ DME, and (3) the mixture of TCNFs and Li2S8/DOL/DME; (b) TEM images of the cycled TCNFs interlayer and corresponding elemental mapping of (c) C, (d) O, (e) Ti, and (f) S; XPS spectra of Ti2p (g) and S2p (h) of the cycled TCNFs interlayer.

(Fig. S4). As we know, the dissolution and diffusion of polysulfides take time, however, the electrochemical process is fairly fast at a high-current density, hence, the TCNFs acts as a fast electron transfer for the rapid redox reaction rather than LiPSs trapping agents. These results indicate the better performance of the cell with TCNF interlayer even at high-current densities. The rate capability of the cells (bare CMK3/S, with CNFs and TCNF interlayers) was estimated (with sulfur loading of 1.2 mg cm−2) (Fig. 5c) by varying current densities from 0.2 to 5 C. It is proved that the cells with the TCNF interlayer have better rate capabilities than the bare CMK3/S and CNFs interlayer. The discharge capacity with the TCNFs interlayer based electrode delivered an initial discharge capacity of 1320 mAh g−1 at 0.2 C, subsequently stabilized at 920 mAh g−1 at 0.5 C, 790 mAh g−1 at 1 C, 680 mAh g−1 at 3 C, and 586 mAh g−1 at 5 C rate, followed by a reversible capacity of 978 mAh g−1 at 0.2 C. However, the electrodes with CNFs interlayer and without interlayer exhibited lower discharge capacities under the same conditions. The excellent rate capabilities of the electrode with TCNFs interlayer even at high current densities are mainly ascribed to the enhanced conductivity and effective capture of polysulfides.

of the Li-S batteries. As high-rate cycling stability is the key factor for the practical applications of the battery, charge/discharge experiments are carried out at high-current densities for various cycles. Fig. 5a shows the ultra-long cycling performance of the bare CMK3/S, CMK3/S with CNFs and TCNFs interlayer at 1 C, in which the discharge capacities after 1000 cycles with the sulfur loading of 1.2 mg cm2 are 187, 207 and 560 mAh g−1, respectively. Consistent with the previous result at 0.2 C, the cell of CMK3/S with the TCNFs interlayer delivers the highest capacity with best cycling stability, which accounts for the TCNFs interlayer’s physical barrier and chemical bonding effects for LiPSs (Li2Sx, 4 ≤ x < 8). Moreover, it exhibited an initial discharging capacity of 793 mAh g−1 and demonstrated good cycling performance even at 3 C rate over 2500 cycles (Fig. 5b) with the capacity decay of just 0.030% per cycle and the extremely stable coulomb efficiency (almost above 99%). These outcomes, especially the high-rate capabilities, are one of the best results reported to date (Table S1). Even at 5 C rate, it presents an initial discharge capacity of 610 mAh g−1, and remained almost 400 mAh g−1 after 300 cycles with a capacity decay of 0.11% per cycle 396

Chemical Engineering Journal 355 (2019) 390–398

Y. Guo et al.

945 mAh g−1 at 0.2 C. More promisingly, during high-rate tests, it shows the excellent discharge capacity of 560 mAh g−1 (1 C after 1000 cycles), 466 mAh g−1 (3 C after 1000 cycles), and 400 mAh g−1 (5 C after 300 cycles). The interlayer still preserved to be unbroken after ultra-long cycling, which meets the sustainable development of our society. What’s more, the facile, scalable and affordable electrospinning with carbothermic reduction fabrication technology is of great potential in the practical use of high performance Li-S batteries.

The charge/discharge curves of the cell with the TCNFs interlayer at 0.2 C rate exhibited two typical common plateaus at 2.3 and 2.1 V (Fig. 5d), which is consistent with the CV profiles. As the current density increases, the curves become more distorted, but the cell could still release a relatively high capacity of 586 mAh g−1 at 5 C rate along with the two distinct plateaus, signifying that the TCNFs inter layer acquires better rate performance and high stability due to the high conductivity and good LiPSs adsorption abilities of Ti4O7. In addition, CMK3/S with higher sulfur areal density of 1.5 and 2.5 mg cm−2 (with the cathode thickness of 450 μm, much higher areal density is limited by the low sulfur loadings due to the relative low porosity of CMK3) were tested at 3 C (Fig. 5e). These cells exhibited the initial discharge capacities of 737 and 670 mAh g−1, capacity remained at 363 and 278 mAh g−1 after 1000 cycles, with capacity decay of just 0.050% and 0.058%, respectively. The electrodes with such low capacity decay make the practical application of lithium sulfur possible [19,33,34]. To visually explain the adsorption capability, the optical pictures of different samples are shown in Fig. 6a. Bare Li2S8 solution, Li2S8 solution with 0.1 g of CNFs powder and Li2S8 solution with 0.1 g TCNFs powder are denoted as sample 1, 2 and 3, respectively. After 1 h, the color of the sample 3 changes from deep yellow to light yellow, and become colorless for 3 h. In contrast, the color of the sample 1 and 2 barely changed, which indicates the strong adsorption of Li2S8 by the polar O-Ti-O groups of Ti4O7 which is sulphiphilic to LiPSs [7,23,35]. The pictures (Fig. S5) of the corresponding cycled separators (the side near lithium) after 2500 cycles at 3 C also validate the result, for the separator with TCNFs shows the lightest yellow color. Fig. S6 depicts the morphology of the TCNFs before and after cycles. It can be seen that the TCNFs are wrapped with a thin layer after long cycling, which is identified as LiPSs by EDS of TEM. To further understand the Ti4O7 adsorption effects, the morphology and elemental mapping of the TCNF interlayer were characterized by TEM, as shown in Fig. 6b–f. In Fig. 6b, the stacked and one-dimensional structure of the CNFs network is well preserved, and lots of Ti4O7 particles are tightly adhered to CNFs even after 2500 cycles at a current density of 3 C. TEM mapping in Fig. 6c–f illuminates the distribution of the elements, including C, Ti, O and S, all of them exhibited the shape of the cross-fibers, indicating the uniform distribution even after the long-cycle charge/discharge process. Remarkably, the sulfur distribution to be regular with a recognizable fibers shapes rather than a chaotic distribution in TiO2-CNFs interlayer. Therefore, it is proved that the adsorption effect of Ti4O7 is much better than TiO2. To further confirm the LiPSs adsorption mechanism of Ti4O7, XPS analysis was performed on the interlayer over 2500 cycles. The Ti 2p spectrum (Fig. 6g) exhibits two characteristic peaks at 459.0 and 464.8 eV, which ascribed to Ti4O7. Besides, the small shoulder peak at 457.4 eV could be referred to the Ti-S bond in the composite. In the S 2p spectrum of cycled CNFs and TCNFs (Fig. S7 and Fig. 6h), the broad peaks at 170 and 168 eV were attributed to sulfate due to residual electrolyte, and the peak around 163 eV was attributed to S-Li. The new peaks in cycled TCNFs centered at 162 and 160.5 eV could be assigned to S-Ti, which is consistent with the corresponding peaks in the Ti 2p spectrum [36,37]. Hence, the strong chemical adsorption effect of Ti4O7 was certified by XPS.

Acknowledgement This work was financially supported by Shenzhen Bureau of Science, Technology and Innovation Commission JCYJ20160525163956782 and JCYJ20170811154527927. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.cej.2018.08.143. References [1] R. Chen, T. Zhao, F. Wu, From a historic review to horizons beyond: lithium-sulphur batteries run on the wheels, Chem. Commun. 51 (2015) 18–33. [2] B. Dunn, H. Kamath, J.-M. Tarascon, electrical energy storage for the grid: a battery of choices, Science 334 (2011) 928–935. [3] X. Ji, L.F. Nazar, Advances in Li–S batteries, J. Mater. Chem. 20 (2010) 9821. [4] X. Ji, K.T. Lee, L.F. Nazar, A highly ordered nanostructured carbon-sulphur cathode for lithium-sulphur batteries, Nat. Mater. 8 (2009) 500–506. [5] C. Hu, H. Chen, Y. Shen, D. Lu, Y. Zhao, A.H. Lu, X. Wu, W. Lu, L. Chen, In situ wrapping of the cathode material in lithium-sulfur batteries, Nat. Commun. 8 (2017) 479. [6] H. Chen, C. Wang, W. Dong, W. Lu, Z. Du, L. Chen, Monodispersed sulfur nanoparticles for lithium-sulfur batteries with theoretical performance, Nano Lett. 15 (2015) 798–802. [7] H. Wei, E.F. Rodriguez, A.S. Best, A.F. Hollenkamp, D. Chen, R.A. Caruso, Chemical bonding and physical trapping of sulfur in mesoporous Magnéli Ti4O7 microspheres for high-performance Li-S battery, Adv. Energy Mater. 7 (2017) 1601616. [8] J. Liang, Z.-H. Sun, F. Li, H.-M. Cheng, Carbon materials for Li–S batteries: functional evolution and performance improvement, Energy Storage Mater. 2 (2016) 76–106. [9] H. Chen, C. Wang, Y. Dai, J. Ge, W. Lu, J. Yang, L. Chen, In-situ activated polycation as a multifunctional additive for Li-S batteries, Nano Energy 26 (2016) 43–49. [10] A. Vizintin, M. Lozinšek, R.K. Chellappan, D. Foix, A. Krajnc, G. Mali, G. Drazic, B. Genorio, R. Dedryvère, R. Dominko, Fluorinated reduced graphene oxide as an interlayer in Li–S batteries, Chem. Mater. 27 (2015) 7070–7081. [11] J. Wang, Y. Yang, F. Kang, Porous carbon nanofiber paper as an effective interlayer for high- performance lithium-sulfur batteries, Electrochim. Acta 168 (2015) 271–276. [12] H.M. Kim, J.Y. Hwang, A. Manthiram, Y.K. Sun, High-performance lithium-sulfur batteries with a self-assembled multiwall carbon nanotube interlayer and a robust electrode-electrolyte interface, ACS Appl. Mater. Interface. 8 (2016) 983–987. [13] J.H. Kim, J. Seo, J. Choi, D. Shin, M. Carter, Y. Jeon, C. Wang, L. Hu, U. Paik, Synergistic ultrathin functional polymer-coated carbon nanotube interlayer for high performance lithium-sulfur batteries, ACS Appl. Mater. Interface 8 (2016) 20092–20099. [14] G. Liang, J. Wu, X. Qin, M. Liu, Q. Li, Y.B. He, J.K. Kim, B. Li, F. Kang, Ultrafine TiO2 decorated carbon nanofibers as multifunctional interlayer for high-performance lithium-sulfur battery, ACS Appl. Mater. Interface 8 (2016) 23105–23113. [15] T. Zhao, Y. Ye, X. Peng, G. Divitini, H.-K. Kim, C.-Y. Lao, P.R. Coxon, K. Xi, Y. Liu, C. Ducati, R. Chen, R.V. Kumar, advanced lithium-sulfur batteries enabled by a bioinspired polysulfide adsorptive brush, Adv. Funct. Mater. 26 (2016) 8418–8426. [16] X. Gu, C. Lai, Recent development of metal compound applications in lithium-sulphur batteries, J. Mater. Res. 33 (2017) 16–31. [17] L. Luo, X. Qin, J. Wu, G. Liang, Q. Li, M. Liu, F. Kang, G. Chen, B. Li, Interwoven [email protected] scaffold interlayer for high-performance lithium-sulfur batteries, J. Mater. Chem. A 6 (2018) 8612–8619. [18] M. Liu, Q. Li, X. Qin, G. Liang, W. Han, D. Zhou, Y. He, B. Li, F. Kang, Suppressing self-discharge and shuttle effect of lithium-sulfur batteries with V2O5-decorated carbon nanofiber interlayer, Small 13 (2017) 1602539. [19] Z. Xiao, Z. Yang, L. Wang, H. Nie, M. Zhong, Q. Lai, X. Xu, L. Zhang, S. Huang, A lightweight TiO(2)/graphene interlayer, applied as a highly effective polysulfide absorbent for Fast, Long-Life Lithium-Sulfur Batteries, Adv. Mater. 27 (2015) 2891–2898. [20] T. Zhou, W. Lv, J. Li, G. Zhou, Y. Zhao, S. Fan, B. Liu, B. Li, F. Kang, Q.-H. Yang, Twinborn TiO2–TiN heterostructures enabling smooth trapping-diffusion-conversion of polysulfides towards ultralong life lithium-sulfur batteries, Energy Environ. Sci. 10 (2017) 1694–1703. [21] H.J. Peng, G. Zhang, X. Chen, Z.W. Zhang, W.T. Xu, J.Q. Huang, Q. Zhang,

4. Conclusions A Ti4O7/CNFs composite was constructed as a flexible and thin interlayer to realize the physical blocking and chemical trapping bifunction for LiPSs in Li-S battery. These composite fibers combine the merits of highly adsorptive Ti4O7 and conducting CNFs, where LiPSs are strongly bonded by Ti4O7 among the Ti4O7-CNFs network, meanwhile, the high conductivity of Ti4O7/CNFs enables LiPSs fast redox and conversion during charge-discharge process thus contribute to high electrochemical performances, even at high rates. The cell with TCNFs interlayer shows a high specific capacity, high-rate capability and ultralong cycling performance. After 100 cycles, it delivers a capacity of 397

Chemical Engineering Journal 355 (2019) 390–398

Y. Guo et al.

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29] J.R. Smith, F.C. Walsh, Electrodes based on MagneÂli phase titanium oxides: the properties and applications of EbonexÒ materials, J. Appl. Electrochem. 28 (1998) 1021–1033. [30] N. Azimi, W. Weng, C. Takoudis, Z. Zhang, Improved performance of lithium-sulfur battery with fluorinated electrolyte, Electrochemistry Commun. 37 (2013) 96–99. [31] J. Liu, Factors impact charging process in lithium/sulfur batteries, Energy Storage Sci. Technol. 1 (2015) 61–65. [32] P. Rosaiah, J. Zhu, D.P. Shaik, O.M. Hussain, Y. Qiu, L. Zhao, Reduced graphene oxide/MN3O4 nanocomposite electrodes with enhanced electrochemical performance for energy storage applications, J. Electroanal. Chem. 794 (2017) 78–85. [33] A. Manthiram, S.H. Chung, C. Zu, Lithium-sulfur batteries: progress and prospects, Adv. Mater. 27 (2015) 1980–2006. [34] S. Zhang, K. Ueno, K. Dokko, M. Watanabe, Recent advances in electrolytes for lithium-sulfur batteries, Adv. Energy Mater. 5 (2015) 1500117. [35] X. Tao, J. Wang, Z. Ying, Q. Cai, G. Zheng, Y. Gan, H. Huang, Y. Xia, C. Liang, W. Zhang, Y. Cui, Strong sulfur binding with conducting magneli-phase Ti(n)O2(n-1) nanomaterials for improving lithium-sulfur batteries, Nano Lett. 14 (2014) 5288–5294. [36] H. Martinez, C. Auriel, D. Gonbeau, G. Pfister-Guillouzo, A. Meerschaut, Electronic Structure of two misfit layer compounds: (PbS)1.18(TiS2) and (PbS)1.18(TiS2)2, J. Electron Spectrosc. Relat. Phenom. 95 (1998) 145–158. [37] H. Martinez, C. Auriel, D. Gonbeau, M. Loudet, G. Pfister-Guillouzo, Studies of 1T TiS, by STM, AFM and XPS: the mechanism of hydrolysis in air, Appl. Surf. Sci. 93 (1996) 231–235.

Enhanced electrochemical kinetics on conductive polar mediators for lithium-sulfur batteries, Angew. Chem. Int. Ed. 55 (2016) 12990–12995. X. Tao, J. Wang, C. Liu, H. Wang, H. Yao, G. Zheng, Z.W. Seh, Q. Cai, W. Li, G. Zhou, C. Zu, Y. Cui, Balancing surface adsorption and diffusion of lithium-polysulfides on nonconductive oxides for lithium-sulfur battery design, Nat. Commun. 7 (2016) 11203. Q. Pang, D. Kundu, M. Cuisinier, L.F. Nazar, Surface-enhanced redox chemistry of polysulphides on a metallic and polar host for lithium-sulphur batteries, Nat. Commun. 5 (2014) 4759. S. Yao, S. Xue, Y. Zhang, X. Shen, X. Qian, T. Li, K. Xiao, S. Qin, J. Xiang, Synthesis, characterization, and electrochemical performance of spherical nanostructure of Magnéli phase Ti4O7, J. Mater. Sci.: Mater. Electron. 28 (2017) 7264–7270. X. Qin, Y. Lu, H. Xiao, Y. Wen, T. Yu, A Comparison of the effect of graphitization on microstructures and properties of polyacrylonitrile and mesophase pitch-based carbon fibers, Carbon 50 (2012) 4459–4469. J. Zhang, M.J. Li, Z.C. Feng, J. Chen, C.U.V. Li, Raman spectroscopic study on TiO2. I. Phase transformation at the surface and in the bulk, J. Phys. Chem. B 110 (2006) 927–935. S. Mei, C.J. Jafta, I. Lauermann, Q. Ran, M. Kärgell, M. Ballauff, Y. Lu, Porous Ti4O7 particles with interconnected-pore structure as a high-efficiency polysulfide mediator for lithium-sulfur batteries, Adv. Funct. Mater. 27 (2017) 1701176. S.S. Zhang, Effect of discharge cutoff voltage on reversibility of lithium/sulfur batteries with LiNO3-contained electrolyte, J. Electrochem. Soc. 159 (2012) A920–A923.

398