Nitrogen-doped holey carbon nanotubes: Dual polysulfides trapping effect towards enhanced lithium-sulfur battery performance

Accepted Manuscript Full Length Article Nitrogen-doped holey carbon nanotubes: dual polysulfides trapping effect towards enhanced lithium-sulfur battery performance Liang Chen, Chenxi Xu, Liming Yang, Minjie Zhou, Binhong He, Zhengu Chen, Zhi Li, Mengting Shi, Zhaohui Hou, Yafei Kuang PII: DOI: Reference:

S0169-4332(18)31484-3 https://doi.org/10.1016/j.apsusc.2018.05.166 APSUSC 39442

To appear in:

Applied Surface Science

Received Date: Revised Date: Accepted Date:

19 March 2018 6 May 2018 21 May 2018

Please cite this article as: L. Chen, C. Xu, L. Yang, M. Zhou, B. He, Z. Chen, Z. Li, M. Shi, Z. Hou, Y. Kuang, Nitrogen-doped holey carbon nanotubes: dual polysulfides trapping effect towards enhanced lithium-sulfur battery performance, Applied Surface Science (2018), doi: https://doi.org/10.1016/j.apsusc.2018.05.166

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Nitrogen-doped holey carbon nanotubes: dual polysulfides trapping effect towards enhanced lithium-sulfur battery performance Liang Chen†

, Chenxi Xu† c, Liming Yang b, Minjie Zhou a, Binhong He a, Zhengu Chen a,

a,c

Zhi Li a, Mengting Shi a, Zhaohui Hou*a, Yafei Kuang*c a

School of Chemistry and Chemical Engineering, Hunan Institute of Science and Technology,

Yueyang 414006, P. R. China b

Key Laboratory of Jiangxi Province for Persistent Pollutants Control and Resources Recycle,

Nanchang Hangkong University, Nanchang 330063, P. R. China c

College of Chemistry and Chemical Engineering, Hunan University, Changsha, 410082, P. R.

China

Abstract: Advanced sulfur hosts decorated with unique structure, optimized composition and excellent conductivity play a vital role in improving lithium-sulfur battery performance. Herein, we first prepare oxidized holey carbon nanotubes (O-H-CNTs) through a controlled air etching strategy, then exploit them as the carbon precursor to produce nitrogen-doped holey carbon nanotubes (N-H-CNTs). The influence of air etching degree on the morphology, structure, composition, conductivity and post-doping effect of O-H-CNTs is discussed in detail. The optimized O-H-CNTs-2 (2 refers to etching time) display unique holey structure and

*Corresponding author, Tel: +86 730 8648502; Fax: +86 730 8640122 E-mail addresses: [email protected] (Z. H. Hou), [email protected] (Y. F. Kuang) L. Chen and C. X. Xu contributed equally to this work 1

obtain a certain amount of structure defects while retaining good conductivity. After N doping, the resultant N-H-CNTs-2 host exhibits improved conductivity and promoted sulfur dispersibility. More importantly, the unique holey structure and effective N doping endow them with strengthened dual trapping capability for polysulfides. Therefore, the N-H-CNTs-2/S cathode shows much enhanced cycling lifespan and rate performance. It delivers a high discharge capacity of over 750 mAh/g even after 500 cycles, and high rate capacity of ~560 mAh/g at 4 C. Clearly, our work provides a good guidance on the design and exploitation of excellent sulfur host by rationally regulating the structure, composition and conductivity. Keywords: carbon nanotubes, structure defects, holey structure, N doping, lithium sulfur batteries

1. Introduction Lithium-sulfur (Li-S) batteries, as one of promising energy storage systems have attracted much attention due to their overwhelming theoretical energy density up to ~2600 Wh/kg, which proves ten times as high as that of commercialized lithium ion batteries [1-2]. Furthermore, the natural abundance, low expenditure and environmental benignity of sulfur make them more compelling to satisfy the increasing demand for extensively prospering energy storage markets [3]. However, current Li-S batteries still suffer from several obstacles for practical application. One major problem is the intrinsic insulativity of sulfur and its partial discharge products (Li2S2 and Li2S), leading to low active materials utilization and poor rate performance. The other crucial trouble is the notorious shuttling effect caused by heavy dissolution 2

of long-chain polysulfides into electrolyte, which results in inferior cycling stability and low coulombic efficiency [4-5]. Many strategies have been put forward to tackle these issues, among which hybridizing sulfur with conductive hosts proves much fruitful. On one hand, the conductive hosts provide a conducting skeleton to support sulfur and make it electrically contacted. On the other hand, the physical adsorption of hosts contributed by van der Waals force mitigates polysulfides dissolution and ameliorates the shuttling effect to some extent [6-7]. Carbon materials, such as carbon nanotubes (CNTs), graphene and so forth, attract much attention owing to their intriguing conductivity [8]. Qiu et al.[9] introduced multi-walled CNTs (MWCNTs) into sulfur through capillarity, and finally fabricated S-coated MWCNTs cathode with much improved battery performance, which mainly ascribed to the remarkably elevated cathode conductivity. Even so, the battery performance of common C/S hybridized cathodes was still far from our demands, and the unsatisfying battery performance mostly resulted from the undesirable sulfur dispersibility and poor polysulfides trapping capability of common carbon hosts. To overcome these drawbacks, rational design and exploitation of carbon hosts decorated with unique morphology and structure have been demonstrated very crucial [10-11]. In our previous work, the unzipped CNTs (UCNTs) were prepared by longitudinally splitting pristine CNTs followed by hydrothermal reduction process [12]. The obtained UCNTs displayed special unzipped shape and much increased specific surface area compared to CNTs, which endowed them with better sulfur

3

dispersibility and certain polysulfides confinement effect. As a result, the UCNTs/S cathode presented enhanced cycling performance. Meanwhile, Yuan et al.[13] successfully synthesized porous CNTs (PCNTs) through KOH activating MWCNTs. The resultant PCNTs with unique mesoporous structure and significant enlarged specific surface area exhibited strengthened sulfur dispersibility and polysulfides trapping capability by encapsulating sulfur inside nanotubes. Besides, the existence of these mesopores boosted electrolyte infiltration and ion transport. Therefore, the as-prepared PCNTs/S cathode showed improved cycling stability and rate capability. Though increased specific surface area and facile accessible channels could be achieved by chemical oxidation or activation, the simultaneously generated unzipped or mesoporous structure resulted in high exposure of polysulfides to electrolyte, and could not guarantee very effective trapping for polysulfides by pure physical confinement [14], as reflected from the imperfect cycling performance of UCNTs/S and PCNTs/S cathodes. Apart from designing unique morphology and structure, regulating surface property and electronic structure of carbon hosts by heteroatoms doping has also been testified very valid [15-19]. In the heteroatoms family, N doping catches the eye of most researchers because it substantially alters surface polarity and improves conductivity of carbon matrix, which is beneficial to sulfur dispersion and polysulfides adsorbing. Zhang et al.[20] produced N-doped CNTs (NCNTs) by chemical vapor deposition (CVD) method, and they found that carbon interface modified by N atoms possessed much enhanced dispersibility and adsorbability for

4

sulfur species, eventually resulting in higher sulfur utilization and weaker shuttling effect. In the meantime, Zhang et al.[21] fabricated N-doped graphene (NG) by thermal nitridation of graphene oxide under ammonia atmosphere, and the resultant S@NG cathode delivered ultra-long cycling life exceeding 2000 cycles and low capacity decay rate. employed it as the carbon host for sulfur cathode, which was due to the strong polysulfides binding capability of the doped N atoms in NG, Generally, to doping N into carbon frameworks, post-doping methods (such as thermal annealing or hydrothermal process) are more preferred due to their simpleness, convenience, high yield and low expenditure [22]. And the structure defects in carbon precursors acting as the reactive sites are always the prerequisite for effective N doping when using these doping methods [23-25]. Pristine carbon materials equipped with ideal sp2 hybridized structure exhibit inert surface, and often fail to complete effective N doping. Hence they usually need to acquire structure defects by way of pre-oxidation before N doping. But if the carbon materials are oxidized overmuch, excessive oxidation will sacrifice much conductivity, ultimately leading to unsatisfactory battery performance. In view of this point, it is essential to take the oxidation degree into account during the introduction of structure defects. In this work, we first controllably oxidized pristine CNTs to oxidized holey carbon nanotubes (O-H-CNTs-2, where 2 represented etching time) through air etching strategy, then exploited them as the carbon precursor to obtain N-doped holey CNTs (N-H-CNTs-2). The O-H-CNTs-2 were utilized as the carbon precursor for post-doping, catching the following advantages. Firstly, the O-H-CNTs-2 got a certain

5

amount of structure defects and retained good conductivity in spite of suffering from high-temperature air etching. Secondly, the as-formed unique holey structure offered accessible pathways for electrolyte infiltration and ion transport. Moreover, this holey structure could serve as a natural barrier to trap polysulfides via physical confinement. After N doping by hydrothermal treatment with urea, improved conductivity and promoted sulfur dispersibility could be achieved synchronously for N-H-CNTs-2. More importantly, the doped N groups mostly around the edge of opening mesopores played a role of efficiently adsorbing polysulfides. Benefiting from these combined merits, the as-prepared N-H-CNTs-2/S cathode exhibited superior cyclability and rate capability. Herein, we mostly discussed the influence of air etching degree on the morphology, structure, composition, conductivity and post-doping effect of O-H-CNTs. Meanwhile, we systematically investigated the dual polysulfides trapping effect caused by physical confinement and chemical adsorbing on the cathode performance.

2. Experimental 2.1. Chemicals and materials All reagents (Sinopharm Chemical Reagent Co., Ltd., China) were of analytical grade without further treatment. Ultrapure water was used in this study. The pristine multi-walled carbon nanotubes (MWCNTs, purity>97%) with 5-15 μm in length and 60-100 nm in diameter were commercially supplied (L-MWNT-60100, Shenzhen Nanotech Port Co., Ltd.). 2.2. Synthesis of O-H-CNTs-t 6

Pristine CNTs were heated at a rate of 15 °C/min to 568 °C in air and remained at this temperature for 0.5 h, 1 h and 2 h, respectively. When the heating program terminated, the product was taken out of the furnace and promptly cooled down to obtain the O-H-CNTs-t, where t stood for heating time. 2.3. Synthesis of N-H-CNTs-t 1.4 g of urea was slowly added into 70 mL of O-H-CNTs-t suspension (1.0 mg/mL). After ultrasonication for a while, the resultant homogeneous suspension was transferred into a 100 mL Teflon-lined autoclave and reacted at 200 °C for 12 h. After that, the black solid product was separated by centrifugation and washed with distilled water and ethanol for several times. The final product labeled as N-H-CNTs-t (t represented heating time) was obtained via drying under vacuum overnight. 2.4. Synthesis of C/S composites The N-H-CNTs-t/S composites were all prepared by a typical melt-diffusion method reported previously [26]. In particular, N-H-CNTs-t were uniformly mixed with sulfur powder in a specific mass ratio, then the resultant C/S mixture was heated to 155 °C and kept for 12 h to obtain the final N-H-CNTs-t/S composite. 2.5. Preparation of polysulfide (Li2S4) solution The Li2S4 solution was synthesized by chemically reacting sulfur powder and an appropriate amount of Li2S (99.9%, Alfa Aesar) in anhydrous tetrahydrofuran solvent [27]. The obtained mixture solution was then stirred in an Ar-filled glove box overnight to get the orange Li2S4 solution. 2.6. Materials characterization

7

Morphology and composition of samples were measured by scanning electron microscope (SEM, Hitachi S-4800) and transmission electron microscope (TEM, Tecnai F20)

equipped

with

an

energy

dispersive

spectrometer

(EDS).

Brunauer-Emmett-Teller (BET) specific surface area was tested based on nitrogen physical sorption (ASAP2020 M+C, USA). Fourier transform infrared spectroscopy (FTIR) analysis was conducted on a Fourier transform infrared spectrometer (nICOLET6700). Raman spectra were detected on a Raman spectrometer (Labram-010, France) from 1200 to 2000 cm−1. X-ray diffraction (XRD) spectra were recorded by an X-ray diffractometer (XRD-6100) with Cu Kα radiation (λ=1.5418 Å) operating at 40 kV and 60 mA. Electrical conductivity was measured on a SZT-2 four probe conductivity meter (Suzhou Tongchuang Electronics Ltd.). X-ray photoelectron spectroscopy (XPS) analysis was carried out on an ESCALAB250 XPS spectrometer with a Mg Kα X-ray source (1350 eV). Thermogravimetric analysis (TGA) was conducted on a simultaneous DSC-TGA analyzer (DTG-60), and the samples were heated from room temperature to 700 °C with a heating rate of 10 °C/min under a continuous flow of N2 or O2 gas. 2.7. Electrochemical measurements Electrochemical measurements were conducted on 2025 type coin cells with lithium foil and polypropylene membrane as counter electrode and separator, respectively. The electrolyte was 1 M lithium bis-trifluoromethanesulfonylimide (LiTFSI) dissolved in a mixed solvent of 1,3-dioxolane and 1,2-dimethoxyethane (DOL/DME, 1:1, v/v) with 1 wt% LiNO3. The C/S composite, Super P and

8

polyvinylidene difluoride (PVDF) were dispersed in N-methyl pyrrolidone (NMP) in the mass ratio of 8:1:1, then the obtained slurry was coated on Al foil and dried at 60 °C overnight under vacuum. All electrodes were punched into round discs with a sulfur loading of about 2 mg/cm2. The specific capacity of cathodes was calculated based on the sulfur mass in the C/S composite. The amount of electrolyte used in every coin cell was ~40 μL/(mg sulfur). Cyclic voltammetry (CV) was tested with an electrochemical workstation (CHI 660E) between 1.7 V and 2.8 V at a scan rate of 0.1 mV/s. The galvanostatic charge-discharge measurement was performed using a battery tester (LAND CT-2001A, Wuhan, China) in the voltage range of 1.7~2.8 V (vs. Li/Li+) at different current densities of 0.1 C, 0.2 C, 0.5 C, 1 C, 2 C, 4 C (1 C=1675 mA/g). Electrochemical impedance spectroscopy (EIS) analysis was carried out on the open circuit potential in the frequency range of 100 kHz to 0.01 Hz. All experiments were accomplished at room temperature.

3. Results and discussion 3.1 Morphology and structure Fig. 1 illustrates the preparation procedure of N-H-CNTs/S composite. Firstly, the O-H-CNTs were synthesized by controlled air etching on pristine CNTs, then they suffered from hydrothermal process with urea, and the N-H-CNTs were obtained. Meanwhile, the corresponding C/S composite (N-H-CNTs/S) was fabricated by typical melt diffusion method. Clearly, the unique holey structure exerted a good confinement for trapping sulfur species.

9

Fig 1. Preparation scheme of N-H-CNTs-2/S composite for sulfur cathode To confirm proper air etching temperature, the thermal gravimetric analysis of CNTs was conducted (Fig. S1), showing that 568 °C became the inflection point, thus 568 °C was regarded as a suitable temperature to achieve effective and controllable etching [28]. Herein, we adjusted the etching degree by changing the etching time, and investigated the influence of etching degree on the morphology, structure, composition, conductivity and post-doping effect of the obtained carbon precursors (O-H-CNTs-t). Fig. 2 shows the SEM and TEM images of O-H-CNTs-t. Similar with pristine CNTs (Fig. S2), both O-H-CNTs-0.5 and O-H-CNTs-1 displayed closed nanotube shape, as discerned from Fig. 2a-b. While for the O-H-CNTs-2, a typical flute-like holey structure was observed, also verified by the TEM image (Fig. 2f). The as-formed flute-like holey structure could be illustrated for effective etching on carbon surface. Fig. S3a exhibits the N2 adsorption/desorption isotherm curves of CNTs and O-H-CNTs-2. It was observed that the specific surface area and pore volume of CNTs increased from 80.8 m2/g and 0.148 cm3/g to 143.9 m2/g and 0.247

10

cm3/g for O-H-CNTs-2, respectively, which further demonstrated the formation of holey structure. As seen from Fig. S3b, pristine CNTs showed an obvious peak at ~39 nm, corresponding to the pores between adjacent tubes and among bundles. While for the O-H-CNTs-2, this peak became lower and two nearly overlapped peaks locating at ~5 nm appeared, which was due to the introduced structure defects and weakened bundles aggregation induced by air etching.

Fig. 2. SEM images of O-H-CNTs-0.5 (a), O-H-CNTs-1 (b) and O-H-CNTs-2 (c); TEM images of O-H-CNTs-0.5 (d), O-H-CNTs-1 (e) and O-H-CNTs-2 (f) FTIR and Raman were conducted to investigate the structure of O-H-CNTs-t. Fig.3a presents the FTIR spectra of CNTs and O-H-CNTs-t (t=0.5, 1 and 2). Clearly, all these O-H-CNTs-t samples exhibited two typical peaks centered at 1700 cm–1 and 3400 cm–1, corresponding to the C=O and ‒OH stretching vibration [29], respectively, while these two peaks were negligible for CNTs, suggesting successful introduction of oxygen groups (structure defects) onto the nanotubes surface. Moreover, the intensity of these two peaks became strengthened with the prolonging of etching time, which indicated increased oxidation degree. Fig. 3b provides the Raman spectra of CNTs 11

and O-H-CNTs-t. It was found that all samples showed two representative peaks (D band and G band). In general, the intensity ratio of these two bands (ID/IG) reflects the structure disorder degree [30]. The ID/IG value for CNTs was 0.86, and it grew with the increase of t, implying higher structural defects, which agreed well with the FTIR results.

Fig. 3. FTIR (a) and Raman (b) spectra of CNTs, O-H-CNTs-0.5, O-H-CNTs-1 and O-H-CNTs-2 XRD was used to further characterize the structure of O-H-CNTs-t. As shown in Fig. 4a, the CNTs exhibited a remarkable (002) peak at 2θ=26.3°, illustrating for a interlayer spacing (d002) of 3.39 Å. After air etching for 0.5 h, the d002 value for O-H-CNTs-0.5 slightly increased to 3.40 Å, reflecting weak oxidation. With the etching time extending to 2 h, the 2θ peak for the obtained O-H-CNTs-2 observably decreased to 25.9°, indicating much more oxygen groups introduced [31]. Despite this, the O-H-CNTs-2 still possessed high crystallinity, revealing good conductivity. To quantitatively analyze the conductivity, the four-probe technique is adopted. As seen from Fig. 4b, the conductivity of CNTs reached as high as 10.6 S/cm, and it slightly reduced to 4.5 S/cm for O-H-CNTs-0.5. With regard to the O-H-CNTs-2, this value still could maintain at a level of 0.2 S/cm, suggesting moderate etching degree, which 12

was consistent with the above XRD analysis. Then, XPS was performed to explore the composition of O-H-CNTs-t. As shown in Fig. S4, it was observed that the oxygen content increased from 5.2 a.t.% for O-H-CNTs-0.5 to 10.6 a.t.% for O-H-CNTs-2, demonstrating prominently increased structure defects with the etching time prolonging to 2 h.

Fig. 4. (a) XRD patterns and electrical conductivity (b) of CNTs, O-H-CNTs-0.5, O-H-CNTs-1 and O-H-CNTs-2 On the basis of the results analyzed above, it is concluded that the morphology, structure, composition and conductivity of the O-H-CNTs-t intimately correlated with the air etching degree. More importantly, the O-H-CNTs-2 was evidenced to display unique holey structure and obtain a certain amount of structure defects while retaining good conductivity. When the O-H-CNTs-t were used as the carbon precursor for subsequent N doping, we also studied the influence of etching degree on the synthesis of N-doped carbon products and their relevant C/S composites. Firstly, XPS is employed to verify successful N doping. Fig. 5a shows the XPS survey spectra of N-H-CNTs-t. Apparently, all these samples displayed a typical N 1s peak locating at ~400 eV, demonstrating successful N doping. Table S1 lists the specific XPS result. Clearly, all of the O contents for N-H-CNTs-t declined 13

dramatically when compared to those for O-H-CNTs-t. The apparent loss of oxygen-containing groups by hydrothermal process mostly ascribed to that urea was a kind of strong reducing agent and effective N source. Meanwhile, it was discerned that the N content for N-H-CNTs-t increased with the growing of etching time, indicating that more structure defects promised higher doping amounts [32]. Besides, we also found that the percentages of pyridinic N and oxygenated N elevated with the increase of etching time, which may ascribe to that the increased edge defects existing around the opening mesopores were more beneficial to the formation of pyridinic N and oxygenated N. As reported previously, the pyridinic N (43.6%) dominating in N-H-CNTs-2 can be regarded as electron-rich donors, and strongly binds with the positively charged Li+ by Lewis acid-base interaction, thus effectively restraining polysulfides shuttling [33].

Fig. 5. XPS survey spectra (a); high resolution N 1s spectra of N-H-CNTs-0.5 (b), N-H-CNTs-1 (c) and N-H-CNTs-2 (d) 14

Afterwards, the morphology of N-H-CNTs-t was measured by SEM. The N-H-CNTs-t displayed analogous shape with O-H-CNTs-t (Fig. S5), reflecting that hydrothermal process imposed little impact on structure stability. In the meantime, the structure of N-H-CNTs-t and their relevant C/S composites was also investigated by Raman and XRD. Fig. 6a presented the Raman spectra. We found that the structure defects of N-H-CNTs-t was more than those of the corresponding carbon precursors (Fig. 3b), which indicated that N doping further increased the disorder degree [22]. Additionally, the effect of etching degree on the structure defects of N-H-CNTs-t was similar to that of O-H-CNTs-t, and the structure defects of N-H-CNTs-2 proved the most among them. After introducing nonpolar sulfur to N-H-CNTs-t, the obtained N-H-CNTs-t/S composites showed reduced ID/IG value [34]. Besides, they all showed three typical sulfur peaks, but these peaks for N-H-CNTs-2/S composite were negligible and much weaker than those of other samples. These results are mainly attributed to the much enhanced sulfur dispersibility caused by unique holey structure and more effective N doping for N-H-CNTs-2. Fig. 6b exhibits the XRD patterns. It was observed that the effect of etching time on the (002) peak for N-H-CNTs-t displayed identical rules with that for O-H-CNTs-t. Typical sulfur diffraction peaks (JCPDS No.08-0247) were detected for all the N-H-CNTs-t/S composites. Nevertheless, the peaks intensity for N-H-CNTs-2/S composite turned out to be much smaller than that of other C/S composites, suggesting better sulfur dispersibility [35], which accorded with the Raman results. To learn about the conductivity of these N-doped products, their conductivity is also

15

tested by four probe technique, as listed in Table S2. Obviously, the conductivity recovered after hydrothermal process, demonstrating that N doping could enhance conductivity by devoting additional free electrons to conduction band [36].

Fig. 6. Raman (a) and XRD (b) spectra of N-H-CNTs-0.5, N-H-CNTs-1, N-H-CNTs-2, N-H-CNTs-0.5/S, N-H-CNTs-1/S and N-H-CNTs-2/S composites To accurately judge the sulfur dispersion state on these N-H-CNTs-t hosts, the morphology of their relevant C/S composites was also measured by SEM. As presented in Fig. 7a, sulfur particles existing in the form of huge blocks unevenly dispersed on the surface of N-H-CNTs-0.5 host, indicating poor sulfur distribution state, while this sulfur distribution state for N-H-CNTs-1/S composite got slightly improved, as seen from Fig. 7b, which may ascribe to better N doping effect. As for N-H-CNTs-2/S composite, no obvious sulfur block was evidenced, and the sulfur particles were homogeneously distributed inside and outside the nanotubes, as shown in Fig. 7c-d. The remarkably promoted sulfur dispersibility for N-H-CNTs-2 was mostly due to the co-effect of special holey structure and effective N doping, which made for excellent sulfur wettability and encapsulation [12-13]. Fig. 7f offers the EDS spectrum of the N-H-CNTs-2/S composite, and the C, S, N and O elements were

16

affirmed in its structure. Fig. 7e presents the STEM image of the N-H-CNTs-2/S composite and the corresponding element mapping images are provided in Fig. 7g-j. The sulfur confinement effect imposed by N-H-CNTs-2 was clearly discerned by the difference between the C and S mappings. Meanwhile, the doped N atoms were certified to situate around sulfur species by comparing the N and S mappings.

Fig. 7. SEM images of N-H-CNTs-0.5/S (a), N-H-CNTs-1/S (b) and N-H-CNTs-2/S (c) composites; TEM image (d), scanning transmission electron microscopy (STEM) image (e), EDS spectrum (f) and the corresponding element mapping images of carbon (g), oxygen (h), sulfur (i) and nitrogen (j) of N-H-CNTs-2/S composite To confirm the sulfur content in the C/S composites, TGA was carried out (Fig.8). A weight loss of ~72% between 250 °C and 350 °C was assigned to sulfur evaporation [37]. Apart from this, it was found that the sulfur in N-H-CNTs-2 host possessed more strengthened thermal stability than other hosts, as evidenced by the much slower sulfur evaporation rate, which might ascribe to the confinement effect of holey structure.

17

Fig. 8. TGA curves of the C/S composites under N2 atmosphere 3.2 Electrochemical performance As analyzed above, it was found that the N-H-CNTs-2 host not only exhibited unique holey structure for sulfur confinement and electrolyte infiltration, but also successfully achieved N doping for enhanced sulfur dispersiblity and polysulfides adsorbing while retaining good conductivity. Benefiting from these advantages, the as-prepared N-H-CNTs-2/S cathode was anticipated to present excellent battery performance. Herein, the electrochemical performance of the sulfur cathodes was evaluated by CV, galvanstatic charge/discharge and EIS. Fig. 9a shows the first cycle CV profiles of the N-H-CNTs-t/S cathodes within the voltage range from 1.7 V to 2.8 V at a scan rate of 0.1 mV/s. Every CV profile displayed two distinct cathodic peaks at ~2.0 V and ~2.3 V and two overlapped anodic peaks at ~2.4 V. As reported previously [38-39], these two cathodic peaks generally corresponded to the reduction of S8 to high-order soluble lithium polysulfides (Li2Sn, 4≤n<8) and then further reduction to insoluble lithium sulfides (Li2S2 and Li2S), while the two almost overlapped anodic peaks was assigned to the conversion of Li2S/Li2S2 to Li2Sn (4≤n<8), and eventually to S8. By contrast, the N-H-CNTs-2/S cathode occupied the

18

largest redox peak current densities and the minimal peak potential separations when compared to the counterparts, indicating the highest sulfur utilization and lowest polarization, which was due to the best sulfur dispersibility contributed by the increased specific surface area and effective N doping [25]. Fig. 9b presents the initial three continuous CV curves of the N-H-CNTs-2/S cathode. No conspicuous changes were detected in the subsequent two cycles, suggesting outstanding reversibility and cycling stability [40].

Fig. 9. (a) CV curves of different N-H-CNTs-t/S cathodes at the first cycle; (b) CV curves of the N-H-CNTs-2/S cathode at the first three cycles (scan rate: 0.1 mV/s ) The charge/discharge voltage profiles of the N-H-CNTs-t/S cathodes at 0.2 C at the 1st and 500th cycles are provided in Fig. S6, showing two typical discharge plateaus and one charge plateau, in good agreement with the CV curves. What’s more, the N-H-CNTs-2/S cathode exhibited much smaller electrochemical polarization (lowest voltage hysteresis, ΔV) than other two cathodes, demonstrating a more facile redox reaction and better cyclic reversibility [41]. Fig. 10a shows the cycling performance of the N-H-CNTs-t/S cathodes at 0.2 C. The N-H-CNTs-0.5/S, N-H-CNTs-1/S and N-H-CNTs-2/S cathodes delivered the initial discharge capacities

19

of 1007, 1066 and 1205 mAh/g, respectively. After consecutive 500 cycles, the N-H-CNTs-2/S cathode still maintained a capacity of over 750 mAh/g, manifesting a capacity decay rate of 0.07%. In contrast, the N-H-CNTs-0.5/S and N-H-CNTs-1/S cathodes exhibited capacities of 370 and 567 mAh/g, respectively, corresponding to capacity degradation rates of 0.13% and 0.10%. Obviously, the N-H-CNTs-2/S cathode enjoyed more superior cycling stability than N-H-CNTs-0.5/S and N-H-CNTs-1/S cathodes, also as proven by its higher coulombic efficiency. The cathode performance of N-H-CNTs-2/S was compared with that of the state-of-the-art carbon-based cathodes (see Table S3), which showed that the N-H-CNTs-2/S cathode possessed the highest initial discharge capacity and most competitive capacity loss rate among these advanced cathodes. The superior cycling performance of N-H-CNTs-2/S cathode chiefly imputed to the joint effect of good conductivity, accessible infiltration and transport pathways, much strengthened sulfur dispersibility and dual polysulfides trapping capability induced by unique holey structure and much effective N doping. To confirm the role of N doping in trapping polysulfides, visualized adsorption experiment was conducted in fresh Li2S4 solution (Fig. S7). The color of Li2S4 solution appeared distinct after adding CNTs, O-H-CNTs-2 and N-H-CNTs-2, and it became light brown more evidently for N-H-CNTs-2 when compared with that for CNTs and O-H-CNTs-2, demonstrating better polysulfides adsorbability. In order to further prove the excellent polysulfides trapping capability of N-H-CNTs-2, the Li-S cells were disassembled after long-term cycling, and the electrolyte color and cathodes morphology were both detected. As shown in Fig. S8,

20

the electrolyte deposited on separator is almost white for N-H-CNTs-2/S cathode, while it became yellow for N-H-CNTs-0.5/S and N-H-CNTs-1/S cathodes, also illustrating for better polysulfides trapping capability. Fig. S9 exhibits the SEM images of N-H-CNTs-t/S cathodes after 500 cycles. It was observed that sulfur species dispersed uniformly as a thin layer on N-H-CNTs-2 host, whereas they agglomerated seriously into huge on N-H-CNTs-0.5 and N-H-CNTs-1 hosts, well explaining for outstanding cycling stability [39]. Fig. 10b provides the rate performance of the N-H-CNTs-t/S cathodes. The N-H-CNTs-2/S cathode delivered a stable discharge capacity of ~1200 mAh/g at 0.1 C, and maintained a reversible capacity as high as ~560 mAh/g at the current rate of 4 C. In addition, this capacity recovered to ~1090 mAh/g when the current rate returned to 0.2 C, indicating good rate capability. In contrast, the N-H-CNTs-0.5/S and N-H-CNTs-1/S cathodes exhibited much lower discharge capacities and inferior stability under the same current rates.

Fig. 10. (a) Cycling performance and the corresponding coulombic efficiency at 0.2C ; (b) the rate capability of the N-H-CNTs-t/S cathodes EIS was conducted to further study the electrochemical performance of different N-H-CNTs-t/S cathodes (Fig. 11). The EIS spectra were analyzed and fitted to the 21

equivalent circuit (Fig. S10), and the specific EIS fitting results were provided in Table S4. The intercept and diameter of the semi-circle in high frequency range correlated with electrolyte resistance (Rs) and charge transfer resistance (Rct), respectively, and the second compressed semi-circle in middle-frequency area corresponded to mass transfer resistance (Rf) that dominated the liquid-to-solid transformation through insoluble discharge products. In addition, the straight line in low-frequency region was assigned to diffusion process of soluble lithium polysulfides [20, 42]. As shown in Fig. 11a, the N-H-CNTs-2/S cathode exhibited much lower Rct value (54.6 Ω) than those of N-H-CNTs-0.5/S (115.6 Ω) and N-H-CNTs-1/S cathodes (98.9 Ω), manifesting faster interfacial charge transfer. Besides, the largest straight line slope in low-frequency region for N-H-CNTs-2/S cathode also suggested the best ion diffusion effect. Fig. 11b presents the Nyquist plots of the N-H-CNTs-2/S cathode at different cycles at 0.2 C. It was found that the Rct value further decreased to 11.2 Ω after 50 cycles, also implying excellent cycling stability.

Fig. 11. (a) Nyquist plots of the N-H-CNTs-t/S cathodes at 0.2 C before cycling; (b) Nyquist plots of the N-H-CNTs-2/S cathode at different cycles at 0.2 C

22

4. Conclusions In a nutshell, N-H-CNTs-2 host could be successfully prepared by a controlled air etching method followed by simple hydrothermal doping process. By this controlled air etching, unique holey structure and a certain amount of structure defects were simultaneously introduced on the surface of O-H-CNTs-2 while retaining good conductivity. After achieving N doping, the resultant N-H-CNTs-2 host showed accessible electrolyte infiltration as well as ion transport pathways, improved conductivity and promoted sulfur dispersibility. Moreover, the unique holey structure and effective N doping endowed them with strengthened dual trapping capability for polysulfides. As a consequence, the as-synthesized N-H-CNTs-2/S cathode possessed excellent cyclability and rate capability. This cathode delivered a high discharge capacity of over 750 mAh/g even after 500 cycles at 0.2 C and high rate capacity of 560 mAh/g at 4 C, superior to most previously reported carbon-based sulfur cathode. Clearly, our work not only provides an outstanding sulfur host, but also offers a good guidance on the design of advanced sulfur host by rationally regulating the structure, composition and conductivity.

Notes The authors declare no competing financial interest. Acknowledgements This work was supported by National Natural Science Foundation of China (Grant No. 51772092), Natural Science Foundation of Hunan Province China (Grant 23

No. 2018JJ3207), Hunan Province College Students Research Learning and Innovative Experiment Project and the Research Foundation of Education Bureau of Hunan Province, China (Grant No. 17A086, 17K039). Appendix A. Supplementary material Supplementary

data

associated

with

this

article

can

be

found

at

http://dx.doi.org/10.1016/j.

References [1] Y. Tang, Y. Zhang, W. Li, B. Ma and X. Chen, Chem. Soc. Rev. 44 (2015) 5926-5940. [2] P. G. Bruce, S. A. Freunberger, L. J. Hardwick and J. M. Tarascon, Nat. Mater. 11 (2012) 19-29. [3] G. Zhou, L. C. Yin, D. W. Wang, L. Li, S. Pei, I. R. Gentle, F. Li and H. M. Cheng, ACS Nano 7 (2013) 5367-5375. [4] A. Manthiram, Y. Fu, S. H. Chung, C. Zu and Y. S. Su, Chem. Rev. 114 (2014) 11751-11787. [5] C. Tang, Q. Zhang, M. Q. Zhao, J. Q. Huang, X. B. Cheng, G. L. Tian, H. J. Peng and F. Wei, Adv. Mater. 26 (2014) 6100-6105. [6] J. Song, M. L. Gordin, T. Xu, S. Chen, Z. Yu, H. Sohn, J. Lu, Y. Ren, Y. Duan and D. Wang, Angew. Chem. Int. Ed. 54 (2015) 4325-4329. [7] S. Evers and L. F. Nazar, Acc. Chem. Res. 46 (2012) 1135-1143. [8] D. W. Wang, Q. Zeng, G. Zhou, L. Yin, F. Li, H. M. Cheng, I. R. Gentle and G. Q. 24

M. Lu, J. Mater. Chem. A 1 (2013) 9382-9394. [9] L. Yuan, H. Yuan, X. Qiu, L. Chen and W. Zhu, J. Power Sources 189 (2009) 1141-1146. [10] K. Mi, Y. Jiang, J. Feng, Y. Qian and S. Xiong, Adv. Funct. Mater. 26 (2016) 1571-1579. [11] Y. Li, W. Zhou, H. Wang, L. Xie, Y. Liang, F. Wei, J. C. Idrobo, S. J. Pennycook and H. Dai, Nat. Nanotechnol. 7 (2012) 394-400. [12] L. Chen, Z. Chen, Z. Huang, Y. Wang, H. Zhou and Y. Kuang, New J. Chem. 39 (2015) 8901-8907. [13] M. He, L. X. Yuan, W. X. Zhang and Y. H. Huang, J. Solid State Electrochem. 17 (2013) 1641-1647. [14] X. Liang, C. Y. Kwok, F. Lodi-Marzano, Q. Pang, M. Cuisinier, H. Huang, C. J. Hart, D. Houtarde, K. Kaup and H. Sommer, Adv. Energy Mater. 6 (2016) 1501636. [15] U. N. Maiti, W. J. Lee, J. M. Lee, Y. Oh, J. Y. Kim, J. E. Kim, J. Shim, T. H. Han and S. O. Kim, Adv. Mater. 26 (2014) 40-67. [16] G. Zhou, E. Paek, G. S. Hwang and A. Manthiram, Nat. Commun. 6 (2015) 7760. [17] A. Manthiram, Y. Fu and Y. S. Su, Acc. Chem. Res. 46 (2012) 1125-1134. [18] H. Yuan, X. Chen, G. Zhou, W. Zhang, J. Luo, H. Huang, Y. Gan, C. Liang, Y. Xia, J. Zhang, J. Wang and X. Tao, ACS Energy Lett. 2 (2017) 1711-1719. [19] H. Yuan, W. Zhang, J. Wang, G. Zhou, Z. Zhuang, J. Luo, H. Huang, Y. Gan, C. Liang, Y. Xia, J. Zhang, X. Tao, Energy Storage Mater. 10 (2018) 1-9. [20] H. J. Peng, T. Z. Hou, Q. Zhang, J. Q. Huang, X. B. Cheng, M. Q. Guo, Z. Yuan,

25

L. Y. He and F. Wei, Adv. Mater. Interfaces 1 (2014) 1400227. [21] Y. Qiu, W. Li, W. Zhao, G. Li, Y. Hou, M. Liu, L. Zhou, F. Ye, H. Li and Z. Wei, Nano Lett. 14 (2014) 4821-4827. [22] L. Chen, Z. Chen, Z. Huang, Z. Huang, Y. Wang, H. Li, H. Zhou and Y. Kuang, J. Phys. Chem. C 119 (2015) 28757-28765. [23] Z. H. Sheng, L. Shao, J. J. Chen, W. J. Bao, F. B. Wang and X. H. Xia, ACS Nano 5 (2011) 4350-4358. [24] X. Li, H. Wang, J. T. Robinson, H. Sanchez, G. Diankov and H. Dai, J. Am. Chem. Soc. 131 (2009) 15939-15944. [25] L. Chen, H. Zhou, C. Fu, Z. Chen, C. Xu and Y. Kuang, Int. J. Hydrogen Energy 41 (2016) 21850-21860. [26] Y. Yang, G. Yu, J. J. Cha, H. Wu, M. Vosgueritchian, Y. Yao, Z. Bao and Y. Cui, ACS Nano 5 (2011) 9187-9193. [27] Z. Ma, Z. Li, K. Hu, D. Liu, J. Huo and S. Wang, J. Power Sources 325 (2016) 71-78. [28] L. Sun, D. Wang, Y. Luo, K. Wang, W. Kong, Y. Wu, L. Zhang, K. Jiang, Q. Li and Y. Zhang, ACS Nano 10 (2015) 1300-1308. [29] L. Chen, H. Zhou, S. Wei, Z. Chen, Z. Huang, Z. Huang, C. Zhang and Y. Kuang, RSC Adv. 5 (2015) 8175-8181. [30] M. Q. Zhao, Q. Zhang, J. Q. Huang, G. L. Tian, J. Q. Nie, H. J. Peng and F. Wei, Nat. Commun. 5 (2014) 3410. [31] A. Manthiram, S. H. Chung and C. Zu, Adv. Mater. 27 (2015) 1980-2006.

26

[32] Z. Huang, Z. Liao, W. Yang, H. Zhou, C. Fu, Y. Gong, L. Chen and Y. Kuang, Electrochim. Acta 245 (2017) 957-966. [33] Y. Qiu, W. Li, W. Zhao, G. Li, Y. Hou, M. Liu, L. Zhou, F. Ye, H. Li and Z. Wei, Nano Lett. 14 (2014) 4821-4827. [34] X. Wang, Z. Zhang, Y. Qu, Y. Lai and J. Li, J. Power Sources 256 (2014) 361-368. [35] L. Xiao, Y. Cao, J. Xiao, B. Schwenzer, M. H. Engelhard, L. V. Saraf, Z. Nie, G. J. Exarhos and J. Liu, Adv. Mater. 24 (2012) 1176-1181. [36] F. Sun, J. Wang, H. Chen, W. Li, W. Qiao, D. Long and L. Ling, ACS Appl. Mater. Interfaces 5 (2013) 5630-5638. [37] J. Z. Wang, L. Lu, M. Choucair, J. A. Stride, X. Xu and H. K. Liu, J. Power Sources 196 (2011) 7030-7034. [38] J. Song, Z. Yu, M. L. Gordin and D. Wang, Nano Lett. 16 (2016) 864-870. [39] L. Chen, J. Feng, H. Zhou, C. Fu, G. Wang, L. Yang, C. Xu, Z. Chen, W. Yang and Y. Kuang, J. Mater. Chem. A 5 (2017) 7403-7415. [40] S. Liu, K. Xie, Z. Chen, Y. Li, X. Hong, J. Xu, L. Zhou, J. Yuan and C. Zheng, J. Mater. Chem. A 3 (2015) 11395-11402. [41] Z. Cui, C. Zu, W. Zhou, A. Manthiram and J. B. Goodenough, Adv. Mater. 28 (2016) 6926-6931. [42] Q. Li, Z. Zhang, Z. Guo, Y. Lai, K. Zhang and J. Li, Carbon 78 (2014) 1-9.

27

Graphic abstract

Unique holey structure and effective N doping for N-H-CNTs-2 host promise outstanding cathode performance.

28

Highlights 1. N-H-CNTs-2 was prepared by a controlled etching strategy followed by hydrothermal process. 2. N-H-CNTs-2 possessed unique holey structure and achieved effective N doping. 3. Holey structure and N doping promised dual polysulfides trapping effect. 4. N-H-CNTs-2/S cathode showed excellent cycling lifespan and rate performance.

29