SnCl2 nanofibers as interlayer for LithiumSulfur batteries

SnCl2 nanofibers as interlayer for LithiumSulfur batteries

Journal of Power Sources 412 (2019) 472–479 Contents lists available at ScienceDirect Journal of Power Sources journal homepage: www.elsevier.com/lo...

2MB Sizes 0 Downloads 54 Views

Journal of Power Sources 412 (2019) 472–479

Contents lists available at ScienceDirect

Journal of Power Sources journal homepage: www.elsevier.com/locate/jpowsour

Freestanding oxidized poly(acrylonitrile-co-vinylpyrrolidone)/SnCl2 nanofibers as interlayer for LithiumeSulfur batteries

T

Elif Ceylan Cengiza,b, Osman Ozturkb,c, Serap Hayat Soytasd,∗∗, Rezan Demir-Cakanb,e,∗ a

Department of Material Science and Engineering, Gebze Technical University, 41400, Gebze, Kocaeli, Turkey Institute of Nanotechnology, Gebze Technical University, 41400, Gebze, Kocaeli, Turkey c Department of Physics, Gebze Technical University, 41400, Gebze, Kocaeli, Turkey d Sabanci University SUNUM Nanotechnology Research Center, 34956, Tuzla, Istanbul, Turkey e Department of Chemical Engineering, Gebze Technical University, 41400, Gebze, Kocaeli, Turkey b

H I GH L IG H T S

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

PANVP/SnCl nanofibers are formed • by electrospinning method. PANVP/SnCl is oxidized at 200 °C • and is not further pyrolyzed. As-obtained oPANVP/SnCl mats are • used as freestanding interlayers. Adsorption effect of oPANVP/SnCl is • proven by X-Ray Photoelectron 2

2

2

2

Spectroscopy.

A R T I C LE I N FO

A B S T R A C T

Keywords: Lithium-sulfur batteries Anode protection Interlayer Polysulfides Polyacrylonitrile

LithiumeSulfur batteries are one of the most promising energy storage systems among the next generation batteries due to their high energy density and the natural abundance of sulfur. Nevertheless, some limitations hinder their implementation to the marketplace which are mostly linked to the shuttle effect resulting fast capacity lost and lithium poisoning by dissolved polysulfides. One of the possible solutions is the use of polysulfide adsorptive interlayers between anode and cathode to inhibit the shuttle effect protecting lithium anode. In this work, oxidized poly(acrylonitrile-co-vinylpyrrolidone) nanofibers containing SnCl2 (oPANVP/SnCl2) are used as an interlayer to enhance the performance of LieS cells. Unlike most of the current literature, the electrospun nanofiber mats are oxidized at 200 °C under air, but not further pyrolyzed to benefit from the functional groups and the partial formation of SnOx. 700 mAh g−1 discharge capacity is obtained at C/5 after 100 cycles by using oPANVP/SnCl2, which is higher than the cells with oPANVP and without interlayer. The improved capacity is mostly associated with the complementary adsorption effect of SnCl2 particles, partially formed SnOx and functional groups of oPANVP. Polysulfide adsorption effect of SnCl2, SnOx, and nitrogen- and oxygen-rich functional groups of oPANVP is proven by X-Ray Photoelectron Spectroscopy.



Corresponding author. Department of Chemical Engineering, Gebze Technical University, 41400, Gebze, Kocaeli, Turkey. Corresponding author. Sabanci University SUNUM Nanotechnology Research Center, 34956, Tuzla, Istanbul, Turkey. E-mail addresses: [email protected] (S. Hayat Soytas), [email protected] (R. Demir-Cakan).

∗∗

https://doi.org/10.1016/j.jpowsour.2018.11.082 Received 16 October 2018; Received in revised form 12 November 2018; Accepted 27 November 2018 0378-7753/ © 2018 Elsevier B.V. All rights reserved.

Journal of Power Sources 412 (2019) 472–479

E.C. Cengiz et al.

1. Introduction

metal oxides, since it binds polysulfides via Van der Waals interaction which does not result of any chemical bond. An effective interlayer should have good electrolyte permeation and Li+ ion transport, low weight to avoid contributing to the reduction of overall energy density and strong affinity to polysulfides for keeping them at cathode side. Herein, a multi-functional interlayer composed of PAN-based nanofibers containing pyridinic nitrogen and carbonyl groups and SnOx/SnCl2 as an effective polysulfide adsorber is proposed. As opposed to most of the studies which are focused on carbon-based interlayers, electrospun poly(acrylonitrile-co-vinylpyrrolidone) (quoted as PANVP) containing tin chloride (SnCl2) nanofibers were only oxidized at 200 °C without further carbonization to preserve the nitrogen and oxygen containing groups. Oxidation of the nanofibrous material at 200 °C is also intended to obtain small amount of SnOx together with SnCl2 particles since metal oxides alone might result in destruction of polysulfide species due to their stronger interaction with polysulfides, as discussed above. While small amount of SnOx particles enables strong attraction with polysulfides, SnCl2 particles soften the effect of polysulfides destruction. Moreover, the pyridinic nitrogen and carbonyls of the oxidized PANVP can act as polar hosts to adsorb polysulfides and prevent their migration and dissolution into electrolyte. This suggested nanofibrous, thin and light-weight interlayer (0.6 mg cm−2) that contains pyridinic nitrogen and carbonyl groups as well as SnOx and SnCl2 particles can act as an efficient host to trap polysulfides. The tested LieS batteries with oPANVP/SnCl2 interlayer exhibit 700 mAh g−1 discharge capacity at C/5 after 100 cycles which is higher than both cells with oPANVP interlayer and without interlayer.

With the commercialization of Li-ion batteries, portable electronics have become an indispensable necessity for our daily life. This commercialization has allowed people to be more liberated starting from the use of walkmans and further spread over to mobile phones, laptops and tablets. Therefore, in parallel with the development of technology, energy demand has also increased. Today, we are facing with the threat of depletion of fossil fuels and correspondingly, the idea of using batteries in automobiles (electric vehicles, EVs) have become a critical issue. However, Li-ion batteries cannot meet the energy density needed for large scale applications with their relatively low energy density (∼230 Wh kg−1) [1]. LithiumeSulfur (LieS) batteries have been considered as a promising energy storage system with their very high theoretical energy density and high theoretical capacity. In addition to these, sulfur is very abundant in nature and as a result of this, the system is very cost-effective. Besides these advantages, LieS batteries suffer from some limitations that hinder their use in practice [1,2]. The first challenge is the insulating nature of sulfur which is simply overcome by making their composite with conductive materials like various porous carbon structures [3,4] and conductive polymers [5]. The second obstacle is the relatively huge volume expansion (80%) at the end-of-discharge which causes electrical disconnection between sulfur and conductive part [6]. The last and the most severe challenges of LieS batteries is the dissolution of polysulfides in aprotic solvents which results in shuttle effect causing loss of active materials and rapid capacity fading. Likewise, the use of lithium metal anode brings safety issues due to the formation of dendritic structures. These two important problems were solved by inserting an interlayer between cathode and separator. Manthiram et al. firstly demonstrated this approach with a microporous carbon paper as interlayer [7]. By using this interlayer, not only shuttle effect was prevented, but also cell resistance decreased owing to the conductive nature of carbon-based interlayer. Starting from this work, various types of interlayers were tried which are mainly based on carbon [8–10], Nafion [11–13] and polymer [14]. Typically, carbon-based interlayers are mostly used in the literature. However, carbon materials can physically adsorb polysulfides which is not sufficient enough for enhancing the performance of the cells. Therefore, nitrogen and/or oxygen doped carbon materials were applied [15–17]. Especially, using nitrogen-doped or nitrogen-rich materials are very prevalent because of the chemical interaction between lithium in polysulfides and nitrogen groups resulting LieN bonds. The ratio of pyridinic N in nitrogen-doped structures is very important, since pyridinic N is very effective to bind polysulfides and it mostly provides better electrochemical performance [18]. Recently, it has been shown that the carbonyl group can also very effectively confine polysulfides by forming strong Lewis acid-based chemical bonding [19]. Polyacrylonitrile (PAN) is a valuable carbon source used in battery studies that has been suggested as electrode [20,21], electrode support material [22], electrolyte [23] or interlayer [18,24,25]. Fibrous structures of PAN nanofibers that are produced via electrospinning provide good electrolyte penetration and Li+ ion diffusion. Moreover, cyano groups of nitrogen-rich PAN can be transformed to pyridinic nitrogen through cyclization and oxygenated groups can be formed with a simple heat treatment at 200–300 °C in air, which shows a great potential to strongly interact with polysulfides [18]. Furthermore, electron transfer can be achieved through conjugated structure of cyclized PAN. Oxides, sulfides and chlorides are known for their ability to interact with polysulfides [26]. Especially, metal oxides are very effective to bind polysulfides. They anchor polysulfides via forming bond between oxygen in metal oxides and lithium in polysulfides and the bond strength increases as the chain length decreases. However, very strong bond may result in decomposition of polysulfide species. Thus, using adsorptive materials that have moderate anchoring ability has been suggested [26]. SnCl2 has comparatively lower adsorption ability than

2. Experimental procedures 2.1. Preparation of SnCl2 containing poly(acrylonitrile-covinylpyrrolidone) nanofibers Poly(acrylonitrile-co-vinylpyrrolidone) [P(AN-co-VP), abbreviated as PANVP in the text for simplicity, was synthesized in DMF:H2O (1:1) in the presence of azobis(isobutyronitrile) (AIBN) as radical initiator at 65 °C under inert atmosphere for 4 h, followed by precipitation in ethanol. The resultant copolymer was dried under vacuum at 50 °C until constant weight. For production of hybrid nanofibers, 7 wt% PANVP copolymer was dissolved in DMF followed by the addition of 20 wt% (with respect to solid content) SnCl2.H2O. Homogenous solution was realized with ultrasonication in water bath at 40 °C and mechanical stirring overnight. The prepared solution was electrospun with a flow rate of 0.5 ml h−1 at a voltage of 15 kV collecting nanofibers as a thin fiber mat on a flat collector covered with aluminum foil. The collected fiber mats were heat treated at 200 °C under air with heating rate 5 °C min−1 and kept at the same temperature for half an hour (referred as oPANVP in the text). A schematic of the process and as-obtained interlayer are shown in Fig. 1b. 2.2. Synthesis of Li2S5 powder Stoichiometric amounts of lithium metal and sulfur powders were put in ethylene glycol diethyl ether and the solution was stirred at 150 °C for 2 days to obtain Li2S5 powders. After drying procedure, 0.03 mol L−1 Li2S5 containing 1,3-Dioxolane (DOL) solution was prepared for adsorption test and 0.1 mol L−1 Li2S5 containing 1,3Dioxolane (DOL) solution was prepared for X-Ray Photoelectron Spectroscopy (XPS) measurements. 2.3. Material characterizations NMR spectra were recorded in DMSO‑d6 with Varian Inova 500 MHz NMR spectrometer and FTIR spectra were recorded with Bruker Equinox 55 FTIR spectrometer using attenuated total reflectance (ATR) attachment. 473

Journal of Power Sources 412 (2019) 472–479

E.C. Cengiz et al.

Fig. 1. a) Reaction scheme for PAN during heat treatment under air, b) schematic of SnCl2 containing PANVP (PANVP/SnCl2) and oxidized SnCl2 containing oPANVP (oPANVP/SnCl2).

Aldrich) solvents with a ratio of 1:1 was used as electrolyte. Electrolyte/ sulfur ratio was 30 μL/mg. Interlayers were placed between anode and cathode together with the commercial glass fiber separator whose weight and thickness are 4.7 mg cm−2 and 185 μm, respectively. The interlayer weighed as 0.6 mg cm−2. In our cell configuration, in comparison with sulfur cathode, 2.7-fold excess lithium was employed as anode electrode to eliminate degradation of lithium. Cells were operated between 1.9 V and 3.0 V vs Li+/Li by using Biologic VMP3 galvanostat/potentiostat. Capacities were calculated based on sulfur mass. Cyclic Voltammetry was conducted with a scan rate of 0.1 mV s−1 between 1.7 V and 3.0 V by using Biologic VMP3 galvanostat/potentiostat. Electrochemical Impedance Spectroscopy (EIS) measurements of lithium electrode were performed with three electrode Swagelok™ cells in which lithium was used as a reference electrode placed on a 316L grade stainless steel wire. Both working and reference electrodes were lithium metal. The measurement was performed within the range of 200 kHz to 2 mHz, with a voltage amplitude of 5 mV around 0 V (vs Li+/Li) in Biologic VMP3 galvanostat/potentiostat. Spectra were collected at open circuit voltage (OCV) for every 2 h of rest. Two cells were prepared for comparison; one of them was the cell with oPANVP/SnCl2 and the other one was the cell without interlayer.

Thermal Gravimetric Analysis (TGA) analysis was conducted in nitrogen atmosphere with the heating rate of 2 °C min−1 from room temperature to 500 °C with Shimadzu DTG-60H. Raman spectrum was recorded with Renishaw Raman Spectrometer with the laser wavelength of 532 nm. The microstructural morphology and elemental detection of nanofibers were investigated by JEOL JSM 6010 Scanning Electron Microscopy equipped with Energy Dispersive X-Ray Spectroscopy (EDS). Adsorption test was applied to see the adsorption ability of oPANVP/SnCl2 interlayer visually. To do so, 0.03 mol L−1 Li2S5 containing DOL solution was prepared, then a piece of oPANVP/SnCl2 was added to the solution and the as-prepared solution left for rest to observe the difference. X-Ray Photoelectron Spectroscopy was applied to oPANVP/SnCl2 samples to investigate the chemical composition and adsorption capability of the sample to catch polysulfides. The measurements were conducted by using a Phoibos 150 Specs charged particle analyzer with a conventional Al Kα radiation (hυ = 1486.61 eV) at 300 W (15 kV and 20 mA). To make this experiment, 0.1 mol L−1 Li2S5 containing DOL solution was poured over the oPANVP/SnCl2 and left for drying overnight. After drying, sample was analyzed by XPS. oPANVP/SnCl2 which did not contain polysulfide solution was also investigated to see the elemental structure of the interlayer and for the comparison. For fitting process of the XPS results, the own adapted macro in Igor 4.0 version was used. The background function like Shirley's was used to provide the background-subtracted photoemission spectra in order to eliminate the effect of inelastic background noise. The Gaussian-type functions were used for the fitting process.

3. RESULTS and DISCUSSION Oxidized PANVP/SnCl2 nanofiber mats were used as an interlayer between the separator and the cathode to trap polysulfides. PANVP copolymer was chosen for this work in order to utilize the stabilization effect of vinylpyrrolidone units to facilitate homogeneous distribution of SnCl2 in the electrospun nanofibers since polyvinylpyrrolidone is known for its ability to make complexes with metal particles or metallic cations [27,28]. We have shown previously that the electrostatic interaction of the carbonyl groups (C=O) of PANVP random copolymer with a metal cation facilitates homogeneous distribution of the metal salt within the polymer matrix; therefore, the distribution of SnCl2 and SnOx can be effectively controlled in the electrospun nanofibers of PANVP [29]. The formation of PANVP copolymer was confirmed by FTIR and 1H NMR analyses as discussed in Supporting Information and shown in Figs. S2 and S3. PAN and PAN-based copolymers go through three main reactions during heat treatment under air up to 300 °C, namely, dehydrogenation, cyclization and oxidation, transforming linear polymer chains into a conjugated ladder structure as shown in Fig. 1a [30]. While the PANVP polymer chains were converted to the cyclic structure forming pyrrole and pyridine rings and gaining oxygenated functional groups with a

2.4. Cell assembly and electrochemical measurements A composite which is composed of sulfur (S) and Carbon Ketjen Black was used as cathode. To prepare the cathode electrode, S and Carbon Ketjen Black were mixed in a mortar. After that, the mixture was heat treated under atmospheric conditions at 155 °C for 6 h (melt diffusion). The as-prepared composite contains 63% sulfur which is measured by Thermogravimetric Analysis (TGA). TGA result can be seen in Fig. S1. Total sulfur content in the prepared cells was around 3 mg cm−2. The cells were prepared with two-electrode Swagelok™ cells in a high purity glove box (Innovative Technology, USA). Lithium foil were used as both counter and reference electrode. 1 mol L−1 LiTFSI (from Sigma Aldrich) with 0.5 mol L−1 LiNO3 (from Acros) containing DOL:DME (1,3-Dioxolane and 1,2-Dimethoxyethane, from Sigma474

Journal of Power Sources 412 (2019) 472–479

E.C. Cengiz et al.

Fig. 2. a) FTIR spectra of PANVP/SnCl2 hybrid nanofibers before and after heat treatment at 200 °C, b) Raman spectra of oPANVP/SnCl2 hybrid nanofibers oxidized at 200 °C.

Fig. 3. Scanning Electron Microscopy (SEM) images of a) PANVP/SnCl2, b) oPANVP/SnCl2. Energy Dispersive X-Ray Spectroscopy (EDS) results of c) PANVP/SnCl2 and d) oPANVP/SnCl2.

structures (G-band), respectively. The oxidized hybrid nanofibers also showed these characteristic bands in Raman spectrum, indicating the formation of polyconjugated ladder structures as shown in Fig. 2b. The microstructure and morphology of PANVP/SnCl2 and oPANVP/ SnCl2 nanofibers along with their EDS analysis can be seen in Fig. 3. Fig. 3a and b show the SEM images of the original and the oxidized PANVP/SnCl2 nanofibers, respectively. The electrospinning process yielded uniform nanofibers without any bead formation. PANVP/SnCl2 nanofibers with approximately 210 nm in diameter preserved their structure and size after oxidation at 200 °C. The interconnected fibrous structures enable good electrolyte penetration, good connection with dissolved active materials in electrolyte and Li+ ion diffusion which contribute to the performance of the cell. Elemental mapping analysis of the interlayer proved the homogenous distribution of SnCl2 in the nanofibers as shown in Figs. S4a–f. The EDX spectra of PANVP/SnCl2 and oPANVP/SnCl2 nanofibers demonstrated the partial decomposition of SnCl2.H2O and the formation of SnOx as the intensity of Cl decreased and the intensity of O increased after the oxidation process as shown in

heat treatment at 200 °C, the electrospun PANVP/SnCl2 hybrid nanofiber mat changed its color from white to dark brown, as depicted in Fig. 1b, but preserved its integrity. FTIR spectra of the hybrid fiber mats before and after heat treatment at 200 °C are presented in Fig. 2a. While the intensity of the C≡N stretching in linear copolymer at 2243 cm−1 decreased, new overlapped multiple absorbance bands appeared at around 1300–1600 cm−1. More specifically, the C=N stretching of the cyclized conjugated structure, the = NeH bending in ring and the CeO stretching of the oxidized ring are observed at 1590 cm−1, 1375 cm−1 and 1170 cm−1, respectively, originated from dehydrogenation, cyclization and oxidation reactions. Formation of the C=N groups after the oxidation reaction is critical, since the C=N provides chemical bonding with the polysulfides. Thus, an efficient interaction can be ensured with the polysulfides to prevent rapid capacity fading. Carbon materials typically have two characteristic bands in Raman spectrum at 1340 cm−1 and 1580 cm−1, which are assigned to disordered turbostratic structures (D-band) and ordered graphitic 475

Journal of Power Sources 412 (2019) 472–479

E.C. Cengiz et al.

Fig. 4. Fitted XPS spectra of oPANVP/SnCl2 and polysulfide dropped oPANVP/SnCl2. a) Cl 2p spectra, b) O 1s spectra, c) N 1s spectra and d) C 1s spectra.

obvious in oPANVP/SnCl2 characterized with the peak at 400.2 eV. After treatment with 0.1 M Li2S5/DOL solution, the peak shifted to 399.3 eV and its intensity lowered due to the strong adsorption capability of pyridinic N. The decreased intensity can be associated with the accumulation of polysulfides on the interlayer. Moreover, functional groups of the oxidized copolymer are very effective to catch polysulfides. For the pristine oPANVP/SnCl2, C 1s peaks were deconvoluted into six characteristic peaks at 285.2 eV, 290.1 eV, 286.05 eV, 286.7 eV, 287.4 eV and 288.0 eV, which correspond to CeC and the related π- π* bonds, C≡N, C=N, CeO and C=O, respectively. These functional groups help polysulfides to stay at cathode side by making LieO and LieN bonds. Indeed, the peaks shifted to lower and higher binding energies upon treatment with polysulfide solution. Also, the intensity of the peaks decreased as a result of adsorption. Sn 3d5/2 peak was deconvoluted into two peaks corresponding to SneCl bond at 496.67 eV and SneO bond at 495.7 eV as shown in Fig. S6, which also confirms the partial SnOx formation that contributes to the adsorption effect of the interlayer. Concordantly, after the addition of the Li2S5 solution, the peak related to SneCl bond was shifted to lower binding energy (496.27 eV) and its intensity decreased with the effect of adsorption of polysulfides by SnCl2. Also, the peak associated with SneO bond disappeared due to the strong polysulfides adsorption. For the untreated oPANVP/SnCl2, the peaks at 201.2 eV and 199.6 eV are

Fig. 3c and d, respectively. Also, a relative increase in the intensity of N compare to that of C was observed due to a possible carbon loss resulting from the degradation of vinylpyrrolidone units during the heat treatment. The developed oPANVP/SnCl2 nanofibers contain a nitrogen rich structure, which would adsorb polysulfides effectively by forming LieN bonds, and SnCl2/SnOx, which are known by their ability to capture polysulfides [26]. To demonstrate these complementary adsorption effects, a piece of oPANVP/SnCl2 nanofiber mat was added to a 0.03 mol L−1 Li2S5/DOL solution and then the color changes were observed. As seen in Fig. S5, the dark red color turned very light orange after 3 days of rest visually proving the effective adsorption ability of oPANVP/SnCl2. To gain a deeper understanding of the relationship between the complementary adsorption effect and the chemical structure of oPANVP/SnCl2, the samples treated with Li2S5 solution were analyzed with XPS. The oPANVP/SnCl2 nanofibers were wetted with 0.1 mol L−1 Li2S5/DOL solution and left at room temperature in a glove box for overnight drying. The pristine oPANVP/SnCl2 nanofibers and the polysulfide-treated dried sample were analyzed by XPS. C 1s, N 1s, Sn 3d, O 1s and Cl 2p spectra were fitted as seen in Fig. 4 and Fig. S6. As mentioned above, cyano groups were transformed to pyridinic N with the effect of the oxidation process. As seen in Fig. 4c, pyridinic N is 476

Journal of Power Sources 412 (2019) 472–479

E.C. Cengiz et al.

located at ∼2.4 V (vs Li+/Li) corresponds to conversion of Li2S/Li2S2 into high chain polysulfides. The cell without interlayer had lower current peaks whose intensity decreases as a function of cycle number because of agglomeration of insulating Li2S in the cell compartments. Moreover, the cell with oPANVP had larger current peaks than the cell without interlayer. However, the current peaks for oPANVP also decreased with proceeded cycles which shows that the effect of oPANVP is not sufficient alone. On the other hand, for the case of the cell with oPANVP/SnCl2 interlayer, the active sulfur rearranges itself to more energetically favorable sites after the first cycle [25,31]. The peaks are larger than that of the cells without interlayer and with oPANVP. They were stabilized and overlapped in the subsequent cycles which indicate the stability, high electrochemical kinetics and better utilization of the active materials owing to the use of oPANVP/SnCl2 nanofibrous interlayer. The galvanostatic test results of the cells without interlayer and the cells with oPANVP and oPANVP/SnCl2 interlayers at C/5 current density are shown in Fig. 6. As seen in the third charge-discharge profiles of the cells in Fig. 6a, all cells showed two plateaus during discharge which is typical for LieS batteries. Sulfur is converted to long chain polysulfides at approximately 2.3 V (vs Li+/Li) and long chain polysulfides is converted to short chain polysulfides at around 2.0 V (vs Li+/ Li). In the first cycles, the cell with oPANVP/SnCl2 performed the best showing a discharge capacity of 1030 mAh g−1. Also, the first and second discharge plateaus of the cell with oPANVP/SnCl2 were longer than that of the others indicating the interaction of the oPANVP/SnCl2 with polysulfides in the electrolyte is stronger than that of the others allowing sulfur to be more effectively used during electrochemical reactions. As a result of these, the cell with oPANVP/SnCl2 interlayer has performed the best compared to the cell with oPANVP interlayer and without interlayer, which can be seen in Fig. 6b. After 100 cycles, the cell with oPANVP/SnCl2 interlayer retained a high discharge capacity of 700 mAh g−1 (68% of its first capacity), which is much higher than that of the cell with oPANVP and without interlayer. On the other hand, while the cell with oPANVP interlayer could reach a discharge capacity of only 466 mAh g−1, discharge capacity of the cell without interlayer reduced below 336 mAh g−1 at 76 cycles, after which the test was discontinued. The remarkable performance improvement of oPANVP/ SnCl2 as an interlayer can be attributed to the complementary polysulfide interaction effects of SnCl2 with a partial amount of SnO2 nanoparticles and N-rich structure of oPANVP resulting in better sulfur utilization. Oxidized PANVP contains oxygen containing functional groups which also help to obtain better electrochemical performance. The cell with oPANVP/SnCl2 interlayer was cycled at C/2 to observe the electrochemical performance at higher current rates for long cycles. The charge-discharge profile showed typical LieS plateaus as seen in Fig. 6c. Discharge capacities as a function of cycle number at C/2 is shown in Fig. 6d. The cell resulted in a discharge capacity of 1091 mAh g−1 at the first cycle and retained its discharge capacity of 440 mAh g−1 after 175 cycles. Coulombic efficiency was in the range of ∼96%. The rate capability test of the cell with oPANVP/SnCl2 interlayer can be seen in Fig. 6e. The cell was first cycled at C/10, then the rate was increased to C/5, C/2 and 1C. At C/10, a high discharge capacity of 1458 mAh g−1 was achieved. When the current rate was increased to C/ 5, C/2 and 1C, 1170 mAh g−1, 757 mAh g−1 and 480 mAh g−1 discharge capacities were obtained, respectively. When the current density was reduced back to C/5 after 20 cycles, the discharge capacity was recovered as 1064 mAh g−1. Moreover, coulombic efficiency approached to nearly 100% when the cell was cycled at 1C. This improved electrochemical performance of the cell indicates the excellent reversibility and efficient trapping of polysulfides by oPANVP/SnCl2 interlayer with its functional groups and N-rich structure. Electrochemical Impedance Spectroscopy (EIS) was conducted to observe the changes on lithium surface in the presence of oPANVP, oPANVP/SnCl2 interlayer and without interlayer. Impedance spectra were recorded with 2 h interval at open circuit voltage (OCV) using

Fig. 5. Cycling Voltammetry (CV) results of a) without interlayer, b) with oPANVP and c) oPANVP/SnCl2 at a scan rate of 0.1 mV s−1.

related to SneCl 2p1/2 and 2p3/2, respectively, as shown in Fig. 4a. After the addition of polysulfides, the peaks were also shifted to lower binding energies and their intensities decreased as a result of adsorption. SnOx formation can also be observed in the O 1s spectrum of the pristine oPANVP/SnCl2 interlayer (Fig. 4b). O 1s peak of the pristine oPANVP/SnCl2 was deconvoluted to three peaks at 531.3 eV, 532.8 eV and 534.3 eV corresponding to SnOx, CeO and C=O, respectively. These peaks were also shifted, and their intensities decreased with the effect of adsorption. These peak shifts and decreased intensities show the strong adsorption ability of the oPANVP/SnCl2 interlayer through its oxygen and nitrogen containing functional groups, and SnCl2 and SnOx particles. The cyclic voltammograms of the cells without interlayer, with oPANVP and oPANVP/SnCl2 interlayers at a scan rate of 0.1 mV s−1 are presented in Fig. 5. All cells showed two main reduction peaks at ∼2.3 V (vs Li+/Li) and ∼2.0 V (vs Li+/Li) which represent the conversion of sulfur into long chain polysulfides and long chain polysulfides into short chain polysulfides, respectively. At anodic scan, peak

477

Journal of Power Sources 412 (2019) 472–479

E.C. Cengiz et al.

Fig. 6. a) Third charge-discharge profile of the cells without interlayer, with oPANVP and oPANVP/SnCl2 at C/5 rate, b) discharge capacities as a function of cycle number for the cells without interlayer, with oPANVP and oPANVP/SnCl2 at C/5 rate. c) First charge-discharge profile of cell with oPANVP/SnCl2 at C/2 rate, d) discharge capacities as a function of cycle number for the cell with oPANVP/SnCl2 at C/2 rate. e) Rate capability test of the cell with oPANVP/SnCl2.

with oPANVP and oPANVP/SnCl2 interlayers were lower than the cell without interlayer after 60 h of rest. The cell with oPANVP/SnCl2 had slightly reduced semicircles than oPANVP, as expected, due to the additional polysulfide adsorption of SnCl2 and SnOx. These results show the improved lithium/electrolyte interface and decreased charge transfer resistance with the strong polysulfide adsorption ability of oPANVP/SnCl2 interlayer.

three electrode Swagelok™ cells. Nyquist plots of the cells at the beginning and after 60 h can be seen in Fig. 7a and Fig. 7b, respectively. The fitting equivalent circuit is shown in Fig. 7c. All cells showed two semicircles at high to medium and medium to low frequencies. Lithium electrodes of all cells showed a depressed semi-circle at high frequency which shows impedance coming from solid electrolyte interphase (RSEI), a second semicircle at medium-to-low frequency presenting resistance from charge transfer resistance (Rct). Rohm represents the ohmic losses. At the beginning, the cells with oPANVP and oPANVP/SnCl2 presented lower resistance with respect to the cell without interlayer (Table 1). This can be due to the sp2 hybridized conjugated structure of oPANVP and oPANVP/SnCl2, which was proven by Raman Spectroscopy showing D and G bands (Fig. 2b), providing relatively high conductivity. After 60 h rest time, both Nyquist plots tended to grow which can be attributed to the interaction of the electrolyte and/or asformed polysulfides with metallic lithium. The semicircles of the cell

4. Conclusions In this work, nanofibrous PANVP/SnCl2 oxidized at 200 °C was used as an interlayer and its effect on the electrochemical performance of LieS batteries was observed. The low-temperature heat treatment of PANVP resulted in the formation of oxygen-containing functional groups and the transformation of nitrogen rich cyano groups into pyridinic nitrogen which ensured better electrochemical performance by 478

Journal of Power Sources 412 (2019) 472–479

E.C. Cengiz et al.

Fig. 7. Electrochemical Impedance Spectra of the cells without interlayer, with oPANVP and oPANVP/SnCl2 a) at the beginning, b) after 60 h, c) the fitting equivalent circuit.

Appendix A. Supplementary data

Table 1 The results obtained from fitted Electrochemical Impedance Spectra (EIS) (from Fig. 6). Sample

Hour

Rohm (Ohm)

RSEI (Ohm)

Rct (Ohm)

Without Interlayer

0 60 0 60 0 60

9.524 7.751 5.762 4.952 12.840 12.440

43.490 199.400 36.110 125.700 25.850 116.100

35.860 99.190 42.270 48 35.110 45.370

oPANVP oPANVP/SnCl2

Supplementary data to this article can be found online at https:// doi.org/10.1016/j.jpowsour.2018.11.082. References [1] [2] [3] [4] [5] [6] [7] [8]

making strong chemical interactions with polysulfides. In addition, the low-temperature heat treatment of PANVP/SnCl2 resulted in the formation of SnOx which contributed to the adsorption of the dissolved polysulfide species without causing any decomposition. The effect of adsorption was proven by XPS measurements showing binding energy shifts and a decrease in the intensity of corresponding peaks in the presence of polysulfides. Owing to these complementary effects, 700 mAh g−1 discharge capacity was obtained after 100 cycles at C/5 with oPANVP/SnCl2 interlayer which was higher than the discharge capacity of the cell with oPANVP interlayer and the cell without interlayer. In addition to the improved performance, the proposed interlayer has a very low weight (0.6 mg cm−2) which does not contribute negatively to the overall energy density. Therefore, this interlayer can be applied to different types of batteries which have working mechanism similar to LieS batteries, such as metal disulfide-Na batteries, room temperature NaeS batteries or LieSe batteries.

[9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21]

Conflicts of interest

[22]

The authors declare no conflict of interest.

[23] [24]

Acknowledgements

[25] [26] [27]

The Authors thanks the partial financial supports from the bilateral project between Gebze Technical University and Sabanci University (project contract no: 2016-GTU-SU-05 and I.A.SN-16-01589). The authors thanks to Ahmet NAZIM for Scanning Electron Microscopy and Ali Ansari Hamedani for Raman measurements. This work is a part of the PhD thesis of Elif Ceylan CENGIZ.

[28] [29] [30] [31]

479

R. Demir-Cakan, World Scientific Publishing Company, (2017) ISBN-10: 9781786342492. O. Salihoglu, R. Demir-Cakan, J. Electrochem. Soc. 164 (2017) A2948–A2955. X.L. Ji, K.T. Lee, L.F. Nazar, Nat. Mater. 8 (2009) 500–506. G. Zheng, Y. Yang, J.J. Cha, S.S. Hong, Y. Cui, Nano Lett. 11 (2011) 4462–4467. Y. Yang, G. Yu, J.J. Cha, H. Wu, M. Vosgueritchian, Y. Yao, Z. Bao, Y. Cui, ACS Nano 5 (2011) 9187–9193. R. Demir-Cakan, J. Power Sources 282 (2015) 437–443. Y.S. Su, A. Manthiram, Nat. Commun. 3 (2012). J. Balach, T. Jaumann, M. Klose, S. Oswald, J. Eckert, L. Giebeler, J. Phys. Chem. C 119 (2015) 4580–4587. J. Zhang, H. Li, Z. Lin, Q. Tang, W. Qi, L. Wang, H. Zheng, K. Zhou, RSC Adv. 7 (2017) 39172–39177. S.-H. Chung, A. Manthiram, Chem. Commun. 50 (2014) 4184–4187. Z. Jin, K. Xie, X. Hong, Z. Hu, X. Liu, J. Power Sources 218 (2012) 163–167. I. Bauer, S. Thieme, J. Brückner, H. Althues, S. Kaskel, J. Power Sources 251 (2014) 417–422. T.-Z. Zhuang, J.-Q. Huang, H.-J. Peng, L.-Y. He, X.-B. Cheng, C.-M. Chen, Q. Zhang, Small 12 (2016) 381–389. G. Ma, Z. Wen, J. Jin, M. Wu, X. Wu, J. Zhang, J. Power Sources 267 (2014) 542–546. L. Wang, Z. Yang, H. Nie, C. Gu, W. Hua, X. Xu, X.a. Chen, Y. Chen, S. Huang, J. Mater. Chem. 4 (2016) 15343–15352. L.-B. Xing, K. Xi, Q. Li, Z. Su, C. Lai, X. Zhao, R.V. Kumar, J. Power Sources 303 (2016) 22–28. C. Reshma, H. Haritha, R. Murugan, G.J. Mary, S.A. Manuel, Chemistry Select 2 (2017) 10484–10495. Q. Li, M. Liu, X. Qin, J. Wu, W. Han, G. Liang, D. Zhou, Y.-B. He, B. Li, F. Kang, J. Mater. Chem. 4 (2016) 12973–12980. G. Li, X. Wang, M.H. Seo, M. Li, L. Ma, Y. Yuan, T. Wu, A. Yu, S. Wang, J. Lu, Z. Chen, Nat. Commun. 9 (2018) 705. M. Li, Y. Chen, Z.-H. Huang, F. Kang, J. Nanomater. (2014) 6. J.K. Lee, K.W. An, J.B. Ju, B.W. Cho, W.I. Cho, D. Park, K.S. Yun, Carbon 39 (2001) 1299–1305. Y. Zhang, Y. Zhao, A. Yermukhambetova, Z. Bakenov, P. Chen, J. Mater. Chem. 1 (2013) 295–301. D. Peramunage, K.M. Abraham, J. Electrochem. Soc. 145 (1998) 2609–2615. J.-Q. Huang, B. Zhang, Z.-L. Xu, S. Abouali, M. Akbari Garakani, J. Huang, J.-K. Kim, J. Power Sources 285 (2015) 43–50. R. Singhal, S.-H. Chung, A. Manthiram, V. Kalra, J. Mater. Chem. 3 (2015) 4530–4538. Q. Zhang, Y. Wang, Z.W. Seh, Z. Fu, R. Zhang, Y. Cui, Nano Lett. 15 (2015) 3780–3786. U. Pal, J.F. Sanchez Ramirez, R. Sivla Gonzalez, J.R. Diaz-Estrada, Appl. Phys. A 80 (2005) 477–481. E. Dı́az, R. Valenciano, P. Landa, J. Luis Arana, J. González, Polym. Test. 21 (2002) 247–251. M. ShakooriOskooie, Y.Z. Menceloglu, S. Unal, S. Hayat Soytas, ACS Appl. Nano Mater. 1 (11) (2018) 6236–6246. Z. Fu, B. Liu, L. Sun, H. Zhang, Polym. Degrad. Stabil. 140 (2017) 104–113. Y.-S. Su, A. Manthiram, Chem. Commun. 48 (2012) 8817–8819.