Electrochimica Acta 247 (2017) 851–859

CuS Microspheres as High-Performance Anode Material for Na-ionBatteries

He Li, Yunhui Wang, Jiali Jiang, Yiyong Zhang, Yueying Peng, Jinbao Zhao*State Key Lab of Physical Chemistry of Solid Surfaces, Collaborative Innovation Centre of Chemistry for Energy Materials, State-Province Joint EngineeringLaboratory of Power Source Technology for New Energy Vehicle, College of Chemistry and Chemical Engineering, Xiamen University, Xiamen, 361005, PRChina

A R T I C L E I N F O

Article history:Received 20 February 2017Received in revised form 24 June 2017Accepted 4 July 2017Available online 8 July 2017

Keywords:Copper sulfideMicrowave synthesisNa-ion batteriesReaction mechanismElectrochemical kinetics

A B S T R A C T

In this work, CuS microspheres are synthesized by a facile microwave synthesis without any template oradditive. As-prepared CuS microspheres display steady reversibility by adopting triethylene glycoldimethyl ether (TEGDME) as the electrolyte solvent and adjusting cut-off voltage to 0.6 � 3.0 V. Thedischarge capacity stands at 162 mAh g�1 after 200 cycles with the capacity retention of 95.8%.Electrochemical kinetics of CuS electrodes is investigated by electrochemical impedance spectroscopy(EIS), cyclic voltammetry (CV) and galvanostatic intermittent titration technique (GITT) tests. Beyondthat, electrochemical reactions of CuS electrodes are explored using ex-situ X-ray diffraction (XRD). As faras we know, it is not only the best performance of CuS electrodes in Na-ion batteries (NIBs) but also thefirst time that the kinetics and reaction mechanism of CuS in NIBs are investigated.

© 2017 Elsevier Ltd. All rights reserved.

Contents lists available at ScienceDirect

Electrochimica Acta

journa l home page : www.e l sev ier .com/ loca te /e le cta cta

1. Introduction

With the rapid consumption of non-renewable fossil fuels andthe issue of environmental pollution, renewable and clean energysuch as wind and solar energy has gained much attention [1]. Largescale energy storage systems are required to store the electricitygenerated from wind turbines and solar cells, which can be used asgreen and renewable energy resources [2,3]. Although lithium-ionbatteries (LIBs) can be regarded as a solution to meet thesechallenges theoretically, yet the insufficiency (the content oflithium in the earth’s crust is only about 20 ppm), maldistribution(lithium is mainly distributed in South America in the earth’s crust)and high price of lithium are blocking LIBs to be used in large scaleenergy storage [4]. Sodium is one of the most abundant elementsin the earth’s crust (over 23 000 ppm) and sodium resources areunlimited everywhere in the earth’s crust [5]. Sodium is the secondlightest alkali metal next to lithium. On the other hand, thestandard electrode potential of Na/Na+ (-2.71 V vs SHE) is close tothat of Li/Li+ (-3.04 V vs SHE) [6]. Due to its abundance andimpressive properties, rechargeable Na-ion batteries (NIBs) areideal substitutes for LIBs [7].

Finding suitable electrode materials for NIBs, especially anodematerials, has always been an important issue for the development

* Corresponding author. Fax: +86 05922186935; Tel: +86 05922186935.E-mail address: jbzhao@xmu.edu.cn (J. Zhao).

http://dx.doi.org/10.1016/j.electacta.2017.07.0180013-4686/© 2017 Elsevier Ltd. All rights reserved.

of NIBs, and a lot of work have been done on this issue [8–12].Natural graphite, which is the most commonly used anodematerial for LIBs, fails to match NIBs on account of the largersize of sodium ion as well as the thermodynamic factors and thehigher plating potential of sodium ion[13,14]. Although disorderedcarbon (hard carbon) is investigated as the anode material for NIBs,the cyclability is unsatisfying for commercial use [15,16]. It remainsvital and challenging to find proper anode materials for NIBs.

Owing to their low cost and high theoretical capacities, metalsulfides have been widely studied as anode materials for NIBs [17–20]. For example, Zhu et al. applied MoS2 nanoplates as the anodematerial for NIBs [21]. Hu et al. prepared long-life NIBs by usingpyrite FeS2 as the anode material [22]. Kim et al. studied thesodium storage property of Cu2S at room temperature [23]. Byadopting ether based electrolyte, the capacity stood at 220 mAhg�1 after 20 cycles and the reaction mechanism was preliminarilyinvestigated. CuS has been widely explored as the anode materialfor LIBs, it shows high discharge capacity and outstanding cycle life[24–28]. CuS could also be one of the potential anode materialcandidates for NIBs. Klein et al. successfully assembled CuS/Nabattery whose initial discharge capacity could reach up to750 mAh g�1 [29]. However, the capacity faded quickly, thedischarge capacity dropped to 85 mAh g�1 after 5 cycles whichmay be caused by the use of carbonate based electrolyte and sidereactions due to the low cut-off voltage (0.01 � 3.0 V). More effortsshould be paid to fulfil the application of CuS in NIBs.

852 H. Li et al. / Electrochimica Acta 247 (2017) 851–859

Herein we report a facile way to synthesize CuS microsphereswith simple raw materials under microwave irradiation. Byadopting triethylene glycol dimethyl ether (TEGDME) as theelectrolyte solvent and adjusting cut-off voltage to 0.6 � 3.0 V, thedischarge capacity of as-prepared CuS microspheres stands at162 mAh g�1 after 200 cycles with the capacity retention of 95.8%.Synthetic CuS shows outstanding cycle performance whichsurpasses previous work by Klein et al. greatly. In addition, theelectrochemical kinetics of the batteries are investigated byelectrochemical impendence spectroscopy (EIS), cyclic voltamme-try (CV) and galvanostatic intermittent titration technique (GITT)tests. Beyond that, the electrochemical reactions of CuS electrodesare primarily investigated by ex-situ X-ray diffraction (XRD).

2. Experimental

2.1. Raw materials and synthesis

Cu(NO3)2�H2O (AR) and Na2S2O3�5H2O (AR) were purchasedfrom Shanghai Chemical Co. Ltd. Sodium metal (99%) and NaClO4

(99.99%, water free) were purchased from Aladdin IndustrialCorporation. Acetylene black (battery grade), polypropylenefluorides (PVDF, battery grade) and N-methyl-2-pyrrolidene(NMP, battery grade) were purchased from Guangzhou SongbaiChemical Industrial Co. Ltd. Ethylene carbonate (EC, battery grade),diethyl carbonate (DEC, battery grade) and TEGDME (batterygrade) were bought from Zhangjiagang Guotai-Huarong NewChemical Materials Co. Ltd. All the chemicals were used withoutfurther purification.

CuS microspheres were prepared under microwave irradiationin this work, where Cu(NO3)2 was operated as copper source,Na2S2O3was used as sulfur source and deionized water was used asthe solvent. 25 mL 2.0 M Cu(NO3)2 solution and 25 mL 2.0 MNa2S2O3 solution were mixed together in a 100 mL flat-bottomedflask to form a transparent emerald solution. This solution wasplaced into domestic microwave reactor (microwave frequency:2450 MHz, output power: 650 W), treated with microwave pulsefor 30 minutes where duty ratio was set to 6/30 (6 seconds on,24 seconds off). Black powders were collected afterwards, washedby deionized water and dehydrate ethanol for several times, thendried overnight in vacuum oven at 60 �C.

2.2. Characterization

The XRD patterns were recorded with Rigaku Ultima IV (RigakuCorporation, Japan) with CuKa radiation operated at 40 kV, 30 mA.Scanning rate was set to 2� min�1 (2 u). Scanning electronmicroscopy (SEM) and energy-dispersive X-ray spectrum (EDS)

Fig. 1. the XRD pattern of as-p

were performed on Hitachi S-4800 (Hitachi Corporation, Japan).Transmission electron microscopy (TEM) and selected areaelectron diffraction (SAED) pattern were tested with the JEM-2100 (JEOL, Japan) at 200 kV. The surface area and pore volumewere measured using a micromeritics ASAP 2020 surface andporosity analyzer at 77 K. Surface areas was calculated using theBrunauer�Emmett�Teller method.

2.3. Electrochemical studies

The electrochemical characteristics were tested using CR2032-type coin cells. The active materials, acetylene black, and PVDFwere mixed in a weight ratio of 70: 15: 15 in NMP. The slurry wascoated on copper foil and dried in vacuum oven at 60 �C for12 hours. The electrode sheet was punched into 12 mm discs, andpressed at 16 MPa to make the active materials contact tighter withthe current collector. The loading mass of active material is 1.82 �2.10 mg per electrode (1.61 � 1.86 mg cm�2). Half cells wereassembled using the prepared electrodes as positive electrode andsodium metal as negative electrode. The electrolyte was preparedby dissolving NaClO4 (1 mol L�1) in the mixture solvent of EC andDEC (1:1, V/V), or dissolving NaClO4 (1 mol L�1) in TEGDME.Celgard 2400 (Celgard, LLC, US) was used as separator. All cellswere fabricated in an argon-filled glove box (Mbraun, Germany).The cyclic voltammograms were tested on the CHI 1030C multi-potentiostat (Shanghai Chenhua Electrochemical Instruments,Shanghai, P. R. China). The galvanostatically charge-discharge testsand GITT test were carried out on the Neware CT-3008W batterytest system (Neware Battery Testing Instruments, Shenzhen, P. R.China). The EIS tests were performed on the Solartron 1260A and1287A impedance/Gain-phase Analyzer (AMETEK, UK) withsodium metal as the reference and counter electrode. All theelectrochemical tests were carried out at room temperature(25 �C).

2.4. Calculation procedure

All calculations reported herein were performed using theGaussian 09 computational package. All geometries were fullyoptimized at the B3PW91 level of DFT. There is no imaginaryfrequency when frequency calculations were performed at theB3PW91. The lanL2DZ basis set was used to describe S, and thestandard 6-311G+(d, p) basis set was used for H, C, and O

3. Results and discussions

In a typical microwave synthesis process, CuS nanoparticles areformed under microwave irradiation firstly, then the nanoparticles

repared CuS microspheres.

H. Li et al. / Electrochimica Acta 247 (2017) 851–859 853

aggregate into microspheres. According to the work by Nafees et al.[30], the chemical equations taken place in this reaction are:

Cu(NO3)2! Cu2+ + 2NO3� (1)

Na2S2O3! 2Na+ + S2O32� (2)

S2O32� + H2O ! HS�+ HSO4

� (3)

Cu2+ + HS�! CuS# (nanoparticles) + H+ (4)

CuS (nanoparticles)! CuS (microspheres) (5)

The structure and morphology information were characterizedby XRD, SEM and TEM. The XRD pattern of as-prepared material isshown in Fig. 1. The diffraction lines correspond well with thestandard PDF card of hexagonal CuS (JCPDS No. 78-0876),indicating pure phase of synthetic CuS. Fig. 2a, 2b display thelow and high magnification SEM images. The particle size of as-prepared CuS microspheres is around 2 mm. The CuS microspheresare assembled by nano-sized primary particles, which also can bedemonstrated from the TEM image (Fig. 2c). The HR-TEM imageand SAED pattern (inset) are shown in Fig. 2d. The SAED patternwith clear rings indicates the polycrystalline structure of CuSmicrospheres formed by nanoparticles. The d-spacing of 0.281 and0.305 nm correspond with the (103) and (102) lattice planes of CuS,respectively. Energy-dispersive X-ray spectrum (EDS) and elemen-tal mapping images of as-prepared CuS are shown in Fig. 3,demonstrating the uniform distribution of Cu and S in the CuS

Fig. 2. (a)&(b) low and high magnification SEM images of as-prepared CuS microsphermicrospheres.

microspheres, the elemental ratio of Cu to S is 1.04, close totheoretical ratio 1.

According to former work by Klein et al., CuS shows high initialdischarge capacity, yet the capacity fades quickly. We have noticedthat in their work, carbonate based electrolyte is used and the cut-off voltage is set to 0.01 � 3.0 V. As is often the case, the mismatchof electrolyte and inappropriate cut-off voltage may cause sidereactions between electrolyte and active materials, further lead topoor battery performance. To optimize the electrochemicalperformance of CuS electrodes, a series of experiments werecarried out. Firstly, batteries with various cut-off voltage weretested using CR-2032 coin-type batteries with the current densityof 50 mA g�1. When the cut-off voltage is adjusted to 0.1 � 3.0 V, thebattery shows high capacity, yet the capacity fades quickly in 30cycles and fails after 40 cycles (Fig. 4a). This can be ascribed to thedecomposition of electrolyte and other side reactions like theformation of solid electrolyte interface (SEI) etc. [29]. Bydischarging the battery to 1.0 V and charging to 3.0 V, the batteryshows much better cycle stability, yet the capacity is much lower(Fig. 4b), which can be owed to insufficient electrochemicalreactions. When the cut-off voltage is set to 0.6 � 3.0 V, a relevantlyhigh capacity and outstanding cycle performance can be obtained(Fig. 4c). The illustrative comparison of CuS electrodes dischargedto different voltage is shown in Fig. 4d.

Different types of electrolytes were tested with the currentdensity of 50 mA g�1 between 0.6 � 3.0 V. Fig. 4e exhibits thecharge and discharge curves of CuS/Na battery using the carbonatebased electrolyte, where NaClO4 is dissolved in the mixture solventof EC and DEC. The initial cycle shows good electrochemicalactivity, yet the capacity declines quickly after several cycles. Thisphenomenon matches with the results of Klein et al. [29]. Fig. 4bdisplays the charge and discharge curves of CuS/Na battery inether-based electrolyte, where NaClO4 is dissolved in TEGDME.

es, (c) TEM image, (d) HR-TEM image and SAED pattern (inset) of as-prepared CuS

Fig. 3. (a) the EDS spectrum of as-prepared CuS microspheres, (b) the normalized weight and atomic percentage of S and Cu in the sample, (c), (d) &(e) the EDS mapping of CuSsample.

854 H. Li et al. / Electrochimica Acta 247 (2017) 851–859

Battery shows impressive reversibility where discharge capacity of168 mAh g�1 can be maintained after 50 cycles. The distinctdifference of cycle performances (Fig. 4f) demonstrates thatTEGDME is a more suitable electrolyte solvent for CuS. Weconjecture that the tremendous difference of cycle performance isrelated to the compatibility of CuS with electrolytes. According tothe research by Klein et al. and our ex-situ XRD results below,conversion reactions take place when CuS is applied in NIBs, thus,sodium sulfide/polysulfide would be generated during electro-chemical reactions [29]. Generated S2�/Sn2� (2 � n � 8) tends toreact with carbonate solvent, resulting in the poor cyclicperformance of CuS/Na batteries [31,32]. On the contrary, theseanionic groups would not react with TEGDME (Scheme S1). On theother hand, CuS microspheres are packed by nano-sized primaryparticles as we have mentioned. The surface is determined to be7.876 m2g�1, and porous structure can be deduced from theadsorption/desorption isotherm (Fig. S1). According to Z. Hu et al.,TEGDME with 1D structure would boost the performance of anodematerials because of lower reaction energy barrier [22]. These mayexplain why CuS shows much better electrochemical performancein ether based electrolyte than carbonate based electrolyte.

Ex-situ XRD tests were carried out to investigate theelectrochemical reactions of the CuS electrodes. Fig. 5a showsthe XRD patterns of CuS electrodes at different charge-dischargestate, which corresponds with the discharge-charge capacity-voltage curve in Fig. 5b. XRD pattern 1 represents the pristine CuSelectrode, where only diffraction lines correspond with hexagonalCuS can be observed, the slope from 20 � 30� (2 theta) is the XRDsignals of conductive carbon black and binder added in theelectrode. Pattern 2 corresponds the CuS sample discharged to1.5 V, where diffraction lines attributed to Cu2-xS (0 � x � 0.2) canbe detected, which means CuS may firstly converts into non-stoichiometric Cu2-xS. Pattern 3 is the diffraction pattern of fullydischarged CuS electrode, where diffraction lines attributed toNa2S and some intermediates (marked with *, mainly composed byNaaCubSg, where a, b, g represent integers) can be observed,diffraction lines corresponded with metallic Cu also can be seen. Itis highly possible that Na+ would react with Cu2-xS to form theintermedia NaaCubSg firstly, then NaaCubSg converts into Na2Sand Cu. Possible electrochemical reactions during dischargeprocess are shown as below:

(2-x)CuS + (2x-2)Na+ + (2x-2)e�! Cu2-xS + (1-x)Na2S (6)

Cu2-xS + Na+ + e�! intermediates ! Cu + 0.5Na2S (7)

the overall reaction is:

CuS + 2Na+ + 2e�! Cu + Na2S (8)

this conjecture matches with the CV curves below. Pattern 4 showsthe electrode charged to 2.0 V after fully discharge, where we cantell that intermediates and some Cu2-xS were formed during thecharge process. Pattern 5 is the diffraction pattern of the CuSelectrode charged to 3.0 V, the main diffraction lines can beattributed to Cu2-xS. The possible electrochemical reactions in thecharge process are listed below:

(2-x)Cu + Na2S ! intermediates ! Cu2-xS + 2Na+ + 2e� (9)

instead of CuS, Cu2-xS are generated after charging. The initialdischarge process can be decomposed into irreversible part (CuS toCu2-xS) and reversible part (Cu2-xS to Cu), the transformation fromCuS to Cu2-xS is irreversible, which mainly leads to the low initialcolumbic efficiency [26]. The second and subsequent cycles areprimarily the reversible transition between Cu2-xS and Cu (Fig. S2).This property is similar with CuO, which has been reported byWang et al. [33].

Cyclic voltammetry (CV) was carried out to investigate thesodium storage property of as-prepared CuS electrode. Theoptimized TEGDME was used as the electrolyte solvent. The CVcurves at scan rate of 0.2 mV s�1 in the voltage range of 0.6 � 3.0 Vare shown in Fig. 6a. In the first cathodic scan, a small reductionpeak at 1.85 V can be observed, which corresponds with theconversion from CuS to Cu2-xS. This reaction is irreversible, so thepeak at 1.85 V disappears gradually in the subsequent scans. Twomajor peaks at 1.12 V and 0.84 V represent the reaction of Cu2-xSinto intermediary NaaCubSg and the reaction from intermediaryNaaCubSg to Na2S and Cu, respectively. In the anodic scan, twomajor oxidation peaks can be observed. The anodic peak at 1.60 Vrepresents the conversion from Cu and Na2S to the intermediaryNaaCubSg, while the peak at 2.10 V corresponds with theconversion reaction from intermediary NaaCubSg to Cu2-xS. Thisreaction mechanism can also be supported by ex-situ XRD carried

Fig. 4. Charge and discharge curves of CuS electrodes in ether based (TEGDME) electrolyte with different cut-off voltage: (a) 0.1 � 3.0 V, (b) 1.0 � 3.0 V, (c) 0.6 � 3.0 V, (d) thecomparison of cyclic performance with different cut-off voltage, (e) Charge and discharge curve of CuS electrode in carbonate based (EC/DEC) electrolyte, (f) the distinctcomparison of cycle performance in different electrolytes.

H. Li et al. / Electrochimica Acta 247 (2017) 851–859 855

out above. In subsequent scans, reduction peaks shift to morepositive potential while oxidation peaks shift to lower potential,which demonstrates smaller polarization of the electrode. Thesmaller polarization indicates better reversibility of the electrode.It is worth noting that a small peak at 1.65 V arouses in thesubsequent anodic scans, which may correspond with the phasetransition of various intermediaries. The dQ/dV curves of CuSelectrode in the initial cycle and the 20th cycle show similar results(Fig. S3).

The electrochemical stability evaluation was carried out withthe current density of 200 mA g�1. The discharge capacity after 200cycles remains at 162 mAh g�1 with the capacity retention of 95.8%

(which is compared with the 2nd discharge capacity, eliminatinginitial irreversible capacity), showing remarkable cyclic stability(Fig. 6b). The charge and discharge curves with different currentdensities are shown in Fig. 6c. The discharge capacities withcurrent density of 50, 100, 200, 400, 800 and 1600 mA g�1 are 180,157, 140, 120, 100 and 90 mAh g�1, respectively. The capacity canreturn to 177 mAh g�1 when the current density is restored to50 mA g�1, indicating good reversibility of as-prepared CuS in thecase of high current density (Fig. 6d). However, the capacity dropssignificantly when operated at high current density, this may becaused by insufficient electronic conductivity, which can bepossibly improved by extra effort such as preparation of CuS/

Fig. 5. (a) Ex-situ XRD patterns of CuS electrodes at different states during the first discharge-charge process, (b) charge-discharge curve of CuS electrode in the initialdischarge-charge process.

Fig. 6. (a) Cyclic voltammograms of CuS electrode between 0.6 � 3.0 V, (b) the cyclic performance of CuS/Na battery at 200 mA g�1, (c) charge-discharge curves, (d) rateperformance of the CuS/Na battery at different current densities from 50 to 1600 mA g�1.

856 H. Li et al. / Electrochimica Acta 247 (2017) 851–859

carbon composites (like CuS/carbon nano tubes or CuS/graphene,etc.) [34,35].

To investigate the electrochemical reaction kinetics of theelectrodes, electrochemical impendence spectroscopy (EIS) wascarried out (Fig. 7a). The high frequency semicircle in the Nyquistplot is related with the charge transfer resistance of the electrode,while the spike at the low frequency is an indication of Warburgimpedance of long-range Na-ion diffusion. The charge transferresistance decreases from 425 to 289 V after 20 cycles (enlarged

Fig. 7b). The reduced resistance can be ascribed to the activation ofthe electrode as well as the better filtration of the electrolyte.

To get further insights, CV test was carried out. Fig. 7c shows CVcurves of CuS/Na battery with different scan rate from 0.2 �1.5 mV s�1. As the scan rate increases, the intensity and area of theredox peaks increases because the capacity equals the area dividedby scan rate, and the capacity should be a constant. Furthermore,the cathodic peaks shift to lower potential while correspondinganodic peaks shift to higher potential, which demonstrates

Fig. 7. (a) Nyquist plots (equivalent circuit diagram inset), (b) enlarged Nyquist plots of CuS/Na batteries before and after 20 cycles, (c) cyclic voltammograms of CuS electrodeat different scan rate, (d) the ip-n1/2 plots of CuS electrode, (e) the comparison of GITT measurement and corresponding constant current (CC) measurement of CuS electrode,(f) the calculated sodium-ion diffusion coefficients from GITT curve.

H. Li et al. / Electrochimica Acta 247 (2017) 851–859 857

enlarged polarization of CuS electrode. The Na-ion apparentdiffusion coefficient (DCV) can be calculated using Randles–Sevcikequation:

ip ¼ 0:4463nFACnFnDRT

� �1=2

ð10Þ

where n represents the number of electrons transferred in theredox event (for single redox peak, n is usually 1), F is FaradayConstant (96485C mol�1), A is the working electrode area

(1.13 cm2), C stands for sodium ion concentration in the electrode(0.048 mol cm�3), n is the scan rate, R is the gas constant(8.314 J mol�1K�1) and T is the temperature in Kelvin (298.15 K).Linear relationship lies between the peak current (ip) and thesquare root of scan rate (n1/2), the DCV can be calculated making useof the slope of the linear fit (Fig. 7d). The DCV calculated throughpeak C1, C2, A3, A4, A5 is 1.20 � 10�11, 1.39 � 10�11, 1.53 � 10�12,2.23 � 10�11 and 2.20 � 10�12 cm2 s�1, respectively. It is obviousthat peak A3 and A5 display much lower diffusion coefficient than

Table 1The calculated diffusion coefficient (D) of CuS electrode at room temperature (25 �C)through CV and GITT tests.

Methods D (cm2 s�1)

CV* 1.53 � 10�12 – 2.33 � 10�11

GITT 4.26 � 10�14 – 1.42 � 10�10

* Contains the DCV calculated through 5 different redox.

858 H. Li et al. / Electrochimica Acta 247 (2017) 851–859

other peaks, which means the transition from Cu back to coppersulfide is more kinetically difficult in the electrochemical reactions[36]. On the contrary, larger DCV derived from peak A4 indicatingthe transition between intermediaries is prone to taking place.

In a typical GITT measurement, the battery was discharged witha current impulse with the current density of 0.1C for an interval tof 600 seconds followed by an open circuit standing for 3600 sec-onds, where 1C was taken as 200 mA g�1 according to previousconstant current discharge test. Fig. 7e exhibits the comparison ofGITT and constant current (CC) measurement of the dischargeprocess of the same CuS electrode. The GITT curve matches wellwith the CC curve, where two plateaus around 1.7 V and 1.2 V canbe observed. According to equation (11), the Na-ion diffusioncoefficients (DGITT) in CuS at various voltages can be determinedusing GITT data:

DGITT ¼ 4pt

mVm

MA

� �2 DEsDEt

� �2

t � L2

DGITT

!ð11Þ

where m and M are the active material mass (1.96 mg) andmolecular weight (96 g mol�1) of CuS, respectively, Vm is the molarvolume of CuS (20.83 cm3mol�1), A is the electrode surface area(1.13 cm2), and L is the thickness of the electrode (13 mm). DEs isthe change of steady-state potential after a discharge pulse andopen circuit standing, while DEt is the voltage change during adischarge pulse. The calculated DGITT values are shown in Fig. 7f.The DGITT values of CuS electrode lie in 4.26 � 10�14 � 1.42 � 10�10

cm2 s�1. We have noticed that the minimum value of DGITT appearsaround 1.64 and 1.15 V, which coincides with the plateaus of thedischarge curve. The smaller diffusion coefficient could be due tothe phase transition during discharge process [37,38]. This result iscomparable with the apparent diffusion coefficients obtained fromCV test (Table 1).

4. Conclusions

In summary, CuS microspheres are synthesized rapidly by afacile microwave synthesis method. By adopting TEGDME as theelectrolyte solvent and adjusting the cut-off voltage to 0.6 � 3.0 V,the cyclic stability of CuS electrodes improve significantly. Whenserved as anode material for NIBs, the synthesized CuS can delivera reversible discharge capacity of 162 mAh g�1 after 200 cycleswith the current density of 200 mA g�1. It can still maintain thedischarge capacity of 101 and 90 mAh g�1 with the current densityof 800 and 1600 mA g�1, respectively. The electrochemical kineticsof CuS electrodes are investigated through EIS, CV and GITT tests.Moreover, the reaction mechanism of CuS in the electrode ispreliminarily studied by ex-situ XRD. To the best of our knowledge,it is not only the first time that CuS is successfully applied in long-life rechargeable NIBs but also the first time that the kinetics andreaction mechanism of CuS/Na batteries are explored.

Acknowledgements

This work was supported by the National Natural ScienceFoundation of China (Grant No. 21273185 and No. 21321062). Weappreciate Prof. Bing-Joe Hwang of National Taiwan University of

Science and Technology for his valuable suggestion on our work. Inaddition, special acknowledgement is delivered to Dr. Binbin Xu atXiamen University for his assistant in TEM test.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.electacta.2017.07.018.

References

[1] J.B. Goodenough, K.-S. Park, The Li-Ion Rechargeable Battery: A Perspective, JAm Chem Soc 135 (2013) 1167–1176.

[2] N. Yabuuchi, K. Kubota, M. Dahbi, S. Komaba, Research development onsodium-ion batteries, Chemical Reviews 114 (2014) 11636–11682.

[3] V. Etacheri, R. Marom, R. Elazari, G. Salitra, D. Aurbach, Challenges in thedevelopment of advanced Li-ion batteries: a review, Energ Environ Sci 4 (2011)3243.

[4] J.-M. Tarascon, Is lithium the new gold? Nat Chem 2 (2010) 510–510.[5] S. Komaba, W. Murata, T. Ishikawa, N. Yabuuchi, T. Ozeki, T. Nakayama, A.

Ogata, K. Gotoh, K. Fujiwara, Electrochemical Na Insertion and Solid ElectrolyteInterphase for Hard-Carbon Electrodes and Application to Na-Ion Batteries,Advanced Functional Materials 21 (2011) 3859–3867.

[6] M.D. Slater, D. Kim, E. Lee, C.S. Johnson, Sodium-Ion Batteries, AdvancedFunctional Materials 23 (2013) 947–958.

[7] S.-W. Kim, D.-H. Seo, X. Ma, G. Ceder, K. Kang, Electrode Materials forRechargeable Sodium-Ion Batteries: Potential Alternatives to Current Lithium-Ion Batteries, Advanced Energy Materials 2 (2012) 710–721.

[8] J. Song, M. Xu, L. Wang, J.B. Goodenough, Exploration of NaVOPO4 as a cathodefor a Na-ion battery, Chem Commun 49 (2013) 5280–5282.

[9] H. Hou, M. Jing, Y. Zhang, J. Chen, Z. Huang, X. Ji, Cypress leaf-like Sb as anodematerial for high-performance sodium-ion batteries, J. Mater. Chem. A 3(2015) 17549–17552.

[10] L. Yu, L.P. Wang, S. Xi, P. Yang, Y. Du, M. Srinivasan, Z.J. Xu, b-FeOOH: An Earth-Abundant High-Capacity Negative Electrode Material for Sodium-IonBatteries, Chemistry of Materials 27 (2015) 5340–5348.

[11] K. Zhang, Z. Hu, X. Liu, Z. Tao, J. Chen, FeSe2 Microspheres as a High-Performance Anode Material for Na-Ion Batteries, Adv Mater 27 (2015) 3305–3309.

[12] L. Liang, Y. Xu, C. Wang, L. Wen, Y. Fang, Y. Mi, M. Zhou, H. Zhao, Y. Lei, Large-scale highly ordered Sb nanorod array anodes with high capacity and ratecapability for sodium-ion batteries, Energ Environ Sci 8 (2015) 2954–2962.

[13] P. Ge, M. Fouletier, Electrochemical intercalation of sodium in graphite, SolidState Ionics 28 (1988) 1172–1175.

[14] H. Hou, X. Qiu, W. Wei, Y. Zhang, X. Ji, Carbon Anode Materials for AdvancedSodium-Ion Batteries, Advanced Energy Materials (2017) 1602898.

[15] D. Stevens, J. Dahn, High Capacity Anode Materials for Rechargeable Sodium-Ion Batteries, Journal of the Electrochemical Society 147 (2000) 1271–1273.

[16] R. Alca’ntara, P. Lavela, G.F. Ortiz, J.L. Tirado, Carbon Microspheres Obtainedfrom Resorcinol-Formaldehyde as High-Capacity Electrodes for Sodium-IonBatteries, Electrochemical and Solid-State Letters 8 (2005) A222.

[17] Y. Xiao, S.H. Lee, Y.-K. Sun, The Application of Metal Sulfides in Sodium IonBatteries, Advanced Energy Materials (2016) 1601329.

[18] Y.N. Ko, Y.C. Kang, Co9S8–carbon composite as anode materials with improvedNa-storage performance, Carbon 94 (2015) 85–90.

[19] M. Krengel, P. Adelhelm, F. Klein, W. Bensch, FeV2S4 as a high capacity electrodematerial for sodium-ion batteries, Chem Commun 51 (2015) 13500–13503.

[20] Z. Shadike, M.-H. Cao, F. Ding, L. Sang, Z.-W. Fu, Improved electrochemicalperformance of CoS2-MWCNT nanocomposites for sodium-ion batteries,Chem Commun 51 (2015) 10486–10489.

[21] C. Zhu, X. Mu, P.A. van Aken, Y. Yu, J. Maier, Single-layered ultrasmallnanoplates of MoS2 embedded in carbon nanofibers with excellentelectrochemical performance for lithium and sodium storage, AngewandteChemie 53 (2014) 2152–2156.

[22] Z. Hu, Z. Zhu, F. Cheng, K. Zhang, J. Wang, C. Chen, J. Chen, Pyrite FeS2 for high-rate and long-life rechargeable sodium batteries, Energy Environ. Sci. 8 (2015)1309–1316.

[23] J.-S. Kim, D.-Y. Kim, G.-B. Cho, T.-H. Nam, K.-W. Kim, H.-S. Ryu, J.-H. Ahn, H.-J.Ahn, The electrochemical properties of copper sulfide as cathode material forrechargeable sodium cell at room temperature, J Power Sources 189 (2009)864–868.

[24] A. Débart, L. Dupont, R. Patrice, J.M. Tarascon, Reactivity of transition metal(Co, Ni, Cu) sulphides versus lithium: The intriguing case of the coppersulphide, Solid State Sci 8 (2006) 640–651.

[25] Y. Wang, X. Zhang, P. Chen, H. Liao, S. Cheng, In situ preparation of CuS cathodewith unique stability and high rate performance for lithium ion batteries,Electrochim Acta 80 (2012) 264–268.

[26] X. Li, X. He, C. Shi, B. Liu, Y. Zhang, S. Wu, Z. Zhu, J. Zhao, Synthesis of one-dimensional copper sulfide nanorods as high-performance anode in lithiumion batteries, ChemSusChem 7 (2014) 3328–3333.

H. Li et al. / Electrochimica Acta 247 (2017) 851–859 859

[27] X. Wang, Y. Wang, X. Li, B. Liu, J. Zhao, A facile synthesis of copper sulfidescomposite with lithium-storage properties, J Power Sources 281 (2015) 185–191.

[28] H. Li, Y. Wang, J. Huang, Y. Zhang, J. Zhao, Microwave-assisted Synthesis of CuS/Graphene Composite for Enhanced Lithium Storage Properties, ElectrochimActa 225 (2017) 443–451.

[29] F. Klein, B. Jache, A. Bhide, P. Adelhelm, Conversion reactions for sodium-ionbatteries, Phys Chem Chem Phys 15 (2013) 15876–15887.

[30] M. Nafees, S. Ali, S. Idrees, K. Rashid, M.A. Shafique, A simple microwave assistsaqueous route to synthesis CuS nanoparticles and further aggregation tospherical shape, Applied Nanoscience 3 (2012) 119–124.

[31] Z. Hu, K. Zhang, Z. Zhu, Z. Tao, J. Chen, FeS2 microspheres with an ether-basedelectrolyte for high-performance rechargeable lithium batteries, J. Mater.Chem. A 3 (2015) 12898–12904.

[32] C. Shi, X. Li, X. He, J. Zhao, New Insight into the Interaction between Carbonate-based Electrolyte and Cuprous Sulfide Electrode Material for Lithium IonBatteries, Electrochim Acta 174 (2015) 1079–1087.

[33] X. Wang, D.-M. Tang, H. Li, W. Yi, T. Zhai, Y. Bando, D. Golberg, Revealing theconversion mechanism of CuO nanowires during lithiation-delithiation by insitu transmission electron microscopy, Chem Commun 48 (2012) 4812–4814.

[34] H. Wang, S. Lu, Y. Chen, L. Han, J. Zhou, X. Wu, W. Qin, Graphene/Co9S8nanocomposite paper as a binder-free and free-standing anode for lithium-ionbatteries, J. Mater. Chem. A 3 (2015) 23677–23683.

[35] C. Feng, L. Zhang, M. Yang, X. Song, H. Zhao, Z. Jia, K. Sun, G. Liu, One-PotSynthesis of Copper Sulfide Nanowires/Reduced Graphene OxideNanocomposites with Excellent Lithium-Storage Properties as AnodeMaterials for Lithium-Ion Batteries, ACS Appl Mater Interfaces 7 (2015) 15726–15734.

[36] X.H. Rui, N. Ding, J. Liu, C. Li, C.H. Chen, Analysis of the chemical diffusioncoefficient of lithium ions in Li3V2(PO4)3 cathode material, Electrochim Acta 55(2010) 2384–2390.

[37] K.M. Shaju, G.V. Subba Rao, B.V.R. Chowdari, Influence of Li-Ion Kinetics in theCathodic Performance of Layered Li(Ni1/3Co1/3Mn1/3)O2, Journal of theElectrochemical Society 151 (2004) A1324.

[38] K. Tang, X. Yu, J. Sun, H. Li, X. Huang, Kinetic analysis on LiFePO4 thin films byCV, GITT, and EIS, Electrochim Acta 56 (2011) 4869–4875.