A1730 Journal of The Electrochemical Society, 163 (8) A1730-A1735 (2016)0013-4651/2016/163(8)/A1730/6/$33.00 © The Electrochemical Society

Amorphous TiS3/S/C Composite Positive Electrodes with HighCapacity for Rechargeable Lithium BatteriesTakuya Matsuyama,a Akitoshi Hayashi,a Connor J. Hart,b Linda F. Nazar,band Masahiro Tatsumisagoa,z

aDepartment of Applied Chemistry, Graduate School of Engineering, Osaka Prefecture University, Naka-ku, Sakai,Osaka 599-8531, JapanbDepartment of Chemistry and the Waterloo Institute for Nanotechnology, University of Waterloo, Waterloo,Ontario N2L 3G1, Canada

Composite electrodes of a-TiS3/S/carbon (Ketjen black; KB) with high capacity were prepared by mechanical milling from a-TiS3and S/KB composites. The composites were fabricated into coin type liquid cells and all-solid-state cells and operated as rechargeablebatteries at room temperature. The reversible capacity of the coin type liquid cells decreased from 484 to 33 mAh g−1 over 50 charge-discharge cycles, because the polysulfides formed from the redox reactions of a-TiS3/S dissolved in the liquid electrolyte. On theother hand, the all-solid-state cells showed higher reversible capacity and better cyclability than the coin type liquid cells. In orderto improve their cycle performance, solid electrolyte (SE) powders were added to the composite electrodes to serve as lithium-ionconduction paths to the active materials. The cell using the a-TiS3/S/KB composite with 30 wt% SE exhibited the highest reversiblecapacity of about 850 mAh g−1 at the 1st cycle and retained a reversible capacity of about 650 mAh g−1 after the 50th cycle.Owing to their relatively high sulfur content, composite positive electrodes of a-TiS3/S/KB are attractive positive electrodes withhigh capacities for all-solid-state lithium secondary batteries.© 2016 The Electrochemical Society. [DOI: 10.1149/2.1061608jes] All rights reserved.

Manuscript submitted February 9, 2016; revised manuscript received May 9, 2016. Published June 14, 2016.

Conventional lithium-ion batteries using organic liquid electrolyteshave been used extensively as power sources for mobile devices andpersonal computers.1,2 Increasingly, they have been scaled up for usein large applications for electric vehicles and smart grids.3 However,traditional lithium ion batteries containing a transition metal oxidecathode and a graphite anode are not able to satisfy energy densitydemands for many applications.4,5

Sulfur is one of the most promising positive electrode materialsbecause of its high theoretical specific capacity of 1672 mAh g−1,which is at least 5 times higher than that of the transition metal oxidessuch as LiCoO2.6,7 In addition, sulfur has many other advantages oflow cost, abundant resource and environmental friendliness. However,pronounced capacity fading during cycling of lithium-sulfur batterieshas been a major challenge that has thwarted their practical use.8,9 Thepoor cycle life is attributable to the dissolution of intermediate lithiumpolysulfides (Li2Sn, n = 4–8)10–12 formed during charge-discharge intoorganic liquid electrolytes. The volumetric expansion and contractionof sulfur during cycling and the insulating nature of sulfur and lithiumsulfide are also drawback for sulfur electrodes. A well-designed sul-fur electrode to solve these issues is required. Current approachesfocus on confining sulfur active materials in porous nanostructures tocapture lithium polysulfides during charge-discharge reactions.13–24

Various sulfur/carbon composites have been studied as positive elec-trodes in Li/S batteries. For example, a sulfur-porous hollow carboncomposite electrode showed the reversible capacity of 974 mAh g−1

after 100 cycles.13 Sulfur/carbon nanotube composites delivered aninitial discharge capacity of about 1100 mAh g−1, and capacity reten-tion of 85% even after the 100th cycle at 1C.24 We have reported thatsulfur/highly ordered nanostructured mesoporous carbon (CMK-3)positive electrodes exhibited a reversible capacity of 1320 mAh g−1.15

Sulfur/porous carbon nanospheres (PCNSs) positive electrode mate-rials with stable cycling capacity for 100 cycles have been reported.23

As described above, the cycle life and utilization of Li/S batteriesare improved by using sulfur/carbon composite positive electrodes,because they maintain an electron conducting path to the sulfur activematerials and physically inhibit the dissolution of polysulfides intoliquid electrolytes.

Transition metal sulfides such as TiS2 have also been studied aspositive electrodes in lithium batteries.25–28 In general, transition metalsulfides exhibit high electronic conductivity and low solubility in elec-trolytes. These properties represent an advantage of transition metal

zE-mail: tatsu@chem.osakafu-u.ac.jp

sulfides over elemental sulfur.25–28 However, reversible capacities ofsuch batteries were lower than those using elemental sulfur. To in-crease the reversible capacity of transition metal sulfides, increasingtheir sulfur content is important because the additional redox capabil-ities can be utilized. For example, crystalline TiS3 (c-TiS3) has beenstudied as a positive electrode in lithium batteries.29,30 In our previousstudies, sulfur-rich transition metal sulfides such as amorphous TiS3

(a-TiS3) and amorphous MoS3 (a-MoS3) were reported.31–33 Theseelectrode active materials exhibited higher reversible capacity thanTiS2 and MoS2 in all-solid-state cells. In addition, it was reportedthat all-solid-state cells using a-TiS3 electrodes maintained better cy-clability than cells using c-TiS3 electrodes.34 Sakuda et al. have alsoshown that amorphous TiS4 electrodes showed reversible capacitiesof about 700 mAh g−1 in lithium batteries using liquid electrolytes.35

This capacity was higher than that of a-TiS3 (560 mAh g−1). These re-sults suggest that amorphous sulfur-rich transition metal sulfides havea high capacity owing to additional sulfur taking part in the redox.Moreover, it is reported that the dissolution of polysulfide into theelectrolyte was significantly suppressed by amorphous TiS4, becauseof the chemical binding between Ti and S.35

For increasing the capacity beyond TiS4, electrode active materialssuch as amorphous TiSx (x > 4) (a-TiSx) are promising. However,although one might expect that the electronic conductivity of a-TiSx

would decrease with increasing sulfur content, we show here thata-TiSx electrodes with high capacity and good cyclability can preparedby combining a-TiS3 and sulfur/carbon composites.

To inhibit the dissolution of lithium polysulfides, the use of in-organic solid electrolytes instead of organic liquid electrolytes is aneffective approach. Among many types of lithium ion conductive solidelectrolytes, sulfide materials are known to exhibit higher ionic con-ductivity than oxide materials because of the higher polarizability ofsulfide ions.36–38 Sulfide-based solid electrolytes such as crystallineLi10GeP2S12

39 and Li2S-P2S5 glass-ceramics40 have a high lithium-ion conductivity of almost 10−2 S cm−1 at room temperature, whichis comparable to the conductivity of conventional organic liquid elec-trolytes. Li/S cells using Li2S-P2S5 solid electrolytes have exhibitedgood cycling performance,41–43 so sulfide electrolytes are suitable forapplication to all-solid-state batteries with a-TiSx electrodes.

In this study, we have prepared a-TiS3/S/carbon composites by me-chanical milling mixtures of a-TiS3 and S/several carbon composites.The electrochemical properties of coin type cells utilizing organicliquid electrolytes, and all-solid-state cells using sulfide solid elec-trolytes with a-TiS3/S/carbon composite electrodes were examined.Four types of carbons encompassing Ketjen black (KB), CMK-3,

) unless CC License in place (see abstract).  ecsdl.org/site/terms_use address. Redistribution subject to ECS terms of use (see 129.97.124.196Downloaded on 2016-07-18 to IP

Journal of The Electrochemical Society, 163 (8) A1730-A1735 (2016) A1731

PCNS and graphite were used. Because the cell with the KB com-posite showed the best cell performance, we primarily report theseresults.

Experimental

Amorphous TiS3 (a-TiS3) electrode active materials were preparedusing mechanical milling of crystalline TiS3 (c-TiS3) prepared bysolid-state reaction.34 The starting material, c-TiS3, was milled me-chanically at ambient temperature using a planetary ball mill apparatus(Pulverisette 7; Fritsch) with an zirconia pot (volume of 45 ml) and500 zirconia balls (4 mm diameter). The rotational speed was set to370 rpm and the milling time was 40 h.

A 70 wt% sulfur containing composite (S/KB) was prepared by viaa melt diffusion technique at 155◦C.15,23 Composites of S/CMK-3,15

S/PCNS23 and S/graphite were prepared in the same manner.Amorphous TiS3/S/KB (a-TiS3/S/KB) electrodes were prepared by

mechanical milling. Prepared a-TiS3 and the S/KB composite wereused as the starting materials in a molar ratio of 1/2. The mixture ofthe materials was milled mechanically at ambient temperature usingthe ball milling parameters were as above and the milling time was20 h.

To examine the crystallinity of the prepared samples, X-ray diffrac-tion was carried out using a X-ray diffractometer (CuKα, UltimaIV;Rigaku Corp.). Surface areas of KB, S/KB and a-TiS3/S/KB weredetermined from nitrogen adsorption and desorption isotherms per-formed on an Autosorb-1; Quantachrome at 77 K. Surface areas werecalculated using the Brunauer–Emmett–Teller (BET) method. Ther-mogravimetric analyses (TGA) of the sample were performed using athermal analyzer (SDT Q600; TA Instruments) with air as the carriergas (100 cc min−1) at a heating rate of 10◦C min−1. The morpholo-gies of the S/KB and a-TiS3/S/KB composites were evaluated usinga field emission scanning electron microscopy (FE-SEM, SU8220;HITACHI).

Coin cells with 2325 type using organic liquid electrolytes werefabricated as follows: the S/KB, a-TiS3 or a-TiS3/S/KB electrodeswere formed by drop casting a slurry onto a carbon-coated aluminumor carbon paper current collector, and drying the electrodes at 60◦Covernight. In these experiments, 80 wt% of S/KB composite, 10 wt%Super P and 10 wt% poly(vinylidene floride) (PVDF) were mixed withN, N-dimethylformamide to form the slurry for drop casting. Differentelectrode ratios were also examined: coin type cells consisting of a-TiS3 and a-TiS3/S/KB were fabricated with electrodes consisting of 70wt% of a-TiS3 or a-TiS3/S/KB composite, 20 wt% Super P and 10 wt%PVDF (also mixed with N, N-dimethylformamide to form a slurry).The mass loading of the electrode was 1 mg cm−2. The electrolyte was1 M lithium bis(trifluoromethanesulfonyl)imide (LiTFSI) in a mixedsolvent of 1,2-dimethoxyethane (DME) and 1,3-dioxolane (DOL) (v/v= 1:1), with 2 wt% of LiNO3. Lithium metal foil was used as the nega-tive electrode, and was electronically isolated using 2 sheets of Celgard3501 separators. All the processes described above were conducted ina dry Ar glove box. The coin type cells with S/KB were dischargedand charged on the 1st cycle in a voltage window between 1.9 and3.0 V at a current density of 0.084 mA cm−2. The voltage win-dow was set between 1.8 and 3.0 V at a current density of0.84 mA cm−2 after the 2nd cycle. Coin cells with a-TiS3 anda-TiS3/S/KB were cycled in a voltage window between 1.8 and 3.0 Vat a current density of 0.064 mA cm−2. Electrochemical studies wereperformed using a cycler (BT-2000; Arbin) in galvanostatic mode atroom temperature.

All-solid-state electrochemical cells were fabricated as follows: an80Li2S · 20P2S5 (mol%) solid electrolyte was prepared by mechani-cal milling and then heated at 210◦C for 1 h.44 The S/KB, a-TiS3

and a-TiS3/S/KB electrodes were used as active materials. The work-ing electrode was prepared by mixing the active materials and thesolid electrolyte. The weight ratios of electrodes were 100/0, 70/30or 40/60. The obtained working electrode (10 mg) and the solid elec-trolyte were placed in a polycarbonate tube (10 mm diameter) andpressed together under 360 MPa of pressure. A Li–In alloy was placed

on the solid electrolyte layer to serve as a counter and reference elec-trode, and a pressure of 120 MPa was applied to the three-layeredpellet. Finally, two electrode cells sandwiched by two stainless-steeldisks that functioned as a current collector were obtained. All theprocesses described above were conducted in a dry Ar glove box. Theelectrochemical tests were conducted at 25◦C in an Ar atmosphereusing a cycler (BTS-2004; Nagano Co.). To analyze the structureof the working electrode, XRD measurements and cross-sectionalFE-SEM observations were carried out with an energy dispersiveX-ray spectroscopy system (EDX, EMAXEvolution X-Max; Horiba,Ltd.) for the working electrodes before and after electrochemical test-ing. The cross sections of the working electrodes were thinned downfor analysis using an ion milling system (IM4000; HITACHI).

Results and Discussion

Figure 1a shows the XRD patterns of S/KB samples before andafter heating, and a-TiS3/S/KB prepared by mechanical milling. Peaksattributable to sulfur were observed for the S/KB sample before heat-ing, but their intensity decreased after heating. This suggests thatelemental sulfur was incorporated into the pores of KB.15 Figure 1bshows the thermogravimetric analysis (TGA) profile of the S/KB sam-ple. The weight loss between 200 to 300◦C indicates a sulfur content of70 wt%. A halo pattern was observed in the XRD pattern of TiS3/S/KBindicating it is amorphous. In addition, specific surface areas of theprepared samples were examined by BET measurements. The resultsare listed in Table I. The specific surface area of KB was 1117 m2

g−1. The specific surface area of S/KB was reduced to 21.6 m2 g−1

10 20 30 40 50 60 70 80

Inte

nsity

(arb

. uni

t)

2θθ / o (CuK α )

S/KB after heating

S/KB before heating

a-TiS3/S/KB

(a)

0

20

40

60

80

100

0 100 200 300 400 500

Wei

ght /

%

Temperature / oC

(b)

Figure 1. (a) XRD patterns of S/KB samples before and after heating and ofa-TiS3/S/KB prepared by mechanical milling and (b) TGA curve of the S/KBsample after heating.

) unless CC License in place (see abstract).  ecsdl.org/site/terms_use address. Redistribution subject to ECS terms of use (see 129.97.124.196Downloaded on 2016-07-18 to IP

A1732 Journal of The Electrochemical Society, 163 (8) A1730-A1735 (2016)

Table I. Specific surface area of KB, S/KB and a-TiS3/S/KB.

Samples Surface area (m2 g−1)

KB 1117S/KB 21.6

a-TiS3/S/KB 4.6

by the inclusion of sulfur within the pores, whereas the specific sur-face area of a-TiS3/S/KB was 4.6 m2 g−1. FE-SEM images of S/KBsample after heating (a) and a-TiS3/S/KB (b) are shown in Fig. 2. TheFE-SEM image of the S/KB sample indicated that the particle sizeswere about 100 nm. Secondary particles of several microns in size,formed by agglomeration of submicron-sized ordinary particles, weremainly observed in the FE-SEM image of the a-TiS3/S/KB composite(Fig. 2b). The agglomeration decreased the specific surface area ofa-TiS3/S/KB.

Figure 3 shows (a) the initial charge-discharge curves and (b)the cycle performance of the liquid electrolyte coin cells with S/KB,a-TiS3 and a-TiS3/S/KB. The cell with S/KB was run at a current den-sity of 0.084 mA cm−2 (at the 1st cycle) and 0.84 mA cm−2 (from the2nd to 100th cycle). The charge–discharge measurements of the cellswith a-TiS3 and a-TiS3/S/KB were conducted at a current density of0.064 mA cm−2 at 25◦C. The cell with S/KB showed reversible capac-ities of about 1200 mAh g−1 and 800 mAh g−1 at the 1st and the 100thcycle. This electrochemical performance is similar to that of previousreports.22 On the other hand, the initial reversible capacities of the cellsusing a-TiS3 and a-TiS3/S/KB were about 400 and 500 mAh g−1, re-spectively. The cell with a-TiS3/S/KB showed higher reversible capac-ity than the cell with a-TiS3, because the additional sulfur in the S/KBreacted with Li during the discharge process. The cycling performanceof the liquid cells with a-TiS3 and a-TiS3/S/KB deteriorated duringthe electrochemical tests, however. We assume this is due to dissolu-

Figure 2. FE-SEM images of (a) S/KB; (b) the a-TiS3/S/KB composite.

0

400

800

1200

1600

0 20 40 60 80 100

Capa

city

/ m

Ah

g-1

Cyc le num ber

S/KB0.084 mA cm-2

0.84 mA cm-2

a-TiS3/S/KB0.064 mA cm-2

a-TiS30.064 mA cm-2

(b)

1.5

2

2.5

3

3.5

0 200 400 600 800 1000 1200

Cell

pote

ntia

l vs.

Li /

V

Capacity / mAh g(S, a -TiS3, a -TiS

3/S)-1

S/KB0.084 mA cm-2

a-TiS3

0.064 mA cm-2

a-TiS3/S/KB0.064 mA cm-2

(a)

Figure 3. (a) Initial charge-discharge curves of liquid electrolyte cells witha-TiS3, S/KB and a-TiS3/S/KB and (b) their cycling performance.

tion of polysulfides produced from a-TiS3 during discharge process.To clarify detailed reaction process, electronic structures of sulfurand titanium elements in a-TiS3 during cycling need to be analyzed.Sakuda et al. have reported that coin cells using liquid electrolytes anda-TiS4/acetylene black (AB) composite showed stable cyclability.35

They explained that the chemical binding between Ti and S formed bythe amorphization procedure suppressed the dissolution of polysul-fide into the electrolyte. Furthermore, the dispersion of AB inside thea-TiS4 was effective in improving the cycle performance.33 In ourstudy, ineffective dispersion of KB inside TiS3 might have led to un-stable cycling performance. We prepared S/carbon composites usingthe other carbon materials such as CMK-3,15 PCNS23 and graphite in asimilar manner. The initial and the 100th charge and discharge capaci-ties of the coin type liquid cells with a-TiS3/S/CMK-3, a-TiS3/S/PCNSand a-TiS3/S/graphite are summarized in Table II. The capacitiesof the cell with a-TiS3/S/KB are also shown for comparison in

Table II. 1st and the 100th discharge and charge capacities of thecoin type liquid cells with a-TiS3/S/KB, a-TiS3/S/CMK-3, a-TiS3/S/PCNS and a-TiS3/S/graphite composite electrodes.

Capacity (mAh g−1)

1st discharge 1st charge 100th discharge 100th charge

a-TiS3/S/KB 673 484 20 12a-TiS3/S/CMK-3 780 666 — —a-TiS3/S/PCNS 774 498 — —

a-TiS3/S/graphite 642 400 113 109

) unless CC License in place (see abstract).  ecsdl.org/site/terms_use address. Redistribution subject to ECS terms of use (see 129.97.124.196Downloaded on 2016-07-18 to IP

Journal of The Electrochemical Society, 163 (8) A1730-A1735 (2016) A1733

0

0.5

1

1.5

2

2.5

3

0 200 400 600 800 1000

1

1.5

2

2.5

3

3.5Ce

ll po

tent

ial v

s. L

i-In

/ V

Cell

pote

ntia

l vs.

Li /

V

Capacity / mAh g(a -TiS3/S)-1

KBgraphiteCMK-3

PCNS

25oC, 0.064 mA cm-2

Figure 4. Initial charge-discharge curves of the all-solid-state cells witha-TiS3/S/KB, a-TiS3/S/CMK-3, a-TiS3/S/PCNS and a-TiS3/S/graphite.

Table II. Capacity degradation of the a-TiS3/S/graphite cell was alsoobserved. The 100th capacities for the cells using a-TiS3/CMK-3/Sand a-TiS3/PCNS/S are not listed because a shuttle phenomenon45

appeared at the 2nd charging process (the charge-discharge curves arenot shown.). On the other hand, the cells using S/CMK-3, S/PCNS andS/graphite, respectively, showed an initial reversible capacity of about1250, 900 and 800 mAh g−1 in a voltage window between 1.9 and3.0 V at a current density of 0.084 mA cm−2. The 2nd reversible capac-ities of the cells with S/CMK-3, S/PCNS and S/graphite in the voltagewindow between 1.8 and 3.0 V at the current density of 0.84 mA cm−2

were about 780, 650 and 470 mAh g−1, respectively. The reversiblecapacities of the cells using S/CMK-3, S/PCNS and S/graphite afterthe 100th cycle were about 750, 480 and 370 mAh g−1, respectively,and the S/carbon samples (carbon: KB, CMK-3, PCNS and graphite)in the cells with liquid electrolytes showed a relatively-stable cycleperformance. Therefore, the degradation of the cells with liquid elec-trolytes and a-TiS3/S/carbon composites is undoubtedly due to thedissolution of polysulfides derived from a-TiS3.

It is expected that a-TiS3/S/carbon composites in all-solid-statecells with sulfide solid electrolytes will have good cyclability be-cause the dissolution of polysulfides is suppressed by using solid elec-trolytes. Figure 4 shows the initial charge-discharge curves of the all-solid-state cells with a-TiS3/S/KB, a-TiS3/S/CMK-3, a-TiS3/S/PCNSand a-TiS3/S/graphite. The Li-In alloy was used as a counter electrode,because it exhibits a stable voltage plateau at 0.62 V vs Li+/Li in anall-solid-state cell using a sulfide solid electrolyte. Charge–dischargestudies of the cells were conducted at a current density of 0.064mA cm−2 at 25◦C, using the a-TiS3/S/carbon composites as a work-ing electrode at room temperature. The results are presented refer-enced to the electrode potential vs. Li+/Li. The all-solid-state cell witha-TiS3/S/KB showed the highest reversible capacity of about 650 mAhg−1 in the cells. Figure 5 shows a comparison of the initial charge-discharge curves (a) and cycle performance (b) of the all-solid-statecells with a-TiS3/S/KB, S/KB, a-TiS3 and a-TiS5 electrodes. Amor-phous TiS5 (a-TiS5) was prepared by mechanical milling from a-TiS3

and sulfur. The cells with S/KB showed an initial discharge capacityof about 150 mAh g−1; the obtained capacity in this study was lowerthan that of our previous results,43 because the Li+ ion conductionpaths were insufficient. The S/KB cell obviously needs the addition ofsulfide solid electrolytes in the working electrode, as it did not exhibita capacity on the 1st charge. On the other hand, a-TiS3 showed a re-versible capacity in the all-solid-state cell using the a-TiS3 electrodewithout the addition of solid electrolyte or conductive additive pow-ders. Its reversible capacity was about 500 mAh g−1. The solid-statecell with a-TiS5 exhibited about the 490 mAh g−1 capacity on initialdischarge, but the cell could not be charged. A lower electronic con-ductivity of a pressed pellet of a-TiS5 (about 10−6 S cm−1 at 25◦C)than that of a-TiS3 (>10−3 S cm−1 at 25◦C) is one reason for theirreversibility. The cell with a-TiS3/S/KB exhibited the highest ini-tial capacity of about 650 mAh g−1. The electronic conductivity of a

0

0.5

1

1.5

2

2.5

3

0 200 400 600 800

1

1.5

2

2.5

3

3.5

Cell

pote

ntia

l vs.

Li-I

n / V

Cell

pote

ntia

l vs.

Li /

V

Capacity / mAh g(S, a -TiS3,a -TiS

5, a -TiS

3/S)-1

a-TiS3/S/KB

a-TiS3

25oC, 0.064 mA cm-2

a-TiS5

(a)

S/KB

0

200

400

600

800

0 10 20 30 40 50Ca

paci

ty /

mA

h g-1

Cycle number

(b)a-TiS3/S/KB

a-TiS3

S/KBa-TiS5

Figure 5. (a) Initial charge-discharge curves and (b) cycling performance ofthe all-solid-state cells with a-TiS3/S/KB, S/KB, a-TiS3 and a-TiS5.

pressed pellet of a-TiS3/S/KB prepared from a-TiS3 and S/KB wasabout 10−3 S cm−1, which is similar to the conductivity of a-TiS3

itself. The a-TiS3/S/KB electrode exhibited a higher capacity than thea-TiS3 electrode, suggesting that additional sulfurs in S/KB were usedas active material in the a-TiS3/S/KB electrode. The additional sulfursin a-TiS3/S/KB electrode thus contribute to a reversible capacity be-cause of conductive feature of a-TiS3. Both all-solid-state cells witha-TiS3 and a-TiS3/S/KB showed better cyclability than the liquid elec-trolyte cells. In particular, the cell comprised of a-TiS3 alone showed apromising reversible capacity of about 450 mAh g−1 at the 50th cyclewithout significant capacity fading. However, capacity degradation ofa-TiS3/S/KB was observed, with a reversible capacity of only 250mAh g−1 being attained on the 50th cycle. This is explained below.

In order to improve the cycle performance of the cell with a-TiS3/S/KB, working electrodes were prepared with a sulfide solid elec-trolyte. These were assembled by hand-mixing a-TiS3/S/KB and the80Li2S · 20P2S5 glass-ceramic solid electrolyte (SE). The weight ra-tios of electrodes were 100/0, 70/30 or 40/60. Figure 6 shows (a) theinitial charge-discharge curves and (b) the cycling performance of theall-solid-state cells with a-TiS3/S/KB composites. The cell using thea-TiS3/S/KB composites with the addition of 30 wt% SE exhibitedthe highest reversible capacity of about 850 mAh g−1 at the 1st cy-cle. Forming favorable contacts among the electrode components ofa-TiS3/S/KB and the sulfide solid electrolytes should clearly be effec-tive in improving the reversible capacity of all-solid-state cells. How-ever, the initial reversible capacity of the cell with the a-TiS3/S/KBcomposite with 60 wt% SE decreased to 650 mAh g−1 over cycling.After the 50th cycle, the all-solid-state cells using the electrodes with0, 30, and 60 wt% SE showed a capacity of about 250, 650 and 370mAh g−1, respectively. Although capacity degradation was observedin all the cells, the cell using a-TiS3/S/KB composite with 30 wt% SEretained higher reversible capacity than the cell using a-TiS3 after the50 cycles.

Figure 7 shows the XRD pattern of the a-TiS3/S/KB working elec-trodes without solid electrolytes before cycling and after the 50th

) unless CC License in place (see abstract).  ecsdl.org/site/terms_use address. Redistribution subject to ECS terms of use (see 129.97.124.196Downloaded on 2016-07-18 to IP

A1734 Journal of The Electrochemical Society, 163 (8) A1730-A1735 (2016)

0

200

400

600

800

1000

0 10 20 30 40 50

Capa

city

/ m

Ah

g-1

Cyc le num ber

30 wt% SE

60 wt% SE

0 wt% SE

(b)

0

0.5

1

1.5

2

2.5

3

0 200 400 600 800 1000

1

1.5

2

2.5

3

3.5Ce

ll po

tent

ial v

s. L

i-In

/ V

Cell

pote

ntia

l vs.

Li /

V

Capacity / mAh g(a -TiS3/S) -1

30 wt% SE

60 wt% SE 0 wt% SE

25oC, 0.064 mA cm-2

(a)

Figure 6. (a) Initial charge-discharge curves and (b) cycle performance of theall-solid-state cells using a-TiS3/S/KB composites with 0, 30 and 60 wt% solidelectrolyte.

charge-discharge measurements. Silicon was used as an internal stan-dard. A halo pattern due to the a-TiS3/S/KB active material was ob-served even after the 50th charge. This implies that the amorphousstate of the a-TiS3/S/KB active materials was maintained during elec-trochemical cycling.

Figure 8 shows the cross-sectional FE-SEM images and their EDXelemental maps for S, Ti and C of the a-TiS3/S/KB working electrodes(without solid electrolytes) (a) before cycling and (b) after the 50thcharge-discharge measurements. Pores and grain boundaries were ob-served in the FE-SEM image (a) for the working electrode beforecycling. Conversely, the working electrode after the 50th cycle (b)was denser. This is due to the volumetric changes of the sulfur com-

10 20 30 40 50 60 70 80

Inte

nsity

(arb

. uni

t)

2θθ / o (CuKα)

Si

Before charge-discharge

After the 50th charge

Figure 7. XRD patterns of the a-TiS3/S/KB working electrodes without elec-trolytes before cycling and after the 50th charge-discharge measurements.

Figure 8. Cross-sectional FE-SEM images and EDX elemental mapping im-ages for S, Ti and C of the a-TiS3/S/KB working electrodes without solidelectrolytes (a) before cycling; and (b) after the 50th charge-discharge.

ponent in the a-TiS3/S/KB composite during charge-discharge. Heet al. have reported that the thickness of the sulfur composite elec-trode expanded during discharging (lithiation process).46 The EDXmaps of all elemental signals of S, Ti and C were observed in the en-tire area of the electrode before and after charge-discharge, suggestingthat a-TiS3/S/KB was homogeneously dispersed during cycling.

In order to understand the capacity degradation of all-solid-statecells with the a-TiS3/S/KB electrode, detailed reaction mechanismof the electrode should be analyzed. Electronic structure changes ofsulfurs in the electrode are effective for clarifying the reaction mech-anism. Very recently, we have reported sulfur electronic structure ina-TiS3 electrodes in all-solid-state cells by X-ray photoelectron spec-troscopy (XPS) and X-ray absorption near edge structure (XANES).47

These analyses for both sulfur and titanium elements will be performedfor the a-TiS3/S/KB electrode before and after charge-discharge cy-cles, and the role of the additional sulfurs in S/KB to a reversiblecapacity will be clarified in the near future.

Conclusions

Novel a-TiS3/S/KB electrode active materials were prepared.The reversible capacity of coin cells using a liquid electrolyte anda a-TiS3/S/KB cathode decreased from 484 to 33 mAh g−1 dur-ing 50 charge-discharge cycles, because polysulfides formed froma-TiS3/S/KB were dissolved in the electrolyte. Capacity degrada-tion of a-TiS3/S/graphite cells was also observed. The cells using

) unless CC License in place (see abstract).  ecsdl.org/site/terms_use address. Redistribution subject to ECS terms of use (see 129.97.124.196Downloaded on 2016-07-18 to IP

Journal of The Electrochemical Society, 163 (8) A1730-A1735 (2016) A1735

a-TiS3/CMK-3/S and a-TiS3/PCNS/S also showed a shuttle phe-nomenon as early as the 2nd charging process. On the other hand,all-solid-state cells using a-TiS3/S/KB composite electrodes showed ahigher reversible capacity and better cyclability. In addition, all-solid-state cells with a-TiS3/KB/S showed a higher reversible capacity (650mAh g−1) than the cell with a-TiS3. We believe this is because theadditional sulfur in KB/S can be utilized as an active material, whilemaintaining the amorphous state of the a-TiS3/KB/S. That makesthis material, with a high sulfur content, attractive as a high capacitypositive electrode for all-solid-state lithium secondary batteries.

Acknowledgments

This research was financially supported by the Japan Science andTechnology Agency (JST), Advanced Low Carbon Technology Re-search and Development Program (ALCA), Specially Promoted Re-search for Innovative Next Generation Batteries (SPRING) Project.LFN thanks the Natural Sciences and Engineering Council (NSERC)for support via their Discovery and Canada Research Chair programs.

References

1. J. M. Trascon and M. Armand, Nature, 414, 359 (2001).2. S. Arico, P. Bruce, B. Scrosati, J. M. Tarascon, and W. VanSchalkwijk, Nature Mater.,

4, 366 (2005).3. B. Dunn, H. Kamath, and J. M. Tarascon, Science, 334, 928 (2011).4. J. B. Goodenough and K. S. Park, J. Am. Chem. Soc., 135, 1167 (2013).5. N. S. Choi, Z. Chen, S. A. Freunberger, X. Ji, Y. K. Sun, K. Amine, G. Yushin,

L. F. Nazar, J. Cho, and P. G. Bruce, Angew. Chem. Int. Ed., 51, 9994 (2012).6. H. Yamin, A. Gorenshtein, J. Penciner, Y. Sternberg, and E. Peled, J. Electrochem.

Soc., 135, 1045 (1988).7. B. H. Jeon, J. H. Yeon, K. M. Kim, and I. J. Chung, J. Power Sources, 109, 89

(2002).8. R. D. Rauh, K. M. Abraham, G. F. Pearson, S. K. Surprenant, and S. B. Brummer, J.

Electrochem. Soc., 126, 523 (1979).9. H. Yamin and E. Peled, J. Power Sources, 9, 281 (1983).

10. A. Manthiram, Y. Fu, and Y. S. Su, Acc. Chem. Res., 46, 1125 (2012).11. S. Evers and L. F. Nazar, Acc. Chem. Res., 46, 1135 (2012).12. D. Bresser, S. Passerini, and B. Scrosati, Chem. Commun., 49, 10545 (2013).13. N. Jayaprakash, J. Shen, S. S. Moganty, A. Corona, and L. A. Archer, Angew. Chem.

Int. Ed., 50, 5904 (2011).14. W. Zheng, Y. W. Liu, X. G. Hu, and C. F. Zhang, Electrochim. Acta, 51, 1330 (2006).15. X. Ji, K. T. Lee, and L. F. Nazar, Nature Mater., 8, 500 (2009).16. C. Liang, N. J. Dudney, and J. Y. Howe, Chem. Mater., 21, 4724 (2009).

17. L. Yuan, H. Yuan, X. Qiu, L. Chen, and W. Zhu, J. Power Sources, 189, 1141 (2009).18. B. Zhang, C. Lai, Z. Zhou, and X. P. Gao, Electrochim. Acta, 54, 3708 (2009).19. X. Ji and L. F. Nazar, J. Mater. Chem., 20, 9821 (2010).20. B. Zhang, X. Qin, G. R. Li, and X. P. Gao, Energy Environ. Sci., 3, 1531 (2010).21. J. Z. Wang, L. Lu, M. Choucair, J. A. Stride, X. Xu, and H. K. Liu, J. Power Sources,

196, 7030 (2011).22. J. Zheng, M. Gu, M. J. Wagner, K. A. Hays, X. Li, P. Zuo, C. Wang, J. G. Zhang,

J. Liu, and J. Xiao, J. Electrochem. Soc., 160, A1624 (2013).23. G. He, S. Evers, X. Liang, M. Cuisinier, A. Garsuch, and L. F. Nazar, ACS Nano, 7,

10920 (2013).24. L. Sun, M. Li, Y. Jiang, W. Kong, K. Jiang, J. Wang, and S. Fan, Nano Lett., 14, 4044

(2014).25. M. S. Whitingham, J. Electrochem. Soc., 123, 315 (1976).26. G. L. Holleck and J. R. Drlscdll, Electrochim. Acta, 22, 647 (1977).27. M. S. Whittingham, Science, 192, 1126 (1976).28. M. S. Whittingham, Prog. Solid State Chem., 12, 41 (1978).29. M. H. Lindic, H. Martinez, A. Benayad, B. Pecquenard, P. Vinatier, A. Levasseur,

and D. Gonbeau, Solid State Ionics, 176, 1529 (2005).30. J. Wu, D. Wang, H. Liu, W. M. Lau, and L. M. Liu, RSC Adv., 5, 21455 (2015).31. A. Hayashi, T. Matsuyama, A. Sakuda, and M. Tatsumisago, Chem. Lett., 41, 886

(2012).32. T. Matsuyama, A. Sakuda, A. Hayashi, Y. Togawa, S. Mori, and M. Tatsumisago, J.

Solid State Electrochem., 17, 2697 (2013).33. T. Matsuyama, T. Ozaki, A. Hayashi, S. Mori, and M. Tatsumisago, J. Mater. Chem.

A, 3, 14142 (2015).34. T. Matsuyama, A. Hayashi, T. Ozaki, S. Mori, and M. Tatsumisago, J. Ceram. Soc.

Jpn., 124, 242 (2016).35. A. Sakuda, N. Taguchi, T. Takeuchi, H. Kobayashi, H. Sakaebe, K. Tatsumi, and

Z. Ogumi, Electrochem. Commun., 31, 71 (2013).36. A. Pradel and M. Ribes, Solid State Ionics, 18/19, 351 (1989).37. F. Mizuno, A. Hayashi, K. Tadanaga, and M. Tatsumisago, Adv. Mater., 17, 918

(2005).38. R. Mercier, J. P. Malugani, B. Fahys, and G. Robert, Solid State Ionics, 5, 663 (1981).39. N. Kamaya, K. Homma, Y. Yamakawa, M. Hirayama, R. Kanno, M. Yonemura,

T. Kamiyama, Y. Kato, S. Hama, K. Kawamoto, and A. Mitsui, Nature Mater., 10,682 (2011).

40. Y. Seino, T. Ohta, K. Takada, A. Hayashi, and M. Tatsumisago, Energy Environ. Sci.,7, 627 (2014).

41. A. Hayashi, T. Ohtomo, F. Mizuno, K. Tadanaga, and M. Tatsumisago, Electrochem.Commun., 5, 701 (2003).

42. A. Hayashi, T. Ohtomo, F. Mizuno, K. Tadanaga, and M. Tatsumisago, Electrochim.Acta, 50, 893 (2004).

43. M. Nagao, A. Hayashi, and M. Tatsumisago, Electrochim. Acta, 56, 6055 (2011).44. A. Hayashi, S. Hama, T. Minami, and M. Tatsumisago, Electrochem. Commun., 5,

111 (2003).45. Y. V. Mikhaylik and J. R. Akridge, J. Electrochem. Soc., 151, A1969 (2004).46. X. He, J. Ren, L. Wang, W. Pu, C. Jiang, and C. Wan, J. Power Sources, 190, 154

(2009).47. T. Matsuyama, M. Deguchi, K. Mitsuhara, T. Ohta, T. Mori, Y. Orikasa, Y. Uchimoto,

Y. Kowada, A. Hayashi, and M. Tatsumisago, J. Power Sources, 313, 104 (2016).

) unless CC License in place (see abstract).  ecsdl.org/site/terms_use address. Redistribution subject to ECS terms of use (see 129.97.124.196Downloaded on 2016-07-18 to IP