当前位置：文档库 › enhancing performance of Li-S cell using a Li-Al alloy anode coating

enhancing performance of Li-S cell using a Li-Al alloy anode coating

Short

communication

Enhancing performance of Li –S cells using a Li –Al alloy anode coating

Hyea Kim a ,b ,Jung Tae Lee a ,Dong-Chan Lee a ,Martin Oschatz c ,Won Il Cho d ,Stefan Kaskel c ,Gleb Yushin a ,?

School of Materials Science and Engineering,Georgia Institute of Technology,Atlanta,GA 30332,USA b

Sila Nanotechnologies,Inc.,Atlanta,GA 30332,USA c

Department of Inorganic Chemistry,Dresden University of Technology,Bergstr.66,Dresden 01069,Germany d

Energy Storage Research Center/National Agenda Research Division,Korea Institute of Science and Technology,Cheong Ryang,Seoul 130-650,South Korea

a b s t r a c t

a r t i c l e i n f o Article history:

Received 8August 2013

Received in revised form 29August 2013Accepted 4September 2013

Available online 13September 2013Keywords:Li –S battery

Cathode dissolution Aluminum Alloy anode Polarization Sulfur cathode

Thin lithium aluminum (Li –Al)alloy layer is formed on the lithium surface in order to mitigate the polysul ?de shuttle phenomenon.The effect of curing temperature of the alloy on electrochemical performance is studied.Electrochemical tests show the ability of the alloy layer to stabilize polarization during plating/deplating of Li ions.Li –S cells with the alloy-coated Li anodes show batter rate capability,lower charge transfer resistance,improved cycle stability and higher coulombic ef ?ciency,compared to bare Li anode.

1.Introduction

Lithium –sulfur (Li –S)battery is an attractive alternative electro-chemical energy storage system due to the very high theoretical

speci ?c and volumetric capacity of S cathodes (1672mAh g S

?1&1930mAh ·cc ?1)[1–3].Furthermore,S is abundant in nature,safe to use and inexpensive.Yet,there are many obstacles that prevent Li –S batteries to become commercially viable on a large scale.Due to the insulating trait of S,it needs to be embedded into electronically conductive matrix,which plays a second role to ef ?ciently con ?ne the S from dissolution.During discharge,the intermediate products,high-order polysul ?des (Li 2S n ,4≤n ≤8),dissolve into the low molarity (1–3M)traditional organic electro-lytes [4,5],resulting in capacity fade.Polysul ?de dissolution causes further signi ?cant problems via polysul ?de shuttle mechanism.The diffused polysul ?des to the anode side are reduced to lithium sul ?de (Li 2S),and then diffuse back to the cathode to be reoxidized,generating a continuous current ?ow without an actual oxidation.

Many approaches have been suggested to resolve these challenges.They include a coating of inorganic shells [6,7]or polymer materials in S cathodes [8]which prevent or reduce polysul ?de out-diffusion into electrolytes (often at the expense of lower capacity and rate perfor-mance)and the use of high molarity electrolytes [5]which greatly reduce polysul ?de solubility in electrolytes at the expense of reduced conductivity.An alternative approach to prevent the polysul ?de shuttle

is protecting Li anodes from the reaction with polysul ?des [9].Previously this has been achieved in-situ by using electrolyte additives,such as LiNO 3[2].

In this work,in order to protect Li anodes,a very thin Li –Al layer is formed on the surface of Li by curing the two laminated foils at elevated temperatures.Polarization test using half-cells shows improved stability of such an alloy layer upon Li plating/deplating in the presence of polysul ?des.The Li –Al-coated Li anode in regular electrolytes showed even better stability than bare Li anode in the electrolyte with LiNO 3additive.The Li –S cells with 90°C cured Li –Al anode coating showed better rate-capability and improved charge-transfer resistance than Li –S cells with bare Li anode.2.Experimental

For preparing LiAl alloy layer,thin Al foil [0.8μm,Alfa Aesar,USA]was laminated on the Li surface [750μm,Alfa Aesar,USA].They were pressed together,cured at 25,60and 90°C for 24h under Ar (termed LiAl(25),LiAl(60)and LiAl(90)).The resulting product is a thin Li –Al layer coated on a bulk Li foil [10].Sulfur cathodes were prepared using ordered mesoporous carbide derived-carbon (OM-CDC)as S hosting materials.The OM-CDCs were synthesized as previously reported [11].The sulfur melted at 120°C diffuses into the CDC pores and the excess sulfur coated on the outside was evaporated at 200°C.The 85wt.%of CDC/S composites and 15wt.%of polyacrylic acid (PAA)[Sigma Aldrich]as a binder [12,13]were mixed in ethanol to prepare slurry.The slurry was stirred at room temperature overnight and casted on a Ni current collector.

Electrochemistry Communications 36(2013)38–41

?Corresponding author.

E-mail address:yushin@https://www.wendangku.net/doc/586618675.html, (G.

Yushin).

https://www.wendangku.net/doc/586618675.html,/10.1016/j.elecom.2013.09.002

Contents lists available at ScienceDirect

Electrochemistry Communications

j o u r n a l h o m e p a g e :w w w.e l s e v i e r.c o m/l o c a t e /e l e c o m

The coin cells were assembled with the 3M bis(tri ?ouro-methanesulfonyl)imide (LiTFSI)in distilled dimethoxyethane (DME):1,3-dioxane(DIOX)(1:1,v:v)as an electrolyte using a Celgard2325(Celgard,USA)separator.The cells were equilibrated for 24h before operation and cycled between 1.2and 3.0V vs.Li/Li +(V)in galvanostatic mode via Arbin battery test system (Arbin Instruments).The half-cells were assembled with Li and LiAl alloys as working electrodes and Li as a counter/reference electrode.Same electrolyte was used with the addition of 0.2M Li 2S 8.The concentration of the polysul ?de additive was stoichiometric-controlled in the reaction of S and Li 2S in the electro-lyte solvent.The half-cells were cycled at 1mAcm ?2for 1h for each plating/deplating.Electrochemical impedance spectroscopy (EIS)mea-surements were carried out using a Gamry Reference 600Potentiostat/Galvanostat/ZRA (Gamry Instruments,Inc.,USA)in the frequency range of 1Hz –106Hz with AC amplitude of 10mV.Scanning electron microsco-py (SEM)and energy-dispersive X-ray spectroscopy were performed using a Zeiss Ultra60FE-SEM (Carl Zeiss,Germany).3.Results and discussion

The purpose of our study was to investigate if a Li –Al alloy layer on the Li surface may allow either reduced or more stable polarization than pure Li during cycling in the presence of polysul ?des in electrolyte.Low temperature alloying of Al foil with Li was selected due to simplic-ity of such a procedure.Because the properties of the resulting alloy could be affected by the diffusion rate of the two metal ions,the effect of the curing temperature within 25–90°C was of interest.All LiAl alloy-coated Li anodes show rough and cracked surface (Fig.1)because of the large volume changes in Al foil upon alloying.More cracks were developed at higher temperatures (Fig.1).In all three cases,X-ray diffraction con ?rmed the presence of crystalline Li –Al alloys having the composition of LiAl (JCPDS#71-0362)and Li 4Al 9(JCPDS#24-0008).

Half-cells were prepared using bare Li,LiAl(60)and LiAl(90)as work-ing electrodes.0.2M Li 2S 8was added in the electrolyte to simulate the polysul ?de dissolution.EIS measurements performed prior to electro-chemical test revealed signi ?cantly reduced charge-transfer resistance (R ct )of LiAl(60)and LiAl(90)compared to bare Li with the electrolyte including polysul ?des (Fig.2).Interestingly,the control samples of bare Li and LiAl(90)without the presence of polysul ?de additive showed comparable impedance results,but the addition of polysul ?des decreased the R ct of LiAl while increased the one of bare Li.This suggests that the LiAl layer forms a fast Li +conducting surface ?lm in the presence of Li 2S 8in the electrolyte.We should also mention that LiAl alloy allows fast Li ion transport,comparable to Li [14].

To see the stability of the anodes during Li +plating/deplating,half-cells were cycled at 1.0mAcm ?2for each plating/deplating for 1h.The voltage pro ?le was monitored to see the polarization during cycling.Both LiAl(60)and LiAl(90)are very stable during 100cycles in spite the presence of polysul ?de (Fig.3a).These samples additionally show lower polarization than pure Li with LiAl(90)showing lower polariza-tion than LiAl(60),likely due to the lower R ct .In contrast,the initial po-larization of the bare Li was the highest and kept increasing till cycle 20.For the next 80cycles,deplating polarization is ?nally stabilized,but plating of Li remained unstable.

Further studies have conducted using the same electrolyte,but without the addition of polysul ?des.Now bare Li has lower polarization value than LiAl(90)and is stable over 100cycles (Fig.3b).As such,the unstable plating polarization with bare Li in the presence of polysul ?des con ?rms that the reduction of dissolved polysul ?des on Li surface competes with Li plating resulting in higher and more unstable polariza-tion.The LiAl coating can evidently protect the anode from continuous reacting of diffused polysul ?de with Li (Fig.3a).

As expected from prior art [2],the addition of LiNO 3into electrolyte decreased the initial Li polarization by forming a solid electrolyte inter-phase (SEI)with improved transport properties (Fig.3c).However,comparing with the LiAl(90),this polarization showed abrupt spikes and was increasing with cycling,which was particularly clearly seen after 60cycles.The increase in polarization is likely caused by the reduced solid state Li +transport through the surface ?lm formed by LiNO 3.Interestingly,while the polarization during Li plating (negative potential vs.Li/Li +)was rather stable,the deplating polariza-tion (positive potential vs.Li/Li +)was not.Since the diffusion coef ?cient

50μm

Fig.1.SEM micrographs show the impact of the alloying temperature on the surface morphology (a)Bare Li,(b-d)LiAl alloy cured at 25,60°C and 90°C respectively.

H.Kim et al./Electrochemistry Communications 36(2013)38–41

of Li +through the SEI should not depend on the Li +traveling direction,we propose that defects (such as voids)form at the Li/SEI interface or within the SEI layer when Li leaves the foil during deplating (Fig.3c).Such results indicate that the commonly used addition of LiNO 3into electrolyte offers an inferior solution compared to the proposed Li –Al alloy layer formation.

In order to con ?rm the useful properties of Li –Al alloy layer on Li for practical applications,we assembled full cells using S/OM-CDC cathodes.At C/10,the cells provided 1100–1300mAh g ?1capacity,

con ?rming the high performance of S/OM-CDC cathodes [4].While the rate performance of LiAl(25)was rather poor,presumably due to the poor quality of the Li/LiAl interface,increasing the alloying temperature resulted in improved capacity retention with increasing current densities (Fig.4a).In fact,LiAl(90)showed better performance than bare Li,likely due to the lower R ct .On a negative note,both LiAl(25)and LiAl(60)anodes exhibited rate performance inferior to pure Li and underwent mechanical failure (delamination of the LiAl ?lm con ?rmed by SEM)at the current densities higher than 1C (Fig.4a).These results emphasize the importance of strong mechanical attachment of the LiAl layer to Li anode foils.

Long-term cycle stability tests on full cells were performed at two different C-rates (C/5and 1C in Fig.4b and c).While Li anode cell showed gradual degradation during 200cycles at C/5,the two LiAl(60)and LiAl(90)cells noticeably slowed down the degradation rate after 100cycles and maintained ~60%of the initial capacity at 200th cycle (Fig.4b).This is nearly 200mAh g ?1higher capacity compared to bare Li.

At 1C,LiAl(90)signi ?cantly outperformed the bare Li for initial 30cycles (Fig.4c).It showed little if any degradation,while the bare Li lost more than 300mAh g ?1during the ?rst 30cycles (Fig.4c).Although LiAl(90)showed faster degradation rate than bare Li for the rest of the cycles,it maintained noticeably higher capacity up to 300th cycle,before the cell capacity reached the same value as the one with the bare Li anode,demonstrating promise for the proposed surface coating technology.The post-mortem analysis of the cycled cells with LiAl(90)anode revealed partial delamination of the alloy layer from the underlying Li.Higher rate tests induced larger stresses within the alloy layer,making the requirements for the mechanical properties of the alloy –Li interface stricter.In spite of these limitations and the signi ?cantly higher surface area of the LiAl alloy layer (compared to a planar Li)available for the reaction with polysul ?des,the

100

120

140

160

I m (Z ) (O h m )

Re(Z) (Ohm)

Fig.2.EIS of the half cells prepared with bare Li,LiAl(60)and LiAl(90)anode and Li counter/reference electrode using 3M LiTFSI in DIOX:DME electrolyte with and without 0.2M Li 2S 8additive.

P o t e n t i a l (V v s . L i /L i +)

Time (h)Time (h)

P o t e n t i a l (V )

Time (h)

2S 8

2S 82S 8

Fig.3.Polarization tests performed in half-cells:(a)comparison of LiAl(60)and LiAl(90)to bare Li in electrolyte containing 0.2M Li 2S 8,(b)comparison of LiAl(90)to bare Li in the electrolyte without 0.2M Li 2S 8additive,and (c)a comparison of LiAl(90)and Li with 0.2M LiNO 3additive.The insets show enlarged view of cycles 93–97.

40H.Kim et al./Electrochemistry Communications 36(2013)38–41

comparison of the degree of overcharge during cycling calculated as (C charge /C discharge ×100)con ?rms that the alloy layer is none-theless capable of effectively protecting Li from polysul ?de

shuttling.While the overcharge of bare Li increases by 4%during 300cycles,the one of LiAl(90)increases nearly 0%(Fig.4c).4.Summary

In order to mitigate the reaction between the diffused polysul ?des and the Li anode which leads to the polysul ?de shuttle,a thin layer of Li –Al alloy material was synthesized on the surface of a Li foil.Curing the alloy layer at different temperatures was found to affect the electro-chemical performance of the Li anode.The Li anodes with an alloy layer cured at the highest temperature of 90°C provided the best rate capa-bility with lowest charge transfer resistance,showing signi ?cant im-provement over bare Li anode.Cycling polarization tests show that the alloy layered Li is stable over 100cycles on plating/deplating,while bare Li cell shows unstable polarization.This alloy layer was found to be more ef ?cient in protecting Li than the SEI formed by LiNO 3additive in electrolyte.The Li –S cells using S/OM-CDC cathodes and Li –Al alloy coated Li foil showed noticeably better cycle stability than similar cells assembled with bare Li.In our future studies we will investigate the impact of the reduced thickness of the alloy layer and the use of com-posites,where one component provides structural reinforcement of the surface layer.Acknowledgments

This work was supported by the US National Science Foundation (NSF)(grant 0954925)and by the Energy Ef ?ciency &Resources program of the Korea Institute of Energy Technology Evaluation and Planning (KETEP)funded by the Korea government Ministry of Knowledge Economy (grant 20118510010030).References

[1] E.Peled,Y.Sternberg,A.Gorenshtein,https://www.wendangku.net/doc/586618675.html,vi,J.Electrochem.Soc.136(1989)

1621–1625.

[2] D.Aurbach,E.Pollak,R.Elazari,G.Salitra,C.S.Kelley,J.Af ?nito,J.Electrochem.Soc.

156(2009)A694–A702.

[3]X.Ji,K.T.Lee,L.F.Nazar,Nat.Mater.8(2009)500–506.

[4]M.Oschatz,S.Thieme,L.Borchardt,M.R.Lohe,T.Biemelt,J.Bruckner,H.Althues,S.

Kaskel,https://www.wendangku.net/doc/586618675.html,mun.49(2013)5832–5834.

[5]J.T.Lee,Y.Zhao,S.Thieme,H.Kim,M.Oschatz,L.Borchardt,A.Magasinski,W.I.Cho,S.

Kaskel,G.Yushin,Adv.Mater.(2013),https://www.wendangku.net/doc/586618675.html,/10.1002/adma.201301579.[6]Z.W.Seh,W.Li,J.J.Cha,G.Zheng,Y.Yang,M.T.McDowell,P.-C.Hsu,Y.Cui,Nat.

Commun.4(2013)1331.

[7]H.Kim,J.T.Lee,D.-C.Lee,A.Magasinski,W.-i Cho,G.Yushin,Adv.Energy Mater.

(2013),https://www.wendangku.net/doc/586618675.html,/10.1002/aenm.201300253.

[8]S.Evers,L.F.Nazar,Acc.Chem.Res.46(2012)1135–1143.

[9]H.Kim,J.T.Lee,G.Yushin,J.Power Sources 226(2012)256–265.[10]Y.S.Nimon,M.-Y.Chu,S.J.Visco,in,Google Patents,2003.

[11]Y.Korenblit,M.Rose,E.Kockrick,L.Borchardt,A.Kvit,S.Kaskel,G.Yushin,Acs Nano

4(2010)1337–1344.

[12]I.T.Kim,A.Magasinski,K.Jacob,G.Yushin,R.Tannenbaum,Carbon 52(2013)

56–64.

[13] A.Magasinki,B.Zdyrko,I.Kovalenko,B.Hertzberg,I.Burtovyy,T.Fuller,I.Luzinov,G.

Yushin,ACS Appl.Mater.Interfaces 2(2010)3004–3010.

[14]T.Brun,J.Jorgensen,M.Misawa,F.Rotella,S.Susman,https://www.wendangku.net/doc/586618675.html,dner,J.Electrochem.

Soc.129(1982)2509–2512.

Cycle Number

Overcharge (%)

Cycle Number

D i s c h a r g e C a p a c i t y (m A h g -1)

Rate

Fig.4.Battery tests of the Li –S cells.(a)rate performance of LiAl(25),LiAl(60)and LiAl(90)in comparison with a bare Li anode,(b)cycle stability of cells with Li,LiAl(60)and LiAl(90)anodes recorded at C/5rate,(c)cycle stability and degree of overcharge of cells with Li and LiAl(90)anodes recorded at 1C rate.

H.Kim et al./Electrochemistry Communications 36(2013)38–41