1

N dordre 208-2014 Anne 2014

Thse prsente devant

LUNIVERSITE CLAUDE BERNARD LYON 1

ECOLE DOCTORALE DE CHIMIE 206

Spcialit Chimie

soutenue publiquement le 24 Octobre 2014

pour lobtention du

DIPLOME DE DOCTORAT (arrt du 7 aot 2006)

Par

La CHANCELIER

DEVELOPPEMENT DE SOLUTIONS INNOVANTES DELECTROLYTES

POUR SECURISER LES ACCUMULATEURS LITHIUM-ION

Directrice de thse Dr Catherine C. SANTINI CNRS

Co-encadrants Dr Sophie MAILLEY CEA, LITEN Dr Thibaut GUTEL CEA, LITEN

JURY Pr Giovanni B. APPETECCHI ENEA, Rome, Italie Rapporteur Dr Corinne LAGROST Universit de Rennes Rapporteur Dr Ccile TESSIER SAFT, Bordeaux Examinateur Pr Bruno ANDRIOLETTI Universit de Lyon 1 Examinateur Dr Guy MARLAIR INERIS Examinateur

2

3

UNIVERSITE CLAUDE BERNARD - LYON 1

Prsident de lUniversit

Vice-prsident du Conseil dAdministration

Vice-prsident du Conseil des Etudes et de la Vie Universitaire

Vice-prsident du Conseil Scientifique

Directeur Gnral des Services

M. Franois-Nol GILLY

M. le Professeur Hamda BEN HADID

M. le Professeur Philippe LALLE

M. le Professeur Germain GILLET

M. Alain HELLEU

COMPOSANTES SANTE Facult de Mdecine Lyon Est Claude Bernard

Facult de Mdecine et de Maeutique Lyon Sud Charles Mrieux

Facult dOdontologie

Institut des Sciences Pharmaceutiques et Biologiques

Institut des Sciences et Techniques de la Radaptation

Dpartement de formation et Centre de Recherche en Biologie Humaine

Directeur: M. le Professeur J. ETIENNE

Directeur: Mme la Professeure C. BURILLON Directeur: M. le Professeur D. BOURGEOIS

Directeur: Mme la Professeure C. VINCIGUERRA

Directeur: M. le Professeur Y. MATILLON

Directeur: M. le Professeur P. FARGE

COMPOSANTES ET DEPARTEMENTS DE SCIENCES ET TECHNOLOGIE

Facult des Sciences et Technologies Dpartement Biologie Dpartement Chimie Biochimie Dpartement GEP Dpartement Informatique Dpartement Mathmatiques Dpartement Mcanique Dpartement Physique Dpartement Sciences de la Terre

UFR Sciences et Techniques des Activits Physiques et Sportives

Observatoire des Sciences de lUnivers de Lyon

Polytech Lyon

Ecole Suprieure de Chimie Physique Electronique

Institut Universitaire de Technologie de Lyon 1

Institut Universitaire de Formation des Matres

Institut de Science Financire et d'Assurances

Directeur: M. le Professeur F. DE MARCHI Directeur: M. le Professeur F. FLEURY Directeur: Mme le Professeur H. PARROT Directeur: M. N. SIAUVE Directeur: M. le Professeur S. AKKOUCHE Directeur: M. le Professeur A. GOLDMAN Directeur: M. le Professeur H. BEN HADID Directeur: Mme S. FLECK Directeur: Mme la Professeure I. DANIEL

Directeur: M. C. COLLIGNON

Directeur: M. B. GUIDERDONI

Directeur: M. P. FOURNIER

Directeur: M. G. PIGNAULT

Directeur: M. C. VITON

Directeur: M. A. MOUGNIOTTE

Administrateur provisoire: M. N. LEBOISNE

4

5

Thse prpare au sein de:

Laboratoire des Composants pour Batteries (LCB)

CEA, LITEN

(Laboratoire dInnovation pour les Technologies des Energies Nouvelles)

17 rue des martyrs 38054 Grenoble

France

et

Laboratoire de Chimie OrganoMtallique de Surface (LCOMS)

C2P2

(Chimie, Catalyse, Polymres et Procds)

UMR 5265 CNRS-Universit Lyon 1-ESCPE

43 Bd du 11 Novembre 1918 69616 Villeurbanne

France

6

7

N'allez pas l o le chemin peut mener ; Allez l o il n'y a pas de chemin et laissez une trace.

Ralph Waldo Emerson

8

9

Acknowledgments

First I would like to thank Giovanni Appetecchi and Corinne Lagrost, who accepted to review this work, for their fruitful questions and remarks. Thank you for helping me to improve this manuscript. Thanks to Ccile Tessier, Guy Marlair and Bruno Andrioletti for their interest in this work, their presence to my defense, and the sincerely interesting discussion. Merci mes encadrants, grce qui je suis heureuse du travail accompli : Catherine, merci de mavoir pousse dans les bonnes directions scientifiques, davoir t lcoute professionnellement et humainement. Merci pour votre aide, votre patience, votre positivisme, votre confiance, et votre suivi. Thibaut, merci pour ta comprhension, ta disponibilit, pour les discussions et conseils. Merci pour ton soutien et tes encouragements tout au long de la thse. Sophie, merci pour ton encadrement, la confiance et la libert accordes. Merci Bernadette Charleux puis Timothy Mckenna pour mavoir accueillie au sein du laboratoire C2P2. Merci Sbastien Patoux et Marlne Rey pour leur accueil chaleureux au sein du LMB, LCB et SCGE. Merci de mavoir permis deffectuer de lenseignement durant ces trois ans, et merci de mavoir offert lopportunit de participer des congrs, expriences combien enrichissantes. Merci maintenant tous les collgues qui sont devenus bien plus :

Tout dabord, ceux avec qui jai commenc mais qui sont partis, plus ou moins loin: Hassan, merci pour ton aide, ton efficacit et tes multiples rponses mes multiples questions en synthse. Merci pour les footings, les discussions, les dlicieux repas et ths libanais. Merci pour le petit dej au Hilton et les selfies Miami ! Inga, merci davoir tant couru et discut avec moi. Merci pour ton coute, ta culture, merci de mavoir fait dcouvrir ton pays, le caf Juliette, et pour ton art de faire de belles soires! Laurent, merci pour les dfoulements au squash, les dcouvertes de bonnes adresses et de mavoir tant fait rire. Merci pour tes mails adorables mme de si loin.

Ensuite, ceux qui ont toujours t l, que jai ador savoir dans les parages: Philippe, merci pour tes conseils vgtariens, merci dtre absolument toujours prt aider. Merci pour les ides de balade jamais ralises et les autres sorties ! Anthony, merci pour les randos, les conseils sportifs, littraires, cinmatographiques... Merci pour les rservations de notre table pour les quiz du Bryans ! Walid, Cherif, merci davoir t toujours dispo pour un coup de main et pour discuter. Sylwia, ma Sylwia Merci ! Pour nos fous rires, nos discussions passionnes, notre persverance, nos sorties Thank you for awesome moments, for example with Hulk and megarich !

10

Enfin les collgues qui ont t suffisamment l pour que a compte Lyon:

Fred, merci pour une belle collaboration cruciverbiste, pour tre parmi ceux qui se bougent au labo et hors labo, ce qui a cr de beaux souvenirs. Merci. Teresa, merci pour ton soutien les jours moyens. Merci pour ta bonne humeur, ta ractivit, ton dynamisme, ton adorabilit La motivation mutuelle fut splendide, mme en Aot Je ne serai jamais bien loin ! Thibault, Ewelina, Phillip, Giuliana, Andreia, Walid, merci pour votre joie de vivre, pour les discussions plus ou moins srieuses, votre enthousiasme pour sortir, pour lambiance multiculturelle que jadore Oui, vous mritez chacun plus que cette phrase, a lot more ! ! ! By the way, special thanks to our lovely Anoocoms team ! Merci Nico pour les BBQ, David, Delphine, Fadila, Guilhem, Iuliia, Stphane, Popoff, Henri, Leila Merci aux stagiaires avec qui jai eu loccasion de travailler : Piotr, Elodie et Cristina. Pierre et Arthur, merci pour la bonne ambiance pendant ma priode dans lopen space. Merci Emmanuelle pour ton aide et ta disponibilit. Mos, Aimery, Emile, Laurent, Alessandra, et le reste de lquipe, merci tous pour lambiance richissime.

Et Grenoble: Greg, merci pour tes conseils en salle anhydre, merci pour les coups de main successifs, merci pour ta bonne humeur, ton coute, tes blagues, ta sincrit ! Djamel, merci pour ton aide, indispensable (Arbin, lectrodes). Merci pour ton humour et pour les sorties au top ! Kim, merci davoir simplifi toutes les complications administratives, avec le sourire ! Et, aussi, merci pour lorganisation des vnements non professionnels hyper sympas ! Merci Charles pour ton assistance rgulire (informatique, technique, hbergement). Merci Isabel (aprs tavoir perdue de vue cest vraiment gnial de te retrouver pour discuter, nager, boire un verre) ! Merci Adriana, Melody, Hanaa, JB, Mathieu, Elise (pour les nombreuses requtes sur mes lectrodes), Cline, Aurlien, Justin, Vincent, Lise, Graldine (et ta grand-mre). Merci Sandra (linattendu) et Marco (htel Suisse et Bordeaux) pour avoir agrment de discussions sympas mes soires Grenoble en solo! Cette thse ma permis de travailler avec beaucoup de personnes que je remercie vivement, et en particulier : En XPS, merci Jean-Luc et Laurence pour votre courage avec mes lectrodes. Merci Anass pour ta patience, ta gentillesse, ta pdagogie. En ATG, merci Olivier pour la formation, et la confiance accorde. En DRIFT-GC-MS, merci Kai pour ton aide, ta disponibilit. En RMN, merci Christine et Frdric. En MS, merci Philippe et Johann pour laide avec le spectro, malgr les fuites En GC-IR, merci Isabel. Hamed, merci infiniment pour ton coute, ta ractivit, ton enthousiasme. Ce fut un rel plaisir de minvestir dans lenseignement sous ta tutelle. Arnaud, merci pour mavoir forme de A Z sur les TPs, merci pour ta patience et ton enthousiasme.

11

I would like also to thank the great people I met in congresses around the world, for their participation to my very very very good time there ! In particular I would like to name Claudiu (ECS), Mei (COIL and London or Berlin soon and for the proofreading mon chou !), Marco (IMLB), Edgard (ECS and Miami), Giuseppe (IMLB), Tom (IMLB), Andreas (XPS in Roscoff), Paul (COIL, Netherlands), Max (COIL), Jakub (IMLB), Sigelinde (XPS in Roscoff), Carmen (EUCHEM, COIL) I really hope to keep in touch with you all !

Claire et Mlissa, merci davoir emmnag Lyon, merci pour les sessions piscine, les longues discussions, profondes ou lgres, et les conseils, ponctus de fous rires. Sur la vie, les mecs, les vernis, la piscine, les cheveux, la vie quoi Merci Margaux, parce que je sais que je peux compter sur toi depuis toujours ! Merci Camille et Kelly retrouves ! Et merci Ebru Parce que 10 ans damiti, cest beau, cest bon, cest drle, cest fort, cest grand, cest magique, cest magnifique ! ! ! Merci Rom, Alex, Denis, Seb, Fred, Tomtom, Marina, Simon Merci dtre toujours motivs pour que lon reste en contact. Merci pour les sorties (ski, soutenances 2014 ne perdons pas le rythme !) et les discussions ! Merci Kevin, pour ton soutien, tes blagues, tes rflexions et tes textos philosophiques mrement rflchis, et merci pour la totale disponibilit de ton luxueux appart. Merci Juliette, Benji, Ben, Bastien, Brice, lautre Brice, Camille, Sami, pour des soires et des vacances au top ! Pierre, merci de mavoir soutenue, voire pousse. Merci pour le succs international des animations powerpoint. Un immense merci pour tant dautres choses. Merci la famille ! Merci au soutien de tous les oncles, tantes, cousines, cousins. De prs, de loin, cest tellement important de vous savoir mes cts ! Merci mes Mamies pour leur force, leur gnrosit, leur coute. Entre autres qualits. Maman, Papa, merci pour votre coute inlassable, votre soutien incommensurablement indfectible, votre confiance inbranlable. Bro, merci pour tout. On parle moins mais ce nen est pas moins fort Merci toutes les personnes qui sont venues assister ma soutenance, et celles qui mont envoy leurs encouragements. Merci tous ceux et toutes celles que je ne citerai pas mais avec qui jai eu plaisir passer ou repasser du temps, avec qui jai pu discuter, rire, danser, pleurer, courir, dlirer, jouer, vivre... Bref, merci toutes les personnes qui ont contribu crer de beaux moments. Merci toutes les personnes qui mont aide me construire jusquici, grce qui je me sens heureuse et panouie, et merci par avance tous ceux et celles qui feront que a dure !

12

Table of contents

13

Abstracts

Summary in french

Introduction

Chapter 1 State of the art General context Lithium-ion batteries Strategy of the work References

Chapter 2 Decomposition temperatures of ionic liquids Introduction General synthesis procedure Experimental parameters for TGA Gathering of decomposition temperatures Stability trends Conclusion Experimental part References

Chapter 3 Thermal stability Introduction Decomposition temperatures Isothermal experiments Maximum operating temperatures Effect of lithium salt concentration Effect of atmosphere Thermal treatment of IL-based electrolytes Combustion behaviour Thermal stability of electrodes Conclusion Experimental part References

Chapter 4 Electrochemical stability Introduction Electrochemical windows Cycling tests in coin cells Lithium diffusion coefficients Lithium insertion Cycling tests in pouch cells Overcharge behaviour Conclusion Experimental part References

Conclusion and outlooks

Appendixes

15

19

31

37 40 41 55 57

65 68 69 70 75 78 91 92 94

101 104 104 106 110 112 115 117 124 128 132 133 135

139 142 143 145 148 151 153 155 166 167 171

175

183

14

15

Dveloppement de solutions innovantes dlectrolytes pour scuriser les accumulateurs lithium-ion

Mots-clefs batterie lithium-ion; lectrolyte; scurit; liquide ionique;

stabilits thermique et lectrochimique; combustion; surcharge; cyclage

Rsum

Les batteries lithium-ion dominent le march des appareils nomades et celui des vhicules

lectriques. Nanmoins elles posent des problmes de scurit lis leur lectrolyte,

contenant des carbonates inflammables et volatils. Pour scuriser ces systmes, les liquides

ioniques (LI) sont tudis comme lectrolytes alternatifs. Ce sont des sels liquides

temprature ambiante, rputs stables thermiquement et non inflammables. Ce caractre

scuritaire des LI, souvent avanc, est pourtant peu tay par des expriences probantes. Les

travaux de cette thse visent comprendre le comportement de ces LI en situations abusives,

telles quun chauffement de la batterie, un feu ou une surcharge. Les tempratures de

dcomposition de LI contenant les cations imidazolium ou pyrrolidinium diffremment

substitus et lanion bis(trifluoromethanesulfonyl)imide ont t dtermines par analyse

thermogravimtrique (ATG). Une analyse critique des donnes (de la littrature et de nos

mesures) a permis de dfinir une procdure optimise, pour obtenir des rsultats

reproductibles et comparables. Des lectrolytes constitus de mlanges de carbonates ou de LI

et de sels de lithium ont t analyss par ATG dynamique et isotherme, et leurs produits de

dcomposition ont t identifis. Leur comportement au feu a t test par la mesure des

chaleurs de combustion, des dlais dinflammation et lidentification des gaz gnrs. Des

tests de cyclage lectrochimique ont t mens avec ces mmes lectrolytes dans des systmes

lithium-ion constitus des lectrodes Li4Ti5O12 et LiNi1/3Mn1/3Co1/3O2. Lvolution des

lectrolytes et des surfaces des lectrodes en situation de surcharge a t examine.

16

17

Development of innovative electrolytes for safer lithium-ion batteries

Keywords lithium-ion; battery; electrolyte; safety; ionic liquid;

thermal and electrochemical stabilities; combustion; overcharge; cycling

Abstract

Lithium-ion batteries are dominating both the nomad device and electric vehicle markets.

However they raise safety concerns related to their electrolyte, which consists of flammable

and volatile carbonate mixtures and toxic salts. The replacement of the latter by ionic liquids

(IL), liquid salts claimed to be thermally stable and non-flammable, could provide a safer

alternative. Yet this often claimed feature has been poorly examined by experiments. The

work of this thesis investigates IL behaviour under abuse conditions such as overheating, fire

or overcharge. Decomposition temperatures of IL based on differently substituted

imidazolium or pyrrolidinium cations and the bis(trifluoromethanesulfonyl)imide anion were

determined by thermogravimetric analysis (TGA). A critical study of gathered data (from

literature and our work) led to the determination of an optimised procedure to obtain

reproducible and comparable results. Electrolytes based on carbonates mixtures or IL and

containing lithium salt were studied by dynamic and isothermal TGA, and their

decomposition products were identified. Their combustion behaviour was also tested by

measuring heats of combustion and ignition delays. Emitted gases were analysed and

quantified. Electrochemical cycling tests were carried out with these electrolytes in

lithium-ion systems based on Li4Ti5O12 and LiNi1/3Mn1/3Co1/3O2 electrodes. The evolution of

the electrolytes and electrodes surface was also examined under overcharge.

18

RSUM SUBSTANTIEL

EN FRANAIS

French summary 20

French summary 21

1. Introduction gnrale

Le stockage de lnergie est au cur des enjeux de notre socit, notamment avec lessor des

nergies renouvelables et des vhicules lectriques. Du fait de leurs performances, les

batteries de technologie lithium-ion sont actuellement les plus utilises, notamment pour les

appareils nomades (63% du march mondial). Si leur dangerosit reste limite pour des

appareils de petite taille, elles peuvent poser des problmes de scurit pour des applications

telles que les vhicules lectriques. Ces accumulateurs doivent en effet pouvoir rsister des

situations de surchauffe, surcharge, surdcharge, court-circuit ou choc.

Les batteries lithium-ion stockent de llectricit par insertions-dsinsertions successives des

ions lithium dans chaque matriau dlectrode. Llectrode positive est gnralement un

oxyde mtallique base de mtaux de transition tels que le cobalt, le fer, le nickel ou le

manganse. Llectrode ngative est constitue de graphite. Ces deux lectrodes sont spares

par un isolant lectronique, imbib dune solution conductrice ionique, appele lectrolyte.

Pendant lutilisation (dcharge), les ions Li+ sinsrent dans llectrode positive, gnrant un

flux dlectrons dans le circuit extrieur, qui alimente lappareil connect, Figure 1.[1] Lors de

la recharge, un courant est impos pour forcer la migration des ions Li+ vers llectrode

ngative. Lalternance de charges et dcharges est appele cyclage lectrochimique.

Figure 1: De gauche droite, Schma dune batterie en dcharge, carbonate dthylne (EC),

carbonate de dithyle (DEC) et hexafluorophosphate de lithium (LiPF6)

Llectrolyte est constitu de mlanges de carbonates, Figure 1, qui solubilisent bien le sel de

lithium et fournissent de bonnes performances lectrochimiques. Dans notre cas llectrolyte

utilis, not [EC:DEC][LiPF6], est un mlange qui-volumique de EC et DEC contenant

1 mol.L-1 de LiPF6. Cependant ces liquides volatils et inflammables peuvent mener des

problmes de scurit (incendie, explosion). De plus le sel LiPF6 mne la formation de

composs toxiques comme lacide fluorhydrique (HF). Pour les remplacer, certains sels

fondus appels liquides ioniques (LI) (sels frquemment liquides temprature ambiante)

French summary 22

sont des candidats potentiels, pouvant prsenter de bonnes performances.[2-5] Ils sont

composs dun cation souvent issu dune amine et dun anion gnralement fluor, et

prsentent une bonne conductivit ionique. Les LI sont liquides sur une large gamme de

temprature,[6] dont la limite suprieure est leur temprature de dcomposition (et non leur

bullition) gnralement assez leve. De plus, ils possdent une pression de vapeur saturante

ngligeable, ce qui leur confre une faible inflammabilit[7] et les rend plus scuritaires.

Cet aspect de suret des LI est un argument souvent avanc,[8] mais peu tay par des

expriences probantes. Les travaux mens dans le cadre de cette thse visent comprendre le

comportement des LI lorsquils sont soumis des conditions dites abusives, telles quun

chauffement de la cellule, un feu, une surcharge etc.

2. Stabilit thermique des lectrolytes

Parmi les plus utiliss, les cations imidazolium et pyrrolidinium combins lanion fluor

bis(trifluoromethanesulfonyl)imide [NTf2] ont t slectionns, Figure 2. Les lectrolytes

correspondants, contenant 1 mol.L-1 de sel de lithium LiNTf2, seront nots [cation][Li][NTf2].

Ces LI, dont la synthse et la purification sont maitrises, prsentent une haute stabilit

thermique et des proprits physicochimiques adaptes leur utilisation en batteries

(viscosit, conductivit).

Figure 2: de gauche droite, les cations 1-butyl-3-methylimidazolium [C1C4Im],

N-butyl-N-methylpyrrolidinium [PYR14], et lanion [NTf2]

La dtermination de la temprature de dcomposition (Td) par Analyse ThermoGravimtrique

(ATG) est couramment utilise pour dfinir la stabilit thermique des LI. Il sagit de suivre la

dcomposition de lchantillon (rvle par une perte de masse) pendant une monte en

temprature dans des conditions contrles (atmosphre, rampe de chauffe). Suivant les

paramtres exprimentaux utiliss, les valeurs de Td pour un mme produit varient de plus de

100 C. Une analyse critique des donnes de la littrature et de nos rsultats nous a mens

dfinir une procdure optimise, permettant dobtenir des rsultats reproductibles et

comparables. Ces lectrolytes ont des tempratures de dcomposition suprieures de plus de

200 C celle des carbonates, Figure 3.

French summary 23

Figure 3: Profils de stabilit thermique tablis par ATG entre 30 et 500 C pour

[C1C4Im][Li][NTf2] (Td: 357 C), [PYR14][Li][NTf2] (Td: 339 C), [C1C1C4Im][Li][NTf2] (Td: 339 C) and [EC:DEC][LiPF6] (Td: 50 C). Echantillons de 10 mg; vitesse de chauffe de 5 C.min-1 sous argon;

creusets en aluminium scells

Nanmoins lATG ne permet pas didentifier les produits de dcomposition. Les deux

lectrolytes [C1C4Im][Li][NTf2] et [PYR14][Li][NTf2] ont t traits sous vide deux heures

350 C et analyss. Pour les deux solutions, des hydrocarbures gazeux inflammables

(typiquement des butnes) issus de llimination des chaines alkyles cationiques ont t

identifis par spectromtrie de masse, rsonance magntique nuclaire et chromatographie en

phase gazeuse, Figure 4. La dcomposition de lanion, contenant du fluor et du soufre, a men

la formation despces toxiques telles que de lacide fluorhydrique et le dioxyde de soufre.[9]

Figure 4: Analyse par chromatographie en phase gaz des constituants de la phase gaz issue de la

dcomposition thermique des lectrolytes (temps de rtention des butnes: 5.85 min)

100 200 300 400 5000

20

40

60

80

100

Mas

se (%

)

Temprature (C)

[C1C

4Im][Li][NTf

2]

[PYR14

][Li][NTf2]

[C1C

1C

4Im][Li][NTf

2]

[EC:DEC][LiPF6]

Td

< 100 C > 330 C

0 1 2 3 4 5 6 7 8 9 10

2,147,64

1,31

Inte

nsit

Temps de rtention (min)

[C1C4Im][Li][NTf2] [PYR14][Li][NTf2]

1,77

3,70

5,70

5,85

6,01

7,48

7,96

French summary 24

3. Comportement en combustion des lectrolytes

Le comportement au feu de ces lectrolytes a t galement test. Les chaleurs de combustion

et les dlais dinflammation de chaque lectrolyte ont t dtermins par calorimtrie incendie

(norme ISO:12136), confirmant que ces LI ont une faible inflammabilit, en particulier celui

bas sur limidazolium. Le dlai dinflammation est denviron cinq minutes pour les

lectrolytes bass sur les LI, alors quil est de trente secondes pour les carbonates, Tableau 2.

Une fois leur combustion amorce, les lectrolytes base de LI dgagent presque deux fois

moins de chaleur (~ 8 vs 14 MJ.kg-1) que les carbonates. A titre de comparaison, la chaleur de

combustion du bois est de 15 MJ.kg-1.

[C1C4Im][Li][NTf2] [PYR14][Li][NTf2] Carbonates[10]

Dlai dinflammation (min) 5 5.5 0.5 Chaleur de combustion (MJ.kg-1) 7.7 8.2 14

Tableau 1: Comportement au feu des diffrents lectrolytes Afin dtablir la toxicit en cas de feu, les produits asphyxiants (HCN, CO) ou irritants (NOx,

SO2, et HF) mis lors de la combustion ont t quantifis. Les espces mises sont les mme

pour les lectrolytes [C1C4Im][Li][NTf2] et [PYR14][Li][NTf2], avec une mission de gaz

toxiques provenant de la dcomposition de lanion NTf2. La dcomposition des cations

produit des espces inflammables.

Facteurs dmission

(mg.g-1) [C1C4Im][Li][NTf2] [PYR14][Li][NTf2]

CO2 552 531 CO 21.6 25.1

Suies 11 37.7 Hydrocarbures 4.8 0.6

SO2 353 317 NO 4.9 3.0 HF 294.8 216.6

HCN 6.9 8.3 Tableau 2: Facteurs dmission de diffrents gaz forms pendant la combustion des lectrolytes

[C1C4Im][Li][NTf2] et [PYR14][Li][NTf2]

French summary 25

4. Caractrisations electrochimiques Les lectrolytes [EC:DEC][LiPF6], [C1C4Im][Li][NTf2], [PYR14][Li][NTf2] et

[C1C1C4Im][Li][NTf2] ont t utiliss en tant qulectrolytes dans des batteries (de formats

pile bouton et sachet souple) avec Li4Ti5O12 (LTO) et LiNi1/3Co1/3Mn1/3O2 (NMC) en tant que

matriaux dlectrodes ngative et positive. Les cyclages galvanostatiques ont t effectus

25 et 60 C, un rgime de C/10 (charges et dcharges en 10 h), Figure 5.

Figure 5: Performances en cyclage des quatre lectrolytes 25 C (gauche) et 60 C (droite);

5 cycles C/20 suivis de 95 cycles C/10 entre 1 et 3.5 V A 25 C, les cellules dmontrent un cyclage rversible et trs stable. Llectrolyte

[EC:DEC][LiPF6] donne la plus haute capacit (181 mAh.g-1) compare celle de

[C1C4Im][Li][NTf2] (110 mAh.g-1) et de [PYR14][Li][NTf2] (33 mAh.g-1).

[C1C1C4Im][Li][NTf2] na pas permis de cycler les cellules. A 60 C, les capacits initiales

sont plus leves qu 25 C, probablement grce une diminution de la viscosit. Cependant

les performances chutent de faon continue pour chaque lectrolyte. En particulier aprs

environ 60 cycles, une diminution brutale est observe pour les lectrolytes

[C1C4Im][Li][NTf2] et [EC:DEC][LiPF6].

Les performances des cellules contenant des liquides ioniques sont infrieures celles des

carbonates pour les deux tempratures et peuvent tre attribues la viscosit plus leve des

liquides ioniques. Pour vrifier ceci, les cintiques de diffusion des ions lithium au sein des

lectrodes et de llectrolyte ont t tudies par voltammtrie cyclique et par RMN en phase

liquide. La diffusion du lithium sest avre dix fois plus rapide au sein des carbonates que

dans les LI, Tableau 3. Pour tous les lectrolytes, la diffusion dans la phase liquide sest

rvle 10 000 fois plus rapide que dans llectrode.

0 20 40 60 80 1000

50

100

150

200

[EC:DEC][LiPF6]

[C1C

4Im][Li][NTf

2]

[PYR14

][Li][NTf2]

[C1C

1C

4Im][Li][NTf

2]

Cap

acit

dc

harg

e (m

Ah.

g-1 )

Numro de cycle

0 20 40 60 80 1000

50

100

150

200

Cap

acit

dc

harg

e (m

Ah.

g-1 )

Numro de cycle

French summary 26

[EC:DEC][LiPF6] [C1C4Im][Li][NTf2] [C1C1C4Im][Li][NTf2] [PYR14][Li][NTf2]

DLi par VC 3.53.10-10 2.68.10-11 1.40.10-11 1.60.10-12 DLi par RMN 3.08.10-6 1.69.10-7 2.24.10-7 1.44.10-7

Tableau 3: Coefficients de diffusion de Li+ (cm.s-1) dtermins 60 C, dans les lectrodes par voltammtrie cyclique (VC) et en solution par RMN du 7Li

Cependant, ces rsultats montrant de meilleures performances des carbonates 60 C

diffrent de la littrature. Gnralement les liquides ioniques sont plus performants que les

carbonates haute temprature, ces derniers ntant pas stables thermiquement.[9, 11-13] Dautre

part, [PYR14][Li][NTf2] sest rvl moins performant que [C1C4Im][Li][NTf2] alors quil

fournit de plus hautes capacits en demi-piles Li // graphite.[14] Enfin, llectrolyte

[C1C1C4Im][Li][NTf2] na pas permis de cycler les batteries bases sur LTO // NMC (capacit

de dcharge nulle aprs 10 cycles) alors quil donne de meilleures performances dans le cas

de batteries graphite // LiFePO4.[15] Ces rsultats laissent penser que llectrode positive NMC

joue un rle dstabilisant. Pour vrifier cette hypothse des tests ont t mens par

voltammtrie cyclique (VC) pour analyser les mcanismes dinsertion du lithium dans les

lectrodes.

Figure 6: Comparaison des seconds cycles par VC des quatre lectrolytes 60C

dans des cellules LTO // NMC Les processus dinsertion et de dsinsertion du lithium sont observs pour tous les lectrolytes

tudis, Figure 6. Les intensits de courant sont plus faibles dans le cas des liquides ioniques

que pour [EC:DEC][LiPF6], donnant lieu un aplatissement des pics. Le dcalage entre le

potentiel dinsertion et de dsinsertion est plus important dans le cas des LI (~ 0.45 V contre

0.14 V). De plus, les pics sont largis dans le cas des LI, ce qui permet de visualiser plusieurs

processus doxydation et de rduction des mtaux de llectrode positive. Daprs la

1,0 1,5 2,0 2,5 3,0-1,5

-1,0

-0,5

0,0

0,5

1,0

1,5

Den

sit

de

cour

ant (

mA

.cm

-2)

Tension (V)

[EC:DEC][LiPF6]

[C1C

4Im][Li][NTf

2]

[C1C

1C

4Im][Li][NTf

2]

[PYR14

][Li][NTf2]

French summary 27

littrature le manganse ne joue pas de rle dans ces procds et les ractions redox se font

dans lordre Ni2+/Ni3+, Ni3+/Ni4+ puis Co3+/Co4+.[16, 17] Il est possible que la prsence dions

mtalliques au degr doxydation (+II) mne la formation dagrgats de type [M(NTf2)n]x-

(M = Ni, Co ou Mn), qui favoriserait la dissolution de la matire active et donc diminuerait la

capacit disponible lors du cyclage.[18]

5. Analyses en surcharge Des cellules en sachets souples LTO // NMC de plus grande capacit (~10 mAh) ont t

assembles avec les lectrolytes [EC:DEC][LiPF6], [C1C4Im][Li][NTf2] et [PYR14][Li][NTf2]

afin didentifier les espces volatiles formes lors de la dcomposition des lectrolytes

pendant une surcharge. Une charge jusqu 4.5 V a t applique (6 V vs Li+/Li) un rgime

de charge de C/10, et cette tension a t maintenue pendant 20 h 60 C. Pendant la phase

tension constante, le courant utilis pour maintenir cette tension tait plus important dans le

cas de [C1C4Im][Li][NTf2] (0.4 vs 0.1 mA, Tableau 4), rvlant une moins bonne stabilit de

cet lectrolyte dans ces conditions. Cela a t confirm par les mesures des volumes des

cellules, qui ont augment de 2.81 mL pour [C1C4Im][Li][NTf2], de 0.30 mL pour

[EC:DEC][LiPF6], et de 0.12 mL pour [PYR14][Li][NTf2].

Courant de fuite (mA)

Augmentation de volume (mL)

[EC:DEC][LiPF6] 0.13 0.30 [C1C4Im][Li][NTf2] 0.40 2.81 [PYR14][Li][NTf2] 0.05 0.12

Tableau 4: Courants de fuite et augmentations de volume des cellules la fin de la surcharge Pour comprendre cette instabilit du liquide ionique contenant le cation imidazolium, les gaz

gnrs dans le sachet souple ont t analyss par chromatographie en phase gaz couple avec

un spectromtre infrarouge (GC-IR). Les gaz forms taient notamment du dioxyde de

carbone, des fragments issus de lanion et des alcanes issus du cation, Figure 7. Ces produits

sont donc diffrents de ceux observs lors de la dcomposition thermique de cet lectrolyte.

French summary 28

Figure 7: Chromatogramme des gaz gnrs dans la cellule contenant llectrolyte

[C1C4Im][Li][NTf2] aprs une surcharge 60 C Des analyses de diffraction des rayons X et microscopie lectronique ont t ralises sur les

lectrodes, aprs leur utilisation en cellules surcharges. La morphologie des deux lectrodes

na pas t modifie, et la structure de LTO est stable. Pour la NMC, la structure

cristallographique a volu vers une phase dlithie, Figure 8.

Figure 8: Diffractogrammes (DRX) des lectrodes ngatives (LTO, gauche) et positive (NMC, droite)

avant utilisation et aprs surcharge; #: dome en polymre utilis pour garder lchantillon sous atmosphere inerte; *: collecteur de courant en aluminium

Une tude dtaille des lectrodes aprs surcharge par Spectroscopie des Photolectrons X

(XPS) a dmontr que les surfaces des lectrodes taient masques. Elles taient recouvertes

de liquide ionique dans le cas de la positive, et de ses produits de dcomposition dans le cas

de llectrode ngative.

10 20 30 40 50 60 70

u.a.

2

(111

)

#

(220

)

(311

)

(400

)

(222

)

(331

)

(333

)

(440

)

*

(531

)

LTO originel (bas) LTO aprs surcharge (haut)

10 20 30 40 50 60 70

#

u.a.

2 ()

*

(003

)

(101

)

(006

)(1

02) (1

04)

(105

)

(107

)

(110

)(1

08) (

113)

NMC originel (bas) NMC aprs surcharge (haut)

French summary 29

6. Conclusion

Les lectrolytes forms par dissolution de 1 mol.L-1 de LiNTf2 au sein de [C1C4Im][NTf2] et

[PYR14][NTf2] prsentent une grande stabilit thermique compars aux carbonates

[EC:DEC][LiPF6], avec des tempratures de dcomposition suprieures 300 C. Les

produits drivs de limidazolium sont plus stables thermiquement que les pyrrolidinium. Ces

deux LI sont des espces trs peu combustibles, avec des dlais dinflammation suprieurs

cinq minutes (augments dun facteur 10 par rapport aux lectrolytes classiques). La

formation de gaz toxiques ou inflammables lors de la combustion est nanmoins prendre en

compte selon les applications vises.

En ce qui concerne les performances lectrochimiques, lutilisation de carbonates conduit de

meilleures capacits, y compris 60 C. La meilleure stabilit lectrochimique dans les

batteries LTO // NMC est observe pour [EC:DEC][LiPF6], suivi de [PYR14][Li][NTf2] puis

[C1C4Im][Li][NTf2], ce qui est un ordre inverse celui de la stabilit thermique. En situation

de surcharge, llectrolyte [C1C4Im][Li][NTf2] sest rvl le moins stable, gnrant 10 fois

plus de gaz que [EC:DEC][LiPF6] et [PYR14][Li][NTf2].

French summary 30

7. Rfrences

[1] J. M. Tarascon, "Vers des accumulateurs plus performants", L'actualit Chimique, 2002, 3(251), p130 [2] A. Lewandowski and A. Swiderska-Mocek, "Ionic liquids as electrolytes for Li-ion batteries-An

overview of electrochemical studies", J. Power Sources, 2009, 194(2), p601 [3] M. Armand, F. Endres, D. R. MacFarlane, H. Ohno and B. Scrosati, "Ionic-liquid materials for the

electrochemical challenges of the future", Nat. Mater., 2009, 8(8), p621 [4] D. R. MacFarlane, N. Tachikawa, M. Forsyth, J. M. Pringle, P. C. Howlett, G. D. Elliott, J. H. Davis,

M. Watanabe, P. Simon and C. A. Angell, "Energy applications of ionic liquids", Energy Environ. Sci., 2014, 7(1), p232

[5] A. Balducci, S. S. Jeong, G. T. Kim, S. Passerini, M. Winter, M. Schmuck, G. B. Appetecchi, R. Marcilla, D. Mecerreyes, V. Barsukov, V. Khomenko, I. Cantero, I. De Meatza, M. Holzapfel and N. Tran, "Development of safe, green and high performance ionic liquids-based batteries (ILLIBATT project)", J. Power Sources, 2011, 196(22), p9719

[6] H. Ohno, Electrochemical aspects of ionic liquid, 2nd ed., Wiley, N. Y., 2011 [7] A. O. Diallo, C. Len, A. B. Morgan and G. Marlair, "Revisiting physico-chemical hazards of ionic

liquids", Sep. Purif. Technol., 2012, 97, p228 [8] M. Galinski, A. Lewandowski and I. Stepniak, "Ionic liquids as electrolytes", Electrochim. Acta, 2006,

51(26), p5567 [9] L. Chancelier, A. O. Diallo, C. C. Santini, G. Marlair, T. Gutel, S. Mailley and C. Len, "Targeting

adequate thermal stability and fire safety in selecting ionic liquid-based electrolytes for energy storage", Phys. Chem. Chem. Phys., 2014, 16(5), p1967

[10] G. G. Eshetu, S. Grugeon, S. Laruelle, S. Boyanov, A. Lecocq, J. P. Bertrand and G. Marlair, "In-depth safety-focused analysis of solvents used in electrolytes for large scale lithium ion batteries", Phys. Chem. Chem. Phys., 2013, 15(23), p9145

[11] S. Menne, R. S. Khnel and A. Balducci, "The influence of the electrochemical and thermal stability of mixtures of ionic liquid and organic carbonate on the performance of high power lithium-ion batteries", Electrochim. Acta, 2013, 90, p641

[12] J. Li, S. Jeong, R. Kloepsch, M. Winter and S. Passerini, "Improved electrochemical performance of LiMO2 (M=Mn, Ni, Co)-Li2MnO3 cathode materials in ionic liquid-based electrolyte", J. Power Sources, 2013, 239(0), p490

[13] N. Wongittharom, T.-C. Lee, C.-H. Hsu, G. Ting-Kuo Fey, K.-P. Huang and J.-K. Chang, "Electrochemical performance of rechargeable Li/LiFePO4 cells with ionic liquid electrolyte: Effects of Li salt at 25 C and 50 C", J. Power Sources, 2013, 240, p676

[14] A. Guerfi, S. Duchesne, Y. Kobayashi, A. Vijh and K. Zaghib, "LiFePO4 and graphite electrodes with ionic liquids based on bis(fluorosulfonyl)imide (FSI)(-) for Li-ion batteries", J. Power Sources, 2008, 175(2), p866

[15] H. Srour, "Dveloppement d'un lectrolyte base de liquide ionique pour accumulateur au Lithium", Thesis from Universit Claude Bernard Lyon 1 (Lyon), 2013

[16] K. M. Shaju, G. V. S. Rao and B. V. R. Chowdari, "Performance of layered Li(Ni1/3Co1/3Mn1/3)O-2 as cathode for Li-ion batteries", Electrochim. Acta, 2002, 48(2), p145

[17] A. Deb, U. Bergmann, S. P. Cramer and E. J. Cairns, "In situ x-ray absorption spectroscopic study of the Li[Ni1/3Co1/3Mn1/3]O2 cathode material", J. Appl. Phys., 2005, 97(11), p113523

[18] H. Zheng, Q. Sun, G. Liu, X. Song and V. S. Battaglia, "Correlation between dissolution behavior and electrochemical cycling performance for LiNi1/3Co1/3Mn1/3O2-based cells", J. Power Sources, 2012, 207(0), p134

GENERAL

INTRODUCTION

General introduction 32

General introduction 33

Both the nomad device and electric vehicle markets are growing, requiring a fast development

of energy storage systems (ESS). These systems must be capable of addressing a number of

technical challenges for the sustainable development of electromobility (terrestrial, maritime

and aerial to some extent). They should efficiently store intermittent renewable sources of

energy (wind, solar, water) and be part of smart grids applications.[1] ESS previously had to

rely on lead-acid, nickel cadmium and nickel metal hydride technologies. Commercialized

since the 90s for the consumer market, lithium-ion (Li-ion) technology and its variants have

taken the lead regarding those emerging applications, as they offer significantly higher power

and energy densities (up to 2 000 W.kg-1 and 120 Wh.kg-1).[2]

However safety has been identified in a number of studies as a potential market restraint.[3, 4]

In the aim of improving the safety of these systems, all components of the cell have to be

considered. Major improvements were already made in the domains of battery management

systems, cell designs and separators. The electrodes were also developed in order to provide

safer batteries, as for the negative one, metallic lithium can be replaced by graphite (Cgr) or a

metallic oxide such as Li4Ti5O12 (LTO). These electrodes could reduce the risk of shortcircuit

due to the growth of lithium dendrites on the surface.[5, 6] In the case of the positive electrode

material, the state of the art Li-ion batteries use mostly LiCoO2 (LCO), but this oxide has a

low thermal stability. Mixed metallic oxides were widely examined with varying

compositions of cobalt, manganese, nickel and aluminium metals. In particular

LiNi1/3Mn1/3Co1/3O2 (NMC) showed good cycling stability and rate capability, higher thermal

stability in the charged state, lower cost and reduced toxicity.[7, 8]

The electrolyte is commonly a mixture of organic solvents such as propylene carbonate (PC),

ethylene carbonate (EC), dimethyl carbonate (DMC) or diethyl carbonate (DEC), containing a

dissolved inorganic lithium salt, typically lithium hexafluorophosphate (LiPF6). These

solvents are very flammable, possess low flash points and are highly volatile.[9] Their

decomposition under moderate thermal stress can lead fast to thermal runaway of the cell,

which can cause the electrolyte ignition or even explosion.

General introduction 34

In order to solve these safety issues, other types of electrolytes are studied, in particular ionic

liquids (IL), widely considered both in literature[10-12] and in industrial projects.[13-15] IL are

defined as salts liquid below 100 C and result from the association of an organic cation with

an organic or inorganic anion. Consequently, they can be designed for a chosen application by

tuning the nature of the anion or/and of the cation. IL exhibit negligible vapour pressure,

similar to solid salts, and unlike most organic solvents, they do not vaporize unless heated to

the point of thermal decomposition, typically 200 C to 300 C.[16] They have flash points

higher than 200 C[17] and are considered as non-flammable.[18, 19]

Contrarily to the current carbonate-based electrolytes,[20] thermal stability of IL for use as

electrolytes in case of abuse conditions (fire, shortcircuit, impact, overcharge or

overdischarge) has been poorly examined by experiments.[19] In the vast domain of IL,

imidazolium and pyrrolidinium cations associated to [NTf2] anion were selected for their high

decomposition temperatures.[21, 22]

The aim of this thesis is to investigate thermal stability up to combustion and electrochemical

behaviour up to overcharge of these IL and their corresponding electrolytes (defined as

solution of lithium salt in IL) for Li-ion cells. The possible routes of degradation of IL and

electrolytes during thermal and electrochemical abuse tests will be investigated under

different experimental conditions. In the first chapter, the state of the art of lithium-ion

batteries will be described. The choice of positive and negative electrodes and electrolytes

will be discussed and the objectives of the work will be described.

In the second chapter, the thermal stability of selected IL will be evaluated. A critical study of

gathered data (from literature and our work) will lead to the determination of an optimised

procedure to obtain reproducible and comparable results. The stability of imidazolium IL

associated to bis(trifluoromethanesulfonyl)imide anion NTf2 will be investigated according to

several modifications of the alkyl chains.

General introduction 35

The third chapter will focus on the thermal stability of the cell components by different

techniques. Decomposition temperatures will be determined by TGA technique. Combustion

behaviours will be investigated by measuring heats of combustion, ignition delays and

analysing emitted gases.

The fourth chapter will be devoted to the study of full Li-ion cells constituted of Li4Ti5O12

and LiNi1/3Mn1/3Co1/3O2 electrodes using electrochemical techniques. The stability of the

systems will be studied under cycling and overcharge. The evolution of the system will be

analysed by volume measurements and surface techniques such as SEM, XRD or XPS.

Finally conclusions of this work will be presented, and some perspectives will be given for

future work.

General introduction 36

References

[1] C. J. Barnhart and S. M. Benson, "On the importance of reducing the energetic and material demands of electrical energy storage", Energy Environ. Sci., 2013, 6(4), p1083

[2] "History of battery invention and development", http://blog.genport.it/?p=133, accessed August 2014 [3] R. S. Khnel, N. Bckenfeld, S. Passerini, M. Winter and A. Balducci, "Mixtures of ionic liquid and

organic carbonate as electrolyte with improved safety and performance for rechargeable lithium batteries", Electrochim. Acta, 2011, 56(11), p4092

[4] "World Hybrid Electric and Electric Vehicle Lithium-ion Battery Market", 2009, report from in Frost & Sullivan, http://www.frost.com (accessed September 24, 2013)

[5] X.-W. Zhang, Y. Li, S. A. Khan and P. S. Fedkiw, "Inhibition of Lithium Dendrites by Fumed Silica-Based Composite Electrolytes", J. Electrochem. Soc., 2004, 151(8), pA1257

[6] C. Brissot, M. Rosso, J. N. Chazalviel and S. Lascaud, "Dendritic growth mechanisms in lithium/polymer cells", J. Power Sources, 1999, 81, p925

[7] A. Deb, U. Bergmann, S. P. Cramer and E. J. Cairns, "In situ x-ray absorption spectroscopic study of the Li[Ni1/3Co1/3Mn1/3]O2 cathode material", J. Appl. Phys., 2005, 97(11), p113523

[8] H. Zheng, Q. Sun, G. Liu, X. Song and V. S. Battaglia, "Correlation between dissolution behavior and electrochemical cycling performance for LiNi1/3Co1/3Mn1/3O2-based cells", J. Power Sources, 2012, 207(0), p134

[9] G. G. Eshetu, S. Grugeon, S. Laruelle, S. Boyanov, A. Lecocq, J. P. Bertrand and G. Marlair, "In-depth safety-focused analysis of solvents used in electrolytes for large scale lithium ion batteries", Phys. Chem. Chem. Phys., 2013, 15(23), p9145

[10] D. R. MacFarlane, N. Tachikawa, M. Forsyth, J. M. Pringle, P. C. Howlett, G. D. Elliott, J. H. Davis, M. Watanabe, P. Simon and C. A. Angell, "Energy applications of ionic liquids", Energy Environ. Sci., 2014, 7(1), p232

[11] B. Scrosati and J. Garche, "Lithium batteries: Status, prospects and future", J. Power Sources, 2010, 195(9), p2419

[12] M. Armand, F. Endres, D. R. MacFarlane, H. Ohno and B. Scrosati, "Ionic-liquid materials for the electrochemical challenges of the future", Nat. Mater., 2009, 8(8), p621

[13] Z. Zheng, B. Gu, H. Wang, L. Ke and Y. Nie, "Lithium ion secondary battery including ionic liquid electrolyte", Microvast New Materials patent, China, US 2013/0029232 A1, 2013

[14] A. Balducci, S. S. Jeong, G. T. Kim, S. Passerini, M. Winter, M. Schmuck, G. B. Appetecchi, R. Marcilla, D. Mecerreyes, V. Barsukov, V. Khomenko, I. Cantero, I. De Meatza, M. Holzapfel and N. Tran, "Development of safe, green and high performance ionic liquids-based batteries (ILLIBATT project)", J. Power Sources, 2011, 196(22), p9719

[15] C. Siret, L. Caratero and P. Biensan, "Lithium-ion battery containing an electrolyte that comprises an ionic liquid", SAFT patent, France, WO 2009/007540, 2009

[16] P. Wasserscheid and T. Welton, Ionic liquids in synthesis, 2nd Ed., Wiley-VCH, Weinheim, 2008 [17] D. M. Fox, J. W. Gilman, A. B. Morgan, J. R. Shields, P. H. Maupin, R. E. Lyon, H. C. De Long and P.

C. Trulove, "Flammability and thermal analysis characterization of imidazolium-based ionic liquids", Ind. Eng. Chem. Res., 2008, 47(16), p6327

[18] A. O. Diallo, C. Len, A. B. Morgan and G. Marlair, "Revisiting physico-chemical hazards of ionic liquids", Sep. Purif. Technol., 2012, 97, p228

[19] C. S. Stefan, D. Lemordant, P. Biensan, C. Siret and B. Claude-Montigny, "Thermal stability and crystallization of N-alkyl-N-alkyl'-pyrrolidinium imides", J. Therm. Anal. Calorim., 2010, 102(2), p685

[20] P. Andersson, P. Blomqvist, A. Lorn and F. Larsson, "Investigation of fire emissions from Li-ion batteries", 2013, report from SP Technical Research Institute of Sweden

[21] C. Maton, N. De Vos and C. V. Stevens, "Ionic liquid thermal stabilities: decomposition mechanisms and analysis tools", Chem. Soc. Rev., 2013, 42(13), p5963

[22] S. A. Forsyth, S. R. Batten, Q. Dai and D. R. MacFarlane, "Ionic Liquids Based on Imidazolium and Pyrrolidinium Salts of the Tricyanomethanide Anion", Aust. J. Chem., 2004, 57(2), p121

CHAPTER 1

STATE OF THE ART

Chapter 1 38

Chapter 1 39

Table of contents

1. General context ............................................................................................................. 40

2. Lithium-ion batteries ..................................................................................................... 41

2.1. Safety issues .............................................................................................................. 43

2.2. Electrolytes ................................................................................................................ 45

2.3. Ionic liquids ............................................................................................................... 47

2.4. Electrodes .................................................................................................................. 51

3. Strategy of this work ..................................................................................................... 55

4. References ..................................................................................................................... 57

Chapter 1 40

1. General context

Energy storage is a crucial driving force to develop major markets. The nomad device one is

growing, in particular for smartphones and laptops, requiring a fast development of energy

storage systems (ESS), Figure 1, left. Indeed lots of functionalities are developed, calling for

more energy and power. Furthermore, pushed by the global warming and in order to limit CO2

emissions, the part of renewable energies (related to wind or sun) in the world is strongly

increasing. But their development is refrained by the need to store their intermittent produced

energy. ESS have to be further developed, to allow the storage of these energy sources, and to

be part of smart grids applications, Table 1.[1] In the same context, electric vehicle market is

also expanding, asking for reliable, safe and long range cars, Figure 1, right. Different types of

ESS are required for these applications, as electric vehicles (EV), hybrid electric vehicles

(HEV) and plug-in electric vehicle (PEV) do not have the same specifications.

Electrochemical energy storage turned out to be very attractive for this kind of applications, as

they convert chemical energy into electric one with high efficiencies. They include various

devices such as batteries, supercapacitors, fuel cells combined with electrolyser etc, batteries

being the most popular among them.

Figure 1: Left: Battery sales worldwide for the phones and laptops between 2000 and 2011[2]

Right: EV, HEV and P-HEV battery needs estimated between 2005 & 2020[2]

Capacities (GW) 2010 2011 2012

Total renewable power 1250 1355 1470

Solar photovoltaic 40 71 100

Wind power 198 238 283 Table 1: Indicators of renewable energy development worldwide[3]

Chapter 1 41

2. Lithium-ion batteries In this context requiring large capacities and stationary setups, secondary batteries are the

most relevant, as they can be charged and discharged several times. Contrarily, primary

batteries mainly based on alkaline cells are used in toys, remote controls, clocks etc since their

price is low and they provide one-time use.[4] Secondary batteries previously relied on lead-

acid (Pb), nickel cadmium (NiCd) and nickel metal hydride (NiMH) technologies. Since

the 90s, lithium-based chemistries were commercialised for the consumer market, and they

took the lead regarding emerging applications, Figure 2.

Figure 2: Estimation of total lithium-ion transportation battery revenue by regions, world markets[5]

The enthusiasm shown for this technology is due to the significantly higher power and energy

densities of Li-ion batteries compared to Ni-MH chemistry (250-360 Wh.L-1 vs

140-300 Wh.L-1 and 100-160 Wh.kg-1 vs 30-80 Wh.kg-1). Also, the specific energy density

(in Wh.kg-1) of a lithium-ion cell is three times the one of a Ni-Cd one, Figure 3.[6, 7] In

addition to this significant increase in energy density, Li-ion cells present several other

advantages, such as reliability, good cycle life and no memory effect, i.e. they do not self-

discharge, Table 2. Li-ion batteries also permit wide design flexibility, they can be designed

for high power or high energy density, and they can be commercialised in various formats

such as cylindrical, coin, flat, and prismatic.[8]

Chapter 1 42

Figure 3: Ragone plot presenting energy and power densities of different battery technologies[9]

Ni-Cd Ni-MH Li-ion Power capability + + + + Energy density - + + + Specific energy o + + +

Cycle life + + o o Calendar life + + ++ +

Price + + + - Self discharge - - + +

Temperature behaviour + + o o Reliability + + + o

Fast charging + + + o Table 2: Comparison of performance parameters by chemistries for use in power tools;

++: very good, +: good, o: neutral, -: disadvantage[10]

Their development is further favoured by the price decrease, from 1250 $ per kWh in 2009, to

210 $ per kWh in 2014 and which is expected to fall to 160 $ per kWh in 2025.[11-13] Still a lot

of research is devoted to increase the energy density delivered by Li-ion batteries, e.g. in the

context of electric vehicles (EV) for extended autonomy, ideally achievable with one fast

charge. For these applications, higher power densities are also sought for initiation,

acceleration and breaking of the vehicle, requiring innovative chemistries for both the

electrode and electrolyte components.

A lithium-ion (Li-ion) battery is the gathering of several cells in parallel or series circuits.

Each cell is constituted of three main components which are the negative electrode, the

positive electrode and the electrolyte. The electrodes are set apart via a separator, soaked with

the electrolyte. The role of the separator is to insulate electronically the positive from the

negative electrode, while conducting lithium ions (Li+). During discharge (when the cell is

used as energy supply), Li+ cations move from the negative electrode to the positive one,

Chapter 1 43

through the electrolyte and the separator, Figure 4. It generates an electron flow in the external

circuit, from the negative electrode to the positive electrode. During the charge of the cell, an

external electrical power source forces an electron flow in the opposite direction. To

compensate these negative charges, the Li+ ions migrate from the positive electrode to get

intercalated into the porous negative electrode, in three steps. First, the solvated Li+ cations

migrate through the liquid electrolyte and separator. Then, they separate from the solvation

shell to penetrate the electrode material, while an electron from the external circuit balances

the charge by reducing metallic elements of the electrode. Finally, Li+ can diffuse into the host

electrode.

Figure 4: Schema of lithium-ion cell during discharge (in use)[14]

2.1. Safety issues

Despite all these advantages, the deployment of this technology for electric and hybrid cars is

restrained by a number of accidents caused by Li-ion batteries, Table 3.

Date Device catching fire Place Fire causes February 2014 Tesla car Canada Unknown

November 2013 Tesla car USA Impact January 2013 Boeing 787 Dreamliner battery Japan Overheated lithium battery

July 2011 EV bus China Overheated LiFePO4 batteries April 2011 EV taxi China 16 Ah LiFePO4 batteries

September 2010 Boeing B747 cargoplace Dubai Overheated lithium battery March 2010 IPod Nano Japan Overheated lithium battery

January 2010 EV buses China Overheated LiFePO4 batteries July 2009 Cargo plane China Spontaneous combustion June 2008 Laptop Japan Overheated battery June 2008 Honda HEV Japan Overheated LiFePO4 batteries

2006 up to now Mobile phone Short-circtuit, overheating Table 3: Examples of lithium ion battery accidents in the past few years inspired from reference[15]

Chapter 1 44

The rate of charge or discharge and the engineering of the battery pack also influence its

safety. When one cell undergoes thermal runaway, the adjacent ones may also heat up and

fail, causing the entire battery to ignite or stop, Figure 5.[15] Thermal management is thus

crucial to limit the propagation of such thermal event, focusing on cell-to-cell thermal

conduction. To overcome this aggravation, many external safety mechanisms exist. At the cell level, a pressure vent can reduce the risk of explosion by lowering the internal pressure, or an

interrupt can stop current to avoid overcharge. Also shut-down separators are developed to

melt at a critical temperature, isolating the cell from further damage and preventing from

thermal runaway.[16, 17] At the battery level, Battery Management System (BMS) can

supervise each cell voltage and temperature, balance the cells, warn the operator or stop the

battery if required.[18]

Figure 5: Thermal management between battery cells to limit thermal runaway[18]

In order to enhance the safety of lithium-ion batteries, active research is carried out on a

variety of approaches, from material design to packaging. In particular, electrolytes are under

strong development.

Chapter 1 45

2.2. Electrolytes

Electrolytes are constituted of a lithium salt dissolved in a solvent or a mixture of solvents. Its

role is to provide a good conduction of lithium ions in the potential range used to cycle the

cell. The required properties are electrochemical stability, embodied by electrochemical

window (EW), thermal stability, represented by decomposition temperature and flash point,

and ionic conductivity, Figure 6. These properties are desired as high as possible in the

application temperature range, e.g. in the case of electric vehicles, from - 40 C to 70 C.

Furthermore electrolytes must be chemically compatible (i.e. inert) with both negative and

positive electrode materials, in the voltage range of the cell.

Figure 6: Electrochemical window and conductivities of several electrolytes families[19]

Electrochemical stability of electrolytes can be measured by linear sweep voltammetry or

cyclic voltammetry. It reveals the limits in oxidation and reduction of the solution. Thermal

stability must be considered, in order to ensure the stability of the system, and can be

represented by decomposition temperature. It is measured by thermogravimetric analyses,

which shows the decomposition observed by mass loss. The flashpoint is the temperature at

which the electrolyte forms an ignitable vapour mixture, i.e. it requires an ignition source. The

Chapter 1 46

autoignition temperature can also be considered, representing the minimum temperature

required to ignite a gas or vapour in air without any ignition source. Ionic conductivity can be

measured by impedance spectroscopy, and it translates the capacity of the lithium cations to

move through the solvent. This property is linked to the viscosity, which influences diffusion

coefficients.[20]

The electrolyte is usually a mixture of carbonates, to which a lithium salt is added. The

mixture combines low viscosity solvent such as diethylcarbonate (DEC, < 1 cP)[21] or

dimethyl carbonate (DMC), and a high dielectric constant solvent such as ethylene carbonate

(EC, > 80)[22] or propylene carbonate (PC), Figure 7.[23, 24] These non-aqueous electrolytes

generally use lithium salts with non-coordinating anions such as hexafluorophosphate (PF6),

perchlorate (ClO4), tetrafluoroborate (BF4) or triflate (CF3SO3). In this work a binary and

equi-volumic mixture of EC and DEC was chosen as a reference, the electrolyte was formed

by adding 1 mol.L-1 of LiPF6, and this solution will be referred to as [EC:DEC][LiPF6].

Figure 7: From left to right, dimethyl carbonate (DMC), diethyl carbonate (DEC),

ethylene carbonate (EC), propylene carbonate (PC) and lithium hexafluorophosphate (LiPF6)

These solvents present inherent drawbacks rising safety concerns, associated to their low

thermal stability, little electrochemical stability at low potentials, and high flammability and

volatility. Thermal runaway is the major malfunction of this type of battery, occurring when

an exothermic reaction goes out of control. It typically happens after exposition to high

temperature or after a short-circuit. First the solid electrolyte interphase (SEI) on the negative

electrode can break from relatively low temperature (~ 70 C), allowing the electrolyte to

react with the carbon electrode and intercalated lithium, generating heat. This energy release

accelerates the reaction, which causes further temperature rise. Gases such as hydrocarbons

from the electrolyte and hydrogen, oxygen, carbon dioxide or carbon monoxide from the

electrodes can be generated, leading to a pressure increase inside the cell.[25, 26] When both

pressure and temperature rise too much, it can cause ignition or even explosion of the cell.

These electrolytes based on carbonates also suffer from low thermal stability, as in particular

Chapter 1 47

their flashpoints are lower than 40 C.[27] The electrolyte can irreversibly deteriorate from

60 C, and performances are diminished below -20 C, close to their freezing point.

To improve the safety of the electrolytes, considerable effort is underway. A wide variety of

additives are designed for specific roles, including flame retardant, overcharge protector, SEI

or lithium salt stabilizer etc.[21, 28-31] Novel lithium salts are also synthesised and tested as

alternatives to LiPF6, to reduce toxicity related to HF formation by hydrolysis.[32-35] Another

approach is the replacement of conventional liquid electrolyte solvents. In this approach

polymers are intensively studied as they provide solid-state safer properties with good

conductivities.[36-40] A new category of solvents is also widely considered and studied, namely

ionic liquids.

2.3. Ionic liquids Ionic liquids (IL) are defined as salts, liquid below 100 C. They result from the association of

an organic cation, Figure 8, with an organic or inorganic anion, Figure 9. Consequently, they

can be theoretically designed for a chosen application by tuning the nature of the anion or/and

of the cation. However this was found to be difficult to predict properties of novel IL.

Figure 8: Common IL cations, from left to right: imidazolium [Im], pyrrolidinium [PYR],

piperidinium, ammonium and phosphonium; R represent alkyl chains

Figure 9: Common IL anions, from left to right: tetrafluoroborate (BF4),

hexafluorophosphate (PF6), bis(trifluoromethanesulfonyl)imide (NTf2), dicyanamide (dca) and acetate (OAc)

One of the first reported IL, ethylammonium nitrate, [N(C2H5)H3][NO3], was synthesised and

reported by Walden in 1914.[41] The first generations of IL were mainly based on

chloroaluminate anions (AlCl4 or Al2Cl7),[42, 43] affording water sensitive, toxic and corrosive

Chapter 1 48

solutions. Then in the 1990s, an important breakthrough was the use of water stable anions

such as BF4 and NO3.[44] Since then, IL were used in a large range of applications such as

catalysts solvents,[45-49] separation media,[50-52] electrolytes,[53-62] or heat transfer fluids.[63-66]

They are generating high interest from the scientific community as shown by an increasing

number of publications, Figure 10.[67, 68]

Figure 10: Evolution of the number of publications containing the term ionic liquid or ionic

liquids in the topic from Web of knowledge from 1988 to 2014

These salts possess low melting points thanks to the large volume and the asymmetry of their

constituting ions. It sterically prevents the formation of a regular network like e.g. in the case

of sodium chloride (NaCl), Figure 11.[69] IL exhibit negligible vapour pressure, similarly to

solid salts,[70] hence they do not vaporize unless heated to the point of thermal decomposition,

typically 200 to 300 C.[53] Their viscosity can be of 30-60 cP,[53, 71-73] they have flash points

higher than 200 C,[74, 75] and are hardly flammable.[76-79]

Figure 11: Structure of NaCl (left) and [Im][PF6] (right) salts;

For the IL, red zones represent ionic parts and green ones represent nonpolar side chains[69]

These properties, associated to a high electrochemical stability (often superior to 4 V) and a

good ionic conductivity (1 to 10 mS.cm-1) highlight their possible use as safe electrolytes.[20,

Chapter 1 49

57, 72, 80-84] This is confirmed by the wide interest for these solvents, showed both in

literature[20, 61, 68, 81, 85-90] and in industrial projects,[91-95] in particular as electrolytes for Li-ion

batteries.

In the case of IL-based electrolytes, many families have been studied, with various electrode

couples.[67, 96] Usually 10 to 100 cycles are reported, and the major part of published cycling

results is carried in half cells. The temperature has a strong effect on the performances of the

batteries containing IL, as the viscosity is a crucial limitation.[57] In the following tables are

listed some of the published results, with IL based on imidazolium (Table 4), pyrrolidinium

(Table 5) and other cations (Table 6). They are classified by increasing temperature and

decreasing capacity, and parameters such as cycling rate and electrode nature are indicated.

Electrolyte solvent based on imidazolium Electrodes

T (C)

Cycle number

Capacity

(mAh.g-1

) C-rate Ref.

[C1C

2Im][NTf

2] LTO / LCO 25 200 106 C [61]

[C1C

nIm][NTf

2],

n=4, 6, 8 Li / LCO 25 120 100 C/8 [56]

[C1C

2Im][BF

4] LTO / LCO 25 50 120 C/5 [97]

[C1C

2Im][FSI] Li / Cgr 25 30 360 C/5 [98]

[C1C

1C

3Im][NTf

2] Li / LMO 30 50 105 C/8 [99]

[C1C

1C

3Im][NTf

2] Li / LCO 30 50 120 C/8 [88]

[C1C

4Im][NTf

2] LTO / LFP 60 100 120 C/10 [100]

[C1C

6Im][NTf

2] LTO / LFP 60 40 130 C/10 [101]

Table 4: Cycling performances of batteries containing ionic liquids based on imidazolium cations as electrolytes in different conditions (cycling rate, temperature, electrodes)

Chapter 1 50

Electrolyte solvent based on pyrrolidinium Electrodes

T (C)

Cycle number

Capacity (mAh.g

-1)

C-rate Ref.

[PYR14

][FSI] Li / LFP 20 220 165 C/10 [93]

[PYR13

][FSI] Li / LFP 25 90 149 C/4 [96]

[PYR14

][NTf2] Li/LiNi0.5Mn1.5O4 25 30 95 C/10 [102]

[PYR13

][FSI] Li / Cgr 30 150 350 C/15 [103]

[PYR14

][PIP13

][FSI] LTO / LFP 60 20 90 C/2 [104]

[PYR14

][NTf2] Cgr / LFP 60 100 80 C/10 [105]

Table 5: Cycling performances of batteries containing ionic liquids based on pyrrolidinion cations as electrolytes in different conditions (cycling rate, temperature, electrodes)

Electrolyte solvent Electrodes T (C) Cycle

number Capacity (mAh.g

-1)

C-rate Ref.

Polymer electrolyte based on [PYR

14][NTf

2] LTO / LFP 20 800 550 C/10

[106]

[PIP13][NTf2] Li / LMNO 20 50 140 C/16 [107]

[ether-functionalised ammonium][NTf

2] Li / LFP 25 50 150 C/10

[108]

Gelled electrolyte based on [C1C2Im][BF4]

LTO / LCO 25 50 120 C/5 [97]

[PIP13][NTf2] Li / LCO 25 28 115 C/10 [109]

[N5555][NTf2] Li / LCO 25 50 120 C/10 [110]

[PIP13][NTf2] Li /LCO 30 then 60 50 125 C/10 [111]

Polymer electrolyte based on [PYR

14][NTf

2] Li / LFP 40 600 450 C/10

[106]

Polymer electrolyte based on [PYR

13][NTf

2] Li / LFP 40 80 120 C/10

[112]

Polymer electrolyte based on [Im][NTf2]

Li / LFP 60 80 160 C/10 [37]

Table 6: Cycling performances of batteries containing ionic liquids as electrolyte components in different conditions (cycling rate, temperature, electrodes)

From this literature survey, we observed that the most used and efficient anion was NTf2. It

can be explained by its high thermal stability, adapted physicochemical properties, i.e. low

viscosity. For the cations, imidazolium and pyrrolidinium are the most used ones. Indeed they

allow easy chemical modification by changing the length, the number and the functionality of

the alkyl chains. Thus one can tune the electrochemical and physicochemical properties, such

Chapter 1 51

as oxidation limits or viscosity. The synthesis of IL combining NTf2 anion with these two

cations is quite easy and provides high yields and purity.

2.4. Electrodes

Lithium-ion batteries can be designed for specific applications by selecting adequate

components, especially intercalation materials. Thick electrodes afford high energy density

cells, while thin electrodes provide powerful ones. Commonly used negative electrode

materials are metallic lithium (Li), graphite (Cgr) or lithium titanate (Li4Ti5O12, LTO). The

positive electrode can be a layered oxide (such as lithium cobalt oxide, LiCoO2, LCO), a

polyanionic framework (such as lithium iron phosphate, LiFePO4, LFP), or a spinel (such as

lithium manganese oxide, LiMn2O4, LMO). The most commonly used positive electrodes are

oxides comprising one or up to three metals, including cobalt, nickel, aluminium or

manganese.[113-117] The choice of the electrode couple determines the nominal voltage of the

cell, Figure 12, which sets its energy density. For example the nominal voltage of cells

constituted of Cgr // LFP or LTO // LCO electrodes is 3.2 V or 2.5 V respectively. A brief

presentation of the most common electrode materials is given below.

Figure 12: Voltage profiles of different Li-ion battery electrodes

Lithium metal is attractive for high capacity batteries (3.86 Ah.g-1), as its potential is very low

(-3.04 V vs SHE) and it is the lightest metal (M = 6.94 g.mol-1 and = 0.53 g.cm-3).[118] In this

case the battery is called half-cell, not Li-ion, referring to the absence of metallic lithium. The

successive lithium depositions induce morphological changes on the electrode surface. They

can generate cracks, which break the passivation layer and promote development of lithium

dendrites, as observed by in situ scanning electron microscopy (SEM), Figure 13.[7, 119, 120]

Chapter 1 52

Figure 13: Left: Dendrite growth mechanism via lithium deposition underneath the surface film,

volume change, surface film crack and dendrite formation[119] Right: SEM picture of lithium dendrite formed on the surface[120]

Common negative electrodes are based on graphite, which theoretical capacity is

372 mAh.g-1. This electrode overcomes the problem of dendrites, as lithium ions get inserted

in a reversible way in the structure. However the major failure mechanism is caused by the

co-intercalation of solvent molecules, pushed together with Li+ cations between the graphene

planes.[121] It makes the graphite exfoliate and decompose into a dust of graphene sheets.[119]

To prevent this process a protective layer is required, called Solid Electrolyte Interphase

(SEI).[122] This organic coating, formed by electrolyte decomposition products, should allow

easy transport of lithium ions, should present low resistance, and should be homogeneous and

stable upon cycling. In the case of electrolytes based on ionic liquids, this layer is not

efficiently formed and cycling is inhibited.[123, 124] Hence additives such as vinylene carbonate

(VC) or vinyl ethylene carbonate (VEC), known to create a good protection layer, can be

added, Figure 14.[125-127] These additives allow reaching high capacities and good

reversibility. However, the complexity of the system is increased, and its safety is depleted as

these additives are flammable and volatile.

Figure 14: Left: Vinylene carbonate; Right: Vinyl ethylene carbonate

Lithium titanium oxide Li4Ti5O12 (LTO) has a theoretical capacity of 175 mAh.g-1, and

approximately 165 mAh.g-1 in practice. LTO electrodes exhibit reduced energy density, about

half that of Cgr, because they operate at higher potential (1.5 V instead of less than

0.2 V vs Li+/Li). The interest of such material is that thanks to its high potential, there is no

need for SEI layer so they do not require electrolyte additives, and lithium plating is avoided.

Chapter 1 53

Contrarily to Cgr, LTO materials allow the use of aluminium current collector, cheaper than

copper. Moreover they can achieve high power (high cycling rates) and improve the cell

safety since they present almost zero strain insertion.[61, 93, 128-130] They are also thermally

stable, generating approximately ten times less heat than graphite during decomposition,

Figure 15, left.[131]

State of the art Li-ion batteries use mostly LiCoO2 (LCO) as positive electrode material. But

the thermal instability of this oxide, its toxicity and the high cost of cobalt may limit its

further development.[132, 133] LFP is a safer alternative, but associated to LTO the cell voltage

is quite low, 2 V, Figure 12, vide supra. To increase this value, high voltage materials such as

mixed metallic oxides of cobalt, manganese, nickel and aluminium were widely examined,

with varying compositions. In particular LiNi1/3Mn1/3Co1/3O2 (NMC) showed good cycling

stability and rate capability, lower cost, reduced toxicity and better thermal stability, Figure

15, right.[134-136]

Figure 15: DSC profiles of graphite and LTO electrode materials (left)[131]

LCO and NCM (NMC) electrode materials (right)[137] Taking into account the characteristics of each electrode material, in this work the Li-ion

system will be based on LTO // NMC electrodes, Figure 16. In order to have a good

reproducibility, a single roll of each electrode will be used in all this work, which features are

described in Table 7.

Chapter 1 54

Figure 16: Comparison between six important characteristics of positive and negative electrodes[138]

Feature LTO NMC

Synthesis solvent water N-methyl pyrrolidone

Binder carboxymethylcellulose (CMC) polyvinylidene fluoride

(PVDF)

Capacity (mAh.cm-2) 1.27 1.10

Loading (mg.cm-2) 7.56 6.8

Nominal voltage (V vs Li+/Li) 1.5 4.3

Thickness (m) 60 60

Porosity (%) 35 35

Current collector Aluminium, 20 m Table 7: Characteristics of LTO and NMC electrodes used in this work

Chapter 1 55

3. Strategy of this work In this first chapter, the state of the art of lithium-ion batteries was described. Lithium-ion

batteries dominate both the nomad device and electric vehicle markets, as they provide more

energy per unit of weight than other chemistries. However they raise safety concerns, leading

to a number of accidents. To enhance safety, the positive and negative electrodes

LiNi1/3Mn1/3Co1/3O2 (NMC) and Li4Ti5O12 (LTO) were chosen. The electrolyte, consisting of

flammable and volatile carbonate mixtures, is the most hazardous component. The

replacement of the latter by ionic liquids (IL), liquid salts claimed to be thermally stable and

non-flammable, could provide a safer alternative. A large number of research groups report

their high performances,[56, 61, 101] yet their often claimed safety has been poorly examined by

experiments. In particular studies of IL for use as electrolytes were scarce in case of abuse

conditions such as fire, shortcircuit, or overcharge.

The work of this thesis will investigate IL behaviour under abuse conditions. It will include

thermal stability up to combustion and electrochemical behaviour up to overcharge. It targets

to help defining the safety of IL-containing cells, in particular the thermal stability of

electrolytes. Imidazolium and pyrrolidinium cations associated to

bis(trifluoromethylsulfonyl)imide anion will be selected from the vast domain of IL for their high decomposition temperatures and adapted physicochemical properties (low viscosity, high

ionic conductivity, wide electrochemical window).[80, 139-143]

In the second chapter, the thermal stability of imidazolium-based IL will be evaluated

according to several trends. A critical study of gathered data (from literature and our work)

will lead to determine an optimised procedure to obtain reproducible and comparable results

of decomposition temperatures. The influence of several structures of alkyl chains (such as

length, branching or functionalization) will be studied in order to deduce the most stable

imidazolium IL.

The third chapter will analyse the possible routes of degradation of the selected IL and their

corresponding electrolytes, during thermal abuse tests and under different experimental

conditions. In particular the combustion behaviours will be investigated by measuring heats of

combustion and ignition delays, and analysing emitted gases.

Chapter 1 56

The last chapter will be devoted to the study of the electrochemical stability of full lithium-ion

cells constituted of Li4Ti5O12 and LiNi1/3Mn1/3Co1/3O2 electrode materials and different

electrolytes using electrochemical techniques. The stability of the systems will be studied by

cycling tests and cyclic voltammetry experiments. To provide a good understanding of the

influence of overcharge on the electrochemical behaviour of the cell, the evolution of the

electrolytes and electrodes surface will also be examined under overcharge.

Chapter 1 57

4. References

[1] C. J. Barnhart and S. M. Benson, "On the importance of reducing the energetic and material demands of electrical energy storage", Energy Environ. Sci., 2013, 6(4), p1083

[2] "World Hybrid Electric and Electric Vehicle Lithium-ion Battery Market", 2009, report from in Frost & Sullivan, http://www.frost.com (accessed September 24, 2013)

[3] E. Martinot, "Renewables Global Futures Report", 2013, report from REN21 [4] "How do alkaline batteries work ?" http://www.energizer.com/learning-center/Pages/how-batteries-

work.aspx, accessed August 2014 [5] "Electric Vehicle Batteries Report", 2012, report from Pike Research,

http://www.navigantresearch.com/ [6] C. Glaize and S. Genies, Lithium Batteries and Other Electrochemical Storage Systems, Wiley-ISTE

ed., 2013 [7] B. J. Landi, M. J. Ganter, C. D. Cress, R. A. DiLeo and R. P. Raffaelle, "Carbon nanotubes for lithium

ion batteries", Energ. Environ. Sci., 2009, 2(6), p638 [8] T. B. Reddy and D. Linden, Linden's handbook of batteries, 4th ed., McGraw-Hill, New York, 2011 [9] "History of battery invention and development", http://blog.genport.it/?p=133, accessed August 2014 [10] W. Weydanz, in Encyclopedia of Electrochemical Power Sources, Elsevier, Amsterdam, 2009, p46 [11] S. Jaffe, "Battery safety strategy: Cell or System ?" Navigant Research, during Battery Safety congress,

in San Diego, CA, USA, 2013 [12] "Electric vehicles in Europe: gearing up for a new phase?" 2014, report from McKinsey&Co [13] "Electric and plug-in hybrid vehicle road map", 2010, report from IEA [14] M. M. Thackeray, C. Wolverton and E. D. Isaacs, "Electrical energy storage for transportation-

approaching the limits of, and going beyond, lithium-ion batteries", Energ. Environ. Sci., 2012, 5(7), p7854

[15] Q. Wang, P. Ping, X. Zhao, G. Chu, J. Sun and C. Chen, "Thermal runaway caused fire and explosion of lithium ion battery", J. Power Sources, 2012, 208, p210

[16] D. Ilic, P. Birke, K. Holl, T. Wohrle, P. Haug and F. Birke-Salam, "PoLiFlex (TM), the innovative lithium-polymer battery", J. Power Sources, 2004, 129(1), p34

[17] Y. Zhu, S. Xiao, Y. Shi, Y. Yang and Y. Wu, "A trilayer poly(vinylidene fluoride)/polyborate/poly(vinylidene fluoride) gel polymer electrolyte with good performance for lithium ion batteries", J. Mater. Chem. A, 2013, 1(26), p7790

[18] L. J. Yount, "Battery Level Safety and Safety Validation", Johnson Controls, during Battery safety congress, in 2012

[19] J. B. Goodenough and Y. Lu, "Solid electrolyte batteries", during in DOE Vehicle Technologies Annual Merit Review Meeting, 2011

[20] M. Galinski, A. Lewandowski and I. Stepniak, "Ionic liquids as electrolytes", Electrochim. Acta, 2006, 51(26), p5567

[21] M. Z. Kufian and S. R. Majid, "Performance of lithium-ion cells using 1 M LiPF6 in EC/DEC (1/2) electrolyte with ethyl propionate additive", Ionics, 2010, 16(5), p409

[22] R. P. Seward and E. C. Vieira, "The dielectric constants of ethylene carbonate and of solutions of ethylene carbonate in water, methanol, benzene and propylene carbonate", J. Phys. Chem., 1958, 62(1), p127

[23] S.-I. Tobishima and T. Okada, "Lithium cycling efficiency and conductivity for high dielectric solvent/low viscosity solvent mixed systems", Electrochim. Acta, 1985, 30(12), p1715

[24] K. Xu, "Nonaqueous liquid electrolytes for lithium-based rechargeable batteries", Chem. Rev., 2004, 104(10), p4303

[25] A. W. Golubkov, D. Fuchs, J. Wagner, H. Wiltsche, C. Stangl, G. Fauler, G. Voitic, A. Thaler and V. Hacker, "Thermal-runaway experiments on consumer Li-ion batteries with metal-oxide and olivin-type cathodes", RSC Advances, 2014, 4(7), p3633

[26] F. La Mantia, F. Rosciano, N. Tran and P. Novak, "Quantification of Oxygen Loss from Li(1+x)(Ni(1/3)Mn(1/3)Co(1/3))(1-x)O(2) at High Potentials by Differential Electrochemical Mass Spectrometry", J. Electrochem. Soc., 2009, 156(11), pA823

Chapter 1 58

[27] G. G. Eshetu, S. Grugeon, S. Laruelle, S. Boyanov, A. Lecocq, J. P. Bertrand and G. Marlair, "In-depth safety-focused analysis of solvents used in electrolytes for large scale lithium ion batteries", Phys. Chem. Chem. Phys., 2013, 15(23), p9145

[28] S. S. Zhang, "A review on electrolyte additives for lithium-ion batteries", J. Power Sources, 2006, 162(2), p1379

[29] R. Wagner, S. Brox, J. Kasnatscheew, D. R. Gallus, M. Amereller, I. Cekic-Laskovic and M. Winter, "Vinyl sulfones as SEI-forming additives in propylene carbonate based electrolytes for lithium-ion batteries", Electrochem. Commun., 2014, 40(0), p80

[30] K. Ciosek Hgstrm, H. Lundgren, S. Wilken, T. G. Zavalis, M. Behm, K. Edstrm, P. Jacobsson, P. Johansson and G. Lindbergh, "Impact of the flame retardant additive triphenyl phosphate (TPP) on the performance of graphite/LiFePO4 cells in high power applications", J. Power Sources, 2014, 256(0), p430

[31] I. Otero, E. R. Lopez, M. Reichelt, M. Villanueva, J. Salgado and J. Fernandez, "Ionic liquids based on phosphonium cations as neat lubricants or lubricant additives for a steel/steel contact", ACS Appl. Mater. Inter., 2014, 6(15), p13115

[32] S. Li, W. Zhao, X. Cui, H. Zhang, X. Wang, W. Zhong, H. Feng and H. Liu, "Lithium difluoro(sulfato)borate as a novel electrolyte salt for high-temperature lithium-ion batteries", Electrochim Acta, 2014, 129, p327

[33] D. W. McOwen, S. A. Delp, E. Paillard, C. Herriot, S.-D. Han, P. D. Boyle, R. D. Sommer and W. A. Henderson, "Anion Coordination Interactions in Solvates with the Lithium Salts LiDCTA and LiTDI", J. Phys. Chem. C, 2014, 118(15), p7781

[34] H. Zhang, C. Liu, S. Gong, W. Feng, F. Xu, J. Nie and Z. Zhou, "Preparation, Characterization and Physicochemical Properties of Alkali Bis(polyfluoroalkyloxysulfonyl) imides and Electrochemical Properties of the Lithium Salts", Chem. J. Chinese U., 2014, 35(4), p804

[35] P. Murmann, P. Niehoff, R. Schmitz, S. Nowak, H. Gores, N. Ignatiev, P. Sartori, M. Winter and R. Schmitz, "Investigations on the electrochemical performance and thermal stability of two new lithium electrolyte salts in comparison to LiPF6", Electrochim. Acta, 2013, 114, p658

[36] F. M. Gray, Solid polymer electrolytes: Fundamentals and technological Applications., Vol. 32, VCH, New York, 1991

[37] K. Yin, Z. Zhang, L. Yang and S.-I. Hirano, "An imidazolium-based polymerized ionic liquid via novel synthetic strategy as polymer electrolytes for lithium ion batteries", J. Power Sources, 2014, 258(0), p150

[38] P. Yang, L. Liu, L. Li, J. Hou, Y. Xu, X. Ren, M. An and N. Li, "Gel polymer electrolyte based on polyvinylidenefluoride-co-hexafluoropropylene and ionic liquid for lithium ion battery", Electrochim. Acta, 2014, 115, p454

[39] D. H. C. Wong, J. L. Thelen, Y. Fu, D. Devaux, A. A. Pandya, V. S. Battaglia, N. P. Balsara and J. M. DeSimone, "Nonflammable perfluoropolyether-based electrolytes for lithium batteries", P. Natl A. Sci., 2014, 111(9), p3327

[40] W.-S. Young, W.-F. Kuan and T. H. Epps, III, "Block Copolymer Electrolytes for Rechargeable Lithium Batteries", Journal of Polymer Science Part B-Polymer Physics, 2014, 52(1), p1

[41] P. Walden, "Molecular weights and electrical conductivity of several fused salt." B. Russ. Acad. Sci., 1914, p405

[42] J. S. Wilkes, J. A. Levisky, R. A. Wilson and C. L. Hussey, "Dialkylimidazolium chloroaluminate melts - a new class of room-temperature ionic liquids for electrochemistry, spectroscopy, and synthesis", Inorg. Chem., 1982, 21(3), p1263

[43] C. J. Dymek, C. L. Hussey, J. S. Wilkes and H. A. Oye, "Thermodynamics of 1-methyl-3-ethylimidazolium chloride-aluminum chloride mixtures", J. Electrochem. Soc., 1987, 134(8B), pC510

[44] J. S. Wilkes and M. J. Zaworotko, "Air and water stable 1-ethyl-3-methylimidazolium based ionic liquids", Chem. Commun., 1992(13), p965

[45] C. E. Song, M. Y. Yoon and D. S. Choi, "Significant improvement of catalytic efficiencies in ionic liquids", Bull. Korean Chem. Soc., 2005, 26(9), p1321

[46] S. H. Schofer, N. Kaftzik,