Cyclic quaternary ammonium ionic liquids with perfluoroalkyltrifluoroborates: Synthesis, characterization, and properties

DOI: 10.1002/chem.200500930

Cyclic Quaternary Ammonium Ionic Liquidswith

Perfluoroalkyltrifluoroborates: Synthesis, Characterization, and Properties

[

a]

Zhi-Bin Zhou,* Hajime Matsumoto,* and Kuniaki Tatsumi

ꢀ

Abstract: New cyclic quaternary am-

monium salts, composed of N-alkyl-

[(CF SO ) N]

salts. The structural

pyrrolidinium pentafluoroethyltrifluor-

oborate as the supporting electrolyte.

Such comprehensive studies enhance

the knowledge necessary to design and

optimize ionic liquids for many applica-

tions, including electrolytes. Some of

these new salts show desirable proper-

ties, including low melting points, high

thermal stabilities, low viscosities, high

conductivities, and wide electrochemi-

cal windows, and may thus be potential

candidates for use as electrolytes in

high-energy storage devices. In addi-

tion, many salts are ionic plastic crys-

tals.

3

2 2

effect on all the above properties was

intensively studied in terms of the iden-

tity of the cation and anion, variation

of the side chain in the cation (i.e.,

alkyl versus alkyl ether), and change in

the length of the perfluoroalkyl group

(

-

alkyl ether)-N-methylpyrrolidinium,

oxazolidinium, -piperidinium, or -mor-

pholinium cations (alkyl=nC H , alkyl

4

9

ether=CH OCH , CH OCH CH ) and

3

2

3

2

a

perfluoroalkyltrifluoroborate anion

ꢀ

(

[R BF ] , R =CF , C F , nC F ,

(R ) in the [R BF ] ion. The reduction

F

3

F

3

2

5

3

7

F

3

+

nC F ), were synthesized and character-

of Li ions and reoxidation of Li metal

took place in pure N-butyl-N-methyl-

4

9

ized. Most of these salts are liquids at

room temperature. The key properties

of these salts—phase transitions, ther-

mal stability, density, viscosity, conduc-

tivity, and electrochemical windows—

were measured and compared to those

Keywords:

ammonium salts

·

anions · cations · ionic liquids ·

perfluoroalkyltrifluoroborates

plastic crystals

·

ꢀ

of their corresponding [BF₄] and

Introduction

fore, they are also suitable as solvent-free electrolytes in

[6]

electrochemical devices and processes, such as rechargea-

[

7,8]

[9]

In recent years, ionic liquids (ILs) have become a rapidly

expanding topic of chemical research on account of their

unique properties that include a negligible vapor pressure,

nonflammability, and good ability to dissolve organic and in-

organic compounds, and even polymeric materials. These

unusual properties mean that ionic liquids are superior

media for a broad range of potential uses, for example, as

environmentally friendly solvents for chemical synthesis,

biocatalysis, separation technologies, and as solvents or

all-in-one” solvent/templates for nanomaterial prepara-

ble lithium (Li) batteries,

electrochemical capacitors,

[10]

fuel cells, and the electrodeposition of electropositive

metals from nuclear fuel at room temperature.

[11]

Rechargeable Li batteries are a ubiquitous energy device

that is being used worldwide in many types of portable elec-

tronic equipment, such as cellular phones, laptop computers,

and digital cameras. In the state-of-the-art technologies of

4V-class rechargeable Li batteries, a mixture of organic

aprotic solvents (e.g., ethylene carbonate and diethyl car-

bonate) and the conducting salt lithium hexafluorophos-

[1]

[2]

[

3]

[4]

“

[5]

tion. ILs predominantly consist of ionic species and are

good ionic conducting materials in the liquid state. There-

phate (LiPF ) is generally used as a non-aqueous electrolyte

6

although the safety problem is still an important concern be-

cause of the high vapor pressure and low flash point of the

[12]

^o^r^g^aⁿⁱ^c^s^o^l^v^eⁿ^t^s^aⁿ^d^t^h^e^l^o^w^t^h^e^r^m^a^l^s^t^a^bⁱ^lⁱ^t^y^o^f^Lⁱ^P^F6^.

[

a]Dr. Z.-B. Zhou, Dr. H. Matsumoto, Dr. K. Tatsumi

Research Institute for Ubiquitous Energy Devices

National Institute of Advanced Industrial Science and Technology

Driven by the nonvolatility, nonflammability, and good ther-

mal stability of ILs, there is much current interest in the in-

vestigation of new electrolyte systems composed of ILs with

new stable lithium salts as possible safe replacements for the

1

-8-31 Midorigaoka, Ikeda, Osaka 563-8577 (Japan)

Fax : (+81)72-727-9622

E-mail: zb-zhou@vip.tom.com

h-matsumoto@aist.go.jp

[6–8,13]

conventional materials.

An ionic liquid capable of operating as a supporting elec-

trolyte for 4V-class rechargeable Li batteries (i.e., Li metal

Supporting information for this article is available on the WWW

under http://www.chemeurj.org/ or from the author.

2196

ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2006, 12, 2196 – 2212

FULL PAPER

or carbon as anode material) must have the following prop-

erties: 1) a large electrochemical window (i.e., a low catho-

ing low symmetry, good charge distribution, and high con-

formational degrees of freedom. Another attractive feature

+

ꢀ

dic and high anodic potential versus Li /Li) that allows the

for the [R BF ] ion is the easy permutation of the perfluor-

F

3

+

ꢀ

Li /Li redox reaction taking place in both the anode and

oalkyl (R ) chain in the [R BF ] ion, which not only ena-

F F 3

the cathode sides to be reversible; 2) a weakly coordinating

nature for its anion to depress ion pairing, which is never

absent with the small Li ion in liquid media; 3) low viscosi-

bles the optimization of a broad range of properties of their

ILs, including viscosity and conductivity, but also presents a

rich platform to gain significant insight into the influence of

the different factors in the anion on the properties of ILs,

which were rarely revealed with the common anion in the

literature.

+

ty at room temperature to enhance the mobility of ionic spe-

cies; and 4) a wide range of operating temperatures (i.e.,

low melting point and high thermal stability). Recent studies

have shown that the highly fluid and conductive 1,3-dialkyl-

imidazolium salts cannot be used as electrolytes for 4V-class

Li batteries, because of the very positive cathodic potential

As a continuing part of our program aimed at developing

low-melting, low-viscosity, and chemically and electrochemi-

cally stable ILs as possible safe electrolytes for high-energy

+

[8,19,21]

of the 1,3-dialkylimidazolium cations (ꢁ1 V versus Li /

density devices,

we conducted a systematic study of

[

7,8a,d]

Li).

On the other hand, it was revealed that the ILs

the synthesis and characterization of a large number of new

cyclic quaternary ammonium (QA) ILs in this study. These

based on relatively small saturated quaternary ammonium

QA) cations with the electrochemically stable and weakly

coordinating anion, bis(trifluoromethanesulfonyl)imide

ꢀ

(

ILs are composed of [R BF ] ions and four types of satu-

F

3

rated cyclic QA cations with a relatively small ion size (total

C atomsꢂ10) including N-alkyl(alkyl ether)-N-methylpyrro-

lidinium, -oxazolidinium, -piperidinium, and -morpholinium

ions. The fundamental properties (e.g., melting point, ther-

mal stability, density, viscosity, ionic conductivity, and elec-

trochemical windows) required for evaluating the applicabil-

ity of ILs for electrolytes and other applications were deter-

mined for all these salts. The corresponding salts with two

ꢀ

(

[(CF SO ) N] , [TFSI] ), offer some promising properties

3 2 2

[8]

for this purpose. These promising properties mainly in-

clude i) wide electrochemical windows on account of the low

cathodic potential of the saturated QA cations and the high

ꢀ

anodic potential of the [TFSI] , ii) low viscosities on account

of the high flexibility and good charge distribution of the

ꢀ

[

TFSI] , and iii) wide stable liquid ranges on account of the

ꢀ

low melting point and high thermal stability of the [TFSI]

popular anions, [BF₄] and [TFSI] , were also synthesized

and compared. Although the ILs based on these cationic

[

8,13–16]

salts.

Despite these desirable properties, the inherent

ꢀ

high formula weight and large volume of the [TFSI] is the

evident bottleneck to improve the fluidity and ionic conduc-

tivity of its electrolyte—a binary mixture of Li[TFSI]and

QA[TFSI]. This is an immediate barrier for the application

cores with some common anions (e.g., [TFSI] , [BF ] , and

[PF ] ) have been recently reported in several individual

4

ꢀ

6

[15,22–26]

studies,

the effect of the different characteristic fea-

tures of these cations on the properties of their salts is not

well understood. This is because there is no unified under-

standing of how these cyclic ammonium cations affect the

properties of their salts. However, this fundamental knowl-

edge is vital for the design and ultimate realization of ILs as

“designer solvents” for many applications, including electro-

lytes for electrochemical devices. In this study, considerable

attention is therefore focused toward a comprehensive un-

derstanding of the structural effect on the properties of

these cyclic QA salts.

ꢀ

of the [TFSI] -based electrolyte to Li batteries, since low

conductivity cannot satisfy the requirement of high power.

To further optimize the properties of quaternary ammonium

ILs for use as an electrolyte in high-energy density devices,

there clearly exists an incentive to envision new stable

anions with improved features, such as relatively small for-

mula weight and ion size, to circumvent the drawbacks of

ꢀ

the [TFSI] .

Our recent interest in exploring new stable electrolytes

for high-energy density devices, such as Li batteries and

electrochemical capacitors as reported by Ue and co-work-

[

17–19]

ers,

has led us to introduce a series of weakly coordinat-

Results and Discussion

ing and electrochemically stable fluoro anions, namely per-

ꢀ

fluoroalkyltrifluoroborate ([R BF ] , R =CF , C F , nC F ,

Synthesis and characterization: The structures of the cations

and anions used for preparing the cyclic quaternary ammo-

nium (QA) salts are depicted here; the pyrrolidinium, oxa-

zolidinium, piperidinium, and morpholinium cores are ab-

breviated as PY, OX, PI, and MO, respectively, and each

subscript in the abbreviation denotes the alkyl or alkoxy

group in the cation (i.e., CH₃, nC H , CH OCH ,

F

3

F

3

2

5

3

7

[

19–22]

nC F ), into the field of ILs.

In non-aqueous solvent sys-

4

9

tems, it was shown that one of the lithium salts, Li[C F BF ],

2

5

3

can offer a comparable performance to the industry stan-

dard LiPF , and a much better performance than LiBF , for

6

4

V-class rechargeable Li batteries, indicating that both the

ꢀ

cathodic and anodic stabilities of the [R BF ] ion are suffi-

cient for its application as an electrolyte in Li batteries. In

our research into ILs, we found that the [R BF ] ions can

F

3

4

9

3

2

[18]

CH OCH CH groups in the cations are subscripted as 1, 4,

3 2 2

ꢀ

1O1, and 1O2, respectively), consistent with previous

F

3

[20–22]

form low-melting and low-viscosity ILs with various cations,

including imidazolium, aliphatic quaternary ammonium, and

usage.

A total of 52 QA salts were prepared (Table 1).

All 52 salts in Table 1 were synthesized by neutralizing an

aqueous solution of quaternary ammonium hydroxide with

an equimolar quantity of the acidic solution of the desired

[

20–22]

alicyclic pyrrolidinium ions.

This is mainly attributed to

ꢀ

a number of favorable features of the [R BF ] ion, includ-

F

3

Chem. Eur. J. 2006, 12, 2196 – 2212

ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

2197

Z.-B. Zhou, H. Matsumoto, and K. Tatsumi

gravimetric analysis (TGA). For those salts that are liquid at

room temperature, their density (1), dynamic viscosity (h),

and ionic conductivity (k) were measured at 258C. Electro-

chemical windows were determined for the salts sharing a

ꢀ

common anion, [C F BF ] . All the measurement data are

2

5

3

summarized in Tables 1 and 2. The effect of structural varia-

tions in the cation and anion on all above properties of their

salts will be discussed below based on the results presented

in these tables.

Phase behavior, melting point, and glass transition tempera-

ture: The solid–liquid phase transition behavior for each salt

in Table 1 was examined by DSC. The data for the glass

transition temperature (T ), crystallization temperature (T ),

[21a]

anion, as described in our previous report.

The main ad-

g

c

vantage of utilizing this acid/base neutralization method is

to effectively reduce halide contamination, which is un-

avoidable in the resultant salts prepared by conventional

metathesis reactions (i.e., meta-

solid–solid transition temperature (T ), and melting point

s–s

(T ), where appropriate, are summarized in Table 1.

m

Figure 1 shows DSC traces for six of the salts (entries 1, 4, 6,

thesis of a halide salt of the de-

sired cation with a metal or am-

monium salt of the desired

anion) in our experience. In the

present work, the levels of the

ꢀ

+

residual Br and K ions in all

the resultant liquid salts pre-

pared by this acid/base reaction,

as estimated by X-ray fluores-

cence spectrometry, were re-

vealed to be less than 20 ppm.

ꢀ

The [R BF ] (R =CF , C F ,

F

3

F

3

2

5

ꢀ

nC F , nC F ) and [TFSI] salts

3

7

4 9

are immiscible with water to

some extent owing to the low

polarizability and good charge

distribution of the highly fluori-

nated anions that makes the

anion–water interactions weak,

ꢀ

whereas the [BF₄] salts are

miscible with water in any ratio

as a result of strong anion–

Figure 1. Representative DSC traces recorded at a heating rate of 108C in the second cooling/heating cycle.

a) PI14[CF BF ]: single melting point (T ); b) PY1.1O2[C F BF ]: single solid–solid transition (Ts–s) before melt-

3 3 m 2 5 3

water interactions. The water ing (T ); c) PY₁_.₁_O₁[CF BF ]: multiple solid–solid transitions (T ) before melting (T ); d) OX [CF BF ]: only

m

3

s–s

m

14

3

glass transition (T

g

); e) OX1.1O2[C

2

F

5

BF

3

]: glass transition (T

g

) followed by crystallization (T

c

), and melting

content in the hydrophobic

ꢀ

(T

m

); and f) PI1.1O2[nC

).

3

F

7

BF

3

]: glass transition (T

g

) followed by crystallization (T

c

), solid–solid transition (Ts–s),

[

R BF ] and [TFSI] salts was

F 3

and melting (T

m

<

50 ppm after drying under a

high vacuum at 70–1008C for

2

4 h, and ꢁ100–300 ppm in the

ꢀ

hydrophilic [BF ] salts after drying under the same condi-

12, 14, and 24) as examples that represent six types of phase

transitions observed in all the 52 salts in the present study.

It is worthwhile noting that many cyclic QA salts contain-

4

tions. Therefore, the deviation of the measurement data aris-

ing from the presence of halide impurities and water in the

resultant ILs is rather limited and should not be a concern

in this study.

ꢀ

ing [R BF ] displayed single or multiple solid–solid transi-

F

3

tions (Figure 1b and c), and finally melted with a melting

ꢀ

1

ꢀ1

The structures and compositions of all the 52 salts pre-

pared in Table 1 were confirmed by FAB-MS, elemental

entropy (DS ) usually <40JK mol (see Table 1), as ob-

m

[21]

served in the related aliphatic QA salts. As stated in the

1

19

11

[15,21]

analysis, and by H, F, and B NMR spectroscopy (see the

Experimental Section and the Supporting Information). The

thermal properties of all the prepared salts were character-

ized by differential scanning calorimetry (DSC) and thermal

previous literature,

the salts exhibiting this remarkable

characteristic are ionic plastic crystals, and represent a fasci-

nating class of fast-ion conductors that are attracting consid-

erable attention as all-solid-state electrolytes for high-

2198

www.chemeurj.org

ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2006, 12, 2196 – 2212

Ionic Liquids

FULL PAPER

Table 1. Physicochemical properties of cyclic quaternary ammonium salts.

[

a]

[b]

[c]

s–s

[d]

[e]

[f]

1^[^g^]

h^[^h^]

[i]

Entry

Salts

T

[

g

T

c

T

m

DS

m

T

d

k

ꢀ

1

ꢀ1

8C][

8C][JK

mol

]

[8C][gmL

][cP] [mScm

]

1

2

3

4

5

6

7

8

9

PY1.1O1[CF

3

BF

3

]

3

]

ꢀ33, ꢀ24, ꢀ14

ꢀ4

ꢀ18

ꢀ16

5.7

49.4

61.9

227

236

232

238

226

1.3091

1.2153

1.2841

1.2889

1.3604

46

137

87

165

134

7.7

3.3

4.3

2.3

2.4

PY14[CF

PY1.1O2[CF

OX14[CF

OX1.1O2[CF

PI14[CF

PI1.1O2[CF

3

BF

]12

BF

3

]

ꢀ94

ꢀ86

3

]

3

BF

3

23.0

244

1.2059

ꢀ46

456

1.0

ꢀ16

3

]

41.9

234

181

232

299

311

289

292

293

1.2707

1.2728

1.3477

1.3776

1.2877

1.3490

1.3525

1.4194

203

1035

471

37

71

52

1.8

0.4

0.7

6.6

3.5

4.5

2.2

2.5

MO14[CF

3

]

ꢀ74

MO1.1O2[CF

3

BF

3

]

ꢀ22, ꢀ5

ꢀ51, ꢀ16

ꢀ93

1

26

22

ꢀ3

ꢀ4

1

26.1

23.7

40.4

53.0

37.0

51.1

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

PY1.1O1[C

PY14[C

PY1.1O2[C

OX14[C

OX1.1O2[C

PI14[C

PI1.1O2[C

MO14[C

MO1.1O2[C

PY14[nC

PY1.1O2[nC

OX14[nC

OX1.1O2[nC

PI14[nC

PI1.1O2[nC

MO14[nC

MO1.1O2[nC

PY14[nC

PY1.1O2[nC

OX14[nC

OX1.1O2[nC

PI14[nC

PI1.1O2[nC

MO14[nC

MO1.1O2[nC

PY1.1O1[BF

PY14[BF

PY1.1O2[BF

OX14[BF

OX1.1O2[BF

PI14[BF

PI1.1O2[BF

MO14[BF

MO1.1O2[BF

2

F

5

BF

]41

BF

]81

BF

3

]

2

F

5

3

]

2

F

5

3

]

ꢀ116

2

F

5

3

]

108

90

2

F

5

3

]

ꢀ88

51.2

ꢀ51

308

2

5 3

F BF

2

F

5

3

]

ꢀ15

80.5

301

303

306

327

283

307

293

1.3349

1.3404

1.4077

112

466

260

2.2

0.5

0.9

2

F

5

3

ꢀ73

ꢀ78

2

F

5

3

]

3

F

7

3

]

ꢀ115, ꢀ86, ꢀ44, 28

53

6

53

10

33.4

60.3

46.1

44.9

3

F

7

3

]

ꢀ65

ꢀ99

1.4059

1.3391

1.3884

1.4600

1.4536

1.5125

1.4341

62

117

131

777

84

3.3

1.6

1.5

0.5

2.1

1.0

0.9

3

F

7

3

]

3

F

7

3

]

ꢀ84

ꢀ30

ꢀ62

3

F

7

BF

F

3

50.4

325

3

7

3

]

ꢀ92

ꢀ75

ꢀ29

ꢀ8

ꢀ9

35.1

35.6

297

317

302

324

284

3

7

3

]

69

3

F

7

3

]

4

F

9

BF

BF ]54

BF

3

]

ꢀ114, ꢀ81

61

ꢀ13

46.9

46.3

4

F

9

3

]

ꢀ100

41.9

ꢀ82

ꢀ63

ꢀ34

4

F

9

3

275

4

F

9

3

]

18

92

20

56.2

51.8

25.1

32.6

14.9

45.1

23.7

49.6

11.7

270

323

298

301

300

290

403

390

322

297

177

187

4

F

9

BF

3

]

ꢀ91, ꢀ84

4

F

9

3

]

ꢀ78, ꢀ28

4

F

9

3

]

ꢀ15, ꢀ8, 38

ꢀ12, ꢀ1, 36

76

4

F

9

BF

3

]

126

ꢀ18

4

]

ꢀ114

ꢀ61

1.2538

100

6.8

[

j]

4

]150

4

]

ꢀ93

ꢀ75

ꢀ70

ꢀ33

ꢀ34

12

ꢀ12

1.2354

1.2387

1.3091

213

731

704

2.9

0.8

4

]

4

]

4

]4, 81

146

22.0

403

4

]

ꢀ77

ꢀ58

387

386

365

269

431

416

327

328

423

417

398

387

1.2162

1240

40

76

53

145

117

182

102

532

310

0.6

4

]

ꢀ54

66

85

25.6

56.6

70.7

46.0

4

]

2

ꢀ54

ꢀ45

PY1.1O1[TFSI]

PY14[TFSI]

PY1.1O2[TFSI]

OX14[TFSI]

OX1.1O2[TFSI]

PI14[TFSI]

PI1.1O2[TFSI]

MO14[TFSI]

MO1.1O2[TFSI]

ꢀ48

ꢀ30

ꢀ21

1.4827

1.3931

1.4539

1.4645

1.5172

1.3786

1.4355

1.4405

1.5055

5.5

2.6

3.7

2.1

1.8

1.1

2.0

0.4

0.6

[

k]

[l]

[m]

[n]

[r]

[o]

ꢀ89

ꢀ91

ꢀ76

ꢀ73

ꢀ77

ꢀ82

ꢀ60

ꢀ66

ꢀ19

12

35

43.1

[

p]

s]

[q]

ꢀ20, 1

80.5

[t]

[u]

[

a]Glass transition. [b]Crystallization temperature. [c]Solid–solid transition. [d]Melting point. [e]Entropy of fusion ( DS

m

m m m

=DH /T , where DH is melt-

m

ing enthalpy at T [K]). [f] Decomposition temperature. [g] Density at 258C. [h]Viscosity at 25 8C. [i]Specific conductivity at 25 8C. [j]Literature value

[

23]

[15a]

ꢀ1 [15a]

[15a]

1

388C. [k]Literature value ꢀ878C.

[l]Literature value ꢀ188C.

[m]Literature value 1.41 gmL

.

[n]Literature value 85 cP.

[o]Literature

[

15a]

ꢀ1 [13]

[25b]

ꢀ1 [25b]

[25b]

values 2.2;

2.6 mScm

.

[p]Literature value 321 8C.

[q]Literature value 1.53 gmL

[u]Literature value 284 cP.

.

[r]Literature value 107 cP.

[s]Literature value

[

25b]

ꢀ1 [25b]

[25b]

3

508C.

[t]Literature value 1.42 gmL

.

[

27,28]

energy density devices.

Therefore, the conductivity be-

which is in turn controlled by three main factors: molecular

symmetry, intermolecular forces, and conformational de-

grees of freedom of the molecule. This principle is also

applicable to the ILs, as intensively described in a very large

number of well-characterized imidazolium and QA

havior of these salts and the mixture with their lithium salt

will be appreciated in future.

The melting point (T ) of an organic molecular com-

[29]

m

pound is determined by the strength of its crystal lattice,

Chem. Eur. J. 2006, 12, 2196 – 2212

ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

2199

Z.-B. Zhou, H. Matsumoto, and K. Tatsumi

Table 2. Cathodic and anodic limits (Ecathodic and Eanodic versus ferrocene

indeed achieved for most salts in our study. With the three

+

ꢀ2

(

Fc)/ferrocenium (Fc ) at 1.0 mAcm ) and electrochemical windows

exceptions of PY [BF ] , PI [BF ], and MO

[nC F BF ],

4 9 3

1

4

14

4

1.1O2

ꢀ

EW) for the [C

2

F

5

BF

3

]

ionic liquids with various cations determined on

all the salts have a melting point of <1008C, and 38 salts

are liquids at room temperature (some of the salts exhibited

only a glass transition at an extremely low temperature

without melting). Most of the low-melting salts are those

a glassy carbon electrode.

Salts

E

cathodic

E

anodic

EW

[V][ V[ V] ]

EMI[C

2

F

5

BF

3

]

ꢀ2.50

ꢀ3.03

ꢀ3.41

ꢀ2.99

ꢀ3.28

ꢀ2.38

ꢀ3.37

ꢀ2.85

ꢀ2.62

ꢀ2.14

2.15

2.24

2.26

2.27

2.15

2.26

2.37

2.34

4.65

5.27

5.64

5.22

5.54

4.65

5.52

5.11

4.99

4.48

^c^oⁿ^t^aⁱⁿⁱⁿ^g^aⁿ^a^l^k^y^l^e^t^h^e^r^sⁱ^d^e^c^h^aⁱⁿ⁽ⁱ^.^e^.^,^C^H^O^C^H2 or

PY1.1O1[C

PY14[C

PY1.1O2[C

OX14[C

OX1.1O2[C

PI14[C

PI1.1O2[C

MO14[C

MO1.1O2[C

2

F

5

3

]

3

F

5

3

]

CH OCH CH ) in the cation, as observed in the aliphatic

3

2

[21,30,31]

2

F

5

QA salts.

It can be seen in Figure 2 that, in most

2

F

5

3

]

cases, the replacement of the N-alkyl (nC H ) in the cation

4

9

2

F

5

3

]

[

a]

with an isoelectronic alkyl ether (CH OCH CH ) group re-

3 2 2

2

F

5

BF

3

]

sults in a lowering of the T , for example, PY versus

F

5

3

]

m

14

2

[

b]

2

F

5

3

]

PY₁_.₁_O₂, OX₁₄versus OX₁_.₁_O₂, and PI₁₄versus PI₁_.₁_O₂. More-

over, the PY, OX, and MO cations containing a much

longer alkyl ether side chain, such as ethoxyethyl

2

F

5

3

]

[

0

a]Measured at 50 8C. [b]The data was measured at a current density of

.7 mAcm .

ꢀ

2

(

CH CH OCH CH ), 2-ethoxyethoxyethyl (CH CH OCH -

3 2 2 2 3 2 2

CH OCH CH ), were also found to produce low-melting

2

salts (T <258C or an extremely low T without melt-

m

[22,25]

g

[14,15,20,21]

salts.

Qualitatively, reducing ion symmetry, increas-

ing),

while the PY salts containing a longer alkyl side

ing the ionꢁs conformational degrees of freedom (e.g., utiliz-

ing flexible substituents), and improving the charge distribu-

tion of the cation and/or anion are effective approaches to

reduce the lattice energy of the salts, thus resulting in low-

melting salts. Based on these empirical principles and a

chain in the cation exhibit melting points above room tem-

perature, namely PY [C F BF ](36 8C), PY [C F BF ]

1

5

2

5

3

16

2

5

3

[22]

(588C), and PY [C F BF ](52 8C). These results suggest

1

7

2

5

3

that the utilization of a flexible substituent to increase the

conformational degrees of freedom of the cyclic QA cation

is an effective approach to lowering the melting points of

their salts.

ꢀ

recent result on the [C F BF ] salts with pyrrolidinium

2

5

3

[22]

bearing various lengths of alkyl or alkyl ether side chains,

our strategy to prepare low-melting and low-viscosity cyclic

QA salts is to incorporate a flexible alkyl or alkyl ether sub-

stituent of a medium size, such as n-butyl (nC H ), methoxy-

The nature of the anion also significantly affects the melt-

ing point of the cyclic QA salts. In the homologous series of

ꢀ

the [R BF ] ions (R =nC F₂_m₊₁, m=0–4) salts, the melt-

4

9

F

3

F

m

methyl (CH OCH ), and methoxyethyl (CH OCH CH )

ing points are usually lower for the salts with the smaller

3

2

3

2

ꢀ

groups to the cyclic QA cation. An additional advantage for

utilizing these three substituents has a beneficial effect on

improving the ionic conductivity of their salts because of

their relatively small formula weight and size.

[R BF ] (R =CF , C F ) than for those with the larger

F

3

F

3

2 5

ꢀ

[R BF ] (R =nC F , nC F ) and high-symmetry [BF ]

F

3

F

3

7

4

9

4

ions (Figure 2). This suggests that an anion with a low sym-

ꢀ

metry and a medium size (e.g., [CF BF ] and [C F BF ] ,

3

2

5

3

Figure 2 shows the melting point (T ) of the cyclic QA

salts in Table 1 that crystallized under the measurement con-

ditions. As shown in Figure 2, low melting points were

in this case reduces the packing efficiency, but without a sig-

nificant increase of the van der Waals interactions in their

m

salts) favors a lowering of the melting point of its salt, while

ꢀ

that with

a

very large size (e.g., [nC F BF ] and

3

7

3

ꢀ

[

nC F BF ] ) or high symmetry (e.g., [BF ] ) tends to in-

4

9

3

4

crease the melting point owing to the increased van der

Waals interactions for the large anions and ordered ion ar-

rangement for the high-symmetry anion. However, all the

ꢀ

salts with the large [TFSI] ion exhibit low melting points

regardless of the cationic species (Figure 2); these results

suggest that the high flexibility and low symmetry in the

ꢀ

[

TFSI] ion are dominant over its large size (i.e., van der

Waals interactions) in determining the melting point. Gener-

ally, the melting point of an ionic liquid is the combined

result of many competing interactions that influence the sta-

bility of its crystal lattice, and a subtle change in structure

has a tremendous impact on its melting point. This is well

manifested by the salts with the isomeric cations, as illustrat-

ꢀ

ed in Figure 2; for example, for the same [nC F BF ] ion,

4

9

3

PY₁_.₁_O₂(ꢀ138C) versus OX (548C), PI₁_.₁_O₂(ꢀ98C) versus

14

MO₁₄(768C). In the present study, no clear correlations are

observed between the structure and chemical composition

of these salts and their melting points.

Figure 2. Melting points of various salts.

2200

www.chemeurj.org

ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2006, 12, 2196 – 2212

Ionic Liquids

FULL PAPER

ꢀ

Figure 3 shows the glass transition temperatures (T ) of

the cyclic QA salts that formed glasses under the experi-

[C F BF ] ,

[CF BF ] <[nC F BF ] ,

[nC F BF ] !

4 9

g

2

5

3

7

3

ꢀ

[TFSI] , [BF ] (Figure 3), which is quite similar to the

4

trend observed in the aliphatic QA salts of these anions.

It seems that the lower T for the [R BF ] salts is essential-

ly attributable to better charge distribution, more conforma-

tional degrees of freedom, and low polarizability residing in

[21]

mental conditions. It seems that the T of these QA salts is

g

ꢀ

mainly determined by the nature of the cation, including ion

size and polarity; and the nature of the anion, including ion

size, flexibility, and the extent of charge distribution. As can

g

F

3

ꢀ

the [R BF ] ion; these factors which are propitious for

F

3

weakening the strength of the ion–ion interactions in the

salts and hence decrease the T , while the higher T for the

g

ꢀ

[

TFSI] salts is probably attributable to its large volume,

which is predominant over its high charge distribution and

flexibility.

Thermal stability: The short-term thermal stability of all

5

2 salts in Table 1 was examined by means of TGA experi-

ments. Figure 4 shows the decomposition temperatures (T_d)

Figure 3. Glass transition temperatures of various salts.

be seen in Figure 3, for a given anion, the T values general-

g

ly increase in the order PY₁_.₁_O₁<PY₁_.₁_O₂<PY <PI

<

1.1O2

14

PI <OX <OX

<MO <MO₁_.₁_O₂, indicating both the

14

1

4

14

1.1O2

cationic core and the alkyl (alkyl ether) side chain affect the

T . In the family of salts containing the alicyclic QA core

g

(

i.e., PY₁_.₁_O₁, PY₁_.₁_O₂, PY , PI₁_.₁_O₂, and PI ), for a common

14 14

anion, the T_gvalues increase with increasing size of the

cation (the size of the CH unit is larger than that of the

2

oxygen atom), as expected by the increased London disper-

sion forces. Likewise, in the family of salts containing an

ether bond in the cationic core (i.e., OX , OX₁_.₁_O₂, MO₁₄,

Figure 4. Decomposition temperatures of various salts.

1

4

and MO₁_.₁_O₂), the T values for the salts with the smaller

OX cation are lower than those with the larger MO cation

of these salts. It was found that the thermal stability of these

salts is significantly affected by the following three factors:

1) the identity of the cationic core, 2) the side chain in the

cation, and 3) the identity of the anion. Figure 5 displays the

g

(

Figure 3). However, one may note that the T values are

g

lower for the OX₁₄salts than for their OX₁_.₁_O₂counterparts,

which is opposite to the cation size; the same trend is also

observed in the MO₁₄versus MO₁_.₁_O₂salts (Figure 3). It ap-

pears that the polarity of the methoxyethyl (CH OCH CH )

ꢀ

TGA profile for the [TFSI] salts with various cations to

demonstrate the impact of the variations in the cationic core

and side chains of the cation on the thermal stability. The

effect of the first two factors on the thermal stability is evi-

3

2

group in the OX and MO cations predominates over its size

and flexibility, thus increasing the dipole–dipole interactions

ꢀ

dent in the series of [BF₄] and [TFSI] salts. As shown in

between the ions, which in turn raises the T . This effect is

Table 1 and Figure 5, for the salts sharing a common

g

ꢀ

more prominent when comparing the T of the OX and MO

[TFSI] ion, the T values decrease in the following order:

g

d

salts with that of the PY and PI salts. The higher T for the

OX and MO salts suggests the high polarity of their cationic

core is the dominant factor in determining the T_g.

PY₁₄(4318C), PI₁₄(4238C)>PY₁_.₁_O₂(4168C), PI₁_.₁_O₂

(4178C)>MO₁₄(3988C), MO₁_.₁_O₂(3878C)>OX₁₄(3278C),

OX₁_.₁_O₂(3288C)>PY₁_.₁_O₁(2698C). The same trend was also

g

ꢀ

For the salts with a common cation, the T of the

observed in the series of [BF₄] salts (Figure 4). This trend

g

ꢀ

[

R BF ] salts is much lower than that of the corresponding

salts of the [BF₄] and [TFSI] ; for example, keeping the

clearly indicates that 1) an ether bond in the cationic core

causes a decrease in the thermal stability, which is much

more noticeable with the OX cation; 2) replacing the alkyl

F

3

ꢀ

OX₁_.₁_O₂constant, the T_gvalues increase in the order

Chem. Eur. J. 2006, 12, 2196 – 2212

ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

2201

Z.-B. Zhou, H. Matsumoto, and K. Tatsumi

OX₁_.₁_O₂, and PY₁_.₁_O₁cations (Figure 4). This may be ex-

plained by the thermal lability of these three cations, as pre-

viously discussed with respect to the influence of the cation-

ic species on the thermal stability.

Density: The density of the 38 room-temperature liquid salts

was measured at 258C (Table 1). Figure 6 shows the density

(

1) of these liquid salts. The values for the density vary be-

ꢀ

Figure 5. Thermogravimetric analysis traces of the [(CF

3

SO

2

)

2

N]

ꢀ

(

[TFSI] ) salts with various cations.

group (nC H ) in the cation with

a

methoxyethyl

4

9

(

CH OCH CH ) group tends to decrease the thermal stabili-

3

2

ty; and 3) a methoxymethyl (CH OCH ) group in the cation

3

2

results in a remarkable decrease (>1008C) in the thermal

ꢀ

stability. Similar results were also observed in the [BF ]

4

ꢀ

and [TFSI] salts with imidazolium cations containing an al-

versus an alkyl side

chain (T >4008C). In addition, the last two effects on the

[32]

koxymethyl (ROCH ) (T <2508C)

2

33]

d

[

d

Figure 6. Density of various liquid salts at 258C.

ꢀ

thermal stability are also observable in the [R BF ] (R =

F

3

F

C F , nC F , nC F ) salts (Figure 4). These findings suggest

2

5

3

7

4

9

that the introduction of ether bonds into the cation, particu-

larly those with an oxygen and a nitrogen atom attached to

ꢀ

1

tween 1.21 and 1.52 gmL . For a given anion, the density

gradually decreases in the order OX₁_.₁_O₂>MO₁_.₁_O₂

PY₁_.₁_O₁>OX >PY >MO >PI >PY >PI . From

the same carbon atom (i.e., containing an -NCH O- moiety,

>

2

such as in the OX and PY₁_.₁_O₁cation), is unsuitable for pro-

ducing thermally stable ILs. It appears that the low thermal

stability of such salts might be partially caused by the rela-

14

1.1O2

14

1.1O2

14

this trend, it is apparent that the density is increased by

1) reducing the size of the cationic core (e.g., MO versus

OX, PI versus PY), and 2) having an ether bond in the cat-

ionic core (e.g., OX versus PY, and MO versus PI) and/or in

the side-chain group of the cation (e.g., OX₁_.₁_O₂versus OX₁₄,

tively high acidity of the protons of the -NCH O- moiety.

2

The identity of the anion also influences the thermal sta-

bility of these cyclic QA salts. The T values for the

d

ꢀ

[

CF BF ] salts (181–2448C) are much lower than those for

the [R BF ] (R =C F , nC F , nC F ), [BF ] , and [TFSI]

PY₁_.₁_O₂versus PY , PI

versus PI ). The observed in-

1.1O2 14

crease in density caused by replacing the CH unit with an

3

14

ꢀ

F

3

F

2

5

3

7

4

9

4

2

salts (283–4318C), irrespective of the cationic species

oxygen atom in the cation (i.e., incorporating an ether bond

to the cationic core or side chain) should be attributed to

the relatively smaller van der Waals volume and higher

(

Figure 4 and Table 1). This further reinforces our previous

ꢀ

conclusion that the lower thermal stability for the [CF BF ]

3

salts is essentially attributed to the thermal lability of the

mass for the oxygen atom than for the CH unit. In addition,

2

ꢀ

[20a,21a]

[

CF BF ] (i.e., [CF BF ] ![BF ] + CF carbene).

for the salts with the isomeric cations, the density is slightly

higher for the salts containing an ether bond in the cationic

core than for those containing an ether bond in the side

chain; for example, OX₁₄versus PY₁_.₁_O₂, and MO₁₄versus

PI₁_.₁_O₂(Figure 6).

3

4

2

ꢀ

In the salts with the [BF ] , [TFSI] , and large [R BF ]

4

F

3

(

R =C F , nC F , nC F ) ions, with the exception of those

F 2 5 3 7 4 9

containing the OX , OX

and PY₁_.₁_O₁cations, for each

14

1.1O2

ꢀ

cation, the T values decrease in the order of [TFSI] (387–

d

ꢀ

4318C)>[BF₄] (365–4038C)@[R BF ] (283–3278C). This

For the salts with a common cation, the density increases

regularly with increasing bulkiness of the fluoro anions in

F

3

trend in T is consistent with that observed in the related

d

[20a,21a]

ꢀ

imidazolium and aliphatic QA salts.

stability for the [R BF ] salts can be ascribed to the pyroly-

sis of the [R BF ] ions at relatively low temperatures. How-

The lower thermal

the order [BF ] <[CF BF ] <[C F BF ] <[nC F BF ] <

4 3 3 2 5 3 3 7

3

ꢀ

3

ꢀ

[nC F BF ] ꢁ[TFSI] (Figure 6), as expected by the large

F

ꢀ

3

4

9

mass and small van der Waals radii of the fluorine atom. In

F

3

ꢀ

ever, such a difference in the thermal stability significantly

diminished in the corresponding salts with the OX ,

the homologous series of the [R BF ] (R =nC F₂_m₊₁, m=

F

3

F

m

0–4) salts, for a common cation, the addition of a CF unit

1

4

2

2202

www.chemeurj.org

ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2006, 12, 2196 – 2212

Ionic Liquids

FULL PAPER

to the perfluoroalkyl (R ) chain in the anion linearly in-

creases the density by ꢁ0.07 gmL .

than for those containing a six-membered cationic core (i.e.,

PI and MO), suggesting that the overall cation-dependent

trend in the viscosity is essentially under the control of van

der Waals interactions.

F

ꢀ

1

Viscosity: Viscosity is a decisive parameter for assessing ILs

with respect to their use as media in chemical and electro-

chemical applications, because of its strong impact on the

rate of mass transport. The electrolytes in many electro-

chemical devices are required to operate in ambient temper-

ature ranges, so the ILs used as supporting electrolytes

should have low viscosities at room temperature. Currently,

one of the largest barriers for the application of the pure

QA liquid salts to electrolytes is their high viscosity at room

temperature. Therefore, the fundamental understanding of

the factors that essentially control the viscosity of the cyclic

QA liquids is of vital importance for the design of new QA

liquids with high fluidity at room temperature.

For the salts with the same cationic core and a common

anion, the introduction of an ether bond to the side-chain

group of the cation (i.e., replacing the nC H group with a

4

9

CH OCH CH group) uniformly reduces the viscosity

3

2

(Figure 7); for example, PY₁_.₁_O₂versus PY , OX

versus

14

1.1O2

OX , PI

versus PI , MO

versus MO , as observed

1.1O2 14

1

4

1.1O2

14

[21]

in the aliphatic QA salts. This effect is more remarkable

for the salts containing a much longer alkyl ether side chain,

for example, PY₁_.₁_O₂_O₂[C F BF ](54 cP), PY

[C F BF ]

2 5 3

2

5

3

1.2O2

22]

[

(49 cP) versus PY [C F BF ](71 cP) at 25 8C. Generally,

1

4

2

5

3

the alkyl ether side chain (e.g., CH OCH CH ) is more flex-

3

2

ible, but more polar than the alkyl chain (e.g., nC H ), the

4

9

Figure 7 shows the dynamic viscosity (h) of the 38 room-

temperature liquid salts at 258C (Table 1). The viscosity

values fall in the wide range of 37–1240 cP. It seems that the

flexibility is beneficial to lowering the viscosity by virtue of

the increase of conformational degrees of freedom, while

the polarity tends to increase the viscosity on account of the

increase of dipole–dipole attractions. The above trend in vis-

cosity suggests that the flexibility of the alkyl ether side

chain predominates over its polarity, thus reducing the vis-

cosity. In contrast, the incorporation of an ether bond into

the alicyclic cationic core (i.e., replacing a CH unit in the

2

PY and PI cores with an oxygen atom) causes a significant

increase in the viscosity (Figure 7); for example, PY₁_.₁_O₂

,

PY₁₄versus OX₁_.₁_O₂, OX , and PI₁_.₁_O₂, PI₁₄versus MO₁_.₁_O₂

,

14

MO . This effect is well displayed by the large difference in

1

4

viscosity of the two salts with the isomeric cations; for ex-

ꢀ

ample, keeping the [CF BF ] constant, PY₁_.₁_O₂(87 cP)

3

versus OX₁₄(165 cP), PI₁_.₁_O₂(203 cP) versus MO₁₄(1035 cP).

As a general trend, changing the oxygen atom from the side

chain to the cationic core increases the viscosity by 2–

3

times for the salts containing the five-membered ring

cation (i.e., OX₁₄versus PY₁_.₁_O₂, Table 1), and by ꢁ5 times

for the salts containing the six-membered ring cation (i.e.,

MO₁₄versus PI₁_.₁_O₂, Table 1). These examples impressively

indicate how a slight change in the cation dramatically

changes the viscosity, which is rarely observed in the report-

ed literature. Since having an ether bond in the alicyclic cat-

ionic core results in an increase in the polarity of the cation,

but not in an increase in its conformation degrees of free-

dom owing to the rigidity of the cationic core, it seems rea-

sonable that the polarity of the OX and MO cations is the

predominant factor to account for the high viscosity of their

salts.

Figure 7. Viscosity of various liquid salts at 258C.

viscosity of these cyclic ILs is largely governed by van der

Waals interactions, the degree of freedom within the ion,

ion shape, and Coulombic forces (mainly in the case of the

salts composed of small ions), while hydrogen-bonding inter-

actions are less significant on account of the very weak ba-

sicity of the fluoro anions used in this study. The van der

Waals interactions in these ILs are mainly contributed from

London dispersion forces, determined by the ion size, and

dipole–dipole attractions, determined by the polarity of the

ion.

A comparison of the viscosity of the alicyclic PY salts

with that of the aliphatic QA ones clearly displays the

impact of the ion shape. Although the PY core has a high ri-

gidity, for a fixed anion, the viscosities of the PY salts are

all lower than those of the aliphatic QA salts with a compa-

rable cation (i.e., containing the same number of carbons in

the cation), and even lower than the salts with a smaller ali-

phatic QA cation, for example, PY [CF BF ](137 cP) <

As shown in Figure 7, for the salts sharing a common

anion, with the exception of PI₁_.₁_O₂[TFSI], the viscosity gen-

erally increases in the order PY₁_.₁_O₁<PY₁_.₁_O₂<PY <

14

3

OX₁_.₁_O₂<OX <PI

<PI <MO

<MO₁₄. From this

N₁₂₂₄[CF BF ](210 cP) (nine carbon atoms in each cation),

3 3

14

1.1O2

14

1.1O2

trend, it is evident that the viscosity is lower for the salts

containing a five-membered cationic core (i.e., PY and OX)

PY₁_.₁_O₂[CF BF ](87 cP, eight carbon atoms) <N₁₁₂_.₁_O₂

-

3

[21]

[CF BF ](97 cP, seven carbon atoms)

<N₁₂₂_.₁_O₂[CF BF ]

3

Chem. Eur. J. 2006, 12, 2196 – 2212

ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

2203

Z.-B. Zhou, H. Matsumoto, and K. Tatsumi

[

21]

ꢀ

(

108 cP, eight carbon atoms).

This result evidently indi-

size of the [TFSI] ion is much larger than that of

ꢀ

cates that the quasi-flat shape of the PY core, as illustrated

[C F BF ] . This may be attributable to the more flexible

2

5

3

[34]

ꢀ

by its crystal structure, plays a dominant role in lowering

the viscosity. However, such an effect is not observed in the

salts with the six-membered PI cation, which should be at-

tributed to the chair conformation and high rigidity of the

PI core. All the above results clearly indicate that 1) among

the four types of cyclic QA cations used in this study, the

PY cation is the best one for producing low-viscosity ILs,

and 2) having an ether bond in the side chain of the QA

cation favors an enhancement of the fluidity of the QA

salts, while that in the cationic core results in a high viscosi-

ty.

nature of the [TFSI] ion, which predominates over its large

size to determine the viscosity. From these findings, it could

be anticipated that the inorganic, weakly coordinating

ꢀ

fluoro anions, such as [PF ] and [AsF ] , form high-viscosi-

ty ILs with the cations used in the present study, because of

the lack of conformational degrees of freedom in these

anions, as in the case of [BF ] . This is also true for the salts

with the large, rigid, and weakly coordinating orthoborate

anions (e.g., bis(oxalato)borate [BOB] ), as already indicat-

ed by the extremely high viscosities for their salts.

the base of the above understanding of the factors in the

anion that affect the viscosity of ILs, we can conclude that,

in the search of low-viscosity ILs, the investigator should

pay more attention to the anion with some conformational

degrees of freedom and a medium size rather than weakly

coordinating ability.

6

ꢀ

4

ꢀ

[31b]

On

In salts with a common cation, the viscosity generally in-

ꢀ

creases in the anion order of [C F BF ] ꢁ[TFSI] <

2

5

3

ꢀ

4

[

nC F BF ] <[CF BF ] , [nC F BF ] <<[BF ] (Figure 7),

3 7 3 3 3 4 9 3

in good agreement with the trend observed in the aliphatic

[

21a]

QA salts with these anions.

This trend in viscosity can be

interpreted in terms of the competing influence of the fol-

lowing three factors: van der Waals interactions, Coulombic

forces, and conformational degrees of freedom in the anion.

Finally, it is worth mentioning that both the cation- and

anion-dependent trends of viscosity at 258C are not exactly

consistent with those for T (see Figures 3 and 7), although

g

ꢀ

The highest viscosity for the [BF₄] ILs is essentially attrib-

it was reported that the T of ILs is a fairly good measure of

g

uted to the lack of conformational degrees of freedom

the ion mobility: the lower the T , the weaker are the cohe-

g

ꢀ

within the rigid [BF₄] ion, and the relatively strong Cou-

sive forces in the salts, hence the more fluid the ILs are in

ꢀ

[14b,31]

lombic forces owing to the smallest size of the [BF₄] ion

many cases.

This result clearly indicates that the T of

g

among the anions in the present study. The much lower vis-

ILs is not the only factor governing the viscosity. Angell and

co-workers recently reported that the fragility of the ILs

(the rate at which the transport properties change with tem-

perature near the glass transition) is another important

factor determining the fluid behavior of ILs in ambient tem-

ꢀ

cosity for the [R BF ] (R =CF , C F , nC F , nC F ) salts

F

3

ꢀ

F

3

2

5

3

7

4

9

than for their [BF₄] counterparts can be explained by the

more conformational degrees of freedom in the [R BF ]

versus [BF ] ions and the decreased Coulombic forces from

ꢀ

F

3

ꢀ

4

[31a]

the improved charge distribution and increased size of the

perature range.

The difference in fragility may exist in

ꢀ

[

R BF ] versus [BF ] ions. Furthermore, it can be inferred

our salts given the diversity of their structure, and this might

be a valid interpretation of the observed discrepancy be-

tween the trend of the viscosity at room temperature (258C)

and glass transitions in the present study.

F

3

4

that the influence on the viscosity of the increase in confor-

mational degrees of freedom and the decrease in Coulombic

forces prevail over the increase in van der Waals interac-

ꢀ

tions caused by the increased the anion size ([R BF ]

F

3

ꢀ

versus [BF ] ). However, the viscosity is dramatically higher

for the salts with the smaller [CF BF ] ion than for those

with the larger [R BF ] ion (R =C F , nC F ), which is op-

Conductivity: The conductivity of an ionic liquid is of vital

importance if it is to be considered as a supporting electro-

lyte in electrochemical devices. It has been reported that the

4

ꢀ

3

ꢀ

F

3

F

2

5

3

7

posite to the trend found for the anion size. It seems that

ionic conductivity of ILs is mainly governed by their viscosi-

ꢀ

[33a]

the higher viscosity for the [CF BF ] salts may be attributa-

ty, formula weight, density, and ion size.

This is also true

3

ble to the lower conformational degrees of freedom and the

for the salts in the present study. Figure 8 shows the specific

conductivity (k) of the 38 room-temperature liquid salts at

258C (Table 1). The conductivity values of these salts cover

ꢀ

stronger Coulombic forces in the [CF BF ] salts owing to

3

ꢀ

the smaller size of the [CF BF ] ion, as in the case of the

3

ꢀ

ꢀ1

[

BF₄] salts. In the series of [R BF ] (R =C F , nC F ,

the range of 0.4–7.7 mScm . For the salts with a common

F

3

F

2

5

3

7

nC F ) salts, the viscosity increases with increasing length of

anion, the conductivity generally decreased in the order

4

9

ꢀ

the perfluoroalkyl (R ) chain in the [R BF ] . This suggests

PY₁_.₁_O₁>PY₁_.₁_O₂>PY >OX

>OX₁₄,

PI₁_.₁_O₂>PI >

14

F

3

14

1.1O2

that the van der Waals interactions have taken over the con-

formational degrees of freedom and the Coulombic forces

MO₁_.₁_O₂>MO . This cation trend is expected by the in-

14

creased viscosity (compare Figure 7 with Figure 8; for a

given anion, the trend in viscosity is exactly opposite to that

of the conductivity), formula weight, and cation size. It is

clear from this trend that the PY cation is the best one to

produce highly conductive ILs within the scope of the cyclic

QA cations investigated in this study. Notably, for a given

anion, the conductivity for the PY salts is evidently higher

than that for the aliphatic QA salts with a comparable

cation, and even higher than that for salts with a smaller ali-

to govern the viscosity. Among the homologous series of

ꢀ

[

nC F

BF₃] (m=0–4) salts with a common cation, the

m

2m+1

ꢀ

C F BF ] salts exhibit the lowest viscosities (37–466 cP at

2 5 3

258C) in all cases on account of a relatively high number of

conformational degrees of freedom in conjunction with a

ꢀ

medium size for the [C F BF ] . The viscosity of the

2

5

3

ꢀ

[

TFSI] salts (40–532 cP at 258C) is comparable to that of

the corresponding [C F BF ] salts (Figure 7), although the

ꢀ

2

5

3

2204

www.chemeurj.org

ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2006, 12, 2196 – 2212

Ionic Liquids

FULL PAPER

Figure 8. Specific conductivity of various liquid salts at 258C.

ꢀ1

Figure 9. Equivalent conductivity (L) versus reciprocal viscosity (h ) at

58C in a common logarithm form for the 38 room-temperature liquid

2

salts given in Table 1 (the numbering is consistent with the entry numbers

in Table 1). The solid line is the “ideal” Walden product line fixed with

1 molL aqueous KCl solutions.

phatic QA cation, for example, PY

[C F BF ]

2 5 3

>N₁₂₂_.₁_O₂-

1

.1O2

ꢀ

1

[31]

ꢀ1

ꢀ1 [21a]

(

4.5 mScm )>N

[C F BF ](3.8 mScm

)

1

12.1O2

2

5

3

ꢀ

1

[21a]

[

C F BF ](3.2 mScm ).

This can be explained by the

2

5

3

lower viscosities for the PY salts (vide supra).

For salts with a common cation, with the exception of the

smallest, PY₁_.₁_O₁, the conductivity generally decreases in the

seems that the Walden rule may be valid for the salts in the

ꢀ

1

present study because a plot of logL versus logh at 258C

for all these liquid salts (Figure 9) lies on or around the so-

called “ideal” Walden product line determined with

ꢀ

order

[C F BF ] ,

[CF BF ] >[TFSI] >[nC F BF ] >

3 3 3 7

2

5

3

ꢀ

[

BF ] , [nC F BF ] . This anion trend is similar to that ob-

4 4 9 3

ꢀ

1

served in the corresponding imidazolium and aliphatic QA

1 molL aqueous KCl solutions, as described by Angell and

[20a,21a]

[31]

salts with these anions,

and is the combined result of

co-workers.

several competing factors that affect the conductivity, in-

cluding the viscosity, formula weight, and ion size, as ex-

Electrochemical stability: One of the reasons for the recent

growing interest in the QA salts is the low cathodic poten-

tials of the QA cations compared to 1,3-dialkylimidazolium

cations, hence the wide electrochemical windows for their

salts, which may allow them to be used as solvent-free sup-

porting electrolytes in high-energy density devices, including

Li batteries and electrochemical capacitors, to improve their

[20a,21a]

plained in previous reports.

The difference in conduc-

tivity becomes smaller with increasing viscosity (see Fig-

ures 8 and 7), and is not noticeable in the most viscous salts

containing the MO cation with various anions (Figure 8).

This suggests that the transport of ions in these highly vis-

ꢀ

cous salts is more dependent on the fluidity. The [C F BF ]

2

5

3

[8,9b]

salts generally show a higher conductivity than the corre-

safety.

Therefore, of particular interest is the relationship

ꢀ

sponding salts of the [TFSI] , although they have compara-

between the cationic and anionic structures of the ILs and

their electrochemical windows. In previous studies, we re-

vealed that the cathodic limit of the QA salts with the

anions used in this study is determined by the reduction of

the respective QA cations, while their anodic limit is gov-

ble viscosities, which clearly indicates that an anion with a

low formula weight and small size is favorable for the pro-

duction of highly conductive ILs. Among the 38 room-tem-

perature liquid salts in Table 1, the salts with the smallest

cation, PY₁_.₁_O₁, show the highest conductivities, for example,

[21a]

erned by the oxidation of the respective anions;

however,

ꢀ

1

ꢀ1

PY₁_.₁_O₁[CF BF ](7.7 mScm

), PY

[BF ](6.8 mScm

),

in the case of the 1,3-dialkylimidazolium salts, both the

cathodic and the anodic decomposition are limited by the

3

1.1O1

4

ꢀ

1

and PY₁_.₁_O₁[C F BF ](6.6 mScm ), mainly owing to their

2

5

3

[19,20a]

lowest viscosities and the lowest formula weight of the

PY₁_.₁_O₁

1,3-dialkylimidazolium cations.

To compare the relative

.

cathodic stability of the cyclic QA cations in the present

ꢀ

It was noted that the empirical Walden rule for dilute

electrolyte solutions is applicable to pure ILs, although the

exact reason for its applicability in ILs is not yet well under-

study, the QA salts with a common anion, [C F BF ] , were

2

5

3

investigated by linear-sweep voltammetry on a glassy carbon

electrode at 258C, except for the salt PI [C F BF ] (T =

1

4

2

5

3

m

[

31]

stood.

Figure 9 shows the equivalent conductivity (L)

418C), which was measured at 508C. Figure 10 shows the

ꢀ1

versus the reciprocal viscosity (h ) in logarithmic form for

the 38 room-temperature liquid salts at 258C (Table 1), in

which L=kM/1, M is the formula weight, k is the specific

conductivity, and 1 is the density of the respective salts. It

linear-sweep voltammograms of the nine cyclic QA salts

ꢀ

with [C F BF ] , and of EMI[C F BF ](EMI =1-ethyl-3-

2

5

3

2

5

3

methylimidazolium) for comparison. The potentials for all

the salts are given versus the ferrocene (Fc)/ferrocenium

Chem. Eur. J. 2006, 12, 2196 – 2212

ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

2205

Z.-B. Zhou, H. Matsumoto, and K. Tatsumi

the present study also show a similar trend in cathodic sta-

bility, as shown in Figure 10, because their cathodic limits

[21a,22a]

are little affected by these anions.

With regard to the anodic stability, as seen in Figure 10,

ꢀ

all these cyclic QA salts with [C F BF ] show nearly equiv-

2

5

3

alent anodic stability at 258C, as indicated by the close

values of their anodic limit potential (Table 2). The excep-

tion is PI [C F BF ], which was measured at a higher tem-

1

4

2

5

3

perature of 508C; it exhibited a slightly lower anodic limit,

as expected from the Nernst equation. All these cyclic QA

salts are more stable against oxidation than EMI[C F BF ]

2

5

3

(

Figure 10). This is supported by our previous results, which

showed that the anodic decomposition of the QA salts are

governed by the oxidation of their anionic species while that

of the 1,3-dialkylimidazolium salts with electrochemical

stable anions are limited by 1,3-dialkylimidazolium cat-

[

20a,21a]

ꢀ

ions.

In other words, the [R BF ] (R =nC F₂_m₊₁,

F F m

3

ꢀ

Figure 10. Linear-sweep voltammograms of various ionic liquids obtained

m=0–4) and [TFSI] ions are more resistant toward oxida-

tion than 1,3-dialkylimidazolium cations.

ꢀ

3

2

on a glassy carbon working electrode (surface area: 7.8510 cm ) in

ꢀ

1

the first scan; scan rate: 50 mVs ; measurement temperature: 258C;

Our key objective is to find highly fluid, conductive, and

electrochemically stable ILs as supporting electrolytes for

4V-class Li batteries. Therefore, electrodeposition investiga-

tions of the very electropositive Li metal from pure QA-

counterelectrode: a platinum (Pt) wire; potentials (V) were referenced to

ferrocene (Fc)/ferrocenium (Fc ) redox couple in each salt; the data for

EMI[C F BF ](EMI =1-ethyl-3-methylimidazolium) were from the liter-

2 5 3

+

[

20a]

ature.

[

R BF ]salts are of fundamental importance. To date,

F 3

among the liquid salts investigated, the electrodeposition of

Li metal in pure room-temperature ILs (without addition of

any additives) was obtained only in the QA salts with

+

(

Fc ) redox couple as the internal standard, because the po-

+

tential of the Fc/Fc couple is little affected by non-aqueous

[35]

ꢀ

[8a,d,13]

solvents. The measurement data for the cathodic (E_c_a_t_h_o_d_i_c

and anodic (E_a_n_o_d_i_c) limits, and electrochemical windows

EW=E_a_n_o_d_i_cꢀE_c_a_t_h_o_d_i_c) of these salts are summarized in

Table 2, in which the values for the respective E and

E_a_n_o_d_i_care defined as the potential at which the current den-

sity reached 1.0 mAcm .

)

[TFSI] .

In the present study, the PY [C F BF ]salt

14 2 5 3

was chosen as an example for this purpose because of its

+

(

very negative cathodic limit (ꢀ3.41 V versus Fc/Fc ) and

low viscosity (71 cP at 258C). Taking into account the fact

cathodic

+

that the redox potential of Li is ꢁꢀ3.2 V versus Fc/Fc in

ꢀ

2

[8d]

propylene carbonate and liquid QA[TFSI]salts,

it seems

As shown in Figure 10, the cathodic stability of these QA

salts is significantly affected by the cationic species, and in-

creases in the order of OX₁_.₁_O₂, MO₁_.₁_O₂<EMI<MO₁₄,

PI₁_.₁_O₂, PY₁_.₁_O₂, PY₁_.₁_O₁<OX <PI , PY , as indicated by

possible that the electrodeposition of Li metal in pure PY -

14

[C F BF ]could be realized. Figure 11 shows the cyclic vol-

2

5

3

14

the values of their cathodic limit potential (Table 2). It is ap-

parent from this order that 1) the cyclic QA cations without

ether bonds in their built-in structure (i.e., PI₁₄and PY₁₄)

show the strongest resistance toward reduction, and

2) having an ether bond in either the side chain or the cat-

ionic core of the QA cation generally results in a decrease

in the cathodic stability to a lesser or greater extent, which

may be ascribed to the reduction of the ether bond of the

QA cation at a more positive potential relative to that of

the positively charged nitrogen atom of the QA, as is the

[21a]

case for the related aliphatic QA salts.

Notably, the effect

of the ether bond on the cathodic stability is much more sig-

nificant for the QA salts containing two ether bonds in the

cations (i.e., OX₁_.₁_O₂and MO₁_.₁_O₂), as indicated by the most

positive cathodic limits for the OX₁_.₁_O₂(2.38 V) and MO₁_.₁_O₂

(

2.14 V) cations, which are even more positive than the EMI

Figure 11. Cyclic voltammograms of a) N-butyl-N-methylpyrrolidinium

cation (2.50 V) (Table 2). This result clearly indicates that

not all saturated QA cations are more resistant toward re-

duction than 1,3-dialkylimidazolium cations. Furthermore, it

can be anticipated that the QA salts of the other anions in

pentafluoroethyltrifluoroborate (PY14[C

2

F

5

BF

3

]), and b) EMI[C

2

F

5

BF

3

]

ꢀ1

(

EMI=1-ethyl-3-methylimidazolium) containing 0.3 molkg of Li[BF

4

ꢀ3

2

obtained on an Ni electrode (surface area: 1.7710 cm ); scan rate:

ꢀ

1

50 mVs ; measurement temperature: 258C.

2206

www.chemeurj.org

ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2006, 12, 2196 – 2212

Ionic Liquids

FULL PAPER

ꢀ

1

ꢀ

tammograms of 0.3 molkg of LiBF in pure PY [C F BF ]

freedom for the [R BF ] . The [C F BF ] salts have the

4

14

2

5

3

F

3

2

5

3

and EMI[C F BF ], respectively, on an Ni electrode (Ni was

lowest viscosities and are comparable to those of their

[(CF SO ) N] analogues. The reduction of Li ions and re-

3 2 2

2

5

3

[35]

ꢀ

+

used to avoid lithium alloy formation). Indeed, a pair of

+

peaks characteristic of the reduction of Li ions and reoxi-

oxidation of Li metal was observed in pure N-butyl-N-meth-

ylpyrrolidinium pentafluoroethyltrifluoroborate (PY₁₄-

[C F BF ]). Based on the characterization and understand-

dation of Li metal were observed at ꢁꢀ3.2 V with PY -

1

4

[

C F BF ], which is in excellent agreement with the redox

2 5 3

2

5

3

[8d]

potential of Li in the QA[TFSI].

suggests that 1) both the PY₁₄cation and [C F BF ] have

This preliminary result

ing of the cyclic QA liquid salts, we may conclude that,

ꢀ

among the new [R BF ] ionic liquids, those having a low

2

5

3

F

3

sufficient cathodic stability to allow the electrodeposition of

melting point, low viscosities, high thermal stabilities, high

conductivities, and wide electrochemical windows may be

potential candidates for use as electrolytes in high-energy

density devices. Moreover, this work also sheds some light

on the future direction in the search for highly fluid and

conductive quaternary ammonium ionic liquids. To further

improve the fluidity and conductivity of the quaternary am-

monium salts, it is believed that more effort should focus on

the development of new stable and weakly coordinating

ꢀ

Li metal, and 2) those of the liquid QA salts with [R BF ]

F

3

ions having a low reduction potential, such as PY -

1

4

[

C F BF ], may be potential candidates for the use of sup-

2 5 3

porting electrolytes for 4V-class Li batteries. However, no

+

electroreduction of Li ions, except for irreversible reduc-

tion of the EMI cations, was observed in pure EMI-

[

C F BF ]as the supporting electrolyte (Figure 11), which

2 5 3

should be attributed to the fact that the cathodic limit of

+

ꢀ

EMI[C F BF ] (ꢀ2.50 V versus Fc/Fc ) is more positive

anions with more flexible features than [C F BF ] and a

2

5

3

2

5

3

+

ꢀ

than the redox potential of Li (ꢀ3.2 V versus Fc/Fc ).

smaller volume than [(CF SO ) N] to produce low-melting

3 2 2

ionic liquids with very small quaternary ammonium cations.

Conclusion

Experimental Section

New cyclic quaternary ammonium salts comprising N-alkyl-

(

alkyl ether)-N-methylpyrrolidinium, -oxazolidinium, -piper-

General: Commercially available reagents were purchased from Wako

Pure Chemicals, Tokyo Kasei, or Aldrich, and used as received. Acidic

cation-exchange and basic anion-exchange resins (ion-exchange capacity

idinium, and -morpholinium (alkyl=nC H , alkyl ether=

CH OCH , CH OCH CH ) with [R BF ] (R =CF , C F ,

nC F , nC F ) were synthesized and characterized. Most of

them are liquids at room temperature. Their key proper-

ties—phase transitions, thermal stability, density, viscosity,

specific conductivity, and electrochemical windows, which

are necessary for evaluating the applicability of ILs for the

use of electrolytes in electrochemical devices—were deter-

mined and were compared to those of the corresponding

4

9

ꢀ

3

2

3

2

F

3

F

3

2

5

ꢀ

1

>

2.0 mmolmL ) were purchased from Mitsubishi Chemicals. Potassium

perfluoroalkyltrifluoroborate (K[R BF ](R =CF nC

and N-methyloxazolidine were synthesized and purified ac-

cording to the literature.

3

7

4

9

F

[38]

3

F

3

,

C

2

F

5

,

3 7

F ,

[17,37]

4 9

nC F ),

1

19

11

H, F, and B NMR spectra were recorded on a JEOL JNMAL400

spectrometer operating at 399.65, 376.05, and 128.15 MHz, respectively.

[

D

4

]Methanol was used as the solvent for the bromide salts, and

ꢀ ꢀ

ꢀ

6

]acetone for the [R BF ] , [BF ] , and [TFSI] salts. Chemical shift

F 3 4

1

values are reported relative to TMS as the internal reference ( H), and to

the external references of CCl F in [D ]acetone ( F) and BF ·Et O in

3 6 3 2

CDCl ( B). FAB-MS were measured on a JEOL JMS-HX110/110A

3

ꢀ

[^B^F4^]and [(CF SO ) N] salts. It was found that all above

19

3

2

1

properties were significantly affected by the identity of the

cation (namely, pyrrolidinium, oxazolidinium, piperidinium,

and morpholinium) and anion (namely, [R BF ] , [BF ] ,

and [(CF SO ) N] ), the variation of the side chain in the

cation (i.e., alkyl versus alkyl ether), and the change in the

spectrometer. Elemental analysis (C, H, and N) was carried out by the

Center for Organic Elemental Microanalysis of Kyoto University.

ꢀ

F

3

4

ꢀ

Water content and impurity: The water content in the room-temperature

liquid salts was detected by means of a Karl-Fischer titration (Mitsubishi

Chemicals, CA-07). The levels of residual halide and potassium ions in

the room-temperature liquid salts were estimated with an X-ray fluores-

cence spectrometer (JEOL, model JSX-3201).

3

2 2

ꢀ

length of the perfluoroalkyl group (R ) in the [R BF ] .

F

3

These systematic studies not only improved the basic under-

standing of the structural effect on the properties of ionic

liquids, but they also enriched the fundamental knowledge

base of ionic liquids, which is beneficial to the design and

optimization of ionic liquids as “designer solvents” for many

applications.

Salts with pyrrolidinium and piperidinium cations general-

ly display lower viscosities and higher conductivities than

those with the respective oxazolidinium and morpholinium

cations. The presence of an ether bond in the side chain of

the cation favors an increase in the fluidity and conductivity;

however, it causes a remarkable decrease in the thermal and

electrochemical (in the cathodic direction) stability of their

Phase transition: Calorimetric measurements were performed on a differ-

ential scanning calorimeter (Perkin–Elmer Pyris1 equipped with a liquid

nitrogen cooling unit). The temperature calibration was performed with

the following standard samples: n-pentane (melting point: ꢀ132.668C),

cyclohexane (solid–solid transition: ꢀ87.068C; melting point: 6.548C), n-

decane (melting point: ꢀ29.648C), benzoic acid (melting point:

1

22.378C), and indium (melting point: 156.68C) used as reference. An

average weight of 5–10 mg of each sample was hermetically sealed in an

aluminum pan in a dry chamber (Daikin, dew point <ꢀ508C), and then

ꢀ1

heated and cooled, both at a rate of 108Cmin , under a flow of helium

gas. The glass transition temperature (T , onset of the heat capacity

g

change), crystallization temperature (Tc, onset of the exothermic peak),

solid–solid transition (Ts–s, onset of the endothermic peak), and melting

point (T , onset of the endothermic peak), where appropriate, were re-

m

corded on heating in the second heating/cooling cycle for each salt.

ꢀ

salts. Liquid salts with [R BF ] show much lower viscosities

than the corresponding salt with [BF ] , owing to a better

charge distribution and more conformational degrees of

F

3

Thermal stability: Thermal gravimetric analysis (TGA) was performed

on a thermal analysis system (Seiko Instruments, TG/DTA6200). An

average sample weight of 5–10 mg was loaded into a platinum pan and

ꢀ

4

Chem. Eur. J. 2006, 12, 2196 – 2212

ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

2207

Z.-B. Zhou, H. Matsumoto, and K. Tatsumi

ꢀ

1

heated at a rate of 108Cmin over a range of approximately 40–6008C

tane (41.1 g, 300 mmol) was heated at 708C for 24 h under an argon at-

mosphere. A yellow viscous oil was obtained, which was evaporated at

708C under a high vacuum for 24 h to evacuate the volatile materials.

After cooling to room temperature, a yellow solid was obtained. The raw

product was purified by recrystallization twice from 2-propanol by addi-

under a slow stream of nitrogen gas. The onset of decomposition was de-

fined as the decomposition temperature (T ).

d

Measurements of the density, viscosity, and specific conductivity: The fol-

lowing measurements for the room-temperature liquid salts were carried

out in a dry chamber (Daikin, HRG-50 A, dew point <ꢀ508C). The

density (1) was measured with a density meter (Anton Paar, DMA-4500)

at 258C. The reported value for the density is the average of three mea-

surements. The dynamic viscosity (h) was examined by a programmable

viscometer (Brookfield, DV-III+) at 258C. The specific conductivity (k)

a conductivity meter (Radiometer Analytical,

CDM230) in a cell (Radiometer Analytical, CDC749, cell constant:

.9 cm ) at 258C. The cell was calibrated with 0.01m KCl solutions and

the data were recorded at a frequency of 23.4 KHz.

m

tion of tetrahydrofuran to afford a white solid (47.5 g, yield 70%). T =

1

4

d

108C; T =2138C; H NMR: d=1.00 (t, J=7.3 Hz, 3H), 1.43 (m, 2H),

.79 (brs, 2H), 3.20 (s, 3H), 3.50 (m, 2H), 3.73–3.80 (2H), 4.36 (m, 2H),

.88 (s, 1H), 4.95 ppm (s, 1H).

N-Methoxyethyl-N-methyloxazolidinium bromide (OX1.1O2Br): The same

procedure was followed as that described for the OX14Br salt, except bro-

moethyl methyl ether (41.7 g, 300 mmol) was used instead of 1-bromobu-

tane. A yellow viscous liquid was obtained that did not crystallize at

room temperature. It was dissolved in deionized water (150 mL), extract-

ed with ethyl acetate (230 mL), and then decolorized with activated

carbon. After filtration, the collected solution was evaporated at 708C

under reduced pressure to remove water, and then dried at 708C under a

was measured with

ꢀ

1

Electrochemical stability: Linear-sweep voltammetry was performed on

an automatic polarization system (ALS model600) interfaced and moni-

tored with a PC. The ionic liquids were deoxygenated with argon gas

prior to the electrochemical measurements. The voltammetric measure-

ments were made in an argon-filled glove box (O

with a standard three-electrode configuration: a glassy carbon electrode

high vacuum for 24 h to yield a slightly pale yellow liquid (57.6 g, yield

1

8

5

5%). T

g

=ꢀ418C; T

d

=1978C; H NMR: d=3.25 (s, 3H), 3.41 (t, J=

2

and water <5 ppm)

.6 Hz, 3H), 3.70–3.90 (m, 6H), 4.34 (m, 2H), 4.89 (d, J=5.6 Hz, 1H),

ꢀ

3

2

5.01 ppm (d, J=5.6 Hz, 1H).

(

surface area: 7.8510 cm ) served as the working electrode, a Pt wire

was used as the counterelectrode, and the reference electrode was a Pt

N-Butyl-N-methylpiperidinium bromide (PI14Br): The same procedure

was followed as described for the PY14Br salt, except N-methylpiperidine

(30.0 g, 300 mmol) was used instead of N-methylpyrrolidine. The raw

product was purified by recrystallization twice from 2-propanol by addi-

ꢀ

3

ꢀ3

wire immersed in the solutions of 0.015 moldm

2

I + 0.060 moldm

[

(nC N]I in EMI[TFSI] (EMI=1-ethyl-3-methylimidazolium) and

3 7 4

H )

contained in a glass tube with a porous Vycor glass tip. All potentials

+

were given versus the ferrocene (Fc)/ferrocenium (Fc ) redox couple as

m

tion of tetrahydrofuran to afford a white solid (61.0 g, yield 86%). T =

1

the internal standard in each salt. The data for each salt were collected in

the first cathodic and anodic scan at 258C. The cathodic and anodic

limits were arbitrarily defined as the potential at which the current densi-

d

2418C; T =2558C; H NMR: d=1.03 (t, J=7.2 Hz, 3H), 1.44 (m, 2H),

1.76 (m, 22H), 1.92 (brs, 22H), 3.08 (s, 3H), 3.40 ppm (m, 32H).

N-Methoxyethyl-N-methylpiperidinium (PI1.1O2Br): The same procedure

was followed as described for the PY1.1O2Br salt, except N-methylpiperi-

dine (30.0 g, 300 mmol) was used instead of N-methylpyrrolidine. The

crude product was purified by recrystallization from 2-butanol by addi-

ꢀ2

ty reached 1.0 mAcm . For Li electrodeposition, an Ni electrode (1.77

ꢀ

3

2

1

0

cm ) was used as the working electrode to avoid lithium alloy forma-

4

tion, and lithium tetrafluoroborate (LiBF ) dissolved in the ILs

1

+

(

ꢁ0.3 molkg^ꢀ) was used as the source of Li ion.

m

tion of tetrahydrofuran to give a white solid (53.6 g, yield 75%). T =

1

Proceduresfor the preparation of quaternary ammonium bromide as lts

1328C; T

d

=2548C; H NMR: d=1.70 (s, 2H), 1.92 (s, 22H), 3.17 (m,

(

[QA][Br])

3H), 3.38 (s, 3H), 3.48 (m, 22H), 3.66 (s, 2H), 3.84 ppm (s, 2H).

N-Methoxymethyl-N-methylpyrrolidinium bromide (PY1.1O1Br): Methoxy-

methyl bromide (25.0 g, 200 mmol) in anhydrous diethyl ether (50 mL)

over 30 min was added dropwise to a stirred solution of N-methylpyrroli-

dine (17.5 g, 205 mmol) in anhydrous diethyl ether (50 mL) at 08C under

an argon atmosphere. The suspension was stirred for 12 h at room tem-

perature. The resulting solid was collected by filtration, washed with di-

ethyl ether (225 mL), and dried. The solid was purified by recrystalliza-

N-Butyl-N-methylmorpholinium bromide (MO14Br): The same procedure

was followed as described for the OX14Br salt, except that N-methylmor-

pholine (31.5 g, 310 mmol) was used instead of N-methyloxazolidine. The

crude product was purified by recrystallization twice from 1-propanol by

addition of tetrahydrofuran to give a white solid (46.4 g, yield 65%).

1

T

m

=2108C; T

d

=2368C; H NMR: d=1.03 (t, J=7.4 Hz, 3H), 1.44 (m,

2H), 1.80 (m, 2H), 3.23 (s, 3H), 3.50 (brs, 32H), 4.01 ppm (brs, 2

2H).

tion from ethanol by addition of diethyl ether to give a white solid.

1

Yield: 82% (34.4 g); T

m

=688C; T

d

=2358C; H NMR: d=2.22 (m, 2

N-Methoxyethyl-N-methylmorpholinium bromide (MO1.1O2Br): The same

procedure was followed as described for the MO14Br salt, except bro-

moethyl methyl ether (41.7 g, 300 mmol) was used instead of 1-bromobu-

tane. The crude product was purified by recrystallization twice from 1-

2

H), 3.48 (s, 2H), 3.68 (m, 3H, 2H), 4.66 ppm (s, 2H).

N-Butyl-N-methylpyrrolidinium bromide (PY14Br): N-Methylpyrrolidine

25.6 g, 301 mmol) and 1-bromobutane (41.1 g, 300 mmol) were dissolved

(

in anhydrous acetone (80 mL) in a 500 mL flask at room temperature.

The mixture was refluxed for 24 h under an argon atmosphere. After

cooling to room temperature, diethyl ether (150 mL) was added to the re-

sulting suspension. The resulting solid was collected by filtration, washed

with diethyl ether (225 mL), and dried under a high vacuum. The solid

was purified by recrystallization from 2-propanol by addition of tetrahy-

propanol by addition of diethyl ether to give a white solid (46.8 g, yield

1

65%). T

m

=1218C; T

d

=2368C; H NMR: d=3.33 (s, 3H), 3.40 (s, 3H),

3.57 (brm, 2H), 3.63 (brm, 2H), 3.82 (s, 2H), 3.89 (s, 2H), 4.12 ppm

(brs, 22H).

Procedure for the preparation of aqueous os lutionsof perfluoroalkyltri-

fluoroborate acid (H[R

[QA][OH]): Aqueous H[R

by the respective cation and anion exchange methods, as described in our

F

BF

3

]) and quaternary ammonium hydroxide

drofuran. White solid; yield: 85% (56 g);

H NMR: d=1.02 (t, J=7.2 Hz, 3H), 1.43 (m, 2H), 1.80 (m, 2H), 2.24

T

m

=2158C;

d

T =2688C;

(

F

3

BF ]and aqueous [QA ][ OH]were prepared

1

(

brs, 22H), 3.09 (s, 3H), 3.40 (m, 2H), 3.57 ppm (brs, 22H).

[17,20]

previous report.

The solutions were collected in a PFA bottle and

N-Methoxyethyl-N-methylpyrrolidinium bromide (PY1.1O2Br): A mixture

of N-methylpyrrolidine (26.4 g, 0.310 mmol), bromoethyl methyl ether

stored in a refrigerator until required.

General procedure for the preparation of hydrophobic quaternary ammo-

nium salts with perfluoroalkyltrifluoroborates and bis(trifluoromethane-

(

41.7 g, 300 mmol), and anhydrous acetone (80 mL) in a 500 mL flask

was refluxed for 24 h under an argon atmosphere. After evaporation at

08C and reduced pressure, the yellow solid was washed with diethyl

ether (225 mL), followed by recrystallization from 2-butanol by addi-

tion of acetone. White solid; yield: 70% (47.0 g); T =668C; T =2648C;

H NMR: d=2.23 (s, 22H), 3.13 (s, 3H), 3.40 (s, 3H), 3.62 (m, 32H),

sulfonyl)imide ([QA][R

F 3

BF ] and [QA][TFSI]): The salts were prepared

7

on a 30 mmol scale. Aqueous [QA][OH] was neutralized with aqueous

ꢀ

H[A]([A] =[(CF

nC

) in a PFA flask until the pH was ꢁ6–7. After concentration to

15 mL at 408C by evaporation under a vacuum, the liquid salt in the

bottom was extracted with CH Cl (50 mL), washed with deionized water

3 2 2 F 3 F 3 2 5 3 7

SO ) N] , and [R BF ] , R =CF , C F , nC F ,

m

d

4 9

F

1

ꢁ

3

.83 ppm (s, 2H).

2

N-Butyl-N-methyloxazolidinium bromide (OX14Br): The salt was pre-

(210 mL), and dried at 70–1008C for 24 h under a high vacuum to

afford a colorless or pale yellow liquid (water content <50 ppm). The

solid salt was collected by filtration, dried under high vacuum, and puri-

[

25b]

pared under neat conditions following a recently published procedure.

[

38]

A mixture of N-methyloxazolidine (27.0 g, 310 mmol) and 1-bromobu-

2208

www.chemeurj.org

ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2006, 12, 2196 – 2212

Ionic Liquids

FULL PAPER

fied by recrystallization from 2-propanol or 2-butanol to give a white

solid. The characterization data for each salt are presented in the follow-

mental analysis calcd (%) for C

4.42; found: C 34.37, H 5.08, N 4.40.

N-Butyl-N-methylpyrrolidinium pentafluoroethyltrifluoroborate (PY14

9 8

H16BF NO (317.03): C 34.10, H 5.09, N

1

9

11

ing section. Satisfactory F and B NMR data for the salts are provided

-

1

9

in the Supporting Information. In general, the chemical shift values of

F

1

2 5 3

[C F BF ]): Colorless liquid; yield: 89%; H NMR: d=0.98 (t, J=7.4 Hz,

1

and B NMR of these cyclic QA salts are nearly the same as those for

3H), 1.43 (m, 2H), 1.91 (m, 2H), 2.32 (brs, 22H), 3.25 (s, 3H), 3.54 (m,

[

20,21]

+

the related imidazolium or aliphatic QA salts.

2H), 3.72 ppm (brs, 22H); FAB-MS: m/z (%): 142 (100) [PY ] , 187

1

4

ꢀ

N-Methoxymethyl-N-methylpyrrolidinium trifluoromethyltrifluoroborate

(100) [C

C 40.15, H 6.13, N 4.26; found: C 39.94, H 5.82, N 4.20.

2

F BF ] ; elemental analysis calcd (%) for C11H20BF N (329.08):

5

3

8

1

(

PY1.1O1[CF

3

BF

3

]): Colorless liquid; yield: 65%; H NMR: d=2.29 (m,

2

2H), 3.24 (s, 3H), 3.61 (m, 2H), 3.70 (m, 3H, 2H), 4.78 ppm (s, 2H);

N-Methoxyethyl-N-methylpyrrolidinium pentafluoroethyltrifluoroborate

+

ꢀ

1

FAB-MS: m/z (%): 130 (100) [PY1.1O1] , 137 (100) [CF

analysis calcd (%) for C NO (267.02): C 35.98, H 6.04, N 5.25;

found: C 35.96, H 5.97, N 5.37.

3

BF

3

] ; elemental

(PY₁_.₁_O₂[C F BF ]): Colorless liquid; yield: 87%; H NMR: d=2.31 (brs,

2

5

3

8

H

16BF

6

22H), 3.30 (s, 3H), 3.39 (s, 3H), 3.77 (m, 32H), 3.94 ppm (s, 2H);

+

ꢀ

3

FAB-MS: m/z (%): 144 (100) [PY₁_.₁_O₂] , 187 (100) [C F BF ] ; elemental

2

5

analysis calcd (%) for C10

found: C 36.10, H 5.47, N 4.24.

N-Butyl-N-methyloxazolidinium pentafluoroethyltrifluoroborate (OX14-

[C F BF ]): Colorless liquid; yield: 85%; H NMR: d=0.99 (t, J=7.2 Hz,

2 5 3

3H), 1.44 (m, 2H), 1.95 (m, 2H), 3.39 (s, 3H), 3.66 (m, 2H), 3.95 (m,

8

H18BF NO (331.05): C 36.28, H 5.48, N 4.23;

N-Butyl-N-methylpyrrolidinium trifluoromethyltrifluoroborate (PY14

-

1

[

CF

3

BF

3

]): Colorless liquid; yield: 75%; H NMR: d=0.98 (t, J=7.4 Hz,

3

2

H), 1.43 (m, 2H), 1.90 (m, 2H), 2.31 (brs, 22H), 3.23 (s, 3H), 3.53 (m,

+

1

H), 3.71 ppm (brs, 22H); FAB-MS: m/z (%): 142 (100) [PY14] , 137

ꢀ

(

100) [CF

3

BF

3

] ; elemental analysis calcd (%) for C10

H

20BF

6

N (279.07):

C 43.04, H 7.22, N 5.02; found: C 42.97, H 7.23, N 5.13.

N-Methoxyethyl-N-methylpyrrolidinium trifluoromethyltrifluoroborate

BF ]): Colorless liquid; yield: 70%; H NMR: d=2.30 (s, 2

H), 3.25 (s, 3H), 3.39 (s, 3H), 3.74 (brs, 32H), 3.92 ppm (s, 2H);

2H), 4.45 (m, 2H), 5.05 (d, J=5.6 Hz, 1H), 5.11 ppm (d, J=5.6 Hz, 1H);

+

ꢀ

2 5 3

F BF ] ; elemental

FAB-MS: m/z (%): 144 (100) [OX14] , 187 (100) [C

analysis calcd (%) for C10 NO (333.03): C 36.28, H 5.48, N 4.23;

found: C 36.39, H 5.43, N 4.18.

N-Methoxyethyl-N-methyloxazolidinium pentafluoroethyltrifluoroborate

1

H18BF

8

(

2

PY1.1O2[CF

3

+

ꢀ

3 3

BF ] ; elemental

FAB-MS: m/z (%): 144 (100) [PY1.1O2] , 137 (100) [CF

analysis calcd (%) for C NO (281.04): C 38.46, H 6.46, N 4.98;

found: C 38.18, H 6.29, N 5.25.

1

(

3

1

OX1.1O2[C

2

F

5

BF

H), 3.42 (s, 3H), 3.90–4.08 (m, 6H), 4.44 (m, 2H), 5.03 (d, J=5.6 Hz,

H), 5.13 ppm (d, J=5.6 Hz, 1H); FAB-MS: m/z (%): 146 (100)

3

]): Pale yellow liquid; yield: 70%; H NMR: d=3.41 (s,

9

H18BF

6

N-Butyl-N-methyloxazolidinium trifluoromethyltrifluoroborate (OX14

-

+

ꢀ

2 5 3

F BF ] ; elemental analysis calcd (%) for

1

[OX1.1O2] , 187 (100) [C

16BF NO (333.03): C 32.46, H 4.84, N 4.21; found: C 32.77, H 4.70,

[

CF

3

BF

3

]): Colorless liquid; yield: 65%; H NMR: d=0.99 (t, J=7.4 Hz,

C

9

H

8

2

3

2

H), 1.44 (m, 2H), 1.93 (m, 2H), 3.36 (s, 3H), 3.64 (m, 2H), 3.94 (m,

N 4.16.

H), 4.45 (m, 2H), 5.02 (d, J=5.6 Hz, 1H), 5.08 ppm (d, J=5.6 Hz, 1H);

+

ꢀ

N-Butyl-N-methylpiperidinium pentafluoroethyltrifluoroborate (PI -

FAB-MS: m/z (%): 144 (100) [OX14] , 137 (100) [CF

analysis calcd (%) for C NO (281.04): C 38.46, H 6.46, N 4.98;

found: C 38.39, H 6.49, N 4.91.

N-Methoxyethyl-N-methyloxazolidinium trifluoromethyltrifluoroborate

3

BF

3

] ; elemental

14

1

[

C

2

F

5

BF

3

]): White solid; yield: 92%; H NMR: d=0.99 (t, J=7.2 Hz,

9

H18BF

6

3

H), 1.44 (m, 2H), 1.74 (m, 2H), 1.87 (m, 2H), 2.00 (brs, 22H), 3.24 (s,

+

H), 3.54 ppm (m, 32H); FAB-MS: m/z (%): 156 (100) [PI14] , 187

ꢀ

(

100) [C

2

F

5

BF

3

] ; elemental analysis calcd (%) for C12

H

22BF

8

N (343.11):

1

(

OX1.1O2[CF

3

BF

H), 3.41 (s, 3H), 3.90–4.08 (m, 6H), 4.45 (m, 2H), 5.03 (d, J=5.60 Hz,

H), 5.12 ppm (d, J=5.60 Hz, 1H); FAB-MS: m/z (%): 146 (100)

3

]): Pale yellow liquid; yield: 58%; H NMR: d=3.39 (s,

C 42.01, H 6.46, N 4.08; found: C 41.90, H 6.16, N 4.06.

3

1

N-Methoxyethyl-N-methylpiperidinium pentafluoroethyltrifluoroborate

1

+

ꢀ

3 3

BF ] ; elemental analysis calcd (%) for

(PI1.1O2[C

2

F

5

BF

3

]): Colorless liquid; yield: 90%; H NMR: d=1.73 (m,

[

C

OX1.1O2] , 137 (100) [CF

16BF NO (283.02): C 33.95, H 5.70, N 4.95; found: C 33.66, H 5.67,

2

H), 1.97 (brs, 22H), 3.26 (s, 3H), 3.38 (s, 3H), 3.60 (m, 22H), 3.72

8

H

6

2

+

(

m, 2H), 3.91 ppm (s, 2H); FAB-MS: m/z (%): 158 (100) [PI1.1O2] , 187

100) [C

345.08): C 38.29, H 5.84, N 4.06; found: C 38.40, H 5.54, N 4.13.

N 4.91.

N-Butyl-N-methylpiperidinium

CF BF ]): Colorless liquid; yield: 85%; H NMR: d=0.99 (t, J=7.2 Hz,

H), 1.44 (m, 2H), 1.74 (m, 2H), 1.87 (brs, 2H), 1.99 (brs, 22H), 3.23

ꢀ

2

F

5

BF

3

] ; elemental analysis calcd (%) for

C

11

H

20BF

8

NO

trifluoromethyltrifluoroborate

(PI14-

1

[

3

N-Butyl-N-methylmorpholinium pentafluoroethyltrifluoroborate (MO14

-

3

1

[

2 5 3

C F BF ]): Colorless liquid; yield: 90%; H NMR: d=0.99 (t, J=7.2 Hz,

+

(

s, 3H), 3.52 ppm (m, 32H); FAB-MS: m/z (%): 156 (100) [PI14] , 137

100) [CF

3

4

H), 1.45 (m, 2H), 1.92 (m, 2H), 3.40 (s, 3H), 3.67 (brs, 32H),

ꢀ

3

BF

3

] ; elemental analysis calcd (%) for C11

H

22BF

6

N (293.10):

+

.11 ppm (brs, 22H); FAB-MS: m/z (%): 158 (100) [MO14] , 187 (100)

C 45.08, H 7.57, N 4.78; found: C 45.23, H 7.60, N 4.96.

N-Methoxyethyl-N-methylpiperidinium trifluoromethyltrifluoroborate

BF ]): Colorless liquid; yield: 78%; H NMR: d=1.73 (m,

H), 2.00 (brs, 22H), 3.30 (s, 3H), 3.39 (s, 3H), 3.58 (m, 22H), 3.78

ꢀ

[

C

2

F

5

BF

3

] ; elemental analysis calcd (%) for C11

H

20BF

8

NO (345.08): C

3

8.29, H 5.84, N 4.06; found: C 38.21, H 5.82, N 3.97.

1

(

PI1.1O2[CF

3

N-Methoxyethyl-N-methylmorpholinium pentafluoroethyltrifluoroborate

2

1

(

MO1.1O2[C

2

F

5

BF

3

]): Colorless liquid; yield: 82%; H NMR: d=3.39 (s,

+

(

m, 2H), 3.95 ppm (s, 2H); FAB-MS: m/z (%): 158 (100) [PI1.1O2] , 137

100) [CF

3

4

H), 3.45 (s, 3H), 3.68 (m, 2H), 3.74 (m, 2H), 3.93 (s, 2H), 3.99 (s, 2H),

ꢀ

3

BF

3

] ; elemental analysis calcd (%) for C10

H

20BF

6

NO (295.07):

+

.10 ppm (brs, 22H); FAB-MS: m/z (%): 160 (100) [MO1.1O2] , 187

C 40.70, H 6.83, N 4.75; found: C 40.75, H 6.95, N 4.94.

ꢀ

(

100) [C

2

F

5

BF

3

] ; elemental analysis calcd (%) for

C

10

H

18BF

8

NO

2

N-Butyl-N-methylmorpholinium trifluoromethyltrifluoroborate (MO14

-

(347.05): C 34.61, H 5.23, N 4.04; found: C 34.31, H 5.14, N 4.05.

N-Butyl-N-methylpyrrolidinium (heptafluoro-n-propyl)trifluoroborate

PY14[nC BF ]): White solid; yield: 92%; H NMR: d=0.98 (t, J=

.2 Hz, 3H), 1.44 (m, 2H), 1.90 (brs, 2H), 2.31 (brs, 22H), 3.23 (s,

H), 3.52 (m, 2H), 3.70 ppm (brs, 22H); FAB-MS: m/z (%): 142 (100)

1

[

CF

3

BF

3

]): Colorless liquid; yield: 80%; H NMR: d=0.99 (t, J=7.2 Hz,

3

4

H), 1.45 (m, 2H), 1.92 (m, 2H), 3.40 (s, 3H), 3.66 (brs, 32H),

1

(

7

3

F

7

3

+

.11 ppm (brs, 22H); FAB-MS: m/z (%): 158 (100) [MO14] , 137 (100)

ꢀ

[

CF

3

BF

3

] ; elemental analysis calcd (%) for C10

H

20BF

6

NO (295.07): C

4

0.70, H 6.83, N 4.75; found: C 40.61, H 6.81, N 4.74.

+

ꢀ

3 7 3

F BF ] ; elemental analysis calcd (%) for

[

PY14] , 237 (100) [C

N-Methoxyethyl-N-methylmorpholinium trifluoromethyltrifluoroborate

C H20BF10N (379.09): C 38.02, H 5.32, N 3.69; found: C 37.84, H 5.20, N

3.80.

12

1

(

MO1.1O2[CF

3

BF

3

]): Colorless liquid; yield: 75%; H NMR: d=3.39 (s,

3

4

H), 3.42 (s, 3H), 3.65 (m, 2H), 3.71 (m, 2H), 3.89 (s, 2H), 3.96 (s, 2H),

N-Methoxyethyl-N-methylpyrrolidinium (heptafluoro-n-propyl)trifluoro-

+

1

.09 ppm (brs, 22H); FAB-MS: m/z (%): 160 (100) [MO1.1O2] , 137

borate (PY₁_.₁_O₂[nC F BF ]): Colorless liquid; yield: 87%; H NMR: d=

3

7

3

ꢀ

(

100) [CF

3

BF

3

] ; elemental analysis calcd (%) for C

9

H18BF

6

NO

2

(297.04):

2.30 (brs, 22H), 3.27 (s, 3H), 3.40 (s, 3H), 3.75 (m, 32H), 3.92 ppm

+

ꢀ

C 36.39, H 6.11, N 4.72; found: C 36.65, H 6.14, N 4.92.

3 7 3

(s, 2H); FAB-MS: m/z (%): 144 (100) [PY₁_.₁_O₂] , 237 (100) [C F BF ] ;

elemental analysis calcd (%) for C11

N 3.68; found: C 34.89, H 4.56, N 3.72.

N-Butyl-N-methyloxazolidinium (heptafluoro-n-propyl)trifluoroborate

(OX14[nC BF ]): White solid; yield: 87%; H NMR: d=0.99 (t, J=

H18BF10NO (381.06): C 34.67, H 4.76,

N-Methoxymethyl-N-methylpyrrolidinium pentafluoroethyltrifluorobo-

1

rate (PY1.1O1[C

2

F

5

BF

3

]): Colorless liquid; yield: 70%; H NMR: d=2.29

(

2

m, 22H), 3.21 (s, 3H), 3.59 (m, 2H), 3.70 (m, 3H + 2H), 4.78 ppm (s,

+

ꢀ

1

H); FAB-MS: m/z (%): 130 (100) [PY1.1O1] , 187 (100) [C

2

F

5

BF

3

] ; ele-

3

F

7

3

Chem. Eur. J. 2006, 12, 2196 – 2212

ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

2209

Z.-B. Zhou, H. Matsumoto, and K. Tatsumi

ꢀ

4 9 3

F BF ] ; elemental analysis calcd (%) for C14H22BF12N (443.13):

7

.4 Hz, 3H), 1.44 (m, 2H), 1.95 (m, 2H), 3.40 (s, 3H), 3.66 (m, 2H), 3.95

(100) [C

C 37.95, H 5.00, N 3.16; found: C 37.70, H 4.94, N 3.19.

N-Methoxyethyl-N-methylpiperidinium (nonafluoro-n-butyl)trifluorobo-

(

m, 2H), 4.48 (m, 2H), 5.05 (d, J=5.6 Hz, 1H), 5.11 ppm (d, J=5.6 Hz,

+

ꢀ

3 7 3

F BF ] ; ele-

1

H); FAB-MS: m/z (%): 144 (100) [OX14] , 237 (100) [C

18BF10NO (381.06): C 34.67, H 4.76, N

.68; found: C 34.40, H 4.69, N 3.66.

N-Methoxyethyl-N-methyloxazolidinium (heptafluoro-n-propyl)trifluoro-

mental analysis calcd (%) for C11

3

H

1

rate (PI1.1O2[nC

4 9 3

F BF ]): Colorless liquid; yield: 93%; H NMR: d=1.73

(

m, 2H), 2.00 (brs, 22H), 3.30 (s, 3H), 3.39 (s, 3H), 3.60 (m, 22H),

+

3.78 (m, 2H), 3.95 ppm (s, 2H); FAB-MS: m/z (%): 158 (100) [PI1.1O2] ,

1

ꢀ

borate (OX1.1O2[nC

3

F

7

BF

3

]): Pale yellow liquid; yield: 82%; H NMR:

287 (100) [C

4

F

9

BF

3

] ; elemental analysis calcd (%) for C13

H

20BF12NO

d=3.39 (s, 3H), 3.41 (s, 3H), 3.90–4.04 (m, 6H), 4.45 (m, 2H), 5.03 (d,

J=5.6 Hz, 1H), 5.12 ppm (d, J=5.6 Hz, 1H); FAB-MS: m/z (%): 146

(445.10): C 35.08, H 4.53, N 3.15; found: C 35.13, H 4.35, N 3.44.

N-Butyl-N-methylmorpholinium (nonafluoro-n-butyl)trifluoroborate

MO14[nC BF ]): White solid; yield: 94%; H NMR: d=1.00 (t, J=

.4 Hz, 3H), 1.45 (m, 2H), 1.91 (m, 2H), 3.40 (s, 3H), 3.67 (m, 32H),

+

ꢀ

(

100) [OX1.1O2] , 237 (100) [C

3

F

7

BF

3

] ; elemental analysis calcd (%) for

1

(

7

4

F

9

3

10 2

C H16BF10NO (383.04): C 31.36, H 4.21, N 3.66; found: C 31.16, H 4.07,

N 3.77.

N-Butyl-N-methylpiperidinium

PI14[nC BF ]): White solid; yield: 95%; H NMR: d=0.99 (t, J=

.2 Hz, 3H), 1.44 (m, 2H), 1.74 (m, 2H), 1.87 (m, 2H), 2.00 (brs, 22H),

+

.11 ppm (brs, 22H); FAB-MS: m/z (%): 158 (100) [MO14] , 287 (100)

ꢀ

(heptafluoro-n-propyl)trifluoroborate

[C

4

F

9

BF

3

] ; elemental analysis calcd (%) for C13

H

20BF12NO (445.10): C

1

(

3

F

7

3

35.08, H 4.53, N 3.15; found: C 34.79, H 4.33, N 3.30.

7

3

2

3

N-Methoxyethyl-N-methylmorpholinium (nonafluoro-n-butyl)trifluorobo-

+

.24 (s, 3H), 3.54 ppm (m, 32H); FAB-MS: m/z (%): 156 (100) [PI14] ,

1

rate (MO1.1O2[nC

4

F

9

BF

3

]): Yield 90%, white solid; H NMR: d=3.39 (s,

ꢀ

37 (100) [C

3

F

7

BF

3

] ; elemental analysis calcd (%) for C13

H

22BF10N: C

3

4

H), 3.44 (s, 3H), 3.67 (m, 2H), 3.73 (m, 2H), 3.91 (s, 2H), 3.97 (s, 2H),

+

9.72, H 5.64, N 3.56; found: C 39.44, H 5.56, N 3.82.

.09 ppm (brs, 22H); FAB-MS: m/z (%): 160 (100) [MO1.1O2] , 287

ꢀ

N-Methoxyethyl-N-methylpiperidinium (heptafluoro-n-propyl)trifluoro-

(100) [C

4

F

9

BF

3

] ; elemental analysis calcd (%) for

C

12

H

18BF12NO

2

1

borate (PI1.1O2[nC

3

F

7

BF

.73 (m, 2H), 1.99 (brs, 22H), 3.30 (s, 3H), 3.38 (s, 3H), 3.60 (m, 2

H), 3.77 (m, 2H), 3.94 ppm (s, 2H); FAB-MS: m/z (%): 158 (100)

3

]): Colorless liquid; yield: 91%; H NMR: d=

(447.07): C 32.24, H 4.06, N 3.13; found: C 32.29, H 4.23, N 3.18.

1

2

N-Methoxymethyl-N-methylpyrrolidinium bis(trifluoromethanesulfonyl)-

1

imide (PY₁_.₁_O₁[TFSI]): Colorless liquid; yield: 85%; H NMR: d=2.29

+

ꢀ

[

PI1.1O2] , 237 (100) [C

3

F

7

BF

3

] ; elemental analysis calcd (%) for

(m, 22H), 3.25 (s, 3H), 3.62 (m, 2H), 3.72 (s, 3H), 3.74 (m, 2H),

+

C

12

H

20BF10NO (393.12): C 36.48, H 5.10, N 3.55; found: C 36.18, H 4.88,

4.79 ppm (s, 2H); FAB-MS: m/z (%): 130 (100) [PY

] , 280 (100)

1

.1O1

ꢀ

N 3.59.

[TFSI] ; elemental analysis calcd (%) for C H F N O S (410.36): C

9

16

6

2

5 2

N-Butyl-N-methylmorpholinium

(heptafluoro-n-propyl)trifluoroborate

26.34, H 3.93, N 6.83; found: C 26.38, H 3.91, N 7.02.

N-Butyl-N-methylpyrrolidinium bis(trifluoromethanesulfonyl)imide

(PY [TFSI]): Colorless liquid; yield: 94%; H NMR: d=0.99 (t, J=

1

(

MO14[nC

3

F

7

BF

3

]): White solid; yield: 90%; H NMR: d=1.00 (t, J=

1

7

4

.2 Hz, 3H), 1.45 (m, 2H), 1.91 (m, 2H), 3.39 (s, 3H), 3.67 (m, 32H),

1

4

+

.10 ppm (brs, 22H); FAB-MS: m/z (%): 158 (100) [MO14] , 237 (100)

7.4 Hz, 3H), 1.43 (m, 2H), 1.90 (m, 2H), 2.32 (brs, 22H), 3.27 (s, 3H),

ꢀ

[

C

3

F

7

BF

3

] ; elemental analysis calcd (%) for C12

H

20BF10NO (395.09): C

3.55 (m, 2H), 3.73 ppm (brs, 22H); FAB-MS: m/z (%): 142 (100)

+

ꢀ

3

6.48, H 5.10, N 3.55; found: C 36.33, H 5.10, N 3.48.

14

[PY ] , 280 (100) [TFSI] ; elemental analysis calcd (%) for

N-Methoxyethyl-N-methylmorpholinium (heptafluoro-n-propyl)trifluoro-

11 20 6 2 4 2

C H F N O S (422.41): C 31.28, H 4.77, N 6.63; found: C 31.30, H 4.47,

N 6.72.

1

borate (MO1.1O2[nC

3

F

7

BF

3

]): Colorless liquid; yield: 90%; H NMR: d=

3

.39 (s, 3H), 3.45 (s, 3H), 3.68 (m, 2H), 3.73 (m, 2H), 3.92 (s, 2H), 3.98

N-Methoxyethyl-N-methylpyrrolidinium bis(trifluoromethanesulfonyl)-

imide (PY₁_.₁_O₂[TFSI]): Colorless liquid; yield: 93%; H NMR: d=2.32

+

1

(

s, 2H), 4.09 ppm (brs, 22H); FAB-MS: m/z (%): 160 (100) [MO1.1O2] ,

ꢀ

2

37 (100) [C

3

F

7

BF

3

] ; elemental analysis calcd (%) for C11

H

18BF10NO

2

(brs, 22H), 3.30 (brs, 3H), 3.40 (s, 3H), 3.79 (brs, 32H), 3.95 ppm (s,

+

ꢀ

(

397.06): C 33.27, H 4.57, N 3.53; found: C 33.01, H 4.31, N 3.61.

2H); FAB-MS: m/z (%): 144 (100) [PY₁_.₁_O₂] , 280 (100) [TFSI] ; ele-

mental analysis calcd (%) for C10 (424.38): C 28.30, H 4.28,

N-Butyl-N-methylpyrrolidinium

(nonafluoro-n-butyl)trifluoroborate

18 6 2 5 2

H F N O S

1

(

7

3

PY14[nC

4

F

9

BF

.4 Hz, 3H), 1.43 (m, 2H), 1.90 (m, 2H), 2.31 (brs, 22H), 3.22 (s, 3H),

.52 (m, 2H), 3.70 ppm (brs, 22H); FAB-MS: m/z (%): 142 (100)

3

]): White solid; yield: 95%; H NMR: d=0.98 (t, J=

N 6.60; found: C 28.41, H 4.20, N 6.60.

N-Butyl-N-methyloxazolidinium

bis(trifluoromethanesulfonyl)imide

1

(OX [TFSI]): Colorless liquid; yield: 90%; H NMR: d=0.99 (t, J=

1

4

+

ꢀ

[

PY14] , 287 (100) [C

20BF12N: C 36.39, H 4.70, N 3.26; found: C 36.11, H 4.66, N 3.47.

N-Methoxyethyl-N-methylpyrrolidinium (nonafluoro-n-butyl)trifluorobo-

4

F

9

BF

3

] ; elemental analysis calcd (%) for

7.6 Hz, 3H), 1.45 (m, 2H), 1.95 (m, 2H), 3.40 (s, 3H), 3.67 (m, 2H), 3.97

(m, 2H), 4.46 (m, 2H), 5.06 (d, J=5.6 Hz, 1H), 5.12 ppm (d, J=5.6 Hz,

C

13

H

1H); elemental analysis calcd (%) for C10

18 6 2 5 2

H F N O S (424.38): C 28.30,

1

rate (PY1.1O2[nC

4 9 3

F BF ]): Colorless liquid; yield: 95%; H NMR: d=2.30

H 4.28, N 6.60; found: C 28.15, H 4.23, N 6.65.

(

2

brs, 22H), 3.27 (s, 3H), 3.40 (s, 3H), 3.75 (brs, 32H), 3.92 ppm (s,

N-Methoxyethyl-N-methyloxazolidinium bis(trifluoromethanesulfonyl)-

imide (OX₁_.₁_O₂[TFSI]): Pale yellow liquid; yield: 86%; H NMR: d=3.42

+

ꢀ

1

H); FAB-MS: m/z (%): 144 (100) [PY1.1O2] , 287 (100) [C

4

F

9

BF

3

] ; ele-

mental analysis calcd (%) for C12

H

18BF12NO (429.10): C 33.44, H 4.21, N

(s, 3H), 3.44 (s, 3H), 3.95–4.08 (m, 6H), 4.45 (m, 2H), 5.05 (d, J=

5.60 Hz, 1H), 5.16 ppm (d, J=5.60 Hz, 1H); elemental analysis calcd

3

.25; found: C 33.18, H 4.02, N 3.45.

N-Butyl-N-methyloxazolidinium

OX14[nC BF

(

%) for C

H 3.75, N 6.63.

N-Butyl-N-methylpiperidinium bis(trifluoromethanesulfonyl)imide (PI -

9 16 6 2 6 2

H F N O S (426.36): C 25.35, H 3.78, N 6.57; found: C 25.38,

(nonafluoro-n-butyl)trifluoroborate

1

(

4

F

9

3

]): White solid; yield: 91%; H NMR: d=0.99 (t, J=

7

.4 Hz, 3H), 1.44 (m, 2H), 1.93 (m, 2H), 3.37 (s, 3H), 3.64 (m, 2H), 3.95

1

4

1

(

m, 2H), 4.45 (m, 2H), 5.03 (d, J=5.6 Hz, 1H), 5.10 ppm (d, J=5.6 Hz,

[TFSI]): Colorless liquid; yield: 96%; H NMR: d=0.99 (t, J=7.2 Hz,

+

ꢀ

1

H); FAB-MS: m/z (%): 144 (100) [OX14] , 287 (100) [C

18BF12NO (431.07): C 33.44, H 4.21, N

.25; found: C 32.45, H 3.97, N 3.19.

N-Methoxyethyl-N-methyloxazolidinium (nonafluoro-n-butyl)trifluorobo-

4

F

9

BF

3

] ; ele-

3H), 1.44 (m, 2H), 1.74 (m, 2H), 1.86 (m, 2H), 2.00 (brs, 22H), 3.24 (s,

+

mental analysis calcd (%) for C12

H

3H), 3.55 ppm (m, 32H); FAB-MS: m/z (%): 156 (100) [PI ] , 280

1

4

ꢀ

3

12 22 6 2 4 2

(100) [TFSI] ; elemental analysis calcd (%) for C H F N O S (436.44):

C 33.02, H 5.08, N 6.42; found: C 33.19, H 4.80, N 6.43.

1

rate (OX1.1O2[nC

4

F

9

BF

.41 (s, 3H), 3.42 (s, 3H), 3.90–4.08 (m, 6H), 4.44 (m, 2H), 5.03 (d, J=

.6 Hz, 1H), 5.13 ppm (d, J=5.6 Hz, 1H); FAB-MS: m/z (%): 146 (100)

3

]): Pale yellow liquid; yield: 85%; H NMR: d=

N-Methoxyethyl-N-methylpiperidinium bis(trifluoromethanesulfonyl)-

1

3

5

imide (PI₁_.₁_O₂[TFSI]): Colorless liquid; yield: 91%; H NMR: d=1.74

(brs, 2H), 1.97 (brs, 22H), 3.33 (s, 3H), 3.39 (s, 3H), 3.64 (m, 22H),

+

ꢀ

+

[

OX1.1O2] , 287 (100) [C

4

F

9

BF

3

] ; elemental analysis calcd (%) for

3.72 (s, 2H), 3.91 ppm (s, 2H); FAB-MS: m/z (%): 158 (100) [PI

] ,

1

.1O2

ꢀ

C

11

H

16BF12NO (433.05): C 30.51, H 3.72, N 3.23; found: C 30.49, H 3.73,

2

280 (100) [TFSI] ; elemental analysis calcd (%) for C H F N O S2

1

20

6

2

5

N 3.35.

(438.41): C 30.14, H 4.60, N 6.39; found: C 30.41, H 4.31, N 6.47.

N-Butyl-N-methylmorpholinium bis(trifluoromethanesulfonyl)imide

(MO [TFSI]): Colorless liquid; yield: 90%; H NMR: d=1.00 (t, J=

N-Butyl-N-methylpiperidinium (nonafluoro-n-butyl)trifluoroborate (PI14

-

1

[

nC

4

F

9

BF

3

]): White solid; yield: 95%; H NMR: d=0.99 (t, J=7.2 Hz,

1

4

3

H), 1.44 (m, 2H), 1.74 (m, 2H), 1.87 (m, 2H), 1.99 (brs, 22H); 3.23

7.2 Hz, 3H), 1.45 (m, 2H), 1.92 (m, 2H), 3.41 (s, 3H), 3.67 (m, 32H),

4.11 ppm (brs, 22H); FAB-MS: m/z (%): 158 (100) [MO14] , 280 (100)

+

(

s, 3H), 3.53 ppm (m, 32H); FAB-MS: m/z (%): 156 (100) [PI14] , 287

2210

www.chemeurj.org

ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2006, 12, 2196 – 2212

Ionic Liquids

FULL PAPER

ꢀ

[

3

TFSI] ; elemental analysis calcd (%) for C11

H

20

F

6

N

2

O

5

S

2

(438.41): C

sis calcd (%) for C

C 44.18, H 8.33, N 5.74.

N-Methoxyethyl-N-methylmorpholinium tetrafluoroborate (MO1.1O2

9

H

20BF

4

NO (245.07): C 44.11, H 8.23, N 5.72; found:

0.14, H 4.60, N 6.39; found: C 30.01, H 4.58, N 6.37.

N-Methoxyethyl-N-methylmorpholinium bis(trifluoromethanesulfonyl)-

imide (MO1.1O2[TFSI]): Colorless liquid; yield: 89%; H NMR: d=3.39

-

1

[BF

4

]): White solid; yield: 88%; H NMR: d=3.39 (s, 3H), 3.43 (s, 3H),

(

2

s, 3H), 3.48 (s, 3H), 3.70 (m, 2H), 3.76 (m, 2H), 3.95 (s, 2H), 3.99 (s,

3.66 (m, 2H), 3.71 (m, 2H), 3.90 (s, 2H), 3.96 (s, 2H), 4.08 ppm (brs, 2

+

ꢀ

4

] ; elemen-

H), 4.11 ppm (brs, 22H); FAB-MS: m/z (%): 160 (100) [MO1.1O2] ,

2H); FAB-MS: m/z (%): 160 (100) [MO1.1O2] , 87 (100) [BF

tal analysis calcd (%) for C NO (247.04): C 38.89, H 7.34, N 5.67;

found: C 38.90, H 7.06, N 5.71.

ꢀ

80 (100) [TFSI] ; elemental analysis calcd (%) for C10

H

18

F

6

N

2

O

6

S

2

8

H

18BF

4

2

(

440.38): C 27.27, H 4.12, N 6.36; found: C 27.15, H 3.84, N 6.44.

General procedure for hydrophilic quaternary ammonium salts with tet-

rafluoroborate ([QA][BF ]): The salts were prepared on a 50 mmol scale

as follows. The aqueous [QA][OH] was neutralized with aqueous H[BF

in a PFA flask to a pH value of 6–7. The solution was concentrated at

0–408C at reduced pressure, giving a viscous liquid or a solid. The vis-

cous liquid was further dried at 408C for 24 h under high vacuum to

remove the volatile impurities and then dissolved in anhydrous CH Cl

60 mL), filtered through a PTFE membrane filter (0.2 mm). The collect-

4

]

3

Acknowledgements

This work was supported by Grant-in-Aid for Scientific Research from

the Japanese Ministry of Education, Science, Sports, and Culture (No.

17073017).

2

(

ed solution was evaporated, and dried at 70–1008C for 48 h under a high

vacuum to give a pale yellow or colorless liquid (water content: ꢁ100–

3

00 ppm). The solid product was dried at 908C and purified by recrystal-

lization from methanol on addition of 2-butanol to give a white solid.

[1]For books, see: a) P. Wasserscheid, T. Welton, Ionic Liquids in Syn-

thesis, Wiley-VCH, Weinheim, 2003; b) R. D. Rogers, K. R. Seddon,

N-Methoxymethyl-N-methylpyrrolidinium tetrafluoroborate (PY1.1O1

-

1

Ionic Liquids: IndustrialAppi lc ations for Green Chemistry

(ACS

[

BF

4

]): Colorless liquid; yield: 95%; H NMR: d=2.26 (m, 22H), 3.20

Symposium Series 818), American Chemical Society, Washington,

DC, 2002; c) R. D. Rogers, K. R. Seddon, Ionic Liquids as Green

Solvents: Progress and Prospects (ACS Symposium Series 856),

American Chemical Society, Washington, DC, 2003; d) R. D.

Rogers, K. R. Seddon, Ionic Liquids IIIA: Fundamentals, Progress,

Challenges, and Opportunities (ACS Symposium Series 901), Ameri-

can Chemical Society, Washington, DC, 2005; e) R. D. Rogers, K. R.

Seddon, Ionic Liquids IIIB: Fundamentals, Progress, Challenges, and

Opportunities (ACS Symposium Series 902), American Chemical So-

ciety, Washington, DC, 2005.

(

s, 3H), 3.59 (m, 2H), 3.71 (m, 3H + 2H), 4.77 ppm (s, 2H); FAB-MS:

+

ꢀ

4

] ; elemental analysis calcd

m/z (%): 130 (100) [PY1.1O1] , 87 (100) [BF

%) for C NO (217.02): C 38.74, H 7.43, N 6.45; found: C 38.94,

H 7.45, N 6.46.

N-Butyl-N-methylpyrrolidinium tetrafluoroborate (PY14[BF

(

7 4

H16BF

4

]): White

solid; yield: 90%; H NMR: d=0.98 (t, J=7.4 Hz, 3H), 1.43 (m, 2H),

.89 (brs, 2H), 2.30 (brs, 22H), 3.24 (s, 3H), 3.52 (m, 2H), 3.71 ppm

1

+

ꢀ

4

] ; ele-

(

brs, 22H); FAB-MS: m/z (%): 142 (100) [PY14] , 87 (100) [BF

mental analysis calcd (%) for C N (229.07): C 47.19, H 8.80, N

.11; found: C 47.02, H 8.95, N 6.14.

N-Methoxyethyl-N-methylpyrrolidinium tetrafluoroborate (PY1.1O2[BF

9

H20BF

4

6

[

2]For selected reviews, see: a) T. Welton, Chem. Rev. 1999, 99, 2071–

4

]):

Pale yellow liquid; yield: 96%; H NMR: d=2.31 (brs, 22H), 3.30 (s,

2

3

084; b) P. Wasserscheid, W. Keim, Angew. Chem. 2000, 112, 3926–

945; Angew. Chem. Int. Ed. 2000, 39, 3772–3789; c) J. Dupont,

1

3

1

H), 3.39 (s, 3H), 3.77 (m, 32H), 3.94 ppm (s, 2H); FAB-MS: m/z (%):

R. F. de Souza, P. A. Z. Suarez, Chem. Rev. 2002, 102, 3667–3692.

3]For selected reviews, see: a) R. A. Sheldon, R. M. Lau, M. J. Sor-

gedrager, F. van-Rantwijk, K. R. Seddon, Green Chem. 2002, 4,

+

ꢀ

44 (100) [PY1.1O2] , 87 (100) [BF

4

] ; elemental analysis calcd (%) for

[

8 4

C H18BF NO (231.04): C 41.59, H 7.85, N 6.06; found: C 41.77, H 8.05, N

6

.34.

1

47–151; b) U. Kragel, M. Eckstein, N. Kaftzik, Curr. Opin. Biotech-

N-Butyl-N-methyloxazolidinium tetrafluoroborate (OX14[BF

4

]): Pale

yellow liquid; yield: 95%; H NMR: d=0.99 (t, J=7.4 Hz, 3H), 1.44 (m,

nol. 2002, 13, 565–571; c) S. Park, R. J. Kazlauskas, Curr. Opin. Bio-

technol. 2003, 14, 432–437.

1

2

H), 1.92 (m, 2H), 3.33 (s, 3H), 3.63 (m, 2H), 3.91 (m, 2H), 4.43 (m,

[4]For selected reviews, see: a) C. C. Tzschucke, C. Markert, W. Bann-

warth, S. Roller, A. Hebel, R. Haag, Angew. Chem. 2002, 114, 4136–

4173; Angew. Chem. Int. Ed. 2002, 41, 3964–4000; b) C. F. Poole, J.

Chromatogr. A 2004, 1037, 49–82.

H), 5.01 (d, J=5.6 Hz, 1H), 5.08 ppm (d, J=5.6 Hz, 1H); FAB-MS:

+

ꢀ

4

] ; elemental analysis calcd

m/z (%): 144 (100) [OX14] , 87 (100) [BF

%) for C NO (231.04): C 41.59, H 7.85, N 6.06; found: C 41.39,

H 7.59, N 6.17.

N-Methoxyethyl-N-methyloxazolidinium (OX1.1O2-

(

8 4

H18BF

[5]a) M. Antonietti, D. Kuang, B. Smarsly, Y. Zhou, Angew. Chem.

2

004, 116, 5096–5100; Angew. Chem. Int. Ed. 2004, 43, 4988–4992;

tetrafluoroborate

]): Pale yellow liquid; yield: 94%; H NMR: d=3.40 (s, 3H), 3.42 (s,

H), 3.88–4.07 (m, 6H), 4.45 (m, 2H), 5.02 (d, J=5.6 Hz, 1H), 5.12 ppm

1

b) E. R. Cooper, C. D. Andrews, P. S. Wheatley, P. B. Webb, P. Wor-

mald, R. E. Morris, Nature 2004, 430, 1012–1016.

[

BF

4

3

+

[6]For books, see: a) H. Ohno, Electrochemical Aspects of Ionic Liq-

uids, Wiley-Interscience, Hoboken/New Jersey, 2005; b) A. Weber,

G. E. Blomgrem in Advances in Lithium-Ion Batteries (Eds.: W. A.

Van Schalkwijk, B. Scrosati), Kluwer Academic/Plenum, New York,

(

d, J=5.6 Hz, 1H); FAB-MS: m/z (%): 146 (100) [OX1.1O2] , 87 (100)

ꢀ

[

BF

4

] ; elemental analysis calcd (%) for C

7

H

16BF

4

NO

2

(233.02): C 36.08,

H 6.92, N 6.01; found: C 35.92, H 6.63, N 6.22.

N-Butyl-N-methylpiperidinium tetrafluoroborate (PI14[BF ]): White

solid; yield: 88%; H NMR: d=0.99 (t, J=7.2 Hz, 3H), 1.43 (m, 2H),

4

2

002, pp. 185–232.

1

[

7]For examples of imidazolium ionic liquids, see: a) N. Nakagawa, S.

1

3

.74 (m, 2H), 1.86 (m, 2H), 1.99 (brs, 22H), 3.23 (s, 3H), 3.53 ppm (m,

Izuchi, K. Kuwana, T. Nukuda, Y. Aihara, J. Electrochem. Soc. 2003,

+

ꢀ

4

] ; elemental

2H); FAB-MS: m/z (%): 156 (100) [PI14] , 87 (100) [BF

N (243.10): C 49.41, H 9.12, N 5.76;

1

50, A695–A700; b) B. Garcia, S. Lavallee, G. Perron, C. Michot,

analysis calcd (%) for C10

found: C 49.38, H 8.82, N 5.74.

H22BF

4

M. Armand, Electrochim. Acta 2004, 49, 4583–4588.

[

8]For examples of quaternary ammonium ionic liquids, see: a) H. Sa-

kaebe, H. Matsumoto, Electrochem. Commun. 2003, 5, 594–598;

b) J. H. Shin, W. A. Henderson, S. Passerini, Electrochem. Commun.

N-Methoxyethyl-N-methylpiperidinium tetrafluoroborate (PI1.1O2[BF

Colorless liquid; yield: 94%; H NMR: d=1.73 (m, 2H), 2.00 (brs, 2

2

4

]):

1

H), 3.30 (s, 3H), 3.39 (s, 3H), 3.60 (m, 22H), 3.78 (m, 2H), 3.95 ppm

2

003, 5, 1016–1020; c) T. Sato, T. Maruo, S. Marukane, K. Takagi, J.

+

ꢀ

4

] ; elemen-

(

s, 2H); FAB-MS: m/z (%): 158 (100) [PI1.1O2] , 87 (100) [BF

tal analysis calcd (%) for C NO (245.07): C 44.11, H 8.23, N 5.72;

found: C 44.15, H 8.01, N 5.75.

N-Butyl-N-methylmorpholinium tetrafluoroborate (MO14[BF

Power Sources 2004, 138, 253–261; d) H. Matsumoto, H. Sakaebe,

K. Tatsumi, J. Power Sources 2005, 146, 45–50.

9

H20BF

4

[

9]a) M. Ue, M. Takeda, A. Toriumi, A. Kominato, R. Hagiwara, Y.

Ito, J. Electrochem. Soc. 2003, 150, 499–502; b) Y. J. Kim, Y. Matsu-

zawa, S. Ozaki, K. C. Park, C. Kim, M. Endo, H. Yoshida, G.

Masuda, T. Sato, M. S. Dresselhausc, J. Electrochem. Soc. 2005, 152,

A710–A715.

4

]): White

solid; yield: 89%; H NMR: d=0.99 (t, J=7.4 Hz, 3H), 1.45 (m, 2H),

.89 (m, 2H), 3.37 (s, 3H), 3.65 (m, 32H), 4.08 ppm (brs, 22H);

1

+

ꢀ

FAB-MS: m/z (%): 158 (100) [MO14] , 87 (100) [BF ] ; elemental analy-

4

Chem. Eur. J. 2006, 12, 2196 – 2212

ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

2211

Z.-B. Zhou, H. Matsumoto, and K. Tatsumi

[

10]a) A. Noda, M. A. B. H. Susan, K. Kudo, S. Mitsushima, K Hayami-

zu, M. Watanabe, J. Phys. Chem. B 2003, 107, 4024–4033; b) M.

Yoshizawa, W. Xu, C. A. Angell, J. Am. Chem. Soc. 2003, 125,

[23]S. Forsyth, J. Golding, D. R. MacFarlane, M. Forsyth, Electrochim.

Acta 2001, 46, 1753–1757.

[24]J. Golding, N. Hamid, D. R. MacFarlane, M. Forsyth, C. Forsyth, C.

Collins, J. Huang, Chem. Mater. 2001, 13, 558–564.

[25]a) J. Kim, R. P. Singh, J. M. Shreeve, Inorg. Chem. 2004, 43, 2960–

2966; b) J. Kim, J. M. Shreeve, Org. Biomol. Chem. 2004, 2, 2728–

2734.

1

5411–15419; c) R. F. de Souza, J. C. Padilha, R. S. Goncalves, J.

Dupont, Electrochem. Commun. 2003, 5, 728–731.

11]a) A. I. Bhatt, I. May, V. A. Volkovich, M. E. Hetherington, B.

Lewin, R. C. Thied, N. Ertokc, J. Chem. Soc. Dalton Trans. 2002,

4

532–4534; b) A. I. Bhatt, I. May, V. A. Volkovich, D. Collison, M.

Helliwell, I. B. Polovov, R. G. Lewin, Inorg. Chem. 2005, 44, 4934–

940.

[26]K. Kim, S. Choi, D. Demberelnyamba, H. Lee, J. Oh, B. Lee, S.

Mun, Chem. Commun. 2004, 828–829.

[27]a) D. MacFarlane, J. Huang, M. Forsyth, Nature 1999, 402, 792–794;

b) D. MacFarlane, M. Forsyth, Adv. Mater. 2001, 13, 957–966.

[28]Y. Abu-Lebdeh, P. J. Alarco, M. Armand, Angew. Chem. 2003, 115,

4637–4639; Angew. Chem. Int. Ed. 2003, 42, 4499–4501.

4

[

12]K. Xu, Chem. Rev. 2004, 104, 4303–4417.

13]P. C. Howlett, D. R. MacFarlane, A. F. Hollenkamp, Electrochem.

Solid-State Lett. 2004, 7, A97–A101.

[

14]a) J. Sun, M. Forsyth, D. R. MacFarlane, Ionics 1997, 3, 356–362;

b) J. Sun, M. Forsyth, D. R. MacFarlane, J. Phys. Chem. B 1998, 102,

[29]a) J. C. Dearden, Sci. TotalEnviron.

1991, 109/110, 59–68;

b) R. J. C. Brown, R. F. C. Brown, J. Chem. Educ. 2000, 77, 724–731.

[30]a) E. I. Cooper, C. A. Angell, Solid State Ionics 1983, 9/10, 617–622;

b) E. I. Cooper, C. A. Angell, Solid State Ionics 1986, 18/19, 570–

576.

[31]a) W. Xu, E. I. Cooper, C. A. Angell, J. Phys. Chem. B 2003, 107,

6170–6178; b) W. Xu, L. M. Wang, R. A. Nieman, C. A. Angell, J.

Phys. Chem. B 2003, 107, 11749–11756.

[32]J. Pernak, K. Sobaszkiewicz, J. Foksowicz-Flaczyk, Chem. Eur. J.

2004, 10, 3479–3485.

[33]a) P. Bonhote, A. P. Dias, N. Papageorgiou, K. Kalyanasundaram, M.

Graetzel, Inorg. Chem. 1996, 35, 1168–1178; b) J. G. Huddleston,

A. E. Visser, W. M. Reichert, H. D. Willauer, G. A. Broker, R. D.

Rogers, Green Chem. 2001, 3, 156–164.

8

858–8864.

[

15]a) D. R. MacFarlane, P. Meakin, J. Sun, N. Amini, M. Forsyth, J.

Phys. Chem. A 1999, 103, 4164–4170; b) D. R. McFarlane, J. Sun, J.

Golding, P. Meakin, M. Forsyth, Electrochim. Acta 2000, 45, 1271–

1

278.

[

16]H. Matsumoto, M. Yanagida, K. Tanimoto, M. Nomura, Y. Kitaga-

wa, Y. Miyazaki, Chem. Lett. 2000, 29, 922–923.

17]Z. -B. Zhou, M. Takeda, M. Ue, J. Fluorine Chem. 2003, 123, 127–

31.

18]Z. -B. Zhou, M. Takeda, T. Fujii, M. Ue, J. Electrochem. Soc. 2005,

52, A351–A356.

19]Z.-B. Zhou, M. Takeda, M. Ue, J. Fluorine Chem. 2004, 125, 471–

76.

20]a) Z.-B. Zhou, H. Matsumoto, K. Tatsumi, Chem. Eur. J. 2004, 10,

1

4

[34]C. M. Forsyth, D. R. MacFarlane, J. J. Golding, J. Huang, J. Sun, M.

Forsyth, Chem. Mater. 2002, 14, 2103–2108.

6

2

581–6591; b) Z.-B. Zhou, H. Matsumoto, K. Tatsumi, Chem. Lett.

004, 33, 680.

[35]G. Gritzner, J. Kuta, Pure Appl. Chem. 1984, 56, 461–466.

[36]A. N. Dey, J. Electrochem. Soc. 1971, 118, 1547–1549.

[37]G. A. Molander, B. P. Hoag, Organometallics 2003, 22, 3313–3315.

[38]H. Heaney, G. Papageorgiou, R. F. Wilkins, Tetrahedron 1997, 53,

14381–14396.

[

21]a) Z.-B. Zhou, H. Matsumoto, K. Tatsumi, Chem. Eur. J. 2005, 11,

7

2

52–766; b) Z.-B. Zhou, H. Matsumoto, K. Tatsumi, Chem. Lett.

004, 33, 886–887.

[

22]Z.-B. Zhou, H. Matsumoto, K. Tatsumi, Chem. Lett. 2004, 33, 1636–

637.

Received: August 1, 2005

Published online: January 3, 2006

1

2212

www.chemeurj.org

ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2006, 12, 2196 – 2212

Article Doi

Cyclic quaternary ammonium ionic liquids with perfluoroalkyltrifluoroborates: Synthesis, characterization, and properties

DOI: 10.1002/chem.200500930

Source and publish data:

Authors:

Article abstract of DOI:10.1002/chem.200500930

Full text of DOI:10.1002/chem.200500930

Products guided by the article

R&D Labs maybe for 223437-11-4

Relevant to this article

Hot Product