New lipophilic 2-Amino-N,N′-dialkyl-4,5-dimethylimidazolium cations: Synthesis, structure, properties, and outstanding thermal stability in alkaline media

Article abstract of DOI:10.1002/chem.200901203

A series of new N,N′-dialkyl-4,5-dimethylimidazolium cations possessing electron-rich 2-imidazolylideneor phosphoranylidene-amino substituents has been efficiently synthesized from common precursors, N,N′-dialkyl-4,5-dimethylimidazol-2-ylidenes. The new l

Full text of DOI:10.1002/chem.200901203

FULL PAPER
DOI: 10.1002/chem.200901203
New Lipophilic 2-Amino-N,N’-dialkyl-4,5-dimethylimidazolium Cations:
Synthesis, Structure, Properties, and Outstanding Thermal Stability in
Alkaline Media
[a]
Roman A. Kunetskiy,[a] Ivana Cꢀsarˇovꢁ,[b] David Saman, and Ilya M. Lyapkalo*[a]
ˇ
Dedicated to Professor H.-U. Reissig on the occasion of his 60th birthday
Abstract: A series of new N,N’-dialkyl-
4,5-dimethylimidazolium cations pos-
sessing electron-rich 2-imidazolylidene-
or phosphoranylidene-amino substitu-
ents has been efficiently synthesized
from common precursors, N,N’-dialkyl-
strong alkali under both biphasic and
homogeneous conditions. Their excep-
tional aqueous base resistance, which
has hitherto only been seen with peral-
kylated polyaminophosphazenium cat-
ions, may be attributed to three fac-
tors: aromatic stabilization, efficient
resonance charge delocalization, and
steric protection of the exocyclic nitro-
gen linkage due to bulky lipophilic N-
alkyl substituents.
Keywords: aromatic stabilization ·
imidazolium cations · lipophilicity ·
4,5-dimethylimidazol-2-ylidenes.
The
phase-transfer catalysis
protection
·
steric
new lipophilic salts obtained have been
found to be highly stable towards
Introduction
important objective with regard to phase-transfer methodol-
ogy.[4,5]
Phase-transfer catalysis by means of lipophilic organic cat-
ions capable of maintaining an efficient concentration of re-
active anionic species in organic media has become an es-
sential tool for both established and state-of-the-art organic
synthesis.[1] Low cost, operational simplicity, and environ-
mental benefits offered by the phase-transfer catalysis tech-
nique are attractive for industrial applications.[2] In order for
the catalysis to be sustainable, the lipophilic cationic catalyst
has to be compatible with the basic/nucleophilic medium
under the reaction conditions. However, conventional or-
ganic cations have limited stability towards concentrated
aqueous alkali hydroxides or powerful nucleophiles, espe-
cially at elevated temperatures.[3,4] For this reason, the
design and synthesis of stable organic cations has become an
The organic phosphazenium cations developed by Schwe-
singer et al.[4,6] set new benchmarks in terms of thermal sta-
bility and base resistance, and hence they attracted consider-
able interest from industrial laboratories.[7] The commercial-
ly available tetrakisACTHNUTRGENUG{N [tris(dimethylamino)phosphoranyliden]-
amino}phosphonium (P5-phosphazenium cation) found in-
dustrial application for the production of high-quality poly-
meric materials.[8] However, excellent stability of the P5-
cation is attained at the expense of the high molecular
weight required for the extensive resonance delocalization
of the positive charge over the large dendritic network of al-
ternating P and N atoms. Moreover, all catalysts of the PN-
type exhibit potential dermal toxicity due to traces of hexa-
alkylphosphoramide.[5f]
It seems reasonable to assume that rational design ex-
ploiting a combination of electronic and steric effects to fa-
cilitate charge delocalization and enhance hydrolytic stabili-
ty could result in the discovery of cationic structures with
comparable or even greater base resistance. In a search for
such structures, we turned our attention to the versatile
tools and immense structural diversity offered by heterocy-
clic chemistry.
ˇ
[a] R. A. Kunetskiy, Dr. D. Saman, Dr. I. M. Lyapkalo
Institute of Organic Chemistry and Biochemistry
Academy of Sciences of the Czech Republic
Flemingovo n. 2, 166 10 Prague 6 (Czech Republic)
Fax : (+420)220-183-578
E-mail: ilya.lyapkalo@uochb.cas.cz
ˇ
[b] Dr. I. Cꢀsarovꢁ
Department of Inorganic Chemistry, Charles University
Hlavova 2030, 128 40 Prague 2 (Czech Republic)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200901203.
Chem. Eur. J. 2009, 15, 9477 – 9485
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
9477

Results and Discussion
In our own approach to the design of stable lipophilic cat-
ions, we assumed that accommodation of the positive charge
within an aromatic system should impart additional stability
and base resistance. Aromatically stabilized cationic systems
appended with appropriately positioned donor substituent(s)
to enable efficient delocalization of positive charge and
bulky alkyl groups to provide steric protection of centres
likely to be susceptible to nucleophilic cleavage were
deemed likely to display the highest thermal stability.
Herein, we report a new family of lipophilic cations incorpo-
rating a fully substituted[9] aromatic imidazolium core,
Scheme 1. Synthesis of the BIMA salts 1·BF4: a) Na/K alloy, glyme,
reflux; b) C2Cl6, THF, À408C; c) see 5 to 7+ in Scheme 2; d) KF in
MeCN, then aq. NaBF4; yields are given for step d).
namely
A
2-[1,3-dialkyl-4,5-dimethyl-1H-imidazol-2(3H)-yl-
5 at C-2 to give 6+ is documented,[14] the amination method
using tert-butyl azide to obtain 7+ is unprecedented in the
literature.[15] We capitalized on the recently described
simple, safe and efficient procedure for large-scale synthesis
of tert-butyl azide.[16] When treated with tert-butyl azide, the
nucleophilic carbenes 5 and tris(dimethylamino)phosphine
quantitatively produce the corresponding tert-butyltriazenes
8 and 9 (Scheme 2).
inafter referred to as bis(N,N’-dialkylimidazolium)amides or
BIMA), 2-[tris(dimethylamino)phosphoranylidenamino]-1,3-
dialkyl-4,5-dimethylimidazoliums 2+, and 2-[bis(dimethyl-
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
Figure 1. New lipophilic cations designed for phase-transfer applications
under drastic conditions: basic/nucleophilic media at high temperature.
Encircled with a dashed line are groups facilitating charge delocalization
due to resonance.
Scheme 2. Amination with tert-butyl azide: a) Me3CN3, THF; b) NH4BF4,
MeOH; c) CF3CO2H, then aqueous NaBF4.
Our choice of donor groups was stipulated by their ability
to facilitate maximum delocalization of a positive charge, as
known for Schwesingerꢃs basicity battery cells (R2N)3P=N[10]
or anticipated by us for the imidazol-2-ylideneamino group
in BIMA 1+. Cations 1+ and 2+ were found to be extremely
base-resistant, with their stability towards the strong alkali
base KOH being no lower than that of industrially applied
[(Me2N)3P=N]4P+ (P5-phosphazenium cation).[11]
The structure of 8a (R=CHMe2) was proven by single-
crystal X-ray diffraction[17] (Figure 2). Compounds 8 and 9
are relatively thermally stable intermediates. Upon mild
acidic treatment, they readily release CH2=CMe2 and N2,
furnishing high yields of the requisite building blocks 7·BF4
and 10·BF4, respectively.
The salts 2·BF4 and 3·BF4 were obtained as shown in
Scheme 3. Here again, KF-promoted coupling proved ad-
vantageous.[18]
The new imidazolium salts (1–3)·BF4 are stable, colour-
less, non-hygroscopic crystalline solids that are readily solu-
ble in chlorinated hydrocarbons, alcohols, and polar non-
protogenic organic solvents.
The synthesis of BIMA 1+ was accomplished in a conver-
gent manner starting from common precursors, stable tet-
raalkylimidazolin-2-ylidenes 5[12] (Scheme 1), which were
prepared on a multigram scale from 1,3-dialkyl-4,5-di-
ACHTUNGTRENNUNG
methyl-1H-imidazole-2(3H)-thiones 4[13] and stored as stock
solutions in hexane at room temperature under an inert at-
mosphere.
The BIMA salts 1·BF4 were obtained by KF-mediated
coupling of the intermediary salts (6a–c)·X with the corre-
sponding derivatives 7·X. We found it convenient to operate
with stable, crystalline, non-hygroscopic building blocks
6·BF4 and 7·BF4. However, the viability of the protocol with-
out isolation of the intermediates 6a·X and 7a·X was exem-
plified by the synthesis of 1a·BF4 from 5a in 65% overall
yield (see the Experimental Section). While chlorination of
In a first attempt to evaluate the stability of BIMA 1+ to-
wards aqueous alkali, a sample of 1a·ONf was heated to
1208C with an excess of 50% aqueous KOH in [D6]DMSO.
Although appreciable H/D exchange in the Me groups at
the 4,5-positions occurred, the N-isopropyl groups remained
intact, and no hydrolytic cleavage of 1a+ was observed
under these conditions. Encouraged by this result, we exam-
ined the base resistances of the cations 1+–3+ under bipha-
sic (PhCl/50% aqueous KOH) and homogeneous (KOH in
9478
www.chemeurj.org
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 9477 – 9485

Novel Lipophilic Highly Stable Imidazolium Cations
FULL PAPER
elevated temperatures, although it was found to be advanta-
geous in the design of cationic phase-transfer catalysts for
nucleophilic fluorinations with KF (Halex process) in aprot-
ic media.[5f]
BIMAs 1a+,c+ and phosphoranylidenaminoimidazolium
2a+ displayed much higher base resistance, their half lives
ranging from 30 to 95 h in a refluxing biphasic mixture of
PhCl/50% aqueous KOH at 1158C (Table 1, entries 1, 3,
and 4). These stabilities are far superior to those of conven-
tional organic cations, and comparable to those of the most
stable peralkylated polyaminophosphazenium cations.[4]
The neopentyl-substituted cations 1b+ and 2b+ and the
P5-phosphazenium cation[4,11] proved to be the most stable
of the studied series. They did not show tangible signs of de-
composition in a refluxing biphasic mixture of PhCl and
50% aqueous KOH (Table 1, entries 2, 5, and 7).[19] Recog-
nizing the temperature limitations of the above biphasic
mixtures at atmospheric pressure, we turned our attention
to homogeneous solutions of KOH in ethylene glycol, b.p.
1978C.
Figure 2. X-ray crystal structure of the triazene 8a. Displacement ellip-
soids are drawn at the 30% probability level. Hydrogen atoms are omit-
À
ted for clarity. Selected bonds lengths and angles [ꢄ, 8]: N1 C1
À
À
À
1.3411(16), N4 C1 1.3638(15), N5 C1 1.3618(16), N4 C2 1.3951(16),
À
À
À
À
C2 C3 1.3537(18), N5 C3 1.4115(17), N1 N2 1.3617(15), N2 N3
À
1.2517(16), N3 C12 1.4856(18); N5-C1-N1 134.36(11), N4-C1-N1
118.74(11), C1-N1-N2 113.83(10), N1-N2-N3 111.92(10).
Prolonged reflux of 1b·BF4, 2b·OTs, or [(Me2N)3P=
N]4P·BF4 in a 1m solution of KOH in ethylene glycol result-
ed in insignificant decomposition (Table 1, entries 8–10).
Based on these data, the half lives of the cations under
these conditions were estimated to be well in excess of
650 h. Thus, the most stable representative of the new class
of lipophilic organic cations having a tetraalkyl imidazolium
core matches the permethylated P5-cation in terms of chemi-
cal resistance towards strong alkali at high temperatures.
To gain better insight into the factors enhancing the sta-
bility of BIMA 1+, we analyzed the alkaline hydrolysis of
the cation 1a+ (Scheme 4), and the X-ray structures of the
salts 1a·OTs (Figure 3) and 1b·I (Figure 4).
Scheme 3. Synthesis of the salts 2·BF4 and 3·BF4: a) 10·BF4, KF in MeCN,
then aq. NaBF4; b) (Me2N)2C=NH, KF in MeCN, then aq. NaBF4.
ethylene glycol) conditions, comparing them with that of
commercially available tetrakisACTHNUGRTNEUNG{[tris(dimethylamino)phos-
phoranyliden]amino}phosphonium (P5-phosphazenium cation),
the exceptional base resistance of which is commonly ac-
knowledged. The results are summarized in Table 1.
The hydrolytic cleavage of 1a+ proceeded cleanly, result-
ing in the imidazolinone derivatives 7a’[9a] and 11. Despite
apparent favourability of the Hofmann degradation pathway
for the cation possessing four N-isopropyl groups, no prod-
ucts arising from base-induced elimination of propylene
could be detected.[20] Evidently, the observed hydrolysis is
initiated by nucleophilic attack of OHÀ at the C centre in
1a+, as shown in Scheme 4. For this to happen, a breakdown
of the 6p-aromaticity of the
Cation 3+ proved to be the least stable. It was found to
be slowly cleaved by 50% aqueous KOH with a loss of one
Me2N group, even at room temperature (Table 1, entry 6).
Apparently, the (Me2N)2C=N group in organic cations is not
compatible with concentrated aqueous alkali, especially at
imidACTHNUGRTNEUNGazolium ring has to occur
Table 1. Half lives of the organic cations in strongly basic media. Conditions A: vigorously stirred biphasic
mixture of PhCl/50% aqueous KOH, reflux (1158C inner temperature, 1458C oil-bath temperature). Condi-
tions B: 1m solution of KOH in ethylene glycol, reflux.
in the rate-limiting step, hence
the increase in the activation
barrier resulting in enhanced
base resistance.
1
Entry
Compound
Conditions
Reaction time [h]
Conversion [%][a]
t = [h]
2
1
2
3
4
5
6
7
8
9
1a·BF4
1b·BF4
1c·BF4
2a·BF4
2b·OTs[b]
3·BF4
A
A
A
A
A
A
A
B
24
48
21
30
48
16
48
24
24
24
42
31
stable
36
The X-ray structure of 1a+ in
1a·OTs[17] (see Figure 3 and the
legend) features two imidazoli-
um rings with nearly identical
structural parameters, bond
lengths, and angles, and an sp2-
hybridized exocyclic N1 atom
trace[a]
33
20
94
trace[a]
18
stable
58[c]
A
trace[a]
<5
stable
>650
>650
>650
1b·BF4
2b·OTs[b]
B
B
<5
<5
À
with two equal C N bonds. It
10
[(Me2N)3P=N]4P·BF4
indicates an even distribution
of the positive charge over both
[a] At the limit of detection by the NMR technique. [b] Obtained according to Scheme 3 followed by treat-
ment with TsOK in 50% aqueous MeOH instead of aqueous NaBF4. [c] At 258C.
Chem. Eur. J. 2009, 15, 9477 – 9485
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9479

I. M. Lyapkalo et al.
imidazolium rings and hence optimum electronic stabiliza-
tion of the system. The spatial arrangement of the core
structure displays significant deviation from planarity with a
dihedral angle of 748 between the planes containing the
imidazolium rings. All four CHMe2 substituents are spatially
arranged in such a way that they provide efficient hydropho-
bic protection of the C1/C4 centres from nucleophilic attack
by the hydroxide anion.[21]
It is reasonable to assume that, while the electronic effects
in BIMA 1+ remain essentially the same, steric shielding by
four bulky neopentyl substituents in 1b+ appears to be the
greatest (X-ray structure of 1b·I;[17] see Figure 4 and the
legend). The structural parameters of the intercyclic C1A-
N1A-C4A junction in 1b+ are almost identical to those in
1a+, but the dihedral angle between the imidazolium rings
(538)[17,22] is smaller than that in 1a+. Surprisingly, both
CMe3 groups of each imidazolium ring are located on the
same face of the ring plane. Impending over the C1A/C4A
centres, the CMe3 groups sterically hinder the approach of
OHÀ from this face, whereas the opposite face of the ring
plane is protected from nucleophilic attack by the neopentyl
group of the neighbouring imidazolium moiety. This ac-
counts for the highest base resistance being observed for
1b+.
Figure 3. X-ray crystal structure of the BIMA salt 1a·OTs. Displacement
ellipsoids are drawn at the 30% probability level. Hydrogen atoms are
À
omitted for clarity. Selected bonds lengths and angles [ꢄ, 8]: N1 C1
À
À
À
À
1.332(2), N1 C4 1.331(2), N2 C1 1.360(2), N2 C2 1.410(3), C2 C3
À
À
À
À
1.351(3), N3 C3 1.403(2), N3 C1 1.357(2), N4 C4 1.355(2), N4 C5
À
À
1.403(2), C5 C6 1.345(3), N5 C6 1.413(2); C4-N1-C1 124.96(13). Dihe-
dral angle between the two imidazolium rings: 73.96(11)8.
Considerable steric crowding is a likely cause of a slow
dynamic process on the NMR timescale, as observed for sol-
utions of the salt 1b·BF4. The process causes line-broaden-
ing and non-equivalence of CMe3 groups and 4,5-Me groups
1
in the H NMR spectrum at 258C. In CD3OD at À338C, the
1H signals of two non-equivalent CH2 groups are split into
AB systems (Figure 5), while the 13C NMR spectrum of 1b+
Figure 4. One of the three symmetrically independent cations observed in
the X-ray crystal structure of 1b·I. Displacement ellipsoids are drawn at
the 30% probability level. Hydrogen atoms are omitted for clarity. Se-
À
À
lected bonds distances and angles [ꢄ, 8]: N1A C1A 1.333(3), N1A C4A
À
À
À
1.336(3), N2A C1A 1.355(3), N2A C2A 1.412(3), C2A C3A 1.343(3),
À
À
À
À
N3A C3A 1.408(3), N3A C1A 1.364(3), N4A C4A 1.359(3), N4A C5A
À
1.407(3), C5A C6A 1.345(4), C1A-N1A-C4A 122.8(2). Dihedral angle
between the two imidazolium rings: 52.9(2)8.
Figure 5. Temperature-dependent 1H NMR spectra of the cation 1b+
measured in CD3OD at a) 240 K, b) 298 K, and c) 313 K.
indicates a structure with two identical imidazolium rings
lacking perpendicular planes of symmetry (Figure 6). At ele-
vated temperature (1008C in [D6]DMSO), the proton signals
within each group become equivalent and appear as typical
singlets (see the Experimental Section). In contrast, no re-
Scheme 4. Hydrolytic cleavage of the BIMA cation.
9480
www.chemeurj.org
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 9477 – 9485

Novel Lipophilic Highly Stable Imidazolium Cations
FULL PAPER
ly stabilized aromatic imidazolium moiety, with bulky N-
alkyl groups providing additional steric protection. Suscepti-
bility to Hofmann degradation is strongly diminished due to
the low charge density on the endocyclic nitrogen atoms or
may be completely precluded, as in the case of the tetra-
neopentyl cation 1b+.
The outstanding stability of the fully substituted imidazo-
lium cations paves the way to applications for which high
base resistance is a prerequisite for sustainable chemical
performance, that is, phase-transfer catalysts under harsh re-
action conditions, anhydrous fluoride sources, anionic reac-
tions in aprotic media, or base-induced polymerization. Syn-
theses and exploratory studies of new salts based on the cat-
ions 1+ and 2+ combined with highly basic anions, specifi-
cally fluorides, alcoholates, and enolates, are currently un-
derway and will be reported in due course.
Figure 6. Temperature-dependent 13C NMR spectra of the cation 1b+
measured in CD3OD at a) 240 K and b) 300 K.
Experimental Section
stricted rotation was observed for the cation 1a+, and its
NMR spectra remained unchanged as a function of temper-
ature down to À208C.
General: Melting points (m.p.; uncorrected): Wagner & Munz PolyTherm
A apparatus; elemental analyses: PE 2400 Series II CHN analyzer. NMR
spectra were measured from solutions in CDCl3 on a Bruker Avance 400
spectrometer at 300 K unless stated otherwise: 1H (400 MHz) referenced
to internal standard SiMe4 in CDCl3 and C6D6 or to residual solvent
peaks[24] in [D6]DMSO and CD3CN; 13C (100.6 MHz) referenced to d-sol-
vent signals;[24] 31P (162 MHz) referenced to internal standard (MeO)3PO
(d=3.0 ppm in CDCl3 or 3.5 ppm in [D6]DMSO);[25] 19F NMR
(376.5 MHz) referenced to internal standard PhCF3 (d=À63.7 ppm),[26]
The observed temperature-dependent dynamics in the
NMR spectra of 1b+ could be rationalized in terms of a
combination of two different processes of sterically hindered
rotation with similar activation barriers (DG): one is due to
À
slow rotation of the Me3CCH2 groups about the CH2
N
bonds causing splitting of the CH2 signals to AB systems,
while the other is due to slow rotation of the imidazolium
À
the signals of BF4 of the imidazolium salts appeared at À155Æ0.5 ppm
À
À
as two singlets due to 10BF4 and 11BF4 (1:4 ratio), the former 0.06 ppm
downfield from the latter. Temperature-dependent NMR spectra were
measured on a Bruker Avance 500 spectrometer (500 MHz for 1H,
125.7 MHz for 13C). IR spectra were recorded on a Bruker Alpha spec-
trometer equipped with an ATR device. ESI mass spectra were recorded
using a Thermo Scientific LCQ Fleet mass spectrometer. High-resolution
mass spectra (HRMS) were obtained with EI or APCI instruments.
À
À
rings about the exocyclic C2 N/N C2’ bonds resulting in a
loss of perpendicular planes of symmetry.
Our failure to obtain an N-methylated product from the
reaction of 1a+ with TfOMe points to substantial steric
crowding created by the isopropyl groups around the C2-N-
C2’ fragment. Instead, N-protonation occurred,[23] giving rise
to the stable symmetrical dication [1aH]2+, the identity of
which was verified independently by treatment of 1a+ with
TfOH (Scheme 5).
3-Hydroxybutan-2-one, N,N’-diisopropylthiourea, pentan-1-ol, tert-buta-
nol, MeOH, CHCl3, Me3CCN, LiAlH4, cyclohexylamine, S=CCl2, C2Cl6,
25% aqueous NH3, 40% aqueous HBF4, CF3CO2H, NaBF4, PACHTUNGTRENNUNG(NMe2)3,
PhCl, ethylene glycol, TfOMe, and TfOH were used as purchased from
commercial suppliers. Et3N was dried and stored over KOH pellets.
(Me2N)2C=NH was freshly distilled in vacuo over BaO. NH4BF4 was pre-
pared by careful neutralization of aqueous NH3 with aqueous HBF4 fol-
lowed by removal of water under reduced pressure and drying under
high vacuum.
tert-Butyl azide,[16] neopentylamine,[27] and N,N’-dineopentyl- and N,N’-di-
cyclohexylthioureas[28] were obtained according to the literature proce-
dures.
All reactions were carried out in heat-gun-dried glassware equipped with
magnetic stirring bars under an atmosphere of dry argon. 1H NMR con-
trol of the reaction mixtures was routinely performed to ensure complete
conversions of the starting materials. Solvents were dried by standard
procedures: THF and 1,2-dimethoxyethane were distilled over Na/K
alloy with added Ph2CO; diglyme was distilled in vacuo over Na; toluene
was distilled over Na; MeCN was distilled over CaH2 and stored over
3 ꢄ molecular sieves; hexane was distilled over P2O5. Commercially
available “dry” KF (ca. 20 g) was placed in a mortar and maintained in
an oven at 2008C for 12 h. Then, whilst still hot, it was thoroughly
ground to a fine powder using a hot pestle (thick leather gloves!). Thus,
pre-dried KF was used for further activation to serve as a heterogeneous
base in the synthesis of the target cations. All the products obtained were
of more than 95% purity (NMR).
Scheme 5. Protonation of 1a·BF4 to give [1aH]·OTf·BF4; no methylation
product was detected in the reaction with TfOMe.
Conclusions
In conclusion, we have designed and synthesized new organ-
ic cations, the alkaline base resistance of which at high tem-
perature far exceeds that of conventional organic cations
and is on a par with that of the permethylated P5-phospha-
zenium cation.[4,11] The enhanced stability may be attributed
to accommodation of the positive charge by an electronical-
Chem. Eur. J. 2009, 15, 9477 – 9485
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9481

I. M. Lyapkalo et al.
N,N’-Dialkyl-4,5-dimethyl-1H-imidazole-2(3H)-thiones (4): General pro-
cedure (GP 1) based on the method described earlier.[13] The requisite
N,N’-dialkylthiourea (100 mmol), 3-hydroxybutanone (110 mmol), and
pentan-1-ol (65 mL) were placed in a 250 mL reaction flask equipped
with a magnetic stirring bar, a Dean–Stark head, and a reflux condenser.
The reaction mixture was heated with an oil bath and maintained under
reflux for about 48 h until no more water was produced, indicating com-
plete conversion. We found that reflux in pentan-1-ol and the use of a
Dean–Stark head provides better reaction control and affords cleaner
conversions. The work-ups of the reaction mixtures and spectral data of
the products were as specified below.
1,3-Dicyclohexyl-4,5-dimethyl-1H-imidazol-2(3H)-ylidene (5c): 1H NMR
(C6D6): d=1.11–1.33 (brm, 6H), 1.44–1.61 (brm, 2H), 1.66–1.86 (brm,
10H), 1.94–2.10 (brm, 4H), 2.24–2.47 (brm, 4H) (all CH2 and Me),
3.66 ppm (tt, J=11.6 Hz, J=3.8 Hz, 2H; NCH); 13C NMR (C6D6): d=8.8
(Me), 26.0, 26.5, 35.5 (all CH2), 58.7 (NCH), 121.5 ppm (C=C); a C-2
signal was not observed due to very low intensity.
2-Chloro-N,N’-dialkyl-4,5-dimethylimidazolium
tetrafluoroborates
(6·BF4): General procedure (GP 3): A stock solution of 5 (15.0 mmol) in
hexane was added to a reaction flask. The hexane was removed in vacuo,
THF (60 mL) was added to the solid residue, and the resulting solution
was cooled to À408C. C2Cl6 (3.800 g, 16.0 mmol) was added in one por-
tion under vigorous stirring, and the reaction mixture was allowed to
gradually warm to ambient temperature and stirred for 24 h. It was then
poured into a separating funnel containing a two-phase mixture of CHCl3
(100 mL)/10% aqueous NaBF4 (100 mL), which was vigorously shaken.
The organic phase was separated, and the aqueous phase was extracted
once with CHCl3 (20 mL). The combined organic phases were washed
with 10% aqueous NaBF4 solution (100 mL), dried (MgSO4), and fil-
tered. The volatiles were removed in vacuo, and the solid residue was
dried under high vacuum at 508C overnight to remove residual traces of
C2Cl6, yielding pure product 6 as a yellowish crystalline solid.
1,3-Diisopropyl-4,5-dimethyl-1H-imidazole-2(3H)-thione (4a): The heat-
ing was turned off, and the reaction mixture was allowed to slowly cool
overnight with stirring while still in the oil bath, whereupon white crys-
tals separated. The mixture was further cooled to À208C, and the precipi-
tated crystals were collected by filtration, washed a few times with cold
(À208C) methanol, and dried in high vacuum to furnish the pure product
4a (18.05 g, 85%) as colourless crystals. 1H and 13C NMR data of 4a
were in good agreement with those reported in the literature.[13]
1,3-Dineopentyl-4,5-dimethyl-1H-imidazole-2(3H)-thione (4b): The reac-
tion mixture was cooled to ambient temperature, the solvent was evapo-
rated in vacuum, and the residual oil was distilled under high vacuum
(1658C/0.3 mbar) to give the product 4b (21.48 g, 80%), which slowly
crystallized to a low-melting yellowish solid. 1H NMR (C6D6): d=1.01 (s,
18H; CMe3), 1.60 ppm (s, 6H; Me); NCH2 signals were not observed due
to very strong broadening; 13C NMR (C6D6): d=10.4 (Me), 29.3 (CMe3),
35.2 (CMe3), 54.9 (NCH2), 121.1 (C=C), 167.6 ppm (C=S); HRMS (EI):
m/z: calcd for C15H28N2S: 268.1973 [M+]; found: 268.1965.
2-Chloro-1,3-diisopropyl-4,5-dimethylimidazolium
tetrafluoroborate
(6a·BF4): 69% yield, m.p. 156–1588C (from AcOEt/MeOH, 20:1);
1H NMR: d=1.64 (d, 3J=7.0 Hz, 12H; CHMe2), 2.35 (s, 6H; Me),
4.85 ppm (sept, 3J=7.0 Hz, 2H; NCH); 13C NMR: d=9.9 (Me), 20.4
(CHMe2), 52.8 (NCH), 126.4 (C-Cl), 128.2 ppm (C=C); these NMR data
are in good agreement with those described earlier for the tetrachloro-
AHCTUNGTERNNUNG ;
aluminate salt;[14] IR: n˜ =899, 1032, 1044, 1093, 1377, 1454, 1506 cmÀ1
HRMS (ESI+): m/z: calcd for C11H2035ClN2+: 215.1310 [M+]; found:
215.1311; elemental analysis calcd (%) for C11H20BClF4N2: C 43.67, H
6.66, N 9.26; found: C 43.77, H 6.69, N 9.12.
1,3-Dicyclohexyl-4,5-dimethyl-1H-imidazole-2(3H)-thione (4c): The reac-
tion mixture was cooled to ambient temperature, the solvent was evapo-
rated in vacuo, and the residual oil was dried under high vacuum (1508C/
0.05 mbar) for 2 h to give the product 4c (28.08 g, 96% yield) as a tawny
2-Chloro-1,3-dineopentyl-4,5-dimethylimidazolium
tetrafluoroborate
(6b·BF4): 95% yield, m.p. 130–1328C (from AcOEt); 1H NMR
(CD3CN): d=1.04 (s, 18H; CMe3), 2.30 (s, 6H; Me), 4.00 ppm (s, 4H;
NCH2); 13C NMR (CD3CN): d=10.8 (Me), 28.1 (CMe3), 35.5 (CMe3),
57.6 (CH2), 130.3 (C=C), 131.6 ppm (C-Cl); IR: n˜ =1034, 1048, 1101,
1484, 1506 cmÀ1; HRMS (ESI+): m/z: calcd for C15H2835ClN2+: 271.1936
[M+]; found: 271.1935; elemental analysis calcd (%) for C15H28BClF4N2:
C 50.23, H 7.87, N 7.81; found: C 50.50, H 7.95, N 7.67.
1
glass. H NMR (C6D6): d=0.59–2.25 (m), 1.78 (s) (26H, all CH2 and Me),
5.86 ppm (brs, 2H; NCH); 13C NMR (C6D6): d=10.6 (Me), 25.9, 26.6,
31.4 (all CH2), 58.0 (NCH), 120.8 ppm (C=C); a C=S signal was not ob-
served due to very low intensity; HRMS (EI): m/z: calcd for C17H28N2S:
292.1973 [M+]; found: 292.1966.
N,N’-Dialkyl-4,5-dimethyl-1H-imidazol-2(3H)-ylidenes (5): General pro-
cedure (GP 2) based on the method described earlier.[13] Na (1.61 g,
70 mmol) and K (5.47 g, 140 mmol) were sequentially added to an argon-
flushed reaction flask equipped with a three-way tap. The flask was then
connected to a diaphragm vacuum pump, and the Na/K eutectic was
formed by cautiously melting the metals with a heat-gun under reduced
pressure while gently swirling the contents of the flask. The reaction
flask was then filled with argon and allowed to cool to ambient tempera-
ture. A glass-coated magnetic stirring bar was placed in the flask, and a
solution of 4 (70 mmol) in 1,2-dimethoxyethane (150 mL) was cautiously
added with vigorous stirring. The reaction flask was fitted with a reflux
condenser, and the reaction mixture was stirred under reflux for 5 h until
full conversion. It was then allowed to cool, the solvent was removed in
vacuo, and the residue was diluted with dry hexane (250 mL) and stirred
for 1 h. The insoluble inorganic residue was allowed to precipitate over-
night. The yields and titres of the products 5 were estimated by with-
drawing aliquots of clear hexane solutions, evaporating the solvent,
drying under high vacuum at RT, and weighing the residues. The result-
ing clear homogeneous stock solutions of the products 5 (> 90% yield in
each case) could be stored for several months at ambient temperature
under an inert atmosphere. Care should be taken to avoid turbidity when
withdrawing aliquots from the stock solutions.
2-Chloro-1,3-dicyclohexyl-4,5-dimethylimidazolium
tetrafluoroborate
(6c·BF4): 65% yield, m.p. 102–1048C (from AcOEt); 1H NMR: d=1.14–
1.33 (m, 2H), 1.35–1.54 (m, 4H), 1.69–1.82 (m, 2H), 1.86–2.16 (m, 12H)
(all CH2), 2.36 (s, 6H; Me), 4.25–4.46 ppm (m, 2H; NCH); 13C NMR: d=
10.3 (Me), 24.8, 25.8, 30.2 (all CH2), 61.0 (NCH), 126.6 (C-Cl), 128.4 ppm
(C=C); IR (CDCl3): n˜ =1026, 1070, 1443, 1496 cmÀ1; HRMS (ESI+): m/z:
calcd for C17H2835ClN2+: 295.1936 [M+]; found: 295.1936; elemental anal-
ysis calcd (%) for C17H28BClF4N2: C 53.36, H 7.38, N 7.32; found: C
53.49, H 7.39, N 7.20.
2-Amino-N,N’-dialkyl-4,5-dimethylimidazolium tetrafluoroborates (7):
General procedure (GP 4): A stock solution of 5 (15.0 mmol) in hexane
was added to a reaction flask. The hexane was removed in vacuo, and the
solid residue was redissolved in THF (15 mL). A solution of tert-butyl
azide (1.640 g, 16.5 mmol) in toluene (16.5 mL) was then slowly added
under vigorous stirring. After 5 min, the reaction mixture was heated to
608C and stirred for 30 min yielding the triazene 8 (NMR control). The
volatiles were removed in vacuo, and the residue was redissolved in
MeOH (15 mL). Solid NH4BF4 (2.360 g, 22.5 mmol) was then slowly
added (Caution: Effervescence due to evolution of N2 and isobutylene!)
and the reaction mixture was stirred for 15 min. After aqueous work-up
as described in GP 3, the pure product 7 was isolated as a colourless crys-
talline solid.
1,3-Diisopropyl-4,5-dimethyl-1H-imidazol-2(3H)-ylidene (5a): 1H and
13C NMR data were in good agreement with those reported in the litera-
ture.[13]
1,3-Dineopentyl-4,5-dimethyl-1H-imidazol-2(3H)-ylidene (5b): 1H NMR
(C6D6): d=1.00 (s, 18H; CMe3), 1.76 (s, 6H; Me), 3.69 ppm (s, 4H;
NCH2); 13C NMR (C6D6): d=10.1 (Me), 28.5 (CMe3), 33.5 (CMe3), 58.3
(NCH2), 122.7 ppm (C=C); a C-2 signal was not observed due to very
low intensity.
2-Amino-1,3-diisopropyl-4,5-dimethylimidazolium
tetrafluoroborate
(7a·BF4): 80% yield, m.p. 119–1228C (from Et2O/AcOEt 1:1); 1H NMR:
d=1.52 (d, 3J=7.0 Hz, 12H; CHMe2), 2.18 (s, 6H; Me), 4.52 (sept, 3J=
7.0 Hz, 2H; NCH), 5.82 ppm (s, 2H; NH2); 13C NMR: d=9.7 (Me), 20.6
(CHMe2), 48.5 (NCH), 119.9 (C=C), 143.3 ppm (C-NH2); IR: n˜ =1020,
1514, 1647, 2941, 2987, 3364, 3445 cmÀ1. Treatment of the sample of 7a
with MeONa in MeOH gave a quantitative yield of the conjugated base,
9482
www.chemeurj.org
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 9477 – 9485

Novel Lipophilic Highly Stable Imidazolium Cations
FULL PAPER
1,3-diisopropyl-4,5-dimethyl-1H-imidazol-2(3H)-imine (7a’), the NMR
data of which matched those reported in the literature.[9a] HRMS (ESI+):
m/z: calcd for C11H22N3+: 196.1808 [M+]; found: 196.1807; elemental
analysis calcd (%) for C11H22BF4N3: C 46.67, H 7.83, N 14.84; found: C
46.81, H 7.88, N 14.69.
Bis(1,3-dineopentyl-4,5-dimethylimidazolium-2-yl)amide tetrafluorobo-
rate (1b·BF4): 1.667 g, 84% yield; recrystallization from hexane/AcOEt
(1:1) gave the inclusion compound 1b·AcOEt, m.p. 145–1488C; after fur-
ther crystallization, the m.p. was increased to 202–2048C; 1H NMR
(CD3OD, 499.84 MHz, 313 K): d=0.95 (brs, 36H; CMe3), 2.21 (s, 12H;
Me), 3.59 ppm (brs, 8H; NCH2); 13C NMR (CD3OD, 125.7 MHz): d=
10.0 (br, Me), 28.7 (br, CMe3), 35.9 (br, CMe3), 54.3 (br, NCH2), 121.4
2-Amino-1,3-dineopentyl-4,5-dimethylimidazolium
tetrafluoroborate
(7b·BF4): 80% yield, m.p. 198–2008C (from AcOEt/MeOH, 50:1);
1H NMR: d=1.03 (s, 18H; CMe3), 2.11 (s, 6H; Me), 3.70 (s, 4H; NCH2),
6.09 ppm (s, 2H; NH2); 13C NMR: d=9.7 (Me), 28.1 (CMe3), 34.9
(CMe3), 53.6 (CH2), 120.3 (C=C), 147.1 ppm (C-NH2); IR: n˜ =1017, 1036,
1
(br, C=C), 148.8 ppm (CNC); H NMR ([D6]DMSO, 499.84 MHz, 373 K):
d=0.92 (s, 36H; CMe3), 2.18 (s, 12H; Me), 2.89 ppm (brs, 8H, NCH2);
1H NMR (CD3OD, 499.84 MHz, 240 K): d=0.80 (s, 18H), 1.09 (s, 18H)
(both CMe3), 2.20 (s, 6H), 2.24 (s, 6H) (both Me), 2.79 (d, 2J=15.3 Hz,
2H; NCH2), 3.60 (d, 2J=14.1 Hz, 2H; NCH2), 3.71 (d, 2J=15.3 Hz, 2H;
NCH2), 3.88 ppm (d, 2J=14.1 Hz, 2H; NCH2); 13C NMR (CD3OD,
125.7 MHz, 240 K): d=9.9 and 10.2 (Me), 28.4 and 28.9 (CMe3), 35.8 and
36.2 (CMe3), 53.6 and 54.4 (NCH2), 121.0 and 121.6 (C=C), 148.3 ppm
(CNC); IR (for 1b·BF4): n˜ =1030, 1048, 1397, 1475, 1580 cmÀ1; HRMS
(ESI+): m/z: calcd for C30H56N5+: 486.4530 [M+]; found: 486.4527; ele-
mental analysis calcd (%) for C30H56BF4N5: C 62.82, H 9.84, N 12.21;
found: C 62.56, H 9.79, N 12.18.
1531, 1656, 2959, 3361, 3424 cmÀ1
;
HRMS (ESI+): m/z: calcd for
C15H30N3+: 252.2434 [M+]; found: 252.2435; elemental analysis calcd (%)
for C15H30BF4N3: C 53.11, H 8.91, N 12.39; found: C 53.25, H 8.97, N
12.27.
2-Amino-1,3-dicyclohexyl-4,5-dimethylimidazolium
tetrafluoroborate
(7c·BF4): 65% yield, m.p. 178–1808C (from AcOEt); 1H NMR: d=1.14–
1.36 (m, 2H), 1.41–1.64 (m, 4H), 1.65–1.79 (m, 2H), 1.79–2.11 (m, 12H)
(all CH2), 2.20 (s, 6H; Me), 3.94–4.12 (m, 2H; NCH), 5.89 ppm (s, 2H;
NH2); 13C NMR: d=10.1 (Me), 24.8, 25.6, 30.6 (all CH2), 56.8 (NCH),
119.9 (C=C), 143.5 ppm (C-NH2); IR: n˜ =1018, 1515, 1656, 2862, 2936,
3352, 3414 cmÀ1; HRMS (ESI+): m/z: calcd for C17H30N3+: 276.2434 [M+];
found: 276.2435; elemental analysis calcd (%) for C17H30BF4N3: C 56.21,
H 8.32, N 11.57; found: C 56.18, H 8.41, N 11.33.
Bis(1,3-dicyclohexyl-4,5-dimethylimidazolium-2-yl)amide tetrafluorobo-
rate (1c·BF4): 1.622 g, 87% yield, m.p. 76–788C; 1H NMR: d=1.05–1.30
(brm) and 1.61–2.12 (brm) (40H, all CH2), 2.21 (s, 12H; Me), 3.68 ppm
(brt, 4H; NCH); 13C NMR: d=10.5 (Me), 25.0, 26.2, 31.1 (all CH2), 57.0
(NCH), 119.4 (C=C), 144.6 ppm (CNC); IR: n˜ =1048, 1378, 1413,
1585 cmÀ1; HRMS (ESI+): m/z: calcd for C34H56N5+: 534.4530 [M+];
found: 534.4529; elemental analysis calcd (%) for C34H56BF4N5: C 65.69,
H 9.08, N 11.27; found: C 65.39, H 8.97, N 10.99.
(E)-2-(tert-Butyltriaz-2-enylidene)-1,3-diisopropyl-4,5-dimethyl-2,3-dihy-
dro-1H-imidazole (8a): A sample of intermediate 8a was obtained from
5a and tert-butyl azide as described in GP 4; >99% yield, beige crystal-
1
line solid that gradually turned dark on exposure to air. H NMR (C6D6):
3
2-[Tris(dimethylamino)phosphoranylidenamino]-1,3-diisopropyl-4,5-di-
methylimidazolium tetrafluoroborate (2a·BF4): 1.266 g, 95% yield, m.p.
99–1018C (from hexane/AcOEt 1:5); 1H NMR: d=1.47 (d, 3J=7.1 Hz,
d=1.21 (brd, J=7.2 Hz, 12H; CHMe2), 1.49 (s, 9H; CMe3), 1.63 (s, 6H;
Me), 5.13 ppm (br, 2H; NCH); 13C NMR: d=10.1 (Me), 21.4 (CHMe2),
29.4 (CMe3), 47.9 (NCH), 60.0 (CMe3), 117.8 (C=C), 153.2 ppm (C=N).
12H; CHMe2), 2.21 (s, 6H; CMe), 2.72 (d, 3J
ACHTUNGTRENNUNG
Tris(dimethylamino)
ACHTUNGTRENNUNG(amino)phosphonium tetrafluoroborate (10·BF4): P-
ACHTUNGTRENNUNG(NMe2)3 (7.670 g, 47.0 mmol) was carefully added dropwise to a solution
AHCTUNGTRENNUNG
of tert-butyl azide (5.520 g, 55.7 mmol) in toluene (5 mL) at À208C under
vigorous stirring. The reaction mixture was gradually allowed to warm to
ambient temperature overnight, yielding the triazene 9.[29] The mixture
was then cooled to À208C, whereupon CF3CO2H (8.04 g, 70.5 mmol) was
added dropwise over a period of 10 min. The cooling bath was then re-
moved, and the reaction mixture was stirred for an additional 15 min.
After aqueous work-up as described in GP 3, the pure product 10 was
isolated as a colourless crystalline solid (11.4 g, 91%). The spectral data
of 10 were in good agreement with those reported in the literature.[10]
AHCTUNGTRENNUNG
n˜ =737, 756, 974, 1035, 1048, 1408, 1568 cmÀ1; HRMS (ESI+): m/z: calcd
for C17H38N6P+: 357.2890 [M+]; found: 357.2889; elemental analysis
calcd (%) for C17H38BF4N6P: C 45.96, H 8.62, N 18.92; found: C 45.80, H
8.43, N 18.90.
2-[Tris(dimethylamino)phosphoranylidenamino]-1,3-dineopentyl-4,5-di-
methylimidazolium tetrafluoroborate (2b·BF4): 1.366 g, 91% yield. Ac-
cording to the procedure given for 1a·OTs (see below), a sample 2b·BF4
was converted into 2b·OTs. 1H NMR: d=0.98 (s, 18H; CMe3), 2.17 (d,
Bis(N,N’-dialkyl-4,5-dimethylimidazolium-2-yl)amide tetrafluoroborates
(1) and 2-[tris(dimethylamino)phosphoranylidenamino]-1,3-dialkyl-4,5-
dimethylimidazolium tetrafluoroborates (2): General procedure (GP 5):
Pre-dried KF (1.400 g, 24.0 mmol) was placed in a reaction flask and
strongly heated with a heat-gun under high vacuum for a few minutes to
remove all traces of moisture. It was then cooled under argon, and the
flask was equipped with a magnetic stirring bar. The requisite salt 6
(3.00 mmol), 7 (3.00 mmol), or 10 (0.800 g, 3.00 mmol) was added, fol-
lowed by MeCN (6 mL), and the reaction mixture was stirred for 48 h at
ambient temperature (synthesis of 1a,c and 2a) or at 608C (synthesis of
ACTHNUTRGNEUNG
6J(H,P)=0.6 Hz, 6H; CMe), 2.30 (s, 3H; Me, OTs), 2.65 (d, 3J=9.9 Hz,
18H; NMe2), NCH2 peaks not observed due to strong broadening, 7.08
(d, 3J=8 Hz, 2H; CHm, OTs), 7.84 ppm (d, 3J=8 Hz, 2H; CHo, OTs);
13C NMR: d=10.4 (Me), 21.3 (Me, OTs), 28.3 (CMe3), 34.8 (CMe3), 37.2
(d, 2J
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
977, 1182, 1549 cmÀ1
;
HRMS (ESI+): m/z: calcd for C21H46N6P+:
413.3516 [M+]; found: 413.3510; elemental analysis calcd (%) for
1
C28H53N6O3PS: C 57.51, H 9.13, N 14.37; found: C 57.36, H 9.00, N 14.11.
1b and 2b). H NMR spectra of the reaction mixtures indicated full con-
version of the starting materials. Each mixture was then poured into a
separating funnel containing a two-phase mixture of CHCl3 (50 mL)/12%
aqueous NaBF4 (110 mL), which was vigorously shaken. The organic
phase was separated, and the aqueous phase was extracted once with
CHCl3 (20 mL). The combined organic phases were washed with 12%
aqueous NaBF4 solution (50 mL), dried (MgSO4), and filtered. The vola-
tiles were removed in vacuo, and the solid residue was dried under high
vacuum to afford the pure product 1 or 2 as a beige crystalline solid.
Bis(1,3-diisopropyl-4,5-dimethylimidazolium-2-yl)amide tetrafluoroborate
(1a·BF4) from 1,3-diisopropyl-4,5-dimethyl-1H-imidazol-2(3H)-ylidene
(5a): One-pot procedure: A stock solution of 5a in hexane was added to
a reaction flask. The hexane was removed in vacuo, and the resulting
crystalline residue (1.140 g, 6.32 mmol) was redissolved in diglyme
(6.3 mL). A solution of tert-butyl azide (0.323 g, 3.16 mmol) in toluene
(3.16 mL) was slowly added under vigorous stirring. After 5 min, the re-
action mixture was heated to 608C and stirred for 30 min, then cooled to
À408C, whereupon C2Cl6 (0.748 g, 3.16 mmol) was added in one portion.
The resulting mixture was gradually allowed to warm to ambient temper-
ature and stirred for 24 h. It was then cooled to À208C, TfOH (0.280 mL,
3.16 mmol) was carefully added, and the resulting mixture was stirred for
15 min before being allowed to gradually warm to ambient temperature.
The volatiles were removed under high vacuum (0.05 mbar), and the re-
sidual mixture of 6a·X and 7a·X was diluted with dry MeCN (6.3 mL)
and combined with KF (1.540 g, 26.0 mmol), activated as described in
Bis(1,3-diisopropyl-4,5-dimethylimidazolium-2-yl)amide tetrafluoroborate
(1a·BF4): 1.080 g, 78% yield, m.p. 151–1538C (from hexane/AcOEt 1:6);
1H NMR: d=1.44 (d, 3J=7.0 Hz, 24H; CHMe2), 2.19 (s, 12H; Me),
4.21 ppm (sept, 3J=7.0 Hz, 4H; NCH); 13C NMR: d=10.2 (Me), 21.2
(CHMe2), 48.0 (NCH), 119.5 (C=C), 144.5 ppm (CNC); IR: n˜ =1051,
+
1096, 1367, 1425, 1573 cmÀ1; HRMS (ESI+): m/z: calcd for C22H40N5
:
374.3278 [M+]; found: 374.3279; elemental analysis calcd (%) for
C22H40BF4N5: C 57.27, H 8.74, N 15.18; found: C 57.06, H 8.56, N 14.78.
Chem. Eur. J. 2009, 15, 9477 – 9485
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9483

I. M. Lyapkalo et al.
GP 5. The resulting mixture was stirred for 48 h at ambient temperature.
Aqueous work-up as described in GP 5 followed by recrystallization of
the solid residue from AcOEt/hexane (6:1) gave the pure product 1a·BF4
(770 mg, 65%) as yellowish crystals.
tered through a 0.2 mm syringe filter. The clear filtrate was concentrated
in vacuo to afford yellowish crystals of 1a·OTs or 1b·I free from tetra-
fluoroborate (19F NMR control).
X-ray structures of the salts 1a·OTs and 1b·I and of the triazene 8a:[17]
Slow cooling of a saturated solution of 1a·OTs in AcOEt or of 1b·I in
AcOEt/hexane (1:1) afforded crystals suitable for X-ray analysis. Suitable
crystals of 8a were obtained by crystallization from Et3N. Crystal data
for 1a·OTs: C22H40N5·C7H7O3S, Mr =545.78, orthorhombic, space group
2-[Bis(dimethylamino)methyleneamino]-1,3-diisopropyl-4,5-dimethylimi-
dazolium tetrafluoroborate (3·BF4): Pre-dried KF (0.232 g, 4.00 mmol)
was placed in a reaction flask and strongly heated with a heat-gun under
high vacuum for a few minutes to remove all traces of moisture. It was
then cooled under argon, and the flask was equipped with a magnetic
stirring bar. MeCN (2 mL), (Me2N)2C=NH (0.126 mL, 1.00 mmol), and
the salt 6a·BF4 (0.303 g, 1.00 mmol) were successively added. The reac-
tion mixture was stirred for 24 h at ambient temperature. Aqueous work-
up as described in GP 5 furnished the pure product 3·BF4 (0.347 g, 91%)
as a yellowish crystalline solid. M.p. 169–1718C; 1H NMR: d=1.46 (d,
3J=7.1 Hz, 12H; CHMe2), 2.24 (s, 6H; Me), 2.88 (s, 12H; NMe2),
4.28 ppm (sept, 3J=7.1 Hz, 2H; NCH); 13C NMR: d=10.1 (Me), 21.0
(CHMe2), 39.7 (NMe2), 48.5 (NCH), 120.6 (C=C), 145.7 (CimN),
Pna21;
2990.96(9) ꢄ3; Z=4; dimensions of colourless crystal 0.5ꢅ0.4ꢅ0.1 mm;
150(2) K; m
(MoKa)=0.15 mmÀ1; R[F2>2s(F2)]=0.036, wR2 (all data)=
a=21.7539(4),
b=8.77290(10),
c=15.6722(3) ꢄ;
V=
A
ACHTUNGTRENNUNG
0.091, S=1.04, reflections/parameters 5606/356; max/min residual elec-
tron density 0.12/À0.28 eꢄÀ3. The hydrogen atoms in all structures were
recalculated in idealized positions (riding model) and assigned tempera-
ture
factors
Hiso(H)=1.2–1.5 Ueq
(pivot
atom);
1b·I:
¯
3ACTHNUTGREN(NNUG C30H56N5)·C4H8O2·3(I), Mr =1929.20, triclinic, space group P1 (No. 2);
a=10.76990(10), b=16.8764(2), c=30.3494(4) ꢄ, a=74.8719(7), b=
89.6370(9), g=83.1406(7)8; V=5285.16(11) ꢄ3; Z=2; colourless crystal
161.4 ppm (N=C
G
;
ACTHNUTRGNEUNG
0.3ꢅ0.2ꢅ0.1 mm; 150(2) K; m (MoKa)=0.94 mmÀ1; R[F2>2s(F2)]=0.035,
+
HRMS (ESI+): m/z: calcd for C16H32N5
294.2652 [M+]; found:
wR2 (all data)=0.079, S=0.89, reflections/parameters 20711/1021; max/
min residual electron density 0.74/À0.76 eꢄÀ3. 8a: C15H29N5, Mr =279.43,
orthorhombic, space group Pna21, a=12.6964(2), b=14.7323(3), c=
9.1166(2) ꢄ; V=1705.23(6) ꢄ3; Z=4; colourless crystal 0.4ꢅ0.4ꢅ
294.2653; elemental analysis calcd (%) for C16H32BF4N5: C 50.40, H 8.46,
N 18.37; found: C 50.15, H 8.43, N 18.08.
Reactions of the salt 1a·BF4 with TfOMe or TfOH: Bis(1,3-diisopropyl-
4,5-dimethylimidazolium-2-yl)amine
tetrafluoroborate
triflate,
0.3 mm; 150(2) K; mACTHNUTRGENNGU ACHTUGNTRENNUGN
(MoKa)=0.07 mmÀ1; R[F2>2s(F2)]=0.036, wR2 (all
[1aH]·OTf·BF4: A sample of 1a·BF4 (0.046 g, 0.10 mmol) in CDCl3 was
treated with TfOMe (0.018 g, 0.11 mmol). No reaction occurred after
24 h at ambient temperature. Heating of the sample for a week at 808C
resulted in disappearance of the signal of TfOMe and the formation of
1a·HOTf, the authenticity of which was verified by matching the NMR
data with those of a reference sample obtained by protonation of 1a·BF4
(0.046 g, 0.10 mmol) with an equimolar amount of TfOH in CHCl3 (ca.
data)=0.097, S=1.05, reflections/parameters 3753/190; max/min residual
electron density 0.19/À0.15 eꢄÀ3
.
Acknowledgements
À808C to RT, then high vacuum) to give
a quantitative yield of
[1aH]·OTf·BF4 as a yellow oil. 1H NMR (CD3CN): d=1.49 (d, 3J=
7.0 Hz, 24H; CHMe2), 2.35 (s, 12H; Me), 4.50 (sept, 3J=7.0 Hz, 4H;
NCH), 8.69 ppm (brs; NH, integral intensity reduced due to H/D ex-
Continuous support of this work by the IOCB under Grant No. Z4 055
0506 and by the Ministry of Education (Grant No. MSM0021620857 for
I.C.) is gratefully acknowledged.
change); 13C NMR (CD3CN): d=10.7 (Me), 21.1 (CHMe2), 52.2 (NCH),
1
121.8 (q, J
N
Stability tests of cations 1+–3+ and P5 under phase-transfer conditions:
Method A: A 50% aqueous KOH solution (1.5 mL) was added to a
Schlenk tube containing a solution of the corresponding tetrafluorobo-
rate salt (0.50 mmol) in chlorobenzene (1.5 mL). As a quantitative NMR
standard, dibenzo-[18]crown-6 or [18]crown-6 was chosen, in order to
avoid overlaps of the NMR signals with those of the cations to be tested.
The resulting two-phase mixture was kept under the conditions specified
in Table 1. The progress of the alkaline cleavage was monitored by 1H
and 31P NMR (if applicable) of aliquots taken from the chlorobenzene
phase and diluted with [D6]DMSO. The authenticities of compounds 7a’
and 11 resulting from the cleavage of 1a+ (see Scheme 4) were proven
by matching the NMR data with those of a reference sample (see the
synthesis of 7a·BF4 above) and HRMS, respectively; 11, HRMS (EI):
m/z: calcd for C11H20N2O: 196.1576 [M+]; found: 196.1572.
+
[1] E. V. Dehmlow, S. S. Dehmlow, Phase Transfer Catalysis, 3rd ed.,
VCH, Weinheim, 1993.
[2] a) B. Langlois, L. Gilbert, G. Forat, in Industrial Chemistry Library,
Vol. 8 (Eds.: J.-R. Desmurs, S. Ratton), Elsevier, Amsterdam, 1996,
pp. 244–292; b) M. E. Halpern, in Process Chemistry in the Pharma-
ceutical Industry (Ed.: K. G. Gadamasetti), Marcel Dekker, New
York, 1999, pp. 283–298; c) M. Makosza, M. Fedorynski, in Interfa-
cial Catalysis (Ed.: A. G. Volkov), Marcel Dekker, New York, 2003,
pp. 159–201; d) D. Albanese, in Interfacial Catalysis (Ed.: A. G.
Volkov), Marcel Dekker, New York, 2003, pp. 203–226.
[3] a) E. V. Dehmlow, V. Knufinke, J. Chem. Res. Synop. 1989, 224–
225; b) J. G. Dawber, J. C. Tebby, A. A. C. Waite, J. Chem. Soc.
Perkin Trans. 2 1983, 1923–1925.
[4] R. Schwesinger, R. Link, P. Wenzl, S. Kossek, M. Keller, Chem. Eur.
J. 2006, 12, 429–437, and references therein.
[5] a) D. J. Brunelle, D. A. Singleton, Tetrahedron Lett. 1984, 25, 3383–
3386; b) D. Albanese, D. Landini, A. Maia, M. Penso, J. Mol. Catal.
A 1999, 150, 113–131; c) D. Albanese, D. Landini, A. Maia, M.
Penso, Ind. Eng. Chem. Res. 2001, 40, 2396–2401; d) M. Makosza,
R. Bujok, Tetrahedron Lett. 2002, 43, 2761–2763; e) M. Henrich, A.
Marhold, A. A. Kolomeitsev, N. Kalinovich, G.-V. Roeschenthaler,
Tetrahedron Lett. 2003, 44, 5795–5798; f) A. Pleschke, A. Marhold,
M. Schneider, A. Kolomeitsev, G.-V. Roeschenthaler, J. Fluorine
Chem. 2004, 125, 1031–1038.
[6] a) R. Schwesinger, R. Link, G. Thiele, H. Rotter, D. Honert, H.-H.
Limbach, F. Maennle, Angew. Chem. 1991, 103, 1376–1378; Angew.
Chem. Int. Ed. Engl. 1991, 30, 1372–1375; b) R. Link, R. Schwesing-
er, Angew. Chem. 1992, 104, 864; Angew. Chem. Int. Ed. Engl. 1992,
31, 850; c) R. Schwesinger, R. Link, P. Wenzl, S. Kossek, Chem. Eur.
J. 2006, 12, 438–445.
+
Stability tests of cations 1b+, 2b+, and P5 under homogeneous condi-
tions: Method B: The salt 1b (0.115 g, 0.20 mmol) and [(Me2N)3P=
N]4P·BF4 were added to a solution of KOH (1 mmol) in ethylene glycol
(1 mL). After 24 h of reflux (see Table 1), a few small side signals in the
characteristic areas of both the 1H and 31P (for P5+) NMR spectra ap-
peared, although no appreciable change in relative ratio of the character-
+
istic 1H signals of the starting cations 1b and P5 was observed. Subject-
ing the salt 2b (0.100 g, 0.20 mmol) to the same conditions led to a simi-
lar degree of decomposition.
Anion exchange in the products 1a·BF4 and 1b·BF4 to obtain salts
1a·OTs and 1b·I for X-ray analysis: A solution of the tetrafluoroborate
salt 1 (1.00 mmol) in MeOH (10 mL) was treated with an at least tenfold
excess of TsOK (concentrated solution in 50% aqueous MeOH) or a sa-
turated KI solution in MeOH and the mixtures were stirred for 1 h at RT
and then set aside for 24 h at 48C (refrigerator). Thereafter, the volatiles
were removed in vacuo, and the residue obtained was dried under high
vacuum and then treated with CHCl3 (10 mL). The resulting suspension
was vigorously stirred for 15 min at ambient temperature and then fil-
[7] a) M. Fischer (BASF AG, Germany), EP 0691334A1, 1996; [Chem.
Abstr. 1996, 124, 264009]; b) P. Hupfield, A. Surgenor, R. Taylor
9484
www.chemeurj.org
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 9477 – 9485

Novel Lipophilic Highly Stable Imidazolium Cations
FULL PAPER
(Dow Corning, USA), EP 1008610A2, 2000; [Chem. Abstr. 2000,
133, 31350]; c) G. Moloney, P. Hupfield, A. Surgenor, R. Taylor
(Dow Corning, USA), EP 1008612A2, 2000; [Chem. Abstr. 2000,
133, 31352]; d) T. Nobori, S. Fujiyoshi, I. Hara, T. Hayashi, A. Shi-
bahara, K. Funaki, K. Mizutani, S. Kiyono (Mitsui Chemicals Inc.,
Japan), US 6469224 B1, 2002; [Chem. Abstr. 2001, 135, 344378]; N.
Yoshimura, S. Kiyono, K. Mizutani, I. Hara, T. Hayashi, T. Nobori,
Y. Yamamoto, M. Konno, Y. Inoue, A. Matsuura, T. Ohkubo
(Mitsui Chemicals Polyurethanes, Japan), EP 1775301A1, 2007;
[Chem. Abstr. 2006, 144, 129398].
R. Fawzi, M. Steimann, J. Wiethoff, Z. Anorg. Allg. Chem. 1997,
623, 554–560; e) N. Kuhn, R. Fawzi, M. Steimann, J. Wiethoff, Z.
Anorg. Allg. Chem. 1997, 623, 769–774; f) N. Kuhn, R. Fawzi, M.
Steimann, J. Wiethoff, Z. Anorg. Allg. Chem. 1997, 623, 1577–1582;
g) N. Kuhn, R. Fawzi, M. Grathwohl, H. Kotowski, M. Steimann, Z.
Anorg. Allg. Chem. 1998, 624, 1937–1939; h) N. Kuhn, H. Kotowski,
J. Wiethoff, Phosphorus Sulfur Silicon Relat. Elem. 1998, 133, 237–
244; ref.[9a]
.
[16] J. C. Bottaro, P. E. Penwell, R. J. Schmitt, Synth. Commun. 1997, 27,
1465–1467.
[8] a) T. Nobori, T. Suzuki, S. Kiyono, M. Kouno, K. Mizutani, Y.
Sonobe, U. Takaki (Mitsui Toatsu Chemicals, Japan), EP
0791600A1, 1997; [Chem. Abstr. 1997, 127, 234743]; b) T. Nobori, T.
Hayashi, S. Kiyono, U. Takaki (Mitsui Chemicals Inc., Japan), EP
0879838A2, 1998; [Chem. Abstr. 1999, 130, 38836]; c) T. Nobori, T.
Hayashi, T. Urakami, S. Yamasaki, Catalysis 2001, 43, 532–537;
[Chem. Abstr. 2001, 135, 372004]; d) T. Nobori, T. Hayashi, S. Fu-
jiyoshi, Petrotech 2002, 25, 516–520; [Chem. Abstr. 2002, 137,
358613]; e) T. Nobori, T. Hayashi, A. Shibahara, S. Fujiyoshi, Chem.
Eng. 2002, 47, 365–370; [Chem. Abstr. 2002, 137, 329748].
[9] The Me groups at the 4- and 5-positions are believed to be essential
to protect the structures from possible side reactions: a) M. Tamm,
D. Petrovic, S. Randoll, S. Beer, T. Bannenberg, P. G. Jones, J. Gru-
nenberg, Org. Biomol. Chem. 2007, 5, 523–530; b) I. Tommasi, F.
Sorrentino, Tetrahedron Lett. 2006, 47, 6453–6456.
[17] CCDC 729001 (8a), 729002 (1a·OTs), and 729003 (1b·I) contain the
supplementary crystallographic data for this paper. These data can
be obtained free of charge from The Cambridge Crystallographic
Data Centre via www.ccdc.cam.ac.uk/data_request/cif
[18] It induces the synthesis of 2a,b from 6a,b and (Me2N)3P=NTMS
equally well.
[19] The half-life of [(Me2N)3P=N]4P·Cl in 50% NaOH/PhCl at 1108C
has been reported to be 33 h.[4]
[20] The insignificance of Hofmann degradation in alkaline cleavage of
cations with extended resonance conjugation was noted earlier by
Schwesinger et al.[4]
[21] Steric repulsion between 4,5-Me and CHMe2 groups seems to con-
tribute to this arrangement. The distances between all but one CH3
carbons of the isopropyl groups and each C1/C4 centre are relatively
short and fall in a narrow range from 3.1 to 4.1 ꢄ.
[22] The dihedral angle ranges from 52.9(2)8 to 58.3(2)8 for three sym-
metrically independent cations observed in the crystal structure of
1b·I.
[23] Similar outcomes resulting in protonation rather than methylation
have been observed in reactions of sterically hindered 2,6-di-tert-bu-
tylpyridine derivatives with MeOSO2F under atmospheric pressure:
a) H. Weber, J. Pant, M. Liedigk, H. Wunderlich, Chem. Ber. 1981,
114, 1455–1463; b) C. J. Hou, Y. Okamato, J. Org. Chem. 1982, 47,
1977–1979. It was conjectured that the protonation occurs due to a-
elimination from the methylating agent or hydrolysis by traces of
stray moisture.
[24] H. E. Gottlieb, V. Kotlyar, A. Nudelman, J. Org. Chem. 1997, 62,
7512–7515.
[25] R. Streck, A. J. Barnes, Spectrochim. Acta Part A 1999, 55, 1049–
1058.
[26] D. Naumann, B. Wilkes, J. Kischkewitz, J. Fluorine Chem. 1985, 30,
73–87.
[27] W. Meister, R. D. Guthrie, J. L. Maxwell, D. A. Jaeger, D. J. Cram, J.
Am. Chem. Soc. 1969, 91, 4452–4458.
[10] R. Schwesinger, H. Schlemper, C. Hasenfratz, J. Willaredt, T. Dam-
bacher, T. Breuer, C. Ottaway, M. Fletschinger, J. Boele, H. Fritz, D.
Putzas, H. W. Rotter, F. G. Bordwell, A. V. Satish, G.-Z. Ji, E.-M.
Peters, K. Peters, H. G. von Schnering, L. Walz, Liebigs Ann. 1996,
1055–1081.
[11] The chloride salt is commercially available from Fluka-Chemie AG
(Switzerland).
[12] The steadily developing chemistry of these compounds has been
summarized in recent review articles: a) N. Kuhn, A. Al-Sheikh,
Coord. Chem. Rev. 2005, 249, 829–857, and references therein;
b) F. E. Hahn, M. C. Jahnke, Angew. Chem. 2008, 120, 3166–3216;
Angew. Chem. Int. Ed. 2008, 47, 3122–3172.
[13] N. Kuhn, T. Kratz, Synthesis 1993, 561–562. For the synthesis of 5
from 4, we found it advantageous to use Na/K alloy in refluxing
glyme instead of K in THF as it provides higher conversions in
shorter reaction times.
[14] a) N. Kuhn, J. Fahl, R. Fawzi, C. Maichle-Moessmer, M. Steimann,
Z. Naturforsch. B 1998, 53, 720–726; b) N. Kuhn, A. Abu-Rayyan,
M. Gçhner, M. Steimann, Z. Anorg. Allg. Chem. 2002, 628, 1721–
1723.
[15] The chemistry and coordinating properties of the conjugated bases,
2-iminoimidazolines, have been reviewed: a) N. Kuhn, M. Gçhner,
M. Grathwohl, J. Wiethoff, G. Frenking, Y. Chen, Z. Anorg. Allg.
Chem. 2003, 629, 793–802. For original experimental results indicat-
ing high basicity and nucleophilicity of the exocyclic N atom, see:
b) N. Kuhn, R. Fawzi, M. Steimann, J. Wiethoff, D. Blꢆser, R. Boese,
Z. Naturforsch. B 1995, 50, 1779–1784; c) N. Kuhn, R. Fawzi, M.
Steimann, J. Wiethoff, Chem. Ber. 1996, 129, 479–482; d) N. Kuhn,
[28] M. Hernꢁndez-Rodrꢀguez, E. Juaristi, Tetrahedron 2007, 63, 7673–
7678.
[29] Spectral data of the intermediate 9 are in good agreement with
those reported in the literature: J. R. Goerlich, M. Farkens, A. Fisch-
er, P. G. Jones, R. Schmutzler, Z. Anorg. Allg. Chem. 1994, 620,
707–715.
Received: May 6, 2009
Published online: August 4, 2009
Chem. Eur. J. 2009, 15, 9477 – 9485
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9485