Convenient, scalable and flexible method for the preparation of imidazolium salts with previously inaccessible substitution patterns

Article abstract of DOI:10.1039/b604236h

A high yielding and modular approach to N,N′-disubstituted imidazolium salts is described, providing access to substitution patterns that are beyond the reach of established methodology. The Royal Society of Chemistry 2006.

Full text of DOI:10.1039/b604236h

View Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Convenient, scalable and flexible method for the preparation of
imidazolium salts with previously inaccessible substitution patterns{
Alois Fu¨rstner,* Manuel Alcarazo, Vincent Ce´sar and Christian W. Lehmann
Received (in Cambridge, UK) 22nd March 2006, Accepted 12th April 2006
First published as an Advance Article on the web 26th April 2006
DOI: 10.1039/b604236h
transformations work well with reactive halides R2X but are
difficult or even impossible with aryl halides,5 tert-alkyl halides,
A high yielding and modular approach to N,N9-disubstituted
imidazolium salts is described, providing access to substitution
and even many sec-alkyl halides, in particular if they are chiral and
non-racemic. To aid further (organometallic) catalyst design,6 a
more general method for the preparation of unsymmetrical
N,N9-disubstituted imidazolium salts is therefore highly desirable.
As part of our ongoing studies on the synthesis, structure and use
of metal–NHC complexes,7–10 we developed such a method that
opens access to a wide variety of substitution patterns that are
beyond reach otherwise. In addition to being highly modular, this
novel protocol is practical, scalable, and high yielding, allowing for
a convenient molecular editing of every substituent on the
imidazolium backbone and even of the escorting counterion.
As outlined in Scheme 1, this novel route to unsymmetrical
N,N9-disubstituted imidazolium salts is based on a heterocycle-
interconversion strategy. In a first attempt, a variant of the Zincke
reaction was considered,11 but the envisaged replacement of the
dinitrophenyl moiety of imidazolium salt 2 even with the fairly
nucleophilic phenethylamine 3 did not proceed (Scheme 2).
Encouraged by previous examples of the formation of N-aryl-
triazolium salts from 1,3,4-oxadiazolium precursors, though
mostly low yielding under the conditions reported in the
literature,12 we envisaged the use of oxazolinium salts F as more
reactive substrates for the same purpose. Such compounds turned
out to be easily accessible by a short sequence of readily scalable
operations (Scheme 3): first, the acid catalyzed reaction of a
(commercially available) a-hydroxyketone of type 4 with the amine
R1–NH2 of choice under azeotropic removal of water afforded the
a-aminoketones 5 which were N-formylated prior to treatment
with acetic anhydride in the presence of a strong mineral acid such
as aqueous HBF4 or HClO4. Surprisingly, the ensuing cyclization
stopped at the intermediate acetal stage 10 and did not lead to the
expected aromatization with formation of an oxazolium salt.13
Compounds 10 could be conveniently precipitated with Et2O and
processed without further purification. The constitution of these
synthetic intermediates was evident from their NMR data and was
unambiguously confirmed by the crystal structure analysis of one
representative case (Fig. 1).{ Access to the 4,5-unsubstituted series
was equally facile, involving N-alkylation of R1–NH2 with
patterns that are beyond the reach of established methodology.
Over the course of the last decade, N-heterocyclic carbenes
(NHC’s) have gained a prominent role in catalysis due to the
favorable properties that they impart as ancillary ligands on many
transition metal templates,1 and because of their ability to act as
powerful organocatalysts in their own right.2 Most studies focused
on N,N9-disubstituted imidazol-2-ylidenes A and their ‘saturated’
analogues B, which are usually prepared by deprotonation of the
corresponding imidazolium or imidazolinium salts C and D,
respectively (Scheme 1).
Although the tremendous success of these 2-electron donor
ligands has led to a huge number of structural variants, it is
surprising to find that several obvious and seemingly trivial
substitution patterns are as yet unknown. Most strikingly,
unsymmetrical imidazolium salts bearing two different aryl groups
on their N-atoms have not been described. Even the number of
unsymmetrical imidazolium salts with one N-aryl- and one
N-alkyl-group is still fairly small and essentially restricted to those
having primary N-alkyl substituents.3,4 These limitations stem from
the commonly practised synthesis routes, which rely on
N-substitutions of appropriate imidazole precursors E. Such
Scheme 1 Formation of NHC’s by deprotonation of (dihydro)imidazo-
lium salts and retrosynthetic analysis for the latter.
Max-Planck-Institut fu¨r Kohlenforschung, D-45470, Mu¨lheim/Ruhr,
Germany. E-mail: fuerstner@mpi-muelheim.mpg.de;
Fax: +49 208 306 2994; Tel: +49 208 306 2342
{ Electronic supplementary information (ESI) available: Experimental
procedures and spectroscopic data of all new compounds; crystal data for
the structures shown in Fig. 1–3. See DOI: 10.1039/b604236h
Scheme 2 Reagents and conditions: [a] 1-chloro-2,4-dinitrobenzene,
toluene, 50 uC; [b] phenethylamine 3, toluene, reflux.
2176 | Chem. Commun., 2006, 2176–2178
This journal is ß The Royal Society of Chemistry 2006

View Online
Fig. 2 Molecular structure of 11 (R1 = Ph, R2 = t-Bu, R4 = R5 = Me)
from single crystal X-ray structure determination. Anisotropic displace-
ment parameter ellipsoids are shown at 50% probability and hydrogen
atoms have been omitted. The tert-butyl group is disordered over two
positions at a ratio of 40 : 60.
Compound 11 was then suspended in toluene and reacted with
Ac2O and a catalytic amount of HX at 80 uC. After evaporation of
the solvent, the residue was triturated with Et2O and the resulting
suspension sonicated in a cleaning bath for ca. 20 min to give the
desired imidazolium salts 12 as solids that can be conveniently
filtered off. As can be seen from Table 1, the isolated yields
obtained over the entire sequence were good to excellent for a
variety of cases.
Scheme 3 Reagents and conditions: [a] R1NH2, HCl cat., toluene, Dean–
Stark; [b] MeC(O)OC(O)H; [c] Ac2O, HX (1.15 eq.); [d] R1NH2, nBuLi,
then bromide 7; [e] (i) HCOOH, Ac2O, THF, 0 uC; (ii) HCOOH; [f]
R2NH2, toluene; [g] Ac2O, HX cat., or: HX (1 eq.), cf. Table 1.
Most noticeable is the fact that the method accommodates
variously substituted anilines as well as secondary and tertiary
amines as the reaction partners, thus giving access to imidazolium
salts that are inaccessible by any of the established routes; the
structure of a representative example falling into this category is
depicted in Fig. 3.{ Even very weakly nucleophilic anilines as well
as sterically hindered amines still provide satisfactory results. We
therefore believe that this practical methodology should find
widespread use and invites applications to the preparation of
structurally novel types of carbene precursors. Investigations along
these lines together with applications of such ligands in catalysis
will be reported shortly.
Generous financial support by the MPG, the Fonds der
Chemischen Industrie, the Spanish Ministerio de Educacio´n y
Fig. 1 Molecular structure of 10 (R1 = 2,6-di-isopropylphenyl, R4 = R5
= H) from single crystal X-ray structure determination. Anisotropic
displacement parameter ellipsoids are shown at 50% probability and
hydrogen atoms have been omitted.
commercial bromoacetaldehyde diethylacetal 7 to give product 8
which was then processed analogously.
Gratifyingly, reaction of the oxazolinium adducts 10 with a
second amine of choice R2–NH2 proceeded smoothly, affording
the hydroxylated imidazolinium salts 11 that precipitated from the
reaction mixture in all cases investigated, thus making the work up
again very straightforward. The constitution of this novel type of
heterocycle featuring a labile hemiaminal motif was confirmed by
the crystal structure depicted in Fig. 2.{
Fig. 3 Molecular structure of 12 (R1 = mesityl, R2 = 2,6-di-isopropyl-
phenyl, R4 = R5 = H) from single crystal X-ray structure determination.
Anisotropic displacement parameter ellipsoids are shown at 50%
probability and hydrogen atoms have been omitted.
This journal is ß The Royal Society of Chemistry 2006
Chem. Commun., 2006, 2176–2178 | 2177

View Online
F. P. Gabba¨ı and G. Bertrand, Chem. Rev., 2000, 100, 39; (d) V. Ce´sar,
Table 1 Unsymmetrical N,N9-disubstituted imidazolium salts pre-
pared by the method depicted in Scheme 3
S. Bellemin-Laponnaz and L. H. Gade, Chem. Soc. Rev., 2004, 33, 619;
(e) W. Kirmse, Angew. Chem., Int. Ed., 2004, 43, 1767; (f) M. F. Lappert,
J. Organomet. Chem., 2005, 690, 5467; (g) C. M. Crudden and
D. P. Allen, Coord. Chem. Rev., 2004, 248, 2247; (h) E. Peris and
R. H. Crabtree, Coord. Chem. Rev., 2004, 248, 2239; (i) M. C.
Perry and K. Burgess, Tetrahedron: Asymmetry, 2003, 14, 951; (j)
L. Jafarpour and S. P. Nolan, Adv. Organomet. Chem., 2001, 46, 181; (k)
A. Fu¨rstner, Angew. Chem., Int. Ed., 2000, 39, 3012; (l) T. M. Trnka and
R. H. Grubbs, Acc. Chem. Res., 2001, 34, 18.
Entry
Product
Yielda
1
2
64% (X = BF4)
59% (X = ClO4)
3
4
75% (X = BF4)
91% (X = ClO4)b
2 (a) D. Enders and T. Balensiefer, Acc. Chem. Res., 2004, 37, 534; (b)
M. Christmann, Angew. Chem., Int. Ed., 2005, 44, 2632; (c) E. F. Connor,
G. W. Nyce, M. Myers, A. Mo¨ck and J. L. Hedrick, J. Am. Chem. Soc.,
2002, 124, 914.
3 Exceptions are known for special ligand architectures, cf.: (a) F. Glorius,
G. Altenhoff, R. Goddard and C. W. Lehmann, Chem. Commun., 2002,
2704; (b) M. Alcarazo, S. J. Roseblade, A. R. Crowley, R. Ferna´ndez,
J. M. Brown and J. M. Lassaletta, J. Am. Chem. Soc., 2005, 127, 3290;
(c) H. Seo, H. Park, B. Y. Kim, J. H. Lee, S. U. Son and Y. K. Chung,
Organometallics, 2003, 22, 618.
4 Unsymmetrical imidazolinium salts D and the saturated N-heterocyclic
carbenes B derived therefrom are more abundant; see the following for
leading references and literature cited therein: (a) J. J. van Veldhuizen,
J. E. Campbell, R. E. Guidici and A. H. Hoveyda, J. Am. Chem. Soc.,
2005, 127, 6877; (b) A. W. Waltman and R. H. Grubbs,
Organometallics, 2004, 23, 3105; (c) F. M. Rivas, U. Riaz, A. Giessert,
J. A. Smulik and S. T. Diver, Org. Lett., 2001, 3, 2673; (d) F. E. Hahn,
M. Paas, D. Le Van and T. Lu¨gger, Angew. Chem., Int. Ed., 2003, 42,
5243; (e) G. Xu and S. R. Gilbertson, Org. Lett., 2005, 7, 4605; (f)
H. Clavier, L. Coutable, L. Toupet, J.-C. Guillemin and M. Mauduit,
J. Organomet. Chem., 2005, 690, 5237; (g) E. Bappert and G. Helmchen,
Synlett, 2004, 1789.
5
6
64%
82%
7
91%
5 An interesting alternative for the N-arylation of imidazoles is the
reaction with aryne intermediates; however, this method is not suitable
for the preparation of unsymmetrical imidazolium salts bearing two
N-aryl groups, cf.: H. Yoshida, S. Sugiura and A. Kunai, Org. Lett.,
2002, 4, 2767.
6 (a) In this context it is worth noting that several transformations are
known in which catalysts with unsaturated carbene ligands A perform
significantly better than those with saturated NHC’s of type B; for
representative examples see ref. 10 and the following: J. W. Sprengers,
J. Wassenaar, N. D. Clement, K. J. Cavell and C. J. Elsevier, Angew.
Chem., Int. Ed., 2005, 44, 2026; (b) in contrast to the prevailing
assumption, recent experimental data indicate that unsaturated NHC’s
might be even stronger electron donors than their saturated counter-
parts, see: R. Dorta, E. D. Stevens, N. M. Scott, C. Costabile,
L. Cavallo, C. D. Hoff and S. P. Nolan, J. Am. Chem. Soc., 2005, 127,
2485.
8
9
84%c
31%
10
11
59%
88%
7 (a) D. Kremzow, G. Seidel, C. W. Lehmann and A. Fu¨rstner, Chem.–
Eur. J., 2005, 11, 1833; (b) A. Fu¨rstner, G. Seidel, D. Kremzow and
C. W. Lehmann, Organometallics, 2003, 22, 907.
8 (a) A. Fu¨rstner, H. Krause, L. Ackermann and C. W. Lehmann, Chem.
Commun., 2001, 2240; (b) A. Fu¨rstner, L. Ackermann, B. Gabor,
R. Goddard, C. W. Lehmann, R. Mynott, F. Stelzer and O. R. Thiel,
Chem.–Eur. J., 2001, 7, 3236; (c) S. Pru¨hs, C. W. Lehmann and
A. Fu¨rstner, Organometallics, 2004, 23, 280; (d) L. Ackermann,
A. Fu¨rstner, T. Weskamp, F. J. Kohl and W. A. Herrmann,
Tetrahedron Lett., 1999, 40, 4787; (e) A. Fu¨rstner, O. R. Thiel,
L. Ackermann, H.-J. Schanz and S. P. Nolan, J. Org. Chem., 2000, 65,
2204; (f) A. Fu¨rstner, O. R. Thiel and C. W. Lehmann, Organometallics,
2002, 21, 331; (g) see also: A. Fu¨rstner, H. Krause and C. W. Lehmann,
Chem. Commun., 2001, 2372.
a
Refers to the overall yield of the reaction sequence shown in
Scheme 3 from compound 6 (or 9) to the final product. Using
b
mesitylamine as R2NH2; 73% yield using 2,6-di-isopropyl-
phenylamine as R2NH2. In racemic form.
c
9 (a) A. Fu¨rstner and A. Leitner, Synlett, 2001, 290; (b) A. Fu¨rstner and
G. Seidel, Org. Lett., 2002, 4, 541.
Ciencia (fellowship to M. A.), and the Alexander-von-Humboldt
Foundation (fellowship to V. C.) is gratefully acknowledged.
10 A. Fu¨rstner and H. Krause, Adv. Synth. Catal., 2001, 343, 343.
11 W. C. Cheng and M. J. Kurth, Org. Prep. Proced. Int., 2002, 34, 585.
12 (a) G. V. Boyd and A. J. H. Summers, J. Chem. Soc. C, 1971, 409; (b)
J. H. Teles, J.-P. Melder, K. Ebel, R. Schneider, E. Gehrer, W. Harder,
S. Brode, D. Enders, K. Breuer and G. Raabe, Helv. Chim. Acta, 1996,
79, 61.
13 Formation of oxazolium salts from a-aminoketones with triethyl
orthoformate was claimed in a patent, cf.: DD 254 0902 A1. In our
hands, however, this procedure is less clean and hence inferior to the
route via the formamide described herein.
Notes and references
{ CCDC 602504–602506. For crystallographic data in CIF or other
electronic format see DOI: 10.1039/b604236h
1 (a) A. J. Arduengo, Acc. Chem. Res., 1999, 32, 913; (b) W. A. Herrmann,
Angew. Chem., Int. Ed., 2002, 41, 1290; (c) D. Bourissou, O. Guerret,
2178 | Chem. Commun., 2006, 2176–2178
This journal is ß The Royal Society of Chemistry 2006