Unexpected oxidative C-C cleavage in the metallation of 2-substituted imidazolium salts to give N-heterocyclic carbene complexes

Article abstract of DOI:10.1039/b409672j

Imidazolium salts blocked at C2 with methyl or benzyl groups unexpectedly react with silver oxide to give N-heterocyclic carbene complexes of silver via an oxidative carbon-carbon bond cleavage.

Full text of DOI:10.1039/b409672j

View Article Online / Journal Homepage / Table of Contents for this issue
Unexpected oxidative C–C cleavage in the metallation of 2-substituted
imidazolium salts to give N-heterocyclic carbene complexes{
Anthony R. Chianese, Brian M. Zeglis and Robert H. Crabtree*
Chemistry Department, Yale University, PO Box 208107, New Haven, Connecticut 06520-8107, USA
Received (in Cambridge, UK) 25th June 2004, Accepted 28th July 2004
First published as an Advance Article on the web 25th August 2004
Imidazolium salts blocked at C2 with methyl or benzyl
the silver–carbene complex 5 in ca. 2 d. A redox process is indicated
by the formation of a silver mirror on the wall of the reaction vessel.
In contrast, a silver mirror is not observed in normal metallation
groups unexpectedly react with silver oxide to give
N-heterocyclic carbene complexes of silver via an oxidative
carbon–carbon bond cleavage.
1
of 2H-imidazolium salts under the same conditions. H NMR
spectroscopy of the reaction mixture from 4a shows only a 6H
singlet at 3.66 ppm for the N-methyl groups, and a 6H singlet at
2.14 ppm for the C-methyl groups. ES-MS in cation mode shows
two peaks of equal intensity at m/z ~ 355 and 357 amu, ascribed to
Ag(NHC)21, arising from 107Ag and 109Ag. A mixture of anions is
originally present, but the complex can be isolated in analytically
pure form as the PF6 salt, by washing the crude product with aq.
NaOH/KPF6. The carbene carbon resonates as a singlet at
177.6 ppm in the 13C NMR spectrum. The lack of coupling to
107Ag and 109Ag is indicative of the usual3 fast exchange of the
NHC between silver atoms on the NMR timescale.
N-Heterocyclic carbenes (NHCs), first isolated by Arduengo,1 have
emerged as an extremely useful class of spectator ligands
for homogeneous catalysis.2 Several methods have now been
developed that avoid the free ligand by in situ metallation of the
imidazolium salts. One such method is the generation of silver–
NHC complexes from silver oxide and imidazolium salts [eqn. (1)].3
The resulting silver complexes are capable of transmetallation to
give NHC complexes of Pd, Au, Rh, Ir, and Cu.4
(1)
In contrast to the normal binding mode via C2, we5 and others6
have shown that abnormal binding of an NHC through C4(5) is
also possible. We are studying the factors controlling the mode of
binding,7 and in the electronic properties of abnormal NHCs versus
normal NHCs.8 To selectively prepare abnormal NHC complexes,
we have blocked the imidazolium ligand precursors by substitution
at C2 with alkyl or aryl groups. We previously reported the
synthesis of an abnormal NHC–iridium(I) complex, using a
2-phenylimidazolium salt.8 In attempting to prepare a rhodium(I)
complex from the tetramethylimidazolium salt 1 by transmetalla-
tion from the silver complex, we were very surprised to isolate by
column chromatography the normal carbene complex 2 in 25%
yield [eqn. (2), cod ~ 1,5-cyclooctadiene). The formation of 2
requires a C–C bond cleavage. The expected abnormal NHC
complex 3 was not isolated but a 1/Ag2O mixture showed
unidentified compounds (1H NMR) after 4 h.
(3)
The series 4b–d was prepared with different blocking groups at
C2. When treated with Ag2O, the benzyl-substituted 4b converted
nearly quantitatively to 5 in one day [eqn. (3)]. Ethyl-substituted 4c
reacted incompletely, even with a large excess of silver oxide and
prolonged reaction time. Isopropyl-substituted 4d did not react at
all, even at reflux temperature. Our previous results8 indicate that
2-phenylimidazolium salts do not undergo C–C cleavage to give
normal silver–NHC complexes upon reaction with silver oxide
under the same conditions. This implies that PhCH2, CH3, and
CH3CH2 are cleaved and are therefore ineffective blocking groups,
but that iPr and Ph are not cleaved and are therefore effective
blocking groups.
The fate of the methyl group (4a) and benzyl group (4b) gives an
important mechanistic clue. With 4b, the only byproduct observed
1
by H NMR is benzoate,10 in about 84% yield. The analogous
byproduct from 4a, formate, is not observed under the original
reaction conditions. However, when the reaction is run with 1 equiv.
of NBu4PF6, we see ca. 54% yield of formate.10 AgOOCH may
precipitate in the absence of the salt.
(2)
Carboxylate formation implies the 2-substituent is oxidized, but
the reaction goes under Ar in dry, degassed THF; ruling out both
air and CH2Cl2 as oxidant. The Ag(0) was isolated by washing the
solid residue with CH2Cl2, then dil. HNO3 and then aq. NH3 to
remove Ag–NHCs, Ag2O and Ag salts. Approximately 3.86
equivalents of metallic silver are isolated in a typical reaction,
relative to the amount of imidazolium salt used.11 This is consistent
with four electrons being required per imidazolium unit (see below).
Initial RCHO or RCH2OH intermediates were not later oxidized
to RCOOH by Ag2O12 because PhCHO and PhCH2OH are
To avoid the competitive formation of abnormal silver–NHC
complexes, we prepared 4a,9 blocked by methyl groups at all
positions [eqn. (3)]. The reaction of 4a with 4 equiv. Ag2O in
CH2Cl2 at room temperature gives nearly quantitative formation of
{ Electronic supplementary information (ESI) available: experimental
details. See http://www.rsc.org/suppdata/cc/b4/b409672j/
2 1 7 6
C h e m . C o m m u n . , 2 0 0 4 , 2 1 7 6 – 2 1 7 7
T h i s j o u r n a l i s ß T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 4

View Article Online
unchanged under our conditions. These observations strongly
suggest that in the reaction of 4b and silver oxide, benzoate is
produced directly as a product of C–C cleavage, and not by a
subsequent oxidation of benzyl alcohol or benzaldehyde.
the reactivity pattern for C–C cleavage is completely different from
the one we find here.
2-Me-, 2-Et-, and 2-PhCH2- can be unreliable blocking groups
for imidazolium/Ag2O reactions while 2-iPr- and 2-Ph- resist
oxidative C–C cleavage. This is relevant in abnormal NHC
synthesis, and in work with imidazolium-based ionic liquids, often
blocked by a C2 Me.
We gratefully acknowledge British Petroleum, Johnson Matthey,
and the U.S. Department of Energy for funding. We also thank
Johnson Matthey for a generous gift of rhodium chloride.
The C–C cleavage probably goes via an initial four-electron
oxidation of the imidazolium 2-substituent to give a 2-formyl (4a)
or 2-benzoyl (4b) imidazolium ion, followed by hydrolytic cleavage
to give the NHC, trapped by silver(I), and the carboxylic acid13
[eqn. (4)]. A variety of oxidants14–16 can convert 2-alkyl imidazoles
to the acyl compounds. The proposed intermediate 2-acylimidazo-
lium salts have previously been shown to hydrolyze to give
carboxylic acids and 2H-imidazolium salts.17 This reaction is
analogous to other substitutions of RCOX derivatives, such as
hydrolysis of an acid chloride or anhydride. In our case, the NHC is
the leaving group, and can be intercepted by silver(I). The complete
mechanism [eqn. (4)] requires four equivalents of silver for oxida-
tion, and one equivalent for complex formation, consistent with
both the measured quantity of silver(0) formed, and the need for a
large excess of silver oxide for the reaction to go to completion.
Notes and references
1 A. J. Arduengo, R. L. Harlow and M. Kline, J. Am. Chem. Soc., 1991,
113, 361–363.
2 (a) W. A. Herrmann and C. Kocher, Angew. Chem., Int. Ed., 1997, 36,
2163–2187; (b) W. A. Herrmann, Angew. Chem., Int. Ed., 2002, 41, 1290–
1309; (c) M. C. Perry and K. Burgess, Tetrahedron: Asymmetry, 2003, 14,
951–961.
3 H. M. J. Wang and I. J. B. Lin, Organometallics, 1998, 17, 972–975.
4 (a) A. M. Magill, D. S. McGuinness, K. J. Cavell, G. J. P. Britovsek,
V. C. Gibson, A. J. P. White, D. J. Williams, A. H. White and
B. W. Skelton, J. Organomet. Chem., 2001, 617, 546–560;
(b) A. R. Chianese, X. W. Li, M. C. Janzen, J. W. Faller and
R. H. Crabtree, Organometallics, 2003, 22, 1663–1667; (c) K. S. Coleman,
H. T. Chamberlayne, S. Turberville, M. L. H. Green and A. R. Cowley,
Dalton Trans., 2003, 2917–2922; (d) R. S. Simons, P. Custer, C. A. Tessier
and W. J. Youngs, Organometallics, 2003, 22, 1979–1982; (e) X. L. Hu,
I. Castro-Rodriguez, K. Olsen and K. Meyer, Organometallics, 2004, 23,
755–764.
(4)
5 S. Gru¨ndemann, A. Kovacevic, M. Albrecht, J. W. Faller and
R. H. Crabtree, Chem. Commun., 2001, 2274–2275.
6 (a) X. Hu, I. Castro-Rodriguez and K. Meyer, Organometallics, 2003, 22,
3016–3018; (b) A. A. Danopoulos, N. Tsoureas, J. A. Wright and
M. E. Light, Organometallics, 2004, 23, 166–168; (c) H. Lebel,
M. K. Janes, A. B. Charette and S. P. Nolan, J. Am. Chem. Soc.,
2004, 126, 5046–5047.
7 (a) S. Gru¨ndemann, A. Kovacevic, M. Albrecht, J. W. Faller and
R. H. Crabtree, J. Am. Chem. Soc., 2002, 124, 10473–10481;
(b) A. Kovacevic, S. Gru¨ndemann, J. R. Miecznikowski, E. Clot,
O. Eisenstein and T. H. Crabtree, Chem. Commun., 2002, 2580–2581.
8 A. R. Chianese, A. Kovacevic, B. M. Zeglis, J. W. Faller and
R. H. Crabtree, Organometallics, 2004, 23, 2461–2468.
To investigate the possible intermediacy of 2-acylimidazolium
salts, we prepared the 2-benzoylimidazolium salt 6. During the
preparation of 6, we observed a small amount (4%) of hydrolysis to
give the 2H-imidazolium salt. When 6 is reacted with silver oxide
under the standard conditions [eqn. (5)], quantitative conversion to
the silver–NHC complex and benzoate ion is observed by 1H
NMR. No metallic silver is produced, as no oxidation is necessary
here.
9 N. Kuhn, G. Henkel and J. Kreutzberg, Z. Naturforsch., B: Chem. Sci.,
1991, 46, 1706–1712.
10 Identification of RCO22 by NMR (comparison with authentic RCO22
and ES-MS (R ~ Ph).
)
11 Average value for two experiments.
12 (a) F. Asinger, Chem. Ber., 1942, 75, 656; (b) H. Fiesselmann, Chem.
Ber., 1942, 75, 881.
13 RCO2Ag, from RCO22/Ag2O, may precipitate when R ~ Ph.
14 W. Kuzmierkiewicz, Acta Pol. Pharm., 1986, 43, 221–226.
15 S. Ram, D. S. Wise and L. B. Townsend, J. Heterocycl. Chem., 1986, 23,
1109–1113.
(5)
We18 and others19 have demonstrated that 2H-imidazolium salts
can oxidatively add to palladium(0). Welton has demonstrated the
non-innocence of 2H-imidazolium-based ionic liquids in chemistry
involving Pd(0).20 The authors also demonstrated that 1,3-dialkyl-
2-phenylimidazolium cation could serve as a source of the aryl
group for the Suzuki reaction, competing with the aryl halide
cosubstrate for coupling with the boronic acid.20 1,2,3-Trialkyl-
imidazolium salts, in contrast, did not give Pd–NHC complexes. So
16 H. Berner and H. Reinshagen, Monatsh. Chem., 1975, 106, 1059–1069.
17 S. Ohta, S. Hayakawa, H. Moriwaki, S. Tsuboi and M. Okamoto,
Heterocycles, 1985, 23, 1759–1764.
18 S. Gru¨ndemann, M. Albrecht, A. Kovacevic, J. W. Faller and
R. H. Crabtree, Dalton Trans., 2002, 2163–2167.
19 D. S. McGuinness, K. J. Cavell, B. F. Yates, B. W. Skelton and
A. H. White, J. Am. Chem. Soc., 2001, 123, 8317–8328.
20 F. McLachlan, C. J. Mathews, P. J. Smith and T. Welton,
Organometallics, 2003, 22, 5350–5357.
C h e m . C o m m u n . , 2 0 0 4 , 2 1 7 6 – 2 1 7 7
2 1 7 7