An air stable carbene and mixed carbene 'dimers'

Article abstract of DOI:10.1021/ja973241o

The air-stable crystalline carbene 1,3-dimesityl-4,5-dichloroimidazol- 2-ylidene (2) is produced from the chlorination of 1,3-dimesityimidazol-2- ylidene (1) by 2 equiv of carbon tetrachloride. The physical and spectroscopic properties of the carbene are

Full text of DOI:10.1021/ja973241o

12742
J. Am. Chem. Soc. 1997, 119, 12742-12749
An Air Stable Carbene and Mixed Carbene “Dimers”
Anthony J. Arduengo, III,* Frederic Davidson, H. V. R. Dias,1 Jens R. Goerlich,2
Dilip Khasnis,3 William J. Marshall, and T. K. Prakasha
Contribution No. 7604 from the DuPont Science and Engineering Laboratory,
Experimental Station, Wilmington, Delaware 19880-0328
ReceiVed September 15, 1997X
Abstract: The air-stable crystalline carbene 1,3-dimesityl-4,5-dichloroimidazol-2-ylidene (2) is produced from the
chlorination of 1,3-dimesityimidazol-2-ylidene (1) by 2 equiv of carbon tetrachloride. The physical and spectroscopic
properties of the carbene are reported. The features which contribute to the exceptional stability of the carbene are
discussed. Further reaction of 2 with carbon tetrachloride leads to reduction of CCl4 to dichlorocarbene. The formation
of the mixed carbene “dimers” (olefins) from in situ generated dichlorocarbene and various imidazol(in)-2-ylidenes
is also reported. The tellurones derived from 1 and 2 are synthesized and compared.
Introduction
essential to produce a stable carbene in this series. Very
recently, we were able to synthesize a stable thiazol-2-ylidene,
demonstrating that substituents other than nitrogen can be
tolerated on the carbene center.17 In this paper we now show
that the electronic features of a σ-electronegativity effect and
π-interactions can be used in concert to further enhance the
stability of the imidazol-2-ylidenes.18
We have previously reported the synthesis and isolation of a
series of stable carbenes, imidazol-2-ylidenes.4,5 These carbenes
are stable at room temperature under a nitrogen atmosphere and
can be kept under these conditions indefinitely. They are
however quite sensitive to moist air. The source of the
remarkable stability of these carbenes has been the focus of
several previous theoretical studies.6-14 The large singlet-
triplet gap in imidazol-2-ylidenes (∼80 kcal/mol), along with
π-interactions in the imidazole ring, electronegativity effects
from the nitrogens, steric effects, and kinetic factors, have all
been offered as explanations of the enhanced stability of these
carbenes. Although all of these effects undoubtedly contribute
somewhat to the stability of these carbenes, it is difficult to
assign one dominant feature that is primarily responsible for
their stability. The role of steric effects seems to be largely
unimportant because a number of sterically unencumbered
imidazol-2-ylidenes have now been prepared. Kinetic factors
are important, but their origin lies primilarily with entropy
effects that tend to favor the monomeric imidazol-2-ylidene over
the corresponding dimer (olefin). A stable saturated imidazolin-
2-ylidene carbene15 and even an acyclic diaminocarbene16 have
been reported which shows that the C4-C5 double bond is not
Results and Discussion
The carbene 1,3-dimesitylimidazol-2-ylidene4b (1) reacts
rapidly with carbon tetrachloride in tetrahydrofuran (thf) at room
temperature to produce 1,3-dimesityl-4,5-dichloroimidazol-2-
ylidene (2) and chloroform (eq 1). This reaction can be easily
1
followed by H NMR where the disappearance of 1 and the
X Abstract published in AdVance ACS Abstracts, December 15, 1997.
(1) DuPont Visiting Research Scientist 1990-2.
formation of 2 and chloroform is observed. The reaction is
complete in about 20 min at room temperature. The enhanced
stability of 2 over 1 is immediately evident since the carbene 1
is such a strong base that it cannot tolerate prolonged exposure
to such acidic solvents as chloroform while 2 is easily isolated
from a reaction mixture that contains chloroform.
Remarkably carbene 2 is even stable to air for limited
exposure times. A solid sample of 2 that was left in a laboratory
hood exposed to air for two days showed no reaction or
decomposition. A benzene solution of 2 that was also exposed
to air overnight showed (by NMR) no evidence for decomposi-
tion. Pure 2 is a slightly off-white solid that melts at 180-2
°C with decomposition. The 1H NMR spectrum of 2 in
benzene-d6 shows resonances at δ 2.09, 2.12, and 6.75 for the
p- and o-methyls and the aromatic ring proton of the mesityl
groups, respectively. The 13C NMR spectrum in benzene-d6
shows a resonance at δ 219.9 for the carbene center. This
carbene resonance is typical of imidazol-2-ylidenes and is very
(2) DuPont Visiting Research Scientist 1995-7.
(3) DuPont Visiting Research Scientist 1994. Deceased.
(4) (a) Arduengo, A. J., III; Harlow, R. L.; Kline, M. J. Am. Chem. Soc.
1991, 113, 361. (b) Arduengo, A. J., III; Dias, H. V. R.; Harlow, R. L.;
Kline, M. J. Am. Chem. Soc. 1992, 114, 5530.
(5) Arduengo, A. J., III; Krafczyk, R. Chem. Unserer Zeit., in press.
(6) Dixon, D. A.; Arduengo, A. J., III. J. Phys. Chem. 1991, 95, 4180.
(7) Cioslowski, J. Int. J. Quantum Chem., Quantum Chem Symp. 1993,
27, 309.
(8) Arduengo, A. J., III; Dias, H. V. R.; Dixon, D. A.; Harlow, R. L.;
Klooster, W. T.; Koetzle, T. F. J. Am. Chem. Soc. 1994, 116, 6812.
(9) Arduengo, A. J., III; Dixon, D. A.; Kumashiro, K. K.; Lee, C.; Power,
W. P.; Zilm, K. W. J. Am. Chem. Soc. 1994, 116, 6361.
(10) Arduengo, A. J., III; Bock, H.; Chen, H.; Dixon, D. A.; Green, J.
C.; Herrmann, W. A.; Jones, N. L.; Wagner, M.; West, R. J. Am. Chem.
Soc. 1994, 116, 6641.
(11) Heinemann, C.; Thiel, W. Chem. Phys. Lett. 1994, 217, 11.
(12) Sauers, R. R. Tetrahedron Lett. 1996, 37, 149.
(13) Heinemann, A. E.; Mu¨ller, T.; Apeloig, Y.; Schwarz, H. J. Am.
Chem. Soc. 1996, 118, 2023.
(14) Boehme, C.; Frenking, G. J. Am. Chem. Soc. 1996, 118, 2039.
(15) Arduengo, A. J., III; Goerlich, J. R.; Marshall, W. J. J. Am. Chem.
Soc. 1995, 117, 11027.
(16) Alder, R. W.; Allen, P. R.; Murray, M.; Orpen, A. G. Angew. Chem.
1996, 108, 1211.
(17) Arduengo, A. J., III; Goerlich, J. R.; Marshall, W. J. Liebigs Ann.
1997, 365.
(18) Arduengo, A. J.; Goerlich, J. R.; Khasnis, D. Appl. U.S. Patent Off.
S0002-7863(97)03241-1 CCC: $14.00 © 1997 American Chemical Society

An Air Stable Carbene and Mixed Carbene “Dimers”
J. Am. Chem. Soc., Vol. 119, No. 52, 1997 12743
similar to the chemical shift of the carbene center of 1 (δ 219.7
in thf-d8).4b
Crystals of 2 that were suitable for X-ray diffraction studies
were grown by cooling a hexane solution. Carbene 2 crystal-
lizes in the monoclinic space group C2/c. Molecules of 2 are
positioned such that a 2-fold symmetry axis passes through the
carbene center and bisects the C4-C5 bond of the imidazole
ring. This arrangement is similar to that observed for 1,3,4,5-
tetramethylimidazol-2-ylidene.4b,8 Selected bond lengths and
angles for 2 and related structures are presented in Table 1.
The solid state structure is depicted by the KANVAS drawing
in Figure 1.19 The imidazole ring in 2 is nearly planar, and
only C4 and C5 deviate above and below the average plane of
the imidazole ring by 0.1 pm. The nitrogen centers are almost
planar. The ipso-carbons of the mesityl substituents deviate
only 6.4 pm from the plane of the imidazole ring. The chlorine
atoms lie almost in the plane of the imidazole ring and deviate
by only 3.8 pm. The valence angle at the carbene center of 2
is 101.9° which is characteristic of singlet carbenes.4,6 The
geometry of the central imidazole ring of 2 is virtually identical
to that of carbene 1 from which it is derived.4b As a result
there are no structural differences between 1 and 2 that betoken
the large difference in stability (reactivity) that is observed for
these carbenes. Imidazolium salts are our usual precursors to
imidazol-2-ylidene, and we have noted that there are charac-
teristic changes between the imidazolium (carbenium) ions and
their corresponding carbenes.4,15,17 The carbenium ion (2‚HCl)
related to 2 can be prepared by the addition of the elements of
HCl to 2. The structural relationship between 2 and 2‚HCl is
similar to those that we have previously observed with other
imidazol-2-ylidenes and their corresponding carbenium ions (cf.
the increase of the C2-N1(3) bond lengths and decrease of the
N-C-N angle on going to the carbene, Table 1).
We believe that the chlorination of 1 by CCl4 to produce 2
proceeds according to Scheme 1. The process begins by the
chlorination of the 2-position of the imidazole with the liberation
of the trichloromethyl anion. The Cl3C- then reacts with the
2-chloro-1,3-dimesitylimidazolium ion by deprotonation at the
4-position to produce chloroform and 2-chloro-1,3-dimesityl-
imidazolium-4-ate. The imidazolium-4-ate then reacts with
additional CCl4 to chlorinate the 4-position and produce another
trichloromethyl anion. This deprotonation/chlorination process
is repeated for the 5-position of the imidazole ring to produce
the 1,3-dimesityl-2,4,5-trichloroimidazolium ion. The 2-position
of this trichloroimidazolium ion is finally dechlorinated by
reaction with 1 or Cl3C- as illustrated to produce carbene 2.
1
When this chlorination reaction is followed by H NMR, only
the initial starting materials and final products are observed
(none of the intermediates are detected). The inability to
observe intermediates by NMR suggests that each successive
step is faster than the preceding step. This accelerated reaction
sequence is consistent with the strong σ-inductive effect of the
chlorines and their acidifing effect on the imidazole ring protons.
A different reaction can be observed when 2 equiv of carbene
2 react with CCl4 (eq 2). This reaction probably initiates along
a pathway analogous to that shown in Scheme 1 in which “Cl+”
is transferred to the 2-position of one molecule of carbene 2 to
form a trichloroimidazolium ion with the liberation of the
trichloromethyl anion (Cl3C-). Reactions of the trichloroimi-
(19) This drawing was made with the KANVAS computer graphics
program. This program is based on the program SCHAKAL of E. Keller
(Kristallographisches Institut der Universita¨t Freiburg, Germany), which
was modified by A. J. Arduengo, III (E. I. du Pont de Nemours & Co.,
Wilmington, DE) to produce the back and shadowed planes. The planes
bear a 50-pm grid, and the lighting source is at infinity so that shadow size
is meaningful.

12744 J. Am. Chem. Soc., Vol. 119, No. 52, 1997
Arduengo et al.
Figure 1. Space filling KANVAS19 drawing of the X-ray structure of
2.
Figure 2. Space-filling KANVAS19 drawing of the X-ray structure of
4.
Scheme 1
of these strongly polarized olefins. Kuhn et al. have reported
a chemical shift of δ 40.2 for the exocyclic methylene of 5
which correlates well with calculated electron densities in other
polarized olefins.22
Crystals of 4 were grown by cooling a hexane solution at
-25 °C. The X-ray structure of 4 is depicted in Figure 2.19
The two CdC double bonds in 4 bear the same heteroatoms
(two chlorines and two nitrogens with different substitution
patterns), and an interesting comparison can be made. The
length of the newly formed CdC double bond is 135.3 pm.
The C4dC5 double bond in the imidazole ring is 130.9 pm.
The exocyclic double bond is twisted (6.9°) slightly more than
the endocyclic double bond (2.4°). This difference is probably
the result of electronic factors that polarize the exocyclic double
bond, in addition to any steric factors. The nitrogens in the
imidazole ring of 4 are noticeably pyramidal with anti-
orientations about the imidazole ring. The two nitrogens are
15 pm (N1) and 23 pm (N3) out of the plane of their three
attached substituents. This anti-pyramidalization is evident in
the shadow of the drawing in Figure 2. The C2-N1(3) bond
distances in 4 are on average about 4.1 pm longer than in the
carbene 2. The C4(5)-N3(1) distances are about 3.2 pm longer
in 4 compared to 2. These small increases could be expected
to arise from a combination of decreased π-interaction in the
imidazole ring and increased p-orbital contribution to the
σ-bonds arising from the pyramidalization of the nitrogens (more
sp3-like in 4). However, the very small change in the N-mesityl
bond distances between 4 and 2 (actually 0.4 pm shorter in 4
than 2) suggests that the contribution from nitrogen rehybrid-
ization to the σ-bond length changes is negligible.
dazolium cation of 3 with Cl3C- are in essence degenerate and
lead directly (or indirectly) to regeneration of 2 and CCl4.
However, after sufficient reaction time, another pathway involv-
ing the R-elimination of Cl- from Cl3C- to form dichlorocar-
bene (:CCl2) becomes important. Under the reaction conditions,
dichlorocarbene is captured by a second equivalent of carbene
2 to produce the mixed carbene “dimer” 4. An analogous
process has been reported to couple in situ generated dichlo-
rocarbene with triphenylphosphine to produce (dichlorometh-
ylene)triphenylphosphorane.20
The mixed “dimer” 4 can be easily separated from the
byproduct trichloroimidazolium chloride salt (3) by extraction
into benzene. Solid 4 is a slightly off-white solid that turns
pink at 147 °C and melts at 172 °C. The 1H NMR spectrum of
4 does not provide much structural information except to indicate
that the adduct is different from the carbene (2) from which it
is derived. The 13C NMR does indicate the loss of the carbene
resonance at δ 219.9 and the appearance of two new resonances
at δ 70.9 and 140.3 for the new olefinic centers derived,
respectively, from :CCl2 and 2. The olefinic resonance at δ
70.9 is an unusually high field position for the CCl2 grouping
(cf. Cl2CdCCl2, δ(13C) ) 120.521) but is actually characteristic
The byproduct from the reduction of CCl4 by 2 is the
trichloroimidazolium chloride 3. The salt 3 is a high melting
solid (>250 °C). It is soluble in polar solvents such as dimethyl
sulfoxide and acetonitrile. This trichloroimidazolium ion shows
a solid state structure (Table 1) that is typical of imidazolium
(21) Sadtler. Standard 13C NMR Spectra; SADTLER Research Labo-
ratories, Inc.: Philiadelphia, 1977; Vol. 10.
(22) Kuhn, N.; Bohnen, H.; Kreutzberg, J.; Bla¨ser, D.; Boese, R. J. Chem.
Soc., Chem. Commun. 1993, 1993, 1136.
(20) Speziale, A. J.; Ratts, K. W. J. Am. Chem. Soc. 1962, 84, 854.

An Air Stable Carbene and Mixed Carbene “Dimers”
J. Am. Chem. Soc., Vol. 119, No. 52, 1997 12745
expected increase in length arising from a decreased π-interac-
tion in the imidazoline ring for 7 relative to 6 seems rather small
and supports a somewhat limited π-delocalization in 6 and/or a
highly polarized olefin structure for 7.
We decided to pursue the syntheses of the monochloro and
nonchlorinated analogs of 7 for further comparisons. These
compounds would be difficult to prepare by a carbene-carbene
coupling route owing to the relatively high and unselective
reactivity of :CHCl and :CH2. However, alkylation/deproto-
nation routes as illustrated in eqs 4 and 5 proved to be useful
pathways to the same olefins.
Figure 3. Space-filling KANVAS19 drawing of the X-ray structure of
7.
ions (e.g., 2‚HCl). The imidazolium ring in 3 exhibits the
characteristically short C2-N1(3) bond distances (133.5 pm) and
wide N1-C2-N3 angle (109.9°). There is also a close approach
of the chloride ion to the chlorine atom in the 2-position of the
imidazole ring (rCl-Cl ≈ 315 pm, θC-Cl-Cl ≈ 175°). More
details of this structure will be discussed in a summary of
carbene-halogen adducts to be published elsewhere.
The mixed carbene coupling of 2 and :CCl2 that leads to the
olefin 4 reveals that some of the chemistry expected for a
carbene is indeed exhibited by the carbene 2 even though it
possesses extraordinary stability. It is useful to compare this
type of carbene coupling reaction with those of other nucleo-
philic carbenes.
The chlorination of the 4- and 5-positions of carbene 1 with
carbon tetrachloride is dependent upon the mild acidity of these
olefinic centers as suggested by Scheme 1. If the C4dC5 double
bond in 1 is removed by saturation, the resulting imidazolin-
2-ylidene (6) still reduces carbon tetrachloride to form dichlo-
rocarbene which couples with the carbene center of 6 to produce
olefin 7 (eq 3).
1
As was the case with 4, the H NMR of 7 does not provide
any significant structural information except to indicate that 7
is different from the starting material (6). The 13C NMR of 7
reveals the nature of this product. The original carbene
resonance of δ 244.5 in 6 is shifted to 145.8 in 7 for the new
olefin carbon, and the CCl2 olefinic center exhibits a resonance
at δ 74.0. This shift pattern is very similar to that observed for
4 (Vide supra).
Surprisingly the reaction of imidazolin-2-ylidene 6 with
methylene chloride was very sluggish at room temperature, and
after seven days no conversion was evident. This is in contrast
to carbene 1 which reacts rapidly with CH2Cl2. However, when
a mixture of 6 and CH2Cl2 in hexane was heated in a sealed
tube at 70 °C for 15 h, conversion to the monochloroolefin 9
and the imidazolinium salt 10 was accomplished. The carbene
6 is sufficiently basic that it can deprotonate the intermediate
2-(chloromethyl)imidazolinium chloride to yield the olefin 9.
The byproduct 10 was identified by comparison with an
authentic sample.15
The olefin 9 proved to be rather unstable and difficult to
handle such that we have not grown satisfactory crystals for
X-ray structure determination. However, NMR studies reveal
the character and structure of the compound. The unique
olefinic proton in 9 shows a 1H NMR resonance at δ 3.81. This
shift is unusually high field for an olefinic center but is
The solid state structure of 7 is depicted in Figure 3.19 The
molecule sits along a crystallographically imposed 2-fold
rotation axis that runs along the CdC double bond. The CdC
double bond in 7 is 133.7 pm long which is slightly shorter
than the corresponding bond in 4. This exocyclic double bond
in 7 is twisted 9.5° (slightly greater than the twist in 4). The
nitrogens in the imidazole ring of 7 are distinctly pyramidal. In
7 the two nitrogens are 18.9 pm out of the plane of their three
attached substituents with an anti-orientation about the imidazole
ring. The C2-N1(3) bond distances in 7 are on average only
about 3.9 pm longer than those in the carbene 6. Thus, the

12746 J. Am. Chem. Soc., Vol. 119, No. 52, 1997
Arduengo et al.
Figure 5. Space-filling KANVAS19 drawing of the X-ray structure of
13.
Figure 4. Space-filling KANVAS19 drawing of the X-ray structure of
11.
We first reported the reaction of an in situ generated
nucleophilic carbene with sulfur, selenium, or tellurium to
produce the corresponding ketone analogs.23 Polarized chal-
cones (O, S, Se, Te) have been more recently reported from
reactions of isolated stable nucleophilic carbenes with chalco-
gens.24-26 We decided to use the reactions of carbenes 1 and
2 with tellurium (eq 6) as another preliminary probe of the
reactivity of the exceptionally stable carbene 2 versus its less
stable parent 1.
consistent with the results of Kuhn on other highly polarized
olefins like 5 (δ CH2 in 5 is 2.77) and strongly resembles that
of olefin 11 (Vide infra).22 Both the two mesityl groups and
the two imidazoline 4- and 5-centers in 9 show slightly different
sets of shifts due to the asymmetry introduced by the chloro-
methylene moiety at position 2. These separated resonances
indicate that the olefin is conformationally locked at room
temperature and does not freely rotate as might have been
expected from a strongly polarized “olefinic” bond. The 13C
NMR of 9 gives a high-field resonance of δ 62.8 for the
chloromethylene unit and δ 146.4 for the imidazoline C2 carbon.
These shifts are similar to those observed for 7 and 4.
The reaction of imidazolin-2-ylidene 6 with methyl iodide
proceeds rapidly to alkylate the carbene center and produce a
2-methylimidazolinium ion (eq 5). This simple 2-methyl residue
of this imidazolinium cation that is intermediate in the process
of eq 5 is not as acidic as the 2-chloromethyl residue on the
intermediate imidazolinium cation that is an intermediate in the
process of eq 4. As a result of this reduced acidity, deproto-
nation by carbene 6 is not satisfactory for the production of the
olefin 11 in good yield. To facilitate the deprotonation that is
required to produce 11, potassium hydride is added to the
reaction mixture as described in the Experimental Section.
The olefin 11 is a colorless solid with a melting range of
135-8 °C. The 1H NMR of 11 shows the terminal methylene
resonance at δ 3.31 which is somewhat higher field than that
observed in 9. This slightly lower field shift in 9 is probably
due to the deshielding effect of the chlorine atom. The terminal
carbon atom in 11 exhibits a 13C resonance at δ 48.7, again
suggesting a polarized olefin structure as found for 9, 7, and 4.
Unlike 9, the olefin 11 was sufficiently stable to allow the
growth of quality crystals for X-ray structure determination. The
solid state structure of 11 is shown in Figure 4.19 The molecule
sits along a crystallographically imposed 2-fold rotation axis
that runs along the CdC double bond as found with 7 although
the two structures are not isomorphic. The CdC double bond
in 11 is 133.0 pm long which is slightly shorter than the
corresponding bond in 7. The nitrogens in the imidazole ring
of 11 are distinctly pyramidal as in the analogs 9, 7, and 4 and
reside 17.6 pm out of the plane of their three attached
substituents. The mesityl groups again adopt the usual anti-
orientation about the imidazole ring. As in 9 the C2-N1(3) bond
distances in 11 are only about 3.9 pm longer than in the carbene
6. As with the previous adducts, the expected increase in the
C2-N1(3) length arising from a decreased π-interaction in the
imidazoline ring for 11 relative to 6 is rather small and supports
a somewhat limited π-delocalization in 6 and/or a highly
polarized olefin structure for 11 (and 9).
Carbene 2 reacts smoothly with tellurium in thf at room
temperature to produce the tellurone 13. The tellurone 13 is a
yellow crystalline solid that melts at 189 °C with decomposition.
The 1H NMR spectrum of 13 provides only sparse information
and indicates that a symmetric adduct, different from the initial
carbene 2, is formed. The 13C NMR spectrum of 13 shows a
resonance for the carbon bound to tellurium of δ 134.26 (86
ppm upfield of the carbene resonance in 2). The 125Te resonance
for tellurone 13 is δ -4. This shift is substantially downfield
of the resonance reported by Kuhn et al. for 1,3-diisopropyl-
4,5-dimethylimidazole-2-tellurone (δ -167.82) and may signal
the strong electron-withdrawing σ-influence of the chlorines.26
Crystals of 13 suitable for X-ray diffraction studies were
grown from a toluene/thf solution on cooling to -25 °C. The
molecule crystallizes in a C2/c space group with the crystal-
lographic 2-fold axis running along the C-Te bond. The solid
state structure of 13 is illustrated in Figure 5, and selected bond
lengths and angles are included in Table 1. The imidazole ring
in 13 is planar with no atom deviating more that 0.1 pm from
the average plane. As required by symmetry, the tellurium atom
lies in the average plane of the imidazole ring. The nitrogen
centers are essentially planar, with the nitrogen rising only 1.5
pm above the plane of its three attached atoms. The ipso-
(23) Benac, B. L.; Burgess, E. M.; Arduengo, A. J., III. Org. Synth. 1986,
64, 92.
(24) Williams, D. J.; Fawcett-Brown, M. R.; Raye, R. R.; VanDerveer,
D.; Jones, R. L. Heteroatom Chem. 1993, 4, 409.
(25) Enders, D.; Breuer, K.; Raabe, G.; Runsink, J.; Teles, J. H.; Melder,
J.-P.; Ebel, K.; Brode, S. Angew. Chem., Int. Ed. Engl. 1995, 34, 1021.
(26) Kuhn, N.; Henkel, G.; Kratz, T. Chem. Ber. 1993, 126, 2047.

An Air Stable Carbene and Mixed Carbene “Dimers”
J. Am. Chem. Soc., Vol. 119, No. 52, 1997 12747
carbons of the mesityl substituents are displaced only 4.5 pm
from the plane of the imidazole ring. The chlorine atoms are
also very nearly planar with the imidazole ring and deviate by
only 3 pm. The N-C-N at the former carbene center of 13 is
105.3° which is intermediate between that found for the carbene
(2) and its corresponding carbenium ion (2‚HCl). The remain-
ing metric parameters (Table 1) in the imidazole ring are also
intermediate between those of the imidazol-2-ylidene and
imidazolium structures. This partial retention of carbene-like
structural characteristics has been noted for other dative bonded
adducts of imidazol-2-ylidenes with some main group elements
and transition metal coordination complexes.27-34
As expected the more reactive carbene, 1, also reacts with
tellurium in thf at room temperature to produce the tellurone
14. Like the tellurone 13, 14 is a yellow crystalline solid. The
1H NMR spectrum of 14 reveals that the imidazole ring protons
in 14 have shifted 0.1 ppm downfield of their positions in the
parent carbene 1. The 13C NMR spectrum of 14 shows a
resonance for the carbon bound to tellurium of δ 133.05 similar
to the corresponding shift that was recorded for 13 (134.26).
The 125Te resonance for tellurone 14 is δ -149.8. This shift is
substantially upfield of the resonance observed for 13 but is
much closer to that reported by Kuhn et al. for 1,3-diisopropyl-
4,5-dimethylimidazole-2-tellurone (δ -167.82).26 Thus, it
appears that the anomalously low field 125Te shift observed in
13 does signal the unique influence of the chlorine substituents.
Crystals of 14 suitable for X-ray diffraction studies were
grown from a hexane/thf solution on cooling to -25 °C. The
solid state structure of 14 is illustrated in Figure 5, and selected
bond lengths and angles are included in Table 1. The imidazole
ring in 14 is essentially planar with no atom deviating more
that 0.2 pm from the average plane. The tellurium atom lies
3.4 pm out of the average plane of the imidazole ring. The
nitrogen centers are essentially planar with the N1 and N3,
respectively, lying 2.9 and 3.4 pm out of the plane of their three
attached substituents. The slight pyramidalization of the nitro-
gens results in a small anti-displacement of the mesityl
substituents above and below the plane of the imidazole ring.
The ipso-carbons of the mesityl substituents deviate 7.3 and
10.4 pm from the plane of the imidazole ring. The gross
structural features (Table 1) of tellurone 14 are quite similar to
those observed for 13. Thus, it is apparent that 14 also retains
some carbene-like structural characteristics as had 13.
Figure 6. Space-filling KANVAS19 drawing of the X-ray structure of
14.
solvents like chloroform and methylene chloride. The structures
of 1,3-dimesitylimidazol-2-ylidene (1) and 4,5-dichloro-1,3-
dimesitylimidazol-2-ylidene (2) are very similar and provide
no suggestion of the dramatically enhanced stability of the latter
1
carbene. There are also no changes in the H or 13C NMR
spectra that reflect the substantial difference in stabilities.
Indeed the 13C resonances for the two carbene centers are less
than 0.2 ppm apart. The introduction of halogen (chlorine) in
the 4- and 5-positions of the imidazole ring appears to have a
special stabilizing effect on the carbene that is not seen from
other positions. For example, when the halogen is introduced
into the para-position of a phenyl substituent on nitrogen as in
1,3-bis(4-chlorophenyl)imidazol-2-ylidene,4b a carbene of such
exceptional stability does not result. In our hands 1,3-bis(4-
chlorophenyl)imidazol-2-ylidene actually appears somewhat
more reactive than 1,3-dimesitylimidazol-2-ylidene.
The first spectral indication of the special influence of the
chlorines is observed in the dative bonded tellurium adducts,
or tellurones, 13 and 14. The tellurone 13 has an unusually
low field shift (δ -4) for the tellurium center compared to 14
(δ -150) and 1,3-diisopropyl-4,5-dimethylimidazole-2-tellurone
(δ -168).26 The low-field shift of the tellurium center in 13
could be the result of a strong σ-electron-withdrawing effect of
the chlorines. Consistent with this intrepretation, 1,3-diisopro-
pyl-4,5-dimethylimidazole-2-tellurone with four electron-releas-
ing alkyl substituents on the imidazole has the highest 125Te
shift observed among these three tellurones.
The influence of the chlorine substituents on carbenes like 2
can include both a π-electron-releasing component by virtue
of the chlorine lone pair electrons and a σ-electron withdrawal
by virtue of chlorine’s high electronegativity. Both of these
effects can act in concert to enhance the stability of a carbene
center at C2 of the imidazole ring. Other than the electronega-
tivity inherent in the nitrogen centers on most stable nucleophilic
carbenes, a σ-electronegativity effect has not previously been a
great aid to enhancing the stability (lowering the reactivity) of
nucleophilic carbenes. The σ-electron withdrawal by the
chlorines substantially reduces the basicity of the σ-lone pair
of electrons at the carbene center. This reduced basicity is
immediately apparent in the tolerance of the halogenated
imidazol-2-ylidenes for acidic solvents like chloroform which
react rapidly with carbenes like 1 and 6. As can been seen for
compound 2, this stabilizing effect can be quite dramatic and
brings the “handability” of such stable carbenes to a new level.
Further studies to understand the enhanced stability and range
of reactivity for carbenes like 2 are currently underway.
In addition to the reaction of CCl4 with imidazol-2-ylidenes
that produces chlorinated carbenes such as 2, we have found
that CCl4 will further react with carbenes like 2 and 6 to reduce
Conclusions
The reactions of imidazol-2-ylidenes that bear hydrogens at
the 4- and 5-positions of the imidazole ring with CCl4 provide
a surprising and convenient way to halogenate the imidazole
ring. The 4,5-dichloroimidazol-2-ylidenes produced by these
reactions show extraordinary stability compared to their unha-
logenated analogs. These dichloroimidazol-2-ylidenes exhibit
enhanced stability toward air, moisture, and acidic halogenated
(27) Arduengo, A. J., III; Kline, M.; Calabrese, J. C.; Davidson, F. J.
Am. Chem. Soc. 1991, 113, 9704.
(28) Arduengo, A. J., III; Dias, H. V. R.; Calabrese, J. C.; Davidson, F.
J. Am. Chem. Soc. 1992, 114, 9724.
(29) Arduengo, A. J., III; Dias, H. V. R.; Calabrese, J. C.; Davidson, F.
Inorg. Chem. 1993, 32, 1541.
(30) Arduengo, A. J., III; Dias, H. V. R.; Davidson, F.; Harlow, R. L. J.
Organomet. Chem. 1993, 462, 13.
(31) Arduengo, A. J., III; Dias, H. V. R.; Calabrese, J. C.; Davidson, F.
Organometallics 1993, 12, 3405.
(32) Arduengo, A. J., III; Gamper, S. F.; Calabrese, J. C.; Davidson, F.
J. Am. Chem. Soc. 1994, 116, 4391.
(33) Arduengo, A. J., III; Tamm, M.; Calabrese, J. C. J. Am. Chem. Soc.
1994, 116, 3625.
(34) Arduengo, A. J., III; Tamm, M.; McLain, S. J.; Calabrese, J. C.;
Davidson, F.; Marshall, W. J. J. Am. Chem. Soc. 1994, 116, 7927.

12748 J. Am. Chem. Soc., Vol. 119, No. 52, 1997
Arduengo et al.
a septum, and the mixture was stirred for two days. The volatiles were
removed under reduced pressure to give a brown residue (240 mg).
This residue was extracted with 20 mL of warm benzene to dissolve
the olefin 4. The benzene insoluble imidazolium salt 3 was dissolved
in 20 mL of CH3CN.
The benzenze extract containing 4 was concentrated to 5 mL, and 3
mL of hexane was added. The solution was cooled to ∼0 °C. Light
yellow crystals of 4 were collected by filtration (115 mg, 93% yield).
Mp: 172 °C. 1H NMR (C6D6, ppm): δ 2.06 (s, 6 H, p-CH3), 2.25 (s,
12 H, o-CH3), 6.69 (s, 4 H, Ar-H). 13C NMR (C6D6): δ 18.14 (s,
o-CH3), 21.01 (s, p-CH3), 70.86 (s, CCl2), 112.06 (s, NCCN), 129.19
(s, m-C), 132.61 (s, ipso-C), 138.35 (s, o-C), 138.98 (s, p-C), 140.25
(s, NCN). Anal. Calcd for C22H22N2Cl4: C, 57.92; H, 4.86; N, 6.14.
Found: C, 57.54; H, 4.67; N, 5.98.
The original acetonitrile extract containing 3 was concentrated to 4
mL in vacuo, and 4 mL of hexane was added. Colorless crystals of
the imidazolium salt 3 monoacetonitrile solvate were obtained when
the solution was cooled to -20 °C (105 mg, 81% yield). 1H NMR
(CD3CN): δ 2.14 (s, 12 H, o-CH3), 2.40 (s, 6 H, p-CH3), 7.27 (s, 4 H,
Ar-H). 13C NMR (CD3CN): δ 17.59 (s, o-CH3), 21.32 (s, p-CH3),
123.60 (s, NCCN), 124.55 (s, NCN), 127.38 (s, ipso-C), 131.37 (s,
m-C), 136.53 (s, o-C), 144.93 (s, p-C). Anal. Calcd for C21H22N2Cl4
(vacuum dried): C, 56.78; H, 4.99; N, 6.31. Found: C, 56.51; H,
4.77; N, 6.25.
Preparation of 1,3-Dimesityl-2-chloroimidazolinium Chloride (8)
and 1,3-Dimesityl-2-(dichloromethyleneimidazoline (7). To a solu-
tion of 0.276 g (0.90 mmol) of 1,3-dimesitylimidazolin-2-ylidene 6 in
30 mL of hexane was added at room temperature 0.14 g (0.9 mmol) of
carbon tetrachloride whereupon 8 precipitated immediately as a colorless
solid which was filtered off after 15 min of stirring at 23 °C. The salt
8 was rinsed with 10 mL of toluene, dried in Vacuo, and recrystallized
from acetonitrile to yield 0.15 g (88%). Mp: >250 °C. 1H NMR
(dmso-d6): δ 2.31 (s, 6 H, p-CH3), 2.33 (s, 12 H, o-CH3), 4.67 (s, 4 H,
NCH2), 7.16 (s, 4 H, Ar-H). 13C NMR (DMSO-d6): δ 16.83 (s, o-CH3),
20.53 (s, p-CH3), 50.52 (s, NCCN), 129.34 (s, ipso-C), 129.75 (s, m-C),
135.33 (s, o-C), 140.65 (s, p-C), 158.51 (s, CCl).
The original mother liquor and toluene washes were combined and
evaporated to dryness. The solid residue was recrystallized from
hexane/toluene by cooling a saturated 23 °C solution to -25 °C. The
olefin 7 was isolated by filtration and dried to yield a light yellow
solid with mp 220-4 °C dec (0.13 g, 67%). 1H NMR (C6D6): δ 2.12
(s, 6 H, p-CH3), 2.29 (s, 12 H, o-CH3), 3.12 (s, 4 H, NCH2), 6.79 (s,
4 H, Ar-H). 13C NMR (C6D6): δ 18.49 (s, o-CH3], 20.94 (s, p-CH3),
50.96 (s, NCCN) 73.97 (s, CCl2), 129.61 (s, m-C), 136.21 (s, p-C),
136.30 (s, o-C), 139.48 (s, ipso-C), 145.76 (s, NCN). MS (EI, 70
eV): m/z (rel intens) 388.1454 (50) [M+, calcd for C22H26Cl2N2
388.1473], 373.1242 (15) [M+ - CH3, 373.1238], 353.1762 (25) [M+
- Cl, 353.1785], 338.1514 (40) [M+ - CH3Cl, 338.1560], 305.2147
(100) [M+ - CHCl2, 305.2018].
Preparation of 1,3-Dimesityl-2-(chloromethylene)imidazolidine
(9). A solution of 0.153 g (0.5 mmol) of the 1,3-dimesitylimidazolin-
2-ylidene (1) in 20 mL of hexane was treated at 23 °C with 1.8 g (21.2
mmol) of dichloromethane. Samples investigated by 1H NMR indicated
that also after seven days no reaction had taken place. The mixture
was transferred to a thick-walled glass tube and heated to 70 °C for 15
h. The formation of a colorless solid was observed. This was filtered
off (0.052 g, identified as imidazolinium salt 10 by 1H NMR), and the
filtrate was evaporated to give 9 as a light yellow solid. The olefin 9
decomposed upon attempted crystallization from hexane. 1H NMR
(C6D6): δ 2.11, 2.14 (s, 6 H, p-CH3), 2.22, 2.41 (s, 12 H, o-CH3),
3.1-3.3 (m, 4 H, 4,5-H), 3.81 (s, 1 H, CHCl), 6.74, 6.82 (s, 4 H, m-CH).
13C NMR (C6D6): δ 17.95, 18.35 (s, o-CH3), 20.96, 21.04 (s, p-CH3),
47.83, 50.17 (s, NCCN), 62.84 (s, CHCl), 129.25 (s, m-C), 129.95 (s,
m-C), 136.06 (s, ipso-C), 136.72 (s, ipso-C), 136.49 (s, p-C), 137.04
(s, p-C), 137.25 (s, o-C), 137.87 (s, o-C), 146.39 (s, NCN).
the CCl4 to a :CCl2 fragment. Under appropriate reaction
conditions the in situ generated :CCl2 can be trapped by the
initial imidazol-2-ylidene or imidazolin-2-ylidene to form mixed
carbene “dimers”. Other trapping reagents are being explored
to further support the postulate that dichlorocarbene is actually
intermediate in this process. This reaction is somewhat
analogous to the reaction of CCl4 with triphenylphosphine that
leads to the in situ generation of (dichloromethylene)tri-
phenylphosphorane.20 Unlike their phosphorus analogs the
olefins 4 and 7 are stable enough to be easily isolated and fully
characterized. The olefins appear to be highly polarized as
indicated by the high-field 13C chemical shifts for the terminal
CCl2 groups (δ 70.86 for 4 and δ 73.97 for 7). A related
nonhalogenated polarized olefin (formally the coupling product
of methylene and 1,3,4,5-tetramethylimidazol-2-ylidene) has
been previously reported by Kuhn et al. and is consistent with
our result.22 Interestingly, this reduction of CCl4 by a carbon-
centered reagent to produce :CCl2 is reminiscent of the report
by Schmeisser and Schro¨ter of the conproportionation of
charcoal and CCl4.35 In our case the reaction proceeds under
much milder conditions. The reducing agent is the imidazol-
(in)-2-ylidene (2 or 6), and the reaction can also be thought of
as a sort of carbene exchange in which the carbene center of an
imidazol(in)-2-ylidene is oxidized to a carbon(IV) center while
the carbon(IV) center in CCl4 is reduced to dichlorocarbene.
The reaction of nucleophilic carbenes with CCl4 has proved
to be a surprisingly rich area for investigation. It has led to a
new class of exceptionally stable carbenes and provided a route
to the synthesis of mixed carbene dimers.
Experimental Section
Reactions and manipulations were carried out under an atmosphere
of dry nitrogen, either in a Vacuum Atmospheres drybox or using
standard Schlenk techniques. Solvents were dried (using standard
procedures),36 distilled, and deoxygenated prior to use, unless otherwise
indicated. Glassware was oven-dried at 160 °C overnight. 1H NMR
spectra were recorded on a General Electric QE-300 spectrometer. 13
C
spectra were recorded on a GE Omega 300WB spectrometer. NMR
references are (CH3)4Si (1H, 13C) and (CH3)2Te and 1 M TeCl4 in THF-
d8 (
125Te). Mass spectra were obtained using a VG-ZAB-E mass
spectrometer. Melting points were obtained on a Thomas-Hoover
capillary apparatus and were not corrected. Elemental analyses were
performed by Micro-Analyses Inc., Wilmington, DE, or Oneida
Research Services, Whitesboro, NY.
Preparation of 4,5-Dichloro-1,3-dimesitylimidazol-2-ylidene (2).
In a drybox, a 200 mL round bottom flask was charged with 4.00 g
(13.2 mmol) of 1,3-dimesitylimidazol-2-ylidene (1) and 80 mL of thf.
To this solution was added 4.05 g (2.50 mL, 26.3 mmol) of CCl4 in 20
mL of thf, and the solution was stirred at 23 °C for 20 min. Subsequent
removal of the volatiles under reduced pressure gave 4.16 g of 4,5-
dichloro-1,3-dimesitylimidazol-2-ylidene (2) as a yellow solid. Further
purification was accomplished by recrystallization from thf to give light
yellow crystals. Mp 180-2°. Yield: 4.16 g (85%). 1H NMR
(C6D6): δ 2.09 (s, 6 H, p-CH3) , 2.12 (s, 12 H, o-CH3), 6.75 (s, 4 H,
Ar-H). 1H NMR (thf-d8): δ 2.07 (s, 12 H, o-CH3) , 2.33 (s, 6 H,
p-CH3), 7.01 (s, 4 H, Ar-H). 13C NMR (C6D6): δ 17.86 (s, o-CH3),
20.99 (s, p-CH3), 116.16 (s, NCCN), 129.3 (s, m-C), 135.67 (s, ipso-
C), 136.05 (s, o-C), 138.53 (s, p-C), 219.89 (s, NCN). Anal. Calcd
for C21H22N2Cl2: C, 67.56; H, 5.94; N, 7.50. Found: C, 66.94; H,
5.74; N, 7.32.
Preparation of 2-(Dichloromethylene)-4,5-dichloro-1,3-dimesi-
tylimidazole (4) and 1,3-Dimesityl-2,4,5-trichloroimidazolium Chlo-
ride (3). In a drybox, a 25 mL round bottom flask was charged with
0.20 g (0.54 mmol) of 4,5-dichloro-1,3-dimesitylimidazol-2-ylidene (2)
and 1.50 mL of CCl4 (2.39 g, 15.5 mmol). The flask was closed with
Preparation of 1,3-Dimesityl-2-methylimidazolinium Iodide (10)
and 1,3-Dimesityl-2-methyleneimidazolidine (11). To a solution of
0.150 g (0.49 mmol) of 1,3-dimesitylylimidazolin-2-ylidene (1) in 30
mL of toluene was added at 23 °C 0.92 g (6.5 mmol) of methyl iodide.
Immediately a colorless solid precipitated which was filtered off after
3 h of stirring at 23 °C. 1H and 13C NMR revealed it to be
1,3-dimesityl-2-methylimidazolinium iodide containing a small amount
(35) Schmeisser, M.; Schro¨ter, H. Angew. Chem. 1960, 72, 349.
(36) Perrin, D. D.; Armarego, W. L. F.; Perrin, D. R. Purification of
Laboratory Chemicals; Pergamon: New York, 1985.

An Air Stable Carbene and Mixed Carbene “Dimers”
J. Am. Chem. Soc., Vol. 119, No. 52, 1997 12749
of 1,3-dimesitylimidazolinium iodide (12). The spectral data for 1,3-
dimesityl-2-methylimidazolinium iodide were as follows. 1H NMR
(dmso-d6): δ 1.81 (s, 3 H, NCCH3), 2.31 (s, 6 H, p-CH3), 2.34 (s, 12
H, o-CH3), 4.38 (s, 4 H, CH2), 7.12 (s, 4 H, Ar-H). 13C NMR (DMSO-
d6): δ 10.84 (s, (NCCH3), 16.91 (s, o-CH3), 20.60 (s, p-CH3), 49.56
(s, NCCN), 129.80 (s, m-C), 130.16 (s, ipso-C), 135.50 (s, o-C), 139.97
(s, p-C), 167.11 (s, NCN). The evaporated mother liquor yielded a
small quantity of the olefin 11. The crude 1,3-dimesityl-2-methylimi-
dazolinium iodide was used without further purification in the depro-
tonation step.
Crude 1,3-dimesityl-2-methylimidazolinium iodide was suspended
in 10 mL of thf, and 0.22 g of a 35% suspension of potassium hydride
in mineral oil (corresponding to 1.9 mmol of potassium hydride) was
added at 23 °C. After the evolution of gas had ceased (ca. 3 h), the
mixture was filtered, and the filtrate was evaporated to give a yellow
oil. Recrystallization from hexane at -25 °C gave colorless crystals
of the olefin 11. Yield: 44 mg (28%). Mp: 135-138 °C. 1H NMR
(C6D6): δ 2.14 (s, 6 H, p-CH3), 2.34 (s, 12 H, o-CH3), 2.59 (s, 2 H,
CdCH2), 3.31 (s, 4 H, CH2), 6.82 (s, 4 H, Ar-H). 13C NMR (C6D6) δ
17.20 (s, o-CH3), 20.30 (s, p-CH3), 47.57 (s, NCCN), 48.70 (s, CdCH2),
129.72 (s, m-C), 136.87 (s, p-C), 137.28 (s, ipso-C), 138.22 (s, o-C),
151.78 (s, NCN). MS (70 eV): m/z (rel intens) 320.2230 (65) [M+,
calcd for C22H28N2 320.2252], 305.2052 (100) [M+ - CH3, 305.2018],
148.1123 (25) [C10H14N+, 148.1126].
Preparation of 1,3-Dimesityl-4,5-dichloroimidazole-2-tellurone
(13). In a drybox, a 100 mL round bottom flask was charged with
0.20 g (0.66 mmol) of carbene 2 and 20 mL of thf. The solution was
stirred for a few minutes, and then 0.084 g (0.66 mmol) of tellurium
powder was added. The solution slowly turned yellow as the reaction
progressed. After the reaction mixture was stirred for 16 h, an
additional 20 mL of thf was added and the solution filtered over Celite
to obtain a clear yellow solution. After concentration of the filtrate to
5 mL, and addition of 4 mL of hexane, the flask was cooled in the
refrigerator to -20 °C. Yellow crystals (280 mg, 100%) were collected
from the cooled solution by filtration. Mp: 189 °C dec. 1H NMR
(thf-d8): δ 2.10 (s, 12 H, o-CH3), 2.35 (s, 6 H, p-CH3), 7.05 (s, 4 H,
m-CH). 13C NMR (thf-d8): δ 18.42 (s, o-CH3), 21.26 (s, p-CH3), 116.59
(s, NCCN), 130.08 (s, m-C), 133.86 (s, ipso-C), 134.26 (s, CdTe),
136.78 (s, o-C), 140.92 (s, p-C). 125Te (94.763 MHz) NMR (thf-d8):
δ -4.08 (ref neat Me2Te); +1730.45 (ref 1 M TeCl4 in thf-d8). Anal.
Calcd for C21H22N2Cl2Te: C, 50.33; H, 4.40; N, 5.59. Found: C, 49.86;
H, 4.77; N, 5.43.
Preparation of 1,3-Dimesitylimidazole-2-tellurone (14). In a
drybox a 200 mL round bottom flask was charged with carbene 1 (1.00
g, 3.28 mmol), tellurium (0.418 g, 3.29 mmol), and thf (100 mL). The
mixture was stirred for 12 h, after which time the solution was slightly
cloudy. This solution was warmed to obtain a clear yellow solution,
filtered over Celite, and concentrated until crystallization began.
Hexane was added, and the crystallization was allowed to continue at
-20 °C. The tellurone 14 was isolated by filtration as yellow crystals
(1.4 g, 100%). A sample that was recrystallized from THF gave the
following data: Mp: 317-318 °C. 1H NMR (thf-d8): δ 2.06 (s, 12
H, o-CH3), 2.36 (s, 6 H, p-CH3), 7.03 (s, 4 H, m-CH), 7.15 (s, 2 H,
NCH). 13C NMR (thf-d8): δ 18.72 (s, o-CH3), 21.20 (s, p-CH3), 123.48
(s, NCCN), 129.66 (s, m-C), 133.05 (s, CdTe), 136.25 (s, o-C), 136.98
(s, ipso-C), 139.53 (s, p-C). 125Te NMR (thf-d8, ref neat Me2Te): δ
-149.8. Anal. Calcd for C21H24N2Te: C, 58.38; H, 5.60; N, 6.48.
Found: C, 58.65; H, 5.62; N, 6.33.
Acknowledgment. A.J.A. is indebted to the Alexander von
Humboldt Stiftung for a senior research prize that made the
completion of this work possible.
Supporting Information Available: A complete description
of the X-ray crystallographic determinations on 2, 2‚HCl,
3‚CH3CN, 4, 7, 9, 11, 13, and 14, including tables of fractional
coordinates, isotropic and anisotropic thermal parameters, bond
distances, and bond angles and ORTEP drawings (60 pages).
See any current masthead page for ordering and Internet access
instructions.
JA973241O