Basicity of a stable carbene, 1,3-di-tert-butylimidazol-2-ylidene, in THF

Article abstract of DOI:10.1021/ja025628j

The basicity of 1,3-di-tert-butylimidazol-2-ylidene (1) was measured in THF against three hydrocarbon indicators. Both ion pairs and free ions were found and the corresponding equilibrium constants were measured. Homoconjugation was not found in either THF or DMSO. The carbene is effectively more basic in DMSO by several pK units, probably because of hydrogen bonding of 1-H⁺ to DMSO. Model ab initio computations are consistent with these results.

Full text of DOI:10.1021/ja025628j

Published on Web 04/25/2002
Basicity of a Stable Carbene,
1,3-Di-tert-butylimidazol-2-ylidene, in THF1
Yeong-Joon Kim and Andrew Streitwieser*
Contribution from the Department of Chemistry, UniVersity of California,
Berkeley, California 94720-1460
Received January 17, 2002
Abstract: The basicity of 1,3-di-tert-butylimidazol-2-ylidene (1) was measured in THF against three
hydrocarbon indicators. Both ion pairs and free ions were found and the corresponding equilibrium constants
were measured. Homoconjugation was not found in either THF or DMSO. The carbene is effectively more
basic in DMSO by several pK units, probably because of hydrogen bonding of 1-H+ to DMSO. Model ab
initio computations are consistent with these results.
Introduction
rene completely but not fluorene; this result suggests that the
carbene is effectively more basic in DMSO than in THF. Some
Study of the structures and chemical reactivities of carbenes
has become an active research area after Arduengo’s first
synthesis and isolation of stable imidazol-2-ylidene carbenes.2
In addition to theoretical and other fundamental studies,3 much
effort has been expended to develop organometallic catalysts
that contain carbenes as ligands.4 For example, ruthenium
alkylidene compounds have been used successfully as olefin
metathesis catalysts.5 Nevertheless, much of the fundamental
chemistry of these carbenes is not known since they are
relatively new. For example, there is only one report dealing
with their basicity. The Alder group has reported the basicity
of 1,3-diisopropyl-4,5-dimethylimidazol-2-ylidene in dimethyl
sulfoxide-d6 (DMSO-d6) as pKa ) 24, which makes it more
basic than such strong nitrogen bases as DBN and DBU.6 They
also tried the same measurement in THF-d8 by an NMR method
and found that the carbene is able to deprotonate 9-phenylfluo-
time later it was found that the carbenes are reactive toward
halogenated and acidic solvents including DMSO.7 8 Thus, we
considered it important to determine the basicity of carbenes
more precisely in inert solvents such as THF.
Results and Discussion
Our group has developed ion pair acidity scales for the lithium
and cesium salts of various hydrocarbons in THF9 relative to a
standard taken as 22.90, the ionic pKa of fluorene per hydrogen
in DMSO.10 The relative acidities apply to the equilibria in eq
1, in which the lithium salts are solvent separated ion pairs
(SSIP) and the cesium salts are contact ion pairs (CIP).
R- M+ + R′H ) RH + R′- M+ (M+ ) Li+ or Cs+) (1)
More recently we have applied such ion pair acidity scales
to the ion pair basicities of amines in THF. Amines are often
used as base catalysts in organic synthesis under ion pairing
conditions.11 The relative ion pair basicities of amines were
reported to differ substantially from the ionic values in polar
solvents such as acetonitrile. We now report the basicity of the
stable carbene, 1,3-di-tert-butylimidazol-2-ylidene (1), in THF
using UV-vis spectroscopy. Three different indicators (2, 3,
and 4) were found to equilibrate with their anions in the presence
of the carbene. The λmax and the concentrations of all species
are listed in Table S1 (Supporting Information). We show below
that both ion pairs and free ions are present at the concentrations
* Address correspondence to this author. E-mail: astreit@socrates.
berkeley.edu.
(1) Carbon Acidity. 117. For paper 116 see: Streitwieser, A.; Kim, Y.-J.; Wang,
D. Z.-R. Org. Lett. 2001, 3, 2599-2601.
(2) Arduengo, A. J., III; Harlow, R. L.; Kline, M. J. Am. Chem. Soc. 1991,
113, 361.
(3) Arduengo, A. J., III Acc. Chem. Res. 1999, 32, 913. Denk, M. K.; Rodezno,
J. M.; Gupta, S.; Lough, A. J. J. Organomet. Chem. 2001, 617-618, 242-
253. Arduengo, A. J., III; Tamm, M.; Calabrese, J. C.; Davidson, F.;
Marshall, W. J. Chem. Lett. 1999, 10, 1021-1022. Lehmann, J. F.;
Urquhart, S. G.; Ennis, L. E.; Hitchcock, A. P.; Hatano, K.; Gupta, S.;
Denk, M. K. Organometallics 1999, 18, 1862-1872. Muchall, H. M.;
Werstiuk, N. H.; Choudhury, B. Can. J. Chem. 1998, 76, 221-227.
Arduengo, A. J., III; Bock, H.; Chen, H.; Denk, M.; Dixon, D. A.; Green,
J. C.; Herrmann, W. A.; Jones, N. L.; Wagner, M.; West, R. J. Am. Chem.
Soc. 1994, 116, 6641-9
(4) Caddick, S.; Cloke, F. G. N.; Clentsmith, G. K. B.; Hitchcock, P. B.;
McKerrecher, D.; Titcomb, L. R.; Williams, M. R. V. J. Organomet. Chem.
2001, 617-618, 635-639. Bohm; Volker, P. W.; Gstottmayr, C. W. K.;
Weskamp, T.; Herrmann, W. A. J. Organomet. Chem. 2000, 595, 186-
190. Arnold, P. L.; Cloke, F.; Geoffrey N.; Geldbach, T.; Hitchcock, P. B.
Organometallics 1999, 18, 3228-3233. Green, J. C.; Scurr, R. G.; Arnold,
P. L.; Cloke, F. G. N. Chem. Commun. 1997, 20, 1963-1964.
(5) Sanford, M. S.; Love, J. A.; Grubbs, R. H. J. Am. Chem. Soc. 2001, 123,
6543. Louie, J.; Grubbs, R. H. Angew. Chem., Int. Ed. 2001, 40, 247.
Weskamp, T.; Kohl, F. J.; Herrmann, W. A. J. Organomet. Chem. 1999,
582, 362-365.
(7) Arduengo, A. J., III; Calabrese, J. C.; Davidson, F.; Dias, H. V. R.; Goerlich,
J. R.; Krafczyk, R.; Marshall, W. J.; Tamm, M.; Schmutzler, R. HelV. Chim.
Acta 1999, 82, 2348.
(8) Denk, M. K.; Rodezno, J. M. J. Organomet. Chem. 2001, 617-618, 737-
740. Denk, M. K.; Rodezno, J. M. J. Organomet. Chem. 2000, 608, 122-
125.
(9) (a) Streitwieser, A.; Ciula, J. C.; Krom, J. A.; Thiele, G. J. Org. Chem.
1991, 56, 1074. (b) Streitwieser, A.; Wang, D. Z.; Stratakis, M.; Facchetti,
A.; Gareyev, R.; Abboto, A.; Krom, J. A.; Kilway, K. V. Can. J. Chem.
1998, 76, 765.
(10) Bordwell, F. G. Acc. Chem. Res. 1988, 21, 456.
(11) Streitwieser, A.; Kim, Y.-J. J. Am. Chem. Soc. 2000, 122, 11783.
(6) Alder, R. W.; Allen, P. R.; Williams, S. J. J. Chem. Soc., Chem. Commun.
1995, 1267.
9
10.1021/ja025628j CCC: $22.00 © 2002 American Chemical Society
J. AM. CHEM. SOC. 2002, 124, 5757-5761
5757

A R T I C L E S
Kim and Streitwieser
Scheme 1
Figure 1. Plot of ([RH][carbene])1/2 vs {R-}. The regression lines shown
are (0.0212 ( 0.0093)x + (24.17 ( 3.47)x2 (R2 ) 0.980) for Ph-2,3-BF
(0); (0.0111 ( 0.0019)x + (7.68 ( 0.15)x2 (R2 ) 0.999) for 3,4-BF (O);
and (0.000499 ( 0.000036)x + (0.066 ( 0.0015)x2 (R2 ) 0.9996) for BnBF
([).
Table 1. Equilibrium Constants at 25 °C Obtained from Plots of
{R-} vs ([RH][carbene])1/2 and Comparison of pK Values with pKLi
K
ip
K
d
K
i
pK a
pK
ip
pK
i
Li
Ph-2,3-BF 24.17
1.87E-05 4.52E-04 17.84 -1.38
3.34
1.58E-05 1.21E-04 19.29 -0.59b 4.22b
used, yet the λmax values are essentially constant over this range.
Accordingly, the extinction coefficients of the Li salts were used
for the calculation of the concentrations of the anion moieties
as in our earlier amine basicity study. These λmax values are
3-5 nm shorter than those of the SSIP lithium salts. Similar
results were found for those amines whose protonated forms
are delocalized cations.
3,4-BF
BnFl
7.68
0.0656 3.79E-06 2.49E-07 21.36
1.18
6.60
a Reference 8b. b Per-hydrogen basis, statistical correction applied (+0.3).
Scheme 2
The reaction of the carbene with the indicator hydrocarbon
produces first an ion pair between the protonated carbene, 1-H+,
and the indicator anion with an equilibrium constant Kip. The
ion pair dissociates to the free ions with an equilibrium constant
Kd (Scheme 1). In treating the data, however, it is convenient
to consider both species as directly equilibrating with the
reactants, that is, the free ions in equilibrium with the carbene
and hydrocarbons with equilibrium constant Ki ()KipKd). Since
the spectra measure both the ion pair and free indicator anion,
the measured anion moiety, expressed with curly brackets as
{R-}, is given by eq 2.
Scheme 3
Arduengo et al. have reported a crystal structure of a carbene
hydrogen bonded to its conjugate acid with C-H-C hydrogen
bonding.13 The rapid proton transfers between carbenes such
as 1 and their conjugate acids on the NMR time scale in THF
suggests that such homoconjugation might apply in the equilibria
in Scheme 1. Two possibilities are that homoconjugation occurs
both in the ion pair and in the free ion (Scheme 2) or that it
applies to the free ion alone (Scheme 3). The corresponding
equations for {R-} are 3 and 4, respectively.
Neither equation provides a satisfactory fit to the data:
negative equilibrium constants result. Scheme 2 and eq 3 were
found to apply to some amines (for example, DMAP) in THF.14
Homoconjugation, therefore, does apply to some cases but it
appears to be negligible for the carbene 1 and its conjugate acid
in THF, at least at the low concentrations involved in our
experiments.
{R-} ) [ion pair] + [free ion] ) Kip[RH][carbene] +
(Ki[RH][carbene])1/2 (2)
Figure 1 shows plots of {R-} vs ([RH][carbene])1/2 for all
three indicator hydrocarbons. The plots are smooth second-order
polynomials (R2 > 0.98) for which straightforward curve-fittings
yield the corresponding equilibrium constants as summarized
in Table 1. The resulting Kd values are of the order of 10-5 M,
which is similar to the values for SSIP Li salts obtained from
conductivity measurements.12 This agreement is evidence for
the validity of this method.
From the Kip and Kd values obtained, the concentrations of
ion pairs and free ions were calculated and are listed in Table
S1. They are of comparable magnitude in the dilute solutions
used, but of course, the ion pairs are much more important at
synthesis concentrations.
If the pKa values were known for the indicator hydrocarbons
in THF then the pKi in Table 1 would provide a value for pKa
for 1-H+ in THF. These pKa values are not known but since
the dissociation constants of the SSIP lithium salts are ap-
(13) Arduengo, A. J., III; Gamper, S. F.; Tamm, M.; Calabrese, J. C.; Davidson,
F.; Craig, H. A. J. Am. Chem. Soc. 1995, 117, 572.
(14) Kim, Y.-J.; Wong, K.; Streitwieser, A. Unpublished results.
(12) Kaufman, M. J.; Gronert, S.; Streitwieser, A. J. Am. Chem. Soc. 1988,
110, 2829.
9
5758 J. AM. CHEM. SOC. VOL. 124, NO. 20, 2002

1,3-Di-tert-butylimidazol-2-ylidene
A R T I C L E S
either THF or DMSO. The derived pKa for 1-H+ in DMSO is
22.7. This value is appreciably lower than the value of 24
reported for a close analogue of 1.6 We note in Table 1 also
the significant ion pairing found in DMSO; Olmstead and
Bordwell have also reported a number of salts that ion pair
appreciably in DMSO.17 Thus, we suggest that Alder et al.6
measured both ion pair and free ion as their anionic species,
which would lead to a pKa that is too high. For example,
combining the ion pair and free ion concentrations in our DMSO
experiment gives a “pKa” of 23.04, which is significantly higher
than that for free ions alone. At the higher concentrations of a
NMR measurement this value would be expected to be still
higher; nevertheless, the two carbenes are different structures
and even the corrected pKa could well be different.
Even with the corrected values, the carbene 1 is still
effectively somewhat more basic in DMSO than in THF. The
most likely explanation is the following. It is well-known that
hydrocarbons with delocalized carbanions have comparable pKa
values in water and in DMSO; neither the hydrocarbon acid
nor the delocalized conjugate base is significantly stabilized by
hydrogen bonding. Hydroxylic acids, however, are generally
less acidic in DMSO than in water; in water, both acid and
conjugate base are stabilized by hydrogen bonding but in DMSO
only the hydroxylic acid is so stabilized. Similarly, the proto-
nated carbene 1-H+ can be stabilized by hydrogen bonding in
DMSO but not the carbene itself.
Figure 2. Plot of {fluorenyl anion} vs ([RH][carbene])1/2 for fluorene and
1 in DMSO at 25 °C. The regression curve shown is (1.30 ( 0.05)x +
(54.36 ( 9.45)x2; R2 ) 0.998. Calculated Kd ) 0.031 M, Ki ) 1.68, and
Kip ) 54 M-1
.
proximately constant for the indicator hydrocarbons, differences
in the ion pair pKLi are equal to the differences in pKa.12
Accordingly, pKLi - pKi in Table 1 should be approximately
constant and it is. The resulting value, 14.9 ( 0.1, has no other
fundamental significance but does demonstrate the self-
consistency of the experiments.
A plot of pKip in Table 1 vs pKLi for the corresponding
indicators gives a linear fit with a slope of 0.74 ( 0.08, a value
similar to that, 0.75, reported earlier for similar equilibria
between DBU and indicator hydrocarbons.11 The compressed
sensitivity for pKip results from changing of the dissociation
constant, Kd. The more basic indicators have tighter ion pairs
and smaller Kd values; the same effect occurs with the cesium
contact ion pair indicator salts.12 This plot gives pKip ) 0 for
pKLi ) 20, which puts a lithium salt of this pKLi and the carbene
1 at the same level of effective basicity, a level several pK units
less basic than that reported in DMSO solution. That is, in THF
the carbene is as basic as the lithium salt of a compound having
an “ion pair pK” of 20 on our lithium scale. To the extent that
this type of comparison is appropriate the carbene is effectively
relatively less basic than in DMSO. One possible explanation
for such a difference is that homoconjugation could be occurring
for the carbene in DMSO; such homocojugation has been
demonstrated for some amines in acetonitrile solution.15 To test
this possibility we applied the method of eqs 2-4 to DMSO
solution.
From the UV-vis spectra we found that fluorene is depro-
tonated when similar amounts of carbene are present as
previously reported.6 The reported extinction coefficients16 in
DMSO were used for the calculation of the concentration of
{R-}. The UV-vis spectra for the anion decreased and
generated unknown peaks to a significant extent after 1 day.
This observation shows that there is ongoing reaction and
gradual decomposition. Thus, the spectral measurements were
limited to a 2-3 h period. The resulting experimental points
(Table S2, Supporting Information) do not fit either eq 3 or eq
4 but do fit eq 2 well (Figure 2). Accordingly, we find no
evidence for homoconjugation of the protonated carbene in
Theory
One of the remaining questions concerns the structure of the
ion pair: is it a type of charge-transfer complex with the two
ring systems parallel or is the central CH of 1-H+ directed
toward the anionic ring system? This question was addressed
with the help of ab initio computations of a model system,
protonated 2,5-dimethylimidazole, 5-H+, and cyclopentadienyl
anion. Ab initio calculations, RHF 6-31+G*, were carried out
with PC Spartan18 and Gaussian98.19 All energies and coordi-
nates for the imidazolium complexes are summarized in Table
S3 (Supporting Information). The energies are summarized in
Table 2. Stationary points were characterized by frequency
calculations.
Starting with 5-H+ in a plane perpendicular to the Cp plane
and with the central C-H pointing at the center of the Cp ring
gave a second-order saddle point with the distance between the
ring center and the H of 5-H+ of 2.089 Å. Relaxing this structure
led to a transition structure, 6A (one imaginary frequency), 0.11
kcal mol-1 lower in energy with the C-H of 5-H+ pointing to
the center of a CC bond in Cp at a distance of 2.210 Å (Figure
3). Starting with a charge-transfer type structure with the 5-H+
and Cp rings parallel and separated by 2.95 Å led to another
transition structure, 6B (Figure 3), in which the central CH bond
(17) Olmstead, W. N.; Bordwell, F. G. J. Org. Chem. 1980, 45, 3299-3305.
(18) PC Spartan Pro; Wavefunction, Inc.; Irvine, CA, 1996-2000.
(19) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M.
A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A., Jr.; Stratmann,
R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin,
K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi, M.; Cammi,
R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.; Ochterski, J.;
Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Malick, D. K.;
Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Cioslowski, J.; Ortiz,
J. V.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.;
Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng,
C. Y.; Nanayakkara, A.; Gonzalez, C.; Challacombe, M.; Gill, P. M. W.;
Johnson, B.; Chen, W.; Wong, M. W.; Andres, J. L.; Gonzalez, C.; Head-
Gordon, M.; Replogle, E. S.; Pople, J. A. Gaussian 98; Gaussian, Inc.:
Pittsburgh, PA, 1998.
(15) Augustin-Nowacka, D.; Chmurzyn˜ski, L. Anal. Chim. Acta 1999, 381, 215-
220. Galezowski, W.; Jarczewski, A.; Stanczyk, M.; Brzezinski, B.; Bartl,
F.; Zundel, G. J. Chem. Soc., Faraday Trans. 1997, 93, 2515-8. Pawlak,
Z.; Urban˜czyk, G. J. Mol. Struct. 1988, 177, 401-406. Coetzee, J. F.;
Padmanabhan, G. R. J. Am. Chem. Soc. 1965, 87, 5005.
(16) Steiner, E. C.; Gilbert, J. M. J. Am. Chem. Soc. 1965, 87, 382.
9
J. AM. CHEM. SOC. VOL. 124, NO. 20, 2002 5759

A R T I C L E S
Kim and Streitwieser
Table 2. Computed Energies at RHF/6-31+G* (Zero-Point
Energies Are Uncorrected)
E, au
ZPE, kcal mol-1
Cp-
-192.208 655
-302.848 178
-303.276 741
-495.628 927
-495.628 960
-551.545 088
-154.069 429
-606.150 052
-854.854 930
-457.362 979
52.98
85.49
carbene, 5
protonated carbene, 5-H+
94.73
protonated carbene-Cp-, 6A
149.15a
149.12a
53.76
protonated carbene-Cp-, 6B
DMSO
Me2O
7
8
9
53.93
181.12
149.39
149.58
a One imaginary frequency.
Figure 4. Hydrogen-bonded complexes of 5-H+ with 5 (7), dimethyl
sulfoxide (8), and dimethyl ether (9).
lengths are 1.082 and 2.314 Å. These values can be compared
to those in the crystal structure (mesityl group instead of methyl,
172.5°, 1.159, and 2.026 Å, respectively).13 The complexation
energy, -15.8 kcal mol-1, is between the values for dimethyl
ether and 5-H+ (9, -10.5 kcal mol-1) and for DMSO and 5-H+
(8, -20.8 kcal mol-1). The stronger hydrogen bond to DMSO
in 8 compared to dimethyl ether in 9 is also reflected in the
two O‚‚‚H bond lengths, 1.966 and 2.057 Å, respectively. The
computational results thus model the experimental findings that
there is no homoconjugation in DMSO and the greater basicity
of the carbene in DMSO compared to THF.
Figure 3. Optimized geometries of two complexes of 5-H+ and cyclo-
pentadienyl anion, 6A and 6B.
of 5-H+ points to a carbon in Cp with a distance of 2.141 Å
and the ring planes at an angle of 117.8°. Transition structure
6B in Figure 3 is lower than 6A by only 0.015 kcal mol-1 and
is less stable than cyclopentadiene plus carbene by 10.6 kcal
mol-1. That is, the computations indicate that in the gas phase
the model carbene is too weakly basic to abstract a proton from
cyclopentadiene, but the results do suggest that the ion pair
structure has the central C-H of the protonated carbene pointing
toward the indicator anion rather than a π-type structure.
Conclusions
The stable carbene, 1,3-di-tert-butylimidazol-2-ylidene (1),
was found to be more basic than any of the amines we have
studied. An accurate value of the effective basicity of the carbene
in THF has been obtained. A method was demonstrated for
deducing the dissociation constant of ion pairs to free ions from
UV-vis data as an alternative to conductivity measurement.
We have also developed methods for determining whether
homoconjugation is important. These new methods are now
being applied in our continuing study of the ion pair basicity
of amines.
To model the role of solvation we compared structures with
5 hydrogen-bonded to the model carbene 5-H+ (7, homocon-
jugation) and to DMSO, 8, and dimethyl ether (as a simple
model of THF), 9. In the homoconjugation model, 7, the
C-H-C hydrogen bond is linear (180.00°) and two CH bond
Experimental Section
The carbene, 1,3-di-tert-butylimidazol-2-ylidene (1), was a generous
gift from Professor Arduengo. The indicators were available from our
previous studies. Solutions of the indicator and the carbene were
9
5760 J. AM. CHEM. SOC. VOL. 124, NO. 20, 2002

1,3-Di-tert-butylimidazol-2-ylidene
A R T I C L E S
prepared in a glovebox and spectroscopic measurements were made at
25 °C using the glovebox-spectrometer system described previously.20
The experimental results are reported in Tables S1 and S2 (Supporting
Information) and plotted in Figures 1 and 2.
Chemical Society, for partial support of this research. We also
thank Professor A. J. Arduengo for his generous gift of
compound 1.
Supporting Information Available: Tables of experimental
data and computational results (PDF). This material is available
free of charge via the Internet at http://pubs.acs.org.
Acknowledgment. The authors thank the donors of the
Petroleum Research Fund, administered by the American
(20) Krom, J. A.; Petty, J. T.; Streitwieser, A. J. Am. Chem. Soc. 1993, 115,
8024-30.
JA025628J
9
J. AM. CHEM. SOC. VOL. 124, NO. 20, 2002 5761