Imidazolium hydrogen carbonates versus imidazolium carboxylates as organic precatalysts for N-heterocyclic carbene catalyzed reactions

Article abstract of DOI:10.1021/jo301597h

Imidazolium-2-carboxylates (NHC-CO₂ adducts, 3) and (benz)imidazolium hydrogen carbonates ([NHC(H)][HCO₃], 4) were independently employed as organic precatalysts for various molecular N-heterocyclic carbene (NHC) catalyzed reactions. NHC-CO₂ adducts were obtained by carboxylation in THF of related free NHCs (2), while the synthesis of [NHC(H)][HCO₃] precursors was directly achieved by anion metathesis of imidazolium halides (1) using potassium hydrogen carbonate (KHCO₃) in methanolic solution, without the need for the prior preparation of free carbenes. Thermogravimetric analysis (TGA) and TGA coupled with mass spectrometry (TGA-MS) of most [NHC(H)][HCO₃] precursors 4 showed a degradation profile in stages, with either a concomitant or a stepwise release of H₂O and CO₂, between 108 and 280 °C, depending on the nature of the azolium and substituents. In solution, NHC generation from both [NHC(H)][HCO₃] salts and NHC-CO₂ adducts could be achieved at room temperature, most likely by a simple solvation effect. Both types of precursors proved efficient for organocatalyzed molecular reactions, including cyanosilylation, benzoin condensation, and transesterification reactions. The catalytic efficiencies of NHC-CO₂ adducts 3 were found to be approximately 3 times higher than those of their [NHC(H)][HCO₃] counterparts 4.

Full text of DOI:10.1021/jo301597h

Article
pubs.acs.org/joc
Imidazolium Hydrogen Carbonates versus Imidazolium Carboxylates
as Organic Precatalysts for N‑Heterocyclic Carbene Catalyzed
Reactions
Mareva Fevre,†,‡ Paul Coupillaud,†,‡ Karinne Miqueu,§ Jean-Marc Sotiropoulos,§ Joan Vignolle,†,‡
́
̀
and Daniel Taton*,†,‡
†CNRS, LCPO, UMR 5629, F-33600 Pessac, France
‡Universite
§Universite
́
́
de Bordeaux, LCPO, UMR 5629, F-33600 Pessac, France
de Pau & des Pays de l’Adour, IPREM, UMR 5254, F-64053 Pau, France
S
* Supporting Information
ABSTRACT: Imidazolium-2-carboxylates (NHC−CO2 adducts, 3) and
(benz)imidazolium hydrogen carbonates ([NHC(H)][HCO3], 4) were
independently employed as organic precatalysts for various molecular N-
heterocyclic carbene (NHC) catalyzed reactions. NHC−CO2 adducts were
obtained by carboxylation in THF of related free NHCs (2), while the
synthesis of [NHC(H)][HCO3] precursors was directly achieved by anion
metathesis of imidazolium halides (1) using potassium hydrogen carbonate
(KHCO3) in methanolic solution, without the need for the prior preparation of
free carbenes. Thermogravimetric analysis (TGA) and TGA coupled with mass
spectrometry (TGA-MS) of most [NHC(H)][HCO3] precursors 4 showed a degradation profile in stages, with either a
concomitant or a stepwise release of H2O and CO2, between 108 and 280 °C, depending on the nature of the azolium and
substituents. In solution, NHC generation from both [NHC(H)][HCO3] salts and NHC−CO2 adducts could be achieved at
room temperature, most likely by a simple solvation effect. Both types of precursors proved efficient for organocatalyzed
molecular reactions, including cyanosilylation, benzoin condensation, and transesterification reactions. The catalytic efficiencies
of NHC−CO2 adducts 3 were found to be approximately 3 times higher than those of their [NHC(H)][HCO3] counterparts 4.
masked NHCs are of practical use, either for synthesizing
NHC−transition metal complexes27 or as precatalysts for
NHC-catalyzed molecular23,28−37 and macromolecular reac-
tions.16−21,38−41
INTRODUCTION
■
In the past few decades, N-heterocyclic carbenes (NHCs) have
emerged not only as powerful ligands for transition metals1−3
but also as true organocatalysts for various molecular
reactions.4−6 This growing interest is explained by the
versatility of NHCs, their steric and/or electronic proper-
tieshence their reactivitybeing finely tuned by varying the
substituents on the heterocyclic ring.7,8 More recently, the
catalytic potential of NHCs has also been exploited for several
chain polymerization reactions, including the ring-opening
polymerization (ROP) of heterocycles such as cyclic esters,
carbonates, carbosiloxanes, N-carboxyanhydrides, epoxides, and
the group transfer polymerization (GTP) of (meth)acrylic
monomers.9,10
Among these masked NHCs, NHC−CO2 adducts, namely
imidazol(in)ium-2-carboxylates, have attracted a great deal of
attention. This is due to the fact that (i) any types of structures
are virtually accessible by reaction of free NHCs with
CO2,24,42−45 (ii) NHC−CO2 adducts are relatively stable in
the solid state under air exposure,46 and (iii) CO2 is the only
byproduct upon NHC generation. However, synthesis of
NHC−CO2 adducts generally requires the corresponding
NHC to be prepared prior to the carboxylation step.24,42−45
Free NHCs are thought to be generated in situ upon heating
such NHC−CO2 adducts, typically between 60 and 90 °C (see
Scheme 1).24,44−51 In the context of organometallic chemistry,
however, Crabtree et al.47,48 and Sauvage et al.51 have noticed
successful NHC transfer onto transition-metal fragments at
room temperature. Such imidazol(in)ium-2-carboxylates, and
some polymer-supported versions, have been utilized as NHC
and poly(NHC) precursors, respectively, for the purpose of
organocatalysis. Related (macro)molecular reactions include
However, the relative instability of NHCs11,12 on exposure to
air makes them difficult to handle. As a matter of fact, NHCs
must generally be stored and manipulated under dry and inert
conditions. A general strategy to circumvent this limitation is to
convert the NHC into a masked form. Various NHC adducts
have thus been derived, the activation of which in situ generates
free NHCs. For instance, NHC-Ag(I) complexes,13,14
alkoxy,15−18 trichloromethyl,19 pentafluorophenyl,19 carboxylic
acid,20,21 thioisocyanates (NCS),22 thiocarboxylates (COS),23
CO2−NHC adducts,24,25 and, very recently, imidazolium
hydrogen carbonates26 can serve as NHC precursors. Such
Received: August 20, 2012
Published: October 23, 2012
© 2012 American Chemical Society
10135
dx.doi.org/10.1021/jo301597h | J. Org. Chem. 2012, 77, 10135−10144

The Journal of Organic Chemistry
Article
are next investigated in several reactions, including cyanosily-
lation, benzoin condensation, and transesterification. Both
solvent and temperature effects have been studied.
Scheme 1. Reversible Generation of NHCs 2 from
[NHC(H)][HCO3] Salts 4 and NHC−CO2 Adducts 3
RESULTS AND DISCUSSION
■
Synthesis and Characterization of NHC Precursors.
Both types of precursors, [NHC(H)][HCO3] salts 4 and
NHC−CO2 adducts 3, were first synthesized following slightly
modified reported procedures.26,44,45 Briefly, 1,3-diisopropyli-
midazolium-2-carboxylate (3a) and 1,3-dimesitylimidazolium-
2-carboxylate (3b) were obtained by direct carboxylation of the
corresponding free NHCs 2 in THF (Scheme 2a) and
characterized by 1H and 13C NMR spectroscopy (Figures
S1−S4). It is worth pointing out, that unlike 2a, 3a was easily
isolated by simple filtration of the reaction mixture. Moreover,
neat 3a was found to be relatively stable when exposed to air at
room temperature, whereas its related NHC had to be stored at
−30 °C under dry and inert conditions. However, 3a,b were
found to progressively hydrolyze into the corresponding
[NHC(H)][HCO3] salts 4a,b, as already reported for similar
compounds (see Scheme 1).45,52
Direct access to the imidazolium hydrogen carbonate salt
precursors ([NHC(H)][HCO3]) 4a−d was achieved by anion
metathesis of imidazol(in)ium bromides 1a−d, using KHCO3
in methanolic solution (Scheme 2b).26,54−58 Characterization of
4a−d by 1H and 13C NMR spectroscopy confirmed the
expected structures (Figures S5−S12).
This synthetic method could also be applied to benzimida-
zolium hydrogen carbonates 4e,f from the corresponding
benzimidazolium bromides 1e,f (see Figures S13−S16 for
characterization by 1H and 13C NMR spectroscopy). In
contrast, an attempt to synthesize a thiazolium hydrogen
carbonate from the thiazolium bromide salt 1g was
unsuccessful: a methanolic solution of 1g in the presence of
KHCO3 turned red within a few minutes, a viscous red-to-black
insoluble paste progressively forming. This might be explained
by the more acidic character of thiazolium salts in comparison
to their imidazolium counterparts (pKa([iPr(H)][Br], 1a) = 22
vs pKa([MeThia(H)][Br], 1g) = 14.5, in DMSO).59 It is thus
likely that, instead of anion metathesis, KHCO3 mediated the
deprotonation of 1g, generating the corresponding NHC,
which further decomposed.
transesterification,31,34 benzoin condensation,34,37 cyanosilyla-
tion,33 cyclotrimerization of isocyanates,28,33 and synthesis of
polyurethanes39,40 and polycarbonates41 by step-growth poly-
merization.
Very recently, we evidenced that imidazol(in)ium hydrogen
carbonates, denoted as [NHC(H)][HCO3], can also behave as
a source of NHCs.26 Upon formal loss of H2CO3 (Scheme 1),
not only were NHC−transition metal complexes synthesized
with no need for inert conditions but also both the NHC-
catalyzed benzoin condensation and ring-opening polymer-
ization of D,L-lactide could be achieved. Of particular interest,
such [NHC(H)][HCO3] salt precursors can be prepared by
the one-step anion metathesis of commercially available
imidazol(in)ium halides (Scheme 2b). In contrast to most
Scheme 2. (a) Synthesis of NHCs 2a,b by Deprotonation of
the Corresponding Imidazolium Bromides (1) with a Strong
Base and of Related Imidazolium-2-carboxylates 3a,b by
Subsequent Carboxylation and (b) Synthesis of
Imidazolium/Benzimidazolium Hydrogen Carbonates 4a−f
by Anion Exchange
synthetic developments to masked NHCs, the latter method
does not involve prior synthesis of the free carbene. Moreover,
imidazol(in)ium hydrogen carbonates can be conveniently
handled and stored in air without any particular precautions. In
solution, they may coexist with their “dehydrated” imidazol-
(in)ium-2-carboxylate counterparts (NHC−CO2 adducts),
depending on the nature and water content of the
solvent.26,52,53 Investigation into the mechanism of the
transformation [NHC(H)][HCO3] (4) → NHC−CO2 (3)
by DFT calculations revealed the facile and reversible formation
of the free NHC from both precursors.26
Given the small energetic barriers obtained in the gas phase
at 25 °C for NHC generation from both [NHC(H)][HCO3]
salts 4 and NHC−CO2 adducts 3, we now wish to compare the
ability of each precursor to generate the NHC 2 in solution,
especially at room temperature and in the solid state.
Characterization by thermogravimetric analysis (TGA) of
[NHC(H)][HCO3] salts 4 is first presented. TGA results
evidence that, in the solid state, most of these precursors release
H2O and CO2 molecules upon heating before they degrade.
The catalytic efficiencies of both types of precursors (3 and 4)
It is well documented that thermal degradation of
imidazolium-based ionic liquids proceeds first by dealkylation
of the imidazole ring.60,61 Likewise, studies of the thermal
behavior of some NHC−CO2 (and NHC−COS) adducts have
shown that decarboxylation (or release of COS) strongly
depends on the bulkiness of the substituents on the nitrogen
atoms.23,45 In contrast, and to the best of our knowledge, no
report has described the thermal properties of [NHC(H)]-
[HCO3] salts. The thermal behavior of 4a−f was thus
investigated simultaneously by thermogravimetric analysis
(TGA) and differential scanning calorimetry (DSC), as
summarized in Table 1.
Except for 4c,d, no phase transition was noted on the DSC
curves before these compounds started to decompose, as
observed by TGA. It should be noted, however, that DSC
results correspond to one heating cycle only. Hence,
endotherms observed by DSC may be due to a weight loss,
as seen by TGA at the same temperatures, rather than to
melting peaks. In the case of 4a, for instance, endotherms
observed between 150 and 200 °C by DSC (Figure 2a) may be
attributed to the melting of 4a, to the melting of degradation
10136
dx.doi.org/10.1021/jo301597h | J. Org. Chem. 2012, 77, 10135−10144

The Journal of Organic Chemistry
Article
Figure 1. [NHC(H)][Br] salt precursors 1, NHCs 2, NHC−CO2 adducts 3, and [NHC(H)][HCO3] salts 4 used in this study.
Table 1. Results of TGA and DSC Analyses of [NHC(H)][HCO3] salts 4a−f
a
b
c
d
e
entry
compd
Tonset (°C)
WLexp (%)
−H2Oth (%)
−CO2,th (%)
Tdeg (°C)
Tm (°C)
1
2
3
4
5
6
4a
4b
4c
4d
4e
4f
110
3.1
8.4
4.9
20.6
12.0
190
310
238
280
232
243
f
134/218
1.8/15.4
g
142
ND
77
42/126
142
108
20.6
18.0
6.2
5.3
15.3
12.9
a
b
Experimental weight loss measured on TGA curves (heating rate 10 °C/min, N2 atmosphere). Theoretical weight loss for the loss of H2O from
c
the initial compounds 4: −H2Oth (%) = MH O/M4 × 100. Theoretical weight loss for the loss of CO2 from the initial compounds 4: −CO2,th (%) =
M
2
d
e
CO /M4 × 100. Degradation temperature measured on TGA curves at the onset of their major weight losses. Experimental endothermic peak
2
f
temperature measured on DSC curves (when observed below Tonset and/or Tdeg, heating rate 10 °C/min, N2 atmosphere). Two weight losses were
observed before degradation. ND = not determined because of the overlapping of the first weight loss with the major degradation.
g
Figure 2. (a) TGA (black) and DSC (red) curves of 4a and (b) TGA (pink) and DSC (red) curves of 4d.
compounds, or to loss of H2O/CO2/alkyl chains. In contrast,
the DSC curve of 4d showed two distinct endothermic peaks
below the degradation temperature (Figure 2b), owing to its
liquid crystalline character. The peak observed at 42 °C can
indeed be ascribed to a solid/liquid crystalline transition (S/
LC), while that at 126 °C is due to a liquid crystalline/liquid
transition (LC/L). A similar thermotropic behavior has already
been reported for imidazolium halide homologues (1,3-
10137
dx.doi.org/10.1021/jo301597h | J. Org. Chem. 2012, 77, 10135−10144

The Journal of Organic Chemistry
Article
dihexadecylimidazolium bromide: TS/LC = 46 °C and TLC/L
=
expected values (WLcalc) by considering a loss of H2O + CO2.
In the case of 4e, for instance, the results were as follows: WLexp
143 °C).62−64 As for precursor 4c, its DSC analysis showed a
single melting peak at 77 °C, similarly to its imidazolium
bromide homologue 1c (Tm = 79 °C). Compounds 4c,d are
thus in their melt state before they decompose, above 142 and
280 °C, respectively.
= 20.6% vs WLcalc(H2O + CO2) = 21.5%, while for 4f WLexp
=
18.0% vs WLcalc(H2O + CO2) = 17.2%.
Of particular interest, a pale yellow solid could eventually be
obtained after heating 4f for 5 min at 130 °C, under a high
dynamic vacuum (<10−7 bar). Analysis of this compound by 1H
NMR and 13C NMR spectroscopy revealed characteristic
signals of both the free benzimidazol-2-ylidene and its dimeric
form, along with other signals likely attributable to degradation
compounds (Figures S17 and S18).65,66 Indeed, as can be seen
on the TGA-MS curve of 4f (Figure 4a), dealkylation of the
benzimidazole backbone (yellow and green curves) starts
before 4f has completely lost its CO2 moiety (blue and orange
curves). This result nonetheless evidences that, for the first
time, a free NHC can be spectroscopically detected upon
heating a benzimidazolium hydrogen carbonate precursor in the
solid state.
All [NHC(H)][HCO3] 4 precursors proved stable at least
up to 100 °C in the solid state (Table 1 and Figure 3). Typical
As anticipated, the size and the nature of the substituents on
the nitrogen atoms significantly influenced the thermal stability
of imidazolium-based [NHC(H)][HCO3] salts 4a−d. In
contrast to the case for 4b, the loss of H2O/CO2 moieties
for 4a,c,d could not be clearly distinguished from their thermal
degradation (Figure 4b, case of precursor 4c). Similar
observations were made with NHC−CO2 adducts 3a−c
undergoing decarboxylation upon heating.45 Loss of CO2
from 3b could thus be discriminated from the degradation
phenomenon. In contrast, 3a,c showed concomitant degrada-
tion and loss of CO2. This difference may be explained by the
aromatic nature of the mesityl groups in 3b, in comparison to
the primary or secondary alkyl substituents in 3c and 3a.
In addition, the initial weight loss of 4a was found to be
lower than the expected value (3.0% vs 8.4% for −H2Oth or
29.0% for −(H2O + CO2)th), which might be explained by the
loss of adsorbed water. Likewise, a decomposition in stages was
not observed for 4c,d. The longer the substituents of the alkyl
chains, the higher the onset temperature: Tonset = 142 vs 280 °C
for 4c,d, respectively. In contrast, due to the higher stability of
secondary cations in comparison to primary ones (iPr+ vs Me+
or Et+), 4a exhibited a lower degradation temperature than 4c,
Figure 3. TGA curves of the thermal degradation of [NHC(H)]-
[HCO3] salts 4a−f.
thermograms obtained by TGA/MS are provided in Figure 4
(in the case of precursors 4c,f). With the exception of 4d,
featuring dodecyl chains, all compounds 4 showed a minor
weight loss prior to their degradation at higher temperature.
In particular, an initial weight loss is observed at 142 and 108
°C, respectively, for both benzimidazolium hydrogen carbo-
nates, 4e,f, followed by their degradation at 230−240 °C. The
initial weight loss can be ascribed to the loss of both H2O and
CO2, as supported by TGA coupled with mass spectrometry
(MS) analysis. In both cases, the loss of H2O molecules
(leading to mass to charge ratios at m/z 18 and 17) occurred at
a slightly lower temperature than that due to the departure of
CO2 (at m/z 44 and 22). In addition, experimental weight loss
(WLexp) results for 4e,f were found to be very close to the
Figure 4. TGA-MS analyses (heating rate 5 °C/min, Ar atmosphere) of (a) 4f (the dotted line denotes the end of the H2O + CO2 loss and the
beginning of loss of alkyl chains) and (b) 4c. log(ion current) was recorded for several mass to charge ratios corresponding to the loss of H2O (pink
and purple), CO2 (orange and blue), and alkyl chains (yellow and green).
10138
dx.doi.org/10.1021/jo301597h | J. Org. Chem. 2012, 77, 10135−10144

The Journal of Organic Chemistry
Article
Table 2. Cyanosilylation of Benzaldehyde or Benzophenone with TMSCN in THF in the Presence of 2a, 3a, or 4a,c−f
a
b
entry
1
catalyst
amt of catalyst (mol %)
0.02
substrate
T (°C)
t (min)
conversn (%)
TOF (h−1
)
2a
benzaldehyde
25
80
25
80
25
80
25
25
25
25
25
80
25
80
10
10
10
10
30
30
10
60
10
10
20
20
20
20
56
95
16 800
28 500
25 800
30 000
8 600
2
3a
4a
0.02
0.02
benzaldehyde
benzaldehyde
86
100
86
c
3
100
93
10 000
5 580
c
4
4c
4d
4e
4f
0.1
0.1
0.1
0.1
0.02
benzaldehyde
benzaldehyde
benzaldehyde
benzaldehyde
benzophenone
c
5
100
99
1 000
c
6
5 940
c
7
100
100
100
100
100
6 000
8
2a
15 000
15 000
15 000
15 000
9
3a
0.02
benzophenone
a
Conversion was calculated by 1H NMR in CDCl3 (Figures S19 and S20) on comparison of the integral value of the aldehyde signal of benzaldehyde
b
c
(δ 10 ppm) to that of the −CH− cyanide product (δ 5.5 ppm). Turnover frequency. 3 Å molecular sieves were added.
consistent with data reported for the imidazolium bromide
homologues 1a,c.60,61,67
It is worth mentioning that compounds 4with the
exception of 4dwere hardly soluble in THF, which was the
main solvent used for the organocatalyzed reactions of the
present study. Complete solubilization of the precursors did
however occur upon addition of the substrate (e.g.,
benzaldehyde, vinyl acetate, and benzyl alcohol; see below).
Catalytic activities of selected precursors (3a and 4a,c−f,
Figure 1) were next evaluated in several molecular reactions,
including cyanosilylation, benzoin condensation, and trans-
esterification reactions. All these reactions were carried out
under different experimental conditions so as to investigate
both solvent and temperature effects. Precursors 4a−f being
thought to generate free NHCs 2 by releasing H2O and CO2
(=H2CO3; Scheme 1),26 molecular sieves (3 Å) were
systematically added in the reaction vessel. Control experiments
were also implemented from the corresponding bare NHC 2a,
which was purposely synthesized. Only 3a and 4a were tested
as precatalysts for transesterification reactions, in order to
highlight the solvent effect on the ability of imidazolium
hydrogen carbonates and their NHC−CO2 counterparts to
generate the corresponding NHCs.
Compound 4b exhibited the highest thermal stability in this
series, which may be attributed to the aromatic nature of its aryl
substituents. Similarly to benzimidazolium hydrogen carbo-
nates, TGA curves of 4b showed a concomitant dehydration/
decarboxylation (WLexp = 15.4% vs WLcalc(H2O + CO2) =
16.9%), corresponding to the weight loss observed at 218 °C,
followed by thermal degradation above 310 °C. The initial
minor weight loss (134 °C, 1.8%) is most probably due to
adsorbed water.
In summary, the solid-state behavior of [NHC(H)][HCO3]
precursors strongly depends on the nature of the azolium
backbone and of the substituents on the nitrogen atoms.
Benzimidazolium backbones, and more bulky substituents, lead
to concomitant or stepwise losses of H2O and CO2 before the
degradation stage, while such changes in the thermal properties
cannot not be differentiated for compounds featuring short
alkyl substituents.
The solution behaviors of the imidazolium/benzimidazolium
hydrogen carbonates 4a−f were also studied. As previously
reported,26 [NHC(H)][HCO3] and their NHC−CO2 counter-
parts can coexist in solution, in particular as a function of the
water content of the solvent analysis. For instance, mixtures of
NHC−CO2 adducts and [NHC(H)][HCO3] salts were
Benzaldehyde and Benzophenone Cyanosilylation
Reaction. Hermann et al.68 and then both Song et al.69 and
Suzuki et al.70 have employed NHCs as organocatalysts for
cyanosilylation reactions of carbonyl-containing substrates.
When aldehydes or ketones react with trimethylsilyl cyanide
(TMSCN) in the presence of catalytic loadings of NHC, high
conversions are obtained. Although the mechanism that may
involve either the activation of the carbonyl group or the silicon
atom remains unclear, this is a general method to access α-
hydroxyamines, α-hydroxy acids, and α-hydroxyaldehydes after
derivatization.71,72 Polymer-supported versions of NHC−CO2
adducts have also been reported for organocatalyzed
cyanosilylation reactions.33 Here, we examined the catalytic
1
observed, roughly in a molar ratio of 1:4, in the H and 13C
NMR spectra of imidazolium hydrogen carbonates 4a−c in
DMSO-d6. Analyses in MeOD showed a single population of
signals attributed to the pure [NHC(H)][HCO3] salt. In
contrast, NMR spectra of imidazolium hydrogen carbonates 4d
and benzimidazolium hydrogen carbonates 4e,f revealed the
presence of the sole [NHC(H)][HCO3] compound, irre-
spective of the solvent analysis (DMSO-d6, MeOD, or CD2Cl2).
10139
dx.doi.org/10.1021/jo301597h | J. Org. Chem. 2012, 77, 10135−10144

The Journal of Organic Chemistry
Article
activity of both molecular imidazolium hydrogen carbonates
4a,c−f and NHC−CO2 adduct 3a toward the cyanosilylation
reaction of benzaldehyde or benzophenone with TMSCN, both
at room temperature and at 80 °C. The most significant results
are collected in Table 2.
In the case of benzaldehyde as substrate, 0.02 mol % of 3a
gave almost quantitative conversions, at both room temperature
and 80 °C, only after 10 min of reaction (entry 2). A control
experiment employing NHC 2a gave similar results (entry 1).
Interestingly, the use of 4a as NHC precursor only divided the
turnover frequency (TOF) by a factor of 3 (entry 3), attesting
to the fact that both 3a and 4a could generate NHC 2a at room
temperature and 80 °C.
(imidazolium salts),37 have been used to catalyze the benzoin
condensation, the in situ generation of poly(NHCs) being
triggered by a temperature increase or the addition of a base,
respectively.
Masked NHCs 3a and 4a,c−f were thus tested as organic
precatalysts for the benzoin condensation reaction (Table 3). In
Table 3. Benzoin Condensation Reaction in THF using 2a,
3a, or 4a−f
In addition to our previous DFT calculations on the
transformation 4a → 2a,26 the reaction 3a → 2a was also
investigated computationally (see the Supporting Information).
As expected, a smaller activation barrier was calculated for the
generation of free NHC from 3a (6.1 kcal mol−1), as compared
to 10.0 kcal mol−1 in the case of the transformation 4a → 2a.
The same trend was noted in the case of 4b → 2b and 3b →
2b, a 2.4 kcal mol−1energy difference in activation barrier being
calculated (7.5 and 5.1 kcal mol−1 for 4b → 2b and 3b → 2b,
respectively). Moreover, all these transformations were found
to be reversible (see the Supporting Information). These results
are compatible with the facile generation of 2 from both
precursors 3 and 4, the formation of 2 being more favorable in
the case of 3 → 2. Moreover, the reversibility of both reactions
3 → 2 and 4 → 2 is in agreement with the absence of an
induction period for the catalyst to be generated.
Imidazolium hydrogen carbonates 4c,d were also efficient in
triggering the cyanosilylation of benzaldehyde at room
temperature (entries 4 and 5), though 4d, bearing more
bulky substituents gave a lower TOF. Benzimidazole-type salts
4e,f were also successfully used as precatalysts at room
temperature (entries 6 and 7) witnessing, for the first time,
that such precursors could generate corresponding NHCs in
solution at room temperature.
Song et al. have noticed that NHC-catalyzed cyanosilylation
of ketonescompared to aldehydesrequires longer reaction
times using 1,3-di-tert-butyl-imidazol-2-ylidene as catalyst, the
reaction being favored in DMF in comparison to THF.69 Here,
both 2a and 3a catalyzed the cyanosilylation of benzophenone
quantitatively, after 20 min in THF, both at room and at higher
temperatures (entries 8 and 9). Hence, high temperatures are
not needed to achieve high yields, suggesting in that case again
that solvation effects, associated with the favorable kinetics and
thermodynamics of the transformations 3a → 2a and 4a →
2a,26 can drive the generation of the NHC catalyst from
NHC−CO2 adducts and [NHC(H)][HCO3] precursors.
Benzoin Condensation Reaction. Pioneer works on
NHC-catalyzed benzoin condensation by Breslow date back
to 1958.73 Since then, various NHCs, including thiazol-2-
ylidenes, imidazol(in)-2-ylidenes, benzimidazol-2-ylidenes, and
triazol-5-ylidenes, as well as NHC−COS adducts23 have served
to catalyze this C−C bond forming reaction that transforms
aromatic aldehyde molecules into benzoin derivatives, the
intermediate of the reaction being a resonance-stabilized
enaminol-type intermediate called the “Breslow intermediate”.5
Application of this elementary reaction to polymer synthesis,
that is, the NHC-catalyzed step-growth polymerization of the
aromatic bis-aldehyde terephthalaldehyde to form polybenzoin,
has also been reported.74 In addition, polymer-supported NHC
precursors, such as poly(NHC−CO2) adducts34 and poly-
amt of
catalyst (mol
%)
b
T
t
conversn TOF
a
entry catalyst
solvent (°C) (h)
(%)
(h−1
)
1
2a
1
THF
25
80
25
24
24
21
48
24
21
48
24
24
24
24
24
24
24
24
24
24
52
72
28
45
72
6
2.17
3
2
3a
1
THF
1.3
0.94
3
80
25
c
3
4a
3
THF
0.09
0.08
1
11
72
88
44
72
17
79
0
80
60
25
80
25
80
25
80
25
80
426
4b
4c
10
10
THF
THF
0.37
0.2
0.3
0.07
0.3
-
c
5
c
6
4d
4e
4f
10
10
10
THF
THF
c
7
0
-
c
8
THF
71
75
0.3
0.3
a
1
Conversion was calculated by H NMR in CDCl3 (Figure S21) on
comparison of the integral value of the aldehyde signal of
benzaldehyde (δ 10 ppm) with that of the −CH− benzoin signal (δ
6 ppm). Turnover frequency. 3 Å molecular sieves were added.
b
c
contrast to cyanosilylation, the temperature has a great impact
on conversion. For instance, a temperature increase, from 25 to
80 °C, led to more than 2 times higher yields upon using 3a as
precursor (Table 3, entry 2). Precatalyst 3a gave approximately
30% conversion after 24 h at room temperature, which was 2
times lower than the result obtained with its free NHC
homologue 2a. The use of 4a at room temperature resulted in
even poorer conversions. At 80 °C, similar kinetics and yields
were noted both with the free NHC 2a and the NHC−CO2
adduct 3a, attesting to the quantitative generation of the NHC
at this temperature. As observed in the cyanosilylation, reaction
using 4a led to similar conversions but divided the TOF by a
factor of 3 at 80 °C.
While high amounts of free NHCs are generally required to
achieve high yields (up to 10−30 mol % for asymmetric
condensations),5 only 1 mol % of precursor 2a gave a rather
good catalytic efficiency at room temperature (entry 1).
However, complete conversion was not achieved at 80 °C
using 2a, 3a, or 4a, after 24 h, even with higher amounts of
10140
dx.doi.org/10.1021/jo301597h | J. Org. Chem. 2012, 77, 10135−10144

The Journal of Organic Chemistry
Article
Table 4. Transesterification between Benzyl Alcohol and Vinyl Acetate using 2a, 3a, or 4a
a
b
entry
1
catalyst
amt of catalyst (mol %)
0.5
solvent
THF
T (°C)
t (min)
conversn (%)
TOF (h−1
)
3a
25
80
25
80
25
80
25
80
25
80
25
80
25
80
25
80
25
80
25
80
25
80
25
80
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
100
100
81
93
30
60
100
100
37
77
94
100
4
1200
1200
980
2c
3
4a
3a
2a
3a
2a
3a
2a
3a
2a
3a
2a
0.5
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
THF
1120
1800
3600
6000
6000
2220
4620
5640
6000
240
THF
4
THF
5
toluene
toluene
cyclohexane
cyclohexane
DMSO
DMSO
none
6
7
15
79
100
0
900
8
4740
6000
9
7
420
300
10
11
12
5
19
7
1140
420
10
32
30
600
0.1
none
1920
1800
a
1
Conversion was calculated by H NMR in CDCl3 (Figure S22) on comparison of the integral value of the −CH2− benzyl alcohol signal (δ 4.5
b
c
ppm) to that of the −CH2− benzyl acetate signal (δ 5 ppm). conversn (%) = (Ibenzyl acetate/(Ibenzyl acetate+ Ibenzyl alcohol)) × 100. Turnover frequency. 3
Å molecular sieves were added.
(pre)catalysts, presumably due to the temperature sensitivity of
2a. Similar results were previously observed upon using 10 mol
% of 4b as NHC precursor (entry 4).26 Changing the
substituents of the imidazole ring to less bulky (4c) or to
more bulky (4d) ones did not improve the TOF values.
Benzimidazolium hydrogen carbonate 4f exhibited catalytic
efficiency similar to that of 4a−d, whereas 4e showed no
activity at all. Indeed, the steric hindrance of tBu substituents as
compared to cyclohexyl substituents (4e vs 4f) may slow the
NHC generation by a solvation effect. Masked NHCs 3a and
4a−d,f are nonetheless capable of catalyzing the benzoin
condensation, by generating NHCs in solution, better results
being achieved at 80 °C.
purposes since, to the best of our knowledge, this particular
NHC has never been tested for transesterification reactions.
When 0.5 mol % of 1,3-diisopropylimidazolium-2-carboxylate
(3a) was used in THF, quantitative yields were achieved within
only 10 min, at 80 °C and, more strikingly, at room
temperature as well (Table 4, entry 1). Interestingly, 4a could
also serve as a precatalyst of transesterification at room
temperature, although a 2 times lower TOF value was noted
in comparison to its carboxylate homologue 3a under the same
conditions (entry 2 vs entry 3). These results are again
consistent with the facile generation of NHC from both types
of precursors at room temperature. Lower TOF values but high
conversions observed with 3a and 4a in comparison to free
NHC 2a are also in agreement with the fact that 2a has to be
generated from its masked forms, prior to entering the catalytic
cycle.
At a precatalyst 3a concentration of 0.1 mol %, increasing the
temperature from 25 to 80 °C gave 2 times higher conversions,
regardless of the nature of the solvent, while almost no
temperature effect was noted with the bare NHC 2a. For
example, TOF values increased from 1800 to 3600 h−1 in THF,
from room temperature to 80 °C, using 3a as precatalyst (entry
3). These results support that NHC 2a is the catalytically active
species, the rate of CO2 release from 3a being increased by
increasing the temperature.
Transesterification Reaction. Free NHCs,75,76 and some
masked NHCs such as imidazol(in)ium-2-thiocarboxylates23 or
polymer-supported versions of imidazolium-2-carboxylates,34
are known to efficiently catalyze transesterification reactions.
However, neither NHC−CO2 adducts nor [NHC(H)][HCO3]
salt precursors have been evaluated as precatalysts for such
reactions at room temperature. The reaction between benzyl
alcohol and vinyl acetate was thus performed in the presence of
3a or 4a, in nonpolar solvents such as cyclohexane and toluene
and in polar solvents such as THF and DMSO, both at room
temperature and at 80 °C. Results obtained with the two types
of NHC precursors are summarized in Table 4. The catalytic
potential of free NHC 2a was also evaluated for comparison
Transesterification reactions were next evaluated both in
nonpolar (entries 5 and 7) and polar solvents (entry 9). Lower
10141
dx.doi.org/10.1021/jo301597h | J. Org. Chem. 2012, 77, 10135−10144

The Journal of Organic Chemistry
Article
was distilled over CaH2 before use following a minimum of 4 h of
reflux. All reagents were purchased from commercial sources with
purity ≥99% unless otherwise stated. TMSCN was used as received.
Vinyl acetate (95%) was purified by distillation over CaH2 and stored
at 0 °C. Benzaldehyde and benzyl alcohol were purified by fractional
distillation. Benzyl alcohol was kept over molecular sieves.
Benzophenone was recrystallized twice in dry warm cyclohexane and
stored in the glovebox.
Methods. NMR spectra were recorded on a 400 MHz
spectrometer. The mass spectra (ESI) instrument is equipped with
an ESI source, and spectra were recorded in the positive mode. The
electrospray needle was maintained at 4500 V and operated at room
temperature. Samples were introduced by injection through a 10 μL
sample loop into a 200 μL/min flow of methanol from the LC pump.
TGA experiments were performed with alumina crucibles. A constant
heating rate of 10 °C/min and gas purging (N2) at a flow rate of 100
mL/min were used for all experiments. The amount of samples used
for TGA was between 9 and 10 mg. The resulting Tonset and Tdeg
values were determined from the step tangent. Thermogravimetric
techniques coupled with mass spectrometry were employed to analyze
the thermal decomposition products in situ. The TGA−MS
experiments were performed from room temperature to 650 °C
with a heating rate of 5 °C/min in argon gas purging. The amount of
sample was ca. 12 mg. The signal intensity of gaseous products during
the heating was measured in situ using mass spectroscopy. DSC
experiments were performed with aluminum sealed pans. A constant
heating/cooling rate of 10 °C/min and gas purging (N2) at a flow rate
of 100 mL/min was used for all experiments.
Synthesis of Imidazolium-2-carboxylates. 1,3-Diisopropylimi-
dazolium bromide (500 mg, 2.14 mmol), synthesized using a
procedure already described,77 and 1,3-dimesitylimidazolium chloride
(50 mg, 0.147 mmol), purchased from a commercial source, were first
deprotonated by exposure to 1 equiv of KHMDS or 1 equiv of NaH/
cat. tBuOK in dry THF. After being stirred overnight, solutions were
filtered under inert conditions to remove metallic salts. CO2
(99.995%) was then introduced into the flask. White precipitates
were observed after a few seconds. The solutions were further exposed
to CO2 (1 atm) for 20 min before being filtered in the glovebox. The
white powders were washed with dry Et2O and dried under vacuum,
affording clean 1,3-diisopropylimidazolium-2-carboxylate (3a; yield
295 mg, 70%) and 1,3-dimesitylimidazolium-2-carboxylate (3b; yield
31 mg, 60%). NMR spectra were in agreement with those already
reported in the literature (Figures S1−S4).45
conversions were noted in THF in comparison to toluene, from
both NHC precursor 3a and free NHC 2a (entries 3 and 4 vs 5
and 6). In addition, NHC−CO2 3a gave almost no conversion
in cyclohexane, most probably due to its insolubility in the
reaction medium. DMSO gave lower yields than all the other
tested solvents despite its high polarity, irrespective of the
catalyst (entries 9 and 10). A control experiment in the absence
of solvent was also attempted, which led to only poor yields,
both at room temperature and at 80 °C, in the presence of
either 2a or 3a (entries 11 and 12). Consistent with
observations made by Delaude et al.51 regarding the NHC
transfer to metal complexes, the solvent may thus help the
decarboxylation of NHC−CO2 adducts, most probably by a
solvation effect. The catalytic activity of 2a in the tested
solvents was thus DMSO < none ≪ cyclohexane ≈ toluene ≈
THF, while in the case of 3a, it varied as DMSO ≈ none ≈
cyclohexane ≪ toluene ≈ THF. The influence of the solvent
on the transesterification reaction was found to be similar upon
using NHC−CO2 3a or its corresponding bare NHC 2a, except
when cyclohexane was employed, owing to the poor solubility
of 3a in this solvent.
CONCLUSION
■
Various (benz)imidazolium hydrogen carbonates, [NHC(H)]-
[HCO3], and imidazolium-2-carboxylates, NHC−CO2 adducts,
are shown to generate related N-heterocyclic carbenes (NHCs)
in situ, allowing for a use in organocatalysis for different
reactions. Formation of the free carbene upon heating a
benzimidazolium-type hydrogen carbonate precursor has been
evidenced for the first time by NMR spectroscopy. In situ
generation of NHC can be readily achieved in solution from all
types of precursors at room temperature, presumably by a
simple solvation effect. Solvents favoring the NHC generation
include mainly THF and toluene, presumably owing to a better
solvation of both the transition state and the free NHC, relative
to the charged NHC−CO2 adduct and the [NHC(H)][HCO3]
salt precursors.
Transesterification reactions can be triggered from NHC−
CO2 adducts, either in THF or in toluene, while [NHC(H)]-
[HCO3] precursors exhibits a good catalytic efficiency in THF.
Both categories of NHC precursors, including those of
benzimidazole type, proved efficient for the cyanosilylation
reaction, giving excellent TOFs at room temperature in THF.
They also allow catalyzing the benzoin condensation,
preferentially at 80 °C.
Although [NHC(H)][HCO3] precursors exhibit a lower
catalytic activity in all reactions tested, a major advantage of
these salts lies in their air stability in the solid state, combined
with an easy one-step access based on anion metathesis. Thus, a
variety of (benz)imidazol(in)ium hydrogen carbonates should
be accessible by simple anion exchange, affording versatile
precursors for various NHC-catalyzed reactions, upon formal
loss of H2CO3. Experiments are currently ongoing in our group
to highlight the potential of such precatalysts in polymer
chemistry. Polymer-supported versions of these [NHC(H)]-
[HCO3] and their use in organocatalysis are also underway.
Synthesis of Imidazolium Hydrogen Carbonates. 1,3-
Diisopropylimidazolium bromide (1a; 500 mg, 2.14 mmol), 1,3-
didodecylimidazolium bromide (1d; 500 mg, 1.03 mmol), synthesized
using a procedure already described,77 1,3-dimesitylimidazolium
chloride (1b; 500 mg, 1.47 mmol), and 1-methyl-3-ethylimidazolium
chloride (1c; 500 mg, 3.41 mmol), purchased from commercial
sources, were dried under vacuum for several hours and then dissolved
in 5 mL of MeOH. A 1.02 equiv portion of dry KHCO3 (0.218 mg,
2.18 mmol with 1a; 0.150 mg, 1.50 mmol with 1b; 0.348 mg, 3.48
mmol with 1c; 0.105 mg, 1.05 mmol with 1d) was added. The solution
was stirred under an inert atmosphere for 2 days to allow complete
precipitation of KBr or KCl. After filtration of the solutions over Celite
to remove the mentioned salt, MeOH was evaporated under vacuum.
The powders were washed with acetone and dried under vacuum
(yield: 4a, 320 mg, 70%; 4b, 402 mg, 75%, 4c, 354 mg, 60%, 4d, 313
mg, 65%). NMR spectra of 4a,b were in agreement with those already
reported in the literature (Figures S5−S8).26
In MeOD (Figures S9 and 10), 4c was the only compound
observed, while in DMSO-d6, 4c equilibrates with 3c in a 1:2.5 ratio, in
1
favor of 3c. H NMR (400 MHz, MeOD): δ 1.62 (t, J = 7.2 Hz, 3H,
NCH2CH3), 3.93 (s, 3H, NCH3), 4.26 (q, J = 7.2 Hz, 2H, NCH2CH3),
7.57 (s, 1H, CHCH), 7.64 (s, 1H, CHCH). The N2CH and
HCO3− protons could not be observed, while the integral of CHCH
was lowered due to rapid exchange with the deuterated solvent on the
NMR time scale. 13C NMR (100 MHz, MeOD): δ 15.6, 36.4, 46.0,
123.2, 124.8, 137.5, 161.4. Mp: 76−78 °C. HRMS (MALDI+): m/z
calcd for C6H11N2 [M]+ 111.0916, found 111.0913.
EXPERIMENTAL SECTION
■
Materials. All experiments were performed under an inert
atmosphere using standard Schlenk techniques. Dry, oxygen-free
solvents and reagents were employed. THF was distilled over Na/
benzophenone, MeOH was distilled over Na, and toluene and
cyclohexane were distilled over polystyryllithium prior to use. DMSO
10142
dx.doi.org/10.1021/jo301597h | J. Org. Chem. 2012, 77, 10135−10144

The Journal of Organic Chemistry
Article
4d was the only compound observed in MeOD and DMSO-d6
(Figures S11 and 12). 1H NMR (400 MHz, DMSO-d6): δ 0.85 (t, J =
6.6 Hz, 6H, CH2CH3), 1.23 (br, 36H, (CH2)9CH3), 1.78 (quin, J = 6.8
Hz, 4H, NCH2CH2), 4.16 (t, J = 7.2 Hz, 4H, NCH2CH2), 7.82 (s, 2H,
CHCH), 9.27 (s, 1H, NCHN). 13C NMR (100 MHz, DMSO-d6): δ
13.9, 22.1, 25.4, 28.3, 28.7, 28.8, 28.9, 29.0, 29.0, 29.2, 31.3, 48.8,
122.5, 136.0, 154.0. Mp: 40−42 °C (solid/liquid crystalline transition).
HRMS (MALDI+): m/z calcd for C27H53N2 [M]+ 405.4203, found
405.4200.
Synthesis of Benzimidazolium Hydrogen Carbonates. 1,3-Di-
tert-butyl-benzimidazolium chloride (1e; 500 mg, 1.71 mmol) and 1,3-
dicyclohexylbenzimidazolium bromide (1f; 500 mg, 1.57 mmol),
purchased from commercial sources, were dried under vacuum for
several hours and then dissolved in 5 mL of MeOH. A 1.02 equiv
portion of dry KHCO3 (0.174 mg, 1.74 mmol and 0.160 mg, 1.60
mmol, respectively) was added. The solution was stirred under an inert
atmosphere for 2 days to allow complete precipitation of KBr or KCl.
After filtration of the solutions over Celite to remove the mentioned
salt, MeOH was evaporated under vacuum. The powders were washed
with acetone and dried under vacuum (yield 209 mg, 40% and 436 mg,
80%, respectively). The obtained products were soluble in MeOD,
DMSO, and CD2Cl2, a single population of signals being observed
whatever the analysis solvent.
mixed in 5 mL of THF. A 4.5 mg portion (25 μmol) of precatalyst 3a
was finally added. After a few seconds at room temperature, complete
solubilization of the catalyst was observed and the solution was divided
1
as described above. After 10 min, conversion was calculated by H
NMR in CDCl3 (Figure S22) to compare the integral value of the
−CH2− benzyl alcohol signal (δ 4.5 ppm) to that of the −CH2−
benzyl acetate signal (δ 5 ppm).
ASSOCIATED CONTENT
■
S
* Supporting Information
Figures, text, and tables giving H and 13C NMR spectra of
1
(benz)imidazolium hydrogen carbonates, H and 13C NMR
1
spectra to evidence the NHC formation after heating of 4f in
the solid state, 1H NMR spectra for the calculation of
conversion of the tested reactions, and optimized structures
of 3 and 2 with the corresponding energies. This material is
available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: taton@enscbp.fr.
4e (Figures S13 and S14): 1H NMR (400 MHz, CD2Cl2) δ 1.97 (s,
18H, NC(CH3)3), 7.62 (dd, J = 3.2 Hz, 6.4 Hz, 2H, Hbenz), 8.00 (dd, J
= 3.2 Hz, 6.8 Hz, 2H, Hbenz), 9.82 (s, 1H, N2CH); the HCO3− proton
could not be observed due to rapid exchange with the deuterated
solvent on the NMR time scale; 13C NMR (100 MHz, CD2Cl2) δ 29.2,
62.3, 116.8, 126.0, 132.0, 140.6, 160.1; mp >152 °C dec; HRMS
(MALDI+) m/z calcd for C15H23N2 [M]+ 231.1859, found 231.1855.
4f (Figures S15 and S16): 1H NMR (400 MHz, CD2Cl2) δ 1.27 (m,
2H, CH2,cyclohexyl), 1.52 (m, 4H, CH2,cyclohexyl), 1.74 (m, 2H,
CH2,cyclohexyl), 1.92 (m, 8H, CH2,cyclohexyl), 2.17 (m, 4H, CH2,cyclohexyl),
4.74 (m, 2H, NCHcyclohexyl), 7.68 (dd, J = 2.6 Hz, 6.6 Hz, 2H, Hbenz),
8.19 (dd, J = 3.2 Hz, 7.2 Hz, 2H, Hbenz), 10.00 (s, 1H, N2CH); the
HCO3− proton could not be observed due to rapid exchange with the
deuterated solvent on the NMR time scale; 13C NMR (100 MHz,
CD2Cl2) δ 24.6, 24.8, 31.8, 48.6, 50.8, 57.0, 114.0, 126.4, 130.6, 139.3,
155.6; mp >140 °C dec; HRMS (MALDI+) m/z calcd for C19H27N2
[M]+ 283.2168, found 283.2177.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We are grateful to the CNRS and Reg
support. Odile Babot (Institut des Sciences Molec
5255, Bordeaux) is acknowledged for the TGA-MS experi-
ments. Christelle Absalon (Cesamo, Bordeaux) is acknowl-
́
ion Aquitaine for financial
ulaires, UMR
́
́
edged for the MALDI-ToF ESI MS experiments. Cedric Le
Coz (LCPO, UMR 5629, Bordeaux) is acknowledged for
fruitful discussions on thermal analysis. Part of the theoretical
work was granted access to the HPC resources of IDRIS under
allocation 2012 (i2012080045) made by the GENCI (Grand
Equipement National de Calcul Intensif).
Catalytic Tests. All catalytic tests were carried out under a dry and
inert atmosphere using Schlenk equipment. To provide exactly similar
conditions for experiments performed at room temperature and at 80
°C, one reaction medium was first prepared mixing catalyst and
reagents. This solution was then equally separated into two different
flasks, one being kept at room temperature and the other being
transferred in an 80 °C preset oil bath. All reactions were stopped by
contact with air, and small samples were immediately withdrawn for
conversion analysis by NMR in CDCl3.
Cyanosilylation (See Table 2). In a typical reaction, 0.5 mL (5
mmol) of benzaldehyde and 0.75 mL of TMSCN (6 mmol) were
mixed with 5 mL of THF. A 23.5 μL portion (1 μmol) of a 42.5 mM
solution (THF/DMSO 5/1) of 3a was finally added. After a few
seconds of homogenization at room temperature, the solution was
divided as described above. Conversion was calculated by 1H NMR in
CDCl3 (Figures S19 and S20) to compare the integral value of the
aldehyde signal of benzaldehyde (δ 10 ppm) to that of the −CH−
cyanide product (δ 5.5 ppm).
Benzoin Condensation (See Table 3). In a typical reaction, 31.5 mg
(150 μmol) of 4a and molecular sieves were introduced into a Schlenk
tube. The powder was submitted to 30 min of vacuum and finally three
Ar/vacuum cycles. A 5 mL portion of THF and then 0.5 mL (5 mmol)
of benzaldehyde were added. After a few seconds at room temperature,
complete solubilization of the catalyst was observed and the solution
was divided as described above. Conversion was calculated by 1H
NMR in CDCl3 (Figure S21) to compare the integral value of the
aldehyde signal of benzaldehyde (δ 10 ppm) with that of the −CH−
benzoin signal (δ 6 ppm).
REFERENCES
(1) Herrmann, W. A. Angew. Chem., Int. Ed. 2002, 41, 1290.
(2) Crabtree, R. H. Coord. Chem. Rev. 2007, 251, 595.
■
(3) Díez-Gonzaal
109, 3612.
(4) Marion, N.; Díez-Gonzal
2007, 46, 2988.
́
ez, S.; Marion, N.; Nolan, S. P. Chem. Rev. 2009,
́
ez, S.; Nolan, S. P. Angew. Chem., Int. Ed.
(5) Enders, D.; Niemeier, O.; Henseler, A. Chem. Rev. 2007, 107,
5606.
(6) Moore, J.; Rovis, T. In Asymmetric Organocatalysis; List, B., Ed.;
Springer: Berlin/Heidelberg, Germany 2009; Vol. 291, p 77.
(7) Glorius, F. In N-Heterocyclic Carbenes in Transition Metal
Catalysis; Springer: Berlin/Heidelberg, Germany, 2007; Vol. 21, p 1.
(8) Droge, T.; Glorius, F. Angew. Chem., Int. Ed. 2010, 49, 6940.
̈
(9) Kiesewetter, M. K.; Shin, E. J.; Hedrick, J. L.; Waymouth, R. M.
Macromolecules 2010, 43, 2093.
(10) Fev
̀
re, M.; Vignolle, J.; Gnanou, Y.; Taton, D., In Polymer
Science: A Comprehensive Reference; Matyjaszewski, K., Moller, M., Eds.;
̈
Elsevier: Amsterdam, 2012; p 67.
(11) Denk, M. K.; Rodezno, J. M.; Gupta, S.; Lough, A. J. J.
Organomet. Chem. 2001, 617−618, 242.
(12) Holloc
́
zki, O.; Terleczky, P.; Szieberth, D.; Mourgas, G.; Gudat,
D.; Nyulas
́
zi, L. J. Am. Chem. Soc. 2010, 133, 780.
(13) Wang, H. M. J.; Lin, I. J. B. Organometallics 1998, 17, 972.
(14) Lin, J. C. Y.; Huang, R. T. W.; Lee, C. S.; Bhattacharyya, A.;
Hwang, W. S.; Lin, I. J. B. Chem. Rev. 2009, 109, 3561.
Transesterification (See Table 4). In a typical reaction, 0.5 mL of
(15) Enders, D.; Breuer, K.; Raabe, G.; Runsink, J.; Teles, J. H.;
benzyl alcohol (5 mmol) and 0.55 mL (6 mmol) of vinyl acetate were
Melder, J.-P.; Ebel, K.; Brode, S. Angew. Chem., Int. Ed. 1995, 34, 1021.
10143
dx.doi.org/10.1021/jo301597h | J. Org. Chem. 2012, 77, 10135−10144

The Journal of Organic Chemistry
Article
(16) Coulembier, O.; Dove, A. P.; Pratt, R. C.; Sentman, A. C.;
Culkin, D. A.; Mespouille, L.; Dubois, P.; Waymouth, R. M.; Hedrick,
J. L. Angew. Chem., Int. Ed. 2005, 44, 4964.
(17) Csihony, S.; Culkin, D. A.; Sentman, A. C.; Dove, A. P.;
Waymouth, R. M.; Hedrick, J. L. J. Am. Chem. Soc. 2005, 127, 9079.
(18) Coulembier, O.; Lohmeijer, B. G. G.; Dove, A. P.; Pratt, R. C.;
Mespouille, L.; Culkin, D. A.; Benight, S. J.; Dubois, P.; Waymouth, R.
M.; Hedrick, J. L. Macromolecules 2006, 39, 5617.
(19) Nyce, G. W.; Csihony, S.; Waymouth, R. M.; Hedrick, J. L.
Chem. Eur. J. 2004, 10, 4073.
(51) Sauvage, X.; Demonceau, A.; Delaude, L. Macromol. Symp. 2010,
293, 28.
(52) Bridges, N. J.; Hines, C. C.; Smiglak, M.; Rogers, R. D. Chem.
Eur. J. 2007, 13, 5207.
(53) Rijksen, C.; Rogers, R. D. J. Org. Chem. 2008, 73, 5582.
(54) Liu, Y.; Li, M.; Lu, Y.; Gao, G.-H.; Yang, Q.; He, M.-Y. Catal.
Commun. 2006, 7, 985.
(55) Ratel, M.; Branca, M.; Breault-Turcot, J.; Zhao, S. S.; Chaurand,
P.; Schmitzer, A. R.; Masson, J.-F. Chem. Commun. 2011, 47.
(56) Choi, Y.-S.; Shim, Y. N.; Lee, J.; Yoon, J. H.; Hong, C. S.;
Cheong, M.; Kim, H. S.; Jang, H. G.; Lee, J. S. Appl. Catal., A 2011,
404, 87.
(20) Coulembier, O.; Delva, X.; Hedrick, J. L.; Waymouth, R. M.;
Dubois, P. Macromolecules 2007, 40, 8560.
(57) Ye, Y.; Elabd, Y. A. Macromolecules 2011, 44, 8494.
(58) Yue, Q. F.; Wang, C. X.; Zhang, L. N.; Ni, Y.; Jin, Y. X. Polym.
Degrad. Stab. 2011, 96, 399.
(59) Magill, A. M.; Cavell, K. J.; Yates, B. F. J. Am. Chem. Soc. 2004,
126, 8717.
(60) Ngo, H. L.; LeCompte, K.; Hargens, L.; McEwen, A. B.
Thermochim. Acta 2000, 357−358, 97.
(61) Zhang, S.; Sun, N.; He, X.; Lu, X.; Zhang, X. J. Phys. Chem. Ref.
Data 2006, 35, 1475.
(62) Ming Lee, K.; Kuan Lee, C.; J. B. Lin, I. Chem. Commun. 1997,
899.
(63) D. Holbrey, J.; R. Seddon, K. J. Chem. Soc., Dalton Trans. 1999,
(21) Coulembier, O.; Moins, S. b.; Dubois, P. Macromolecules 2011,
44, 7493.
(22) Norris, B. C.; Sheppard, D. G.; Henkelman, G.; Bielawski, C. W.
J. Org. Chem. 2010, 76, 301.
(23) Hans, M.; Wouters, J.; Demonceau, A.; Delaude, L. Eur. J. Org.
Chem. 2011, 35, 7083.
(24) Delaude, L. Eur. J. Inorg. Chem. 2009, 2009, 1681.
(25) Delaude, L.; Demonceau, A.; Wouters, J. Eur. J. Inorg. Chem.
2009, 2009, 1882.
̀
(26) Fevre, M.; Pinaud, J.; Leteneur, A.; Gnanou, Y.; Vignolle, J.;
Taton, D.; Miqueu, K.; Sotiropoulos, J.-M. J. Am. Chem. Soc. 2012,
134, 6776.
2133.
(27) Poyatos, M.; Mata, J. A.; Peris, E. Chem. Rev. 2009, 109, 3677.
(28) Duong, H. A.; Cross, M. J.; Louie, J. Org. Lett. 2004, 6, 4679.
(29) Zhou, H.; Zhang, W.-Z.; Liu, C.-H.; Qu, J.-P.; Lu, X.-B. J. Org.
Chem. 2008, 73, 8039.
(64) Sobota, M.; Wang, X.; Fekete, M.; Happel, M.; Meyer, K.;
Wasserscheid, P.; Laurin, M.; Libuda, J. ChemPhysChem 2010, 11,
1632.
(65) Poater, A.; Ragone, F.; Giudice, S.; Costabile, C.; Dorta, R.;
Nolan, S. P.; Cavallo, L. Organometallics 2008, 27, 2679.
(66) Tapu, D.; Dixon, D. A.; Roe, C. Chem. Rev. 2009, 109, 3385.
(67) Hao, Y.; Peng, J.; Hu, S.; Li, J.; Zhai, M. Thermochim. Acta 2010,
501, 78.
(30) Kayaki, Y.; Yamamoto, M.; Ikariya, T. Angew. Chem., Int. Ed.
2009, 48, 4194.
(31) Naik, P. U.; Petitjean, L.; Refes, K.; Picquet, M.; Plasseraud, L.
Adv. Synth. Catal. 2009, 351, 1753.
(32) Tommasi, I.; Sorrentino, F. Tetrahedron Lett. 2009, 50, 104.
(33) Pawar, G. M.; Buchmeiser, M. R. Adv. Synth. Catal. 2010, 352,
917.
(34) Pinaud, J.; Vignolle, J.; Gnanou, Y.; Taton, D. Macromolecules
2011, 44, 1900.
(35) Van Ausdall, B. R.; Poth, N. F.; Kincaid, V. A.; Arif, A. M.;
Louie, J. J. Org. Chem. 2011, 76, 8413.
(36) Zhou, H.; Wang, Y.-M.; Zhang, W.-Z.; Qu, J.-P.; Lu, X.-B. Green
(68) Herrmann, W. A.; Goossen, L. J.; Kocher, C.; Artus, G. R. J.
̈
Angew. Chem., Int. Ed. 1996, 35, 2805.
(69) Song, J. J.; Gallou, F.; Reeves, J. T.; Tan, Z.; Yee, N. K.;
Senanayake, C. H. J. Org. Chem. 2006, 71, 1273.
(70) Suzuki, Y.; Bakar, M. D. A.; Muramatsu, K.; Sato, M.
Tetrahedron 2006, 62, 4227.
(71) Grunewald, G. L.; Brouillette, W. J.; Finney, J. A. Tetrahedron
Lett. 1980, 21, 1219.
(72) Utimoto, K.; Wakabayashi, Y.; Horiie, T.; Inoue, M.;
Shishiyama, Y.; Obayashi, M.; Nozaki, H. Tetrahedron 1983, 39, 967.
(73) Breslow, R. J. Am. Chem. Soc. 1958, 80, 3719.
(74) Pinaud, J.; Vijayakrishna, K.; Taton, D.; Gnanou, Y. Macro-
molecules 2009, 42, 4932.
Chem. 2011, 13, 644.
(37) Powell, A. B.; Suzuki, Y.; Ueda, M.; Bielawski, C. W.; Cowley, A.
H. J. Am. Chem. Soc. 2011, 133, 5218.
(38) Coulembier, O.; Kiesewetter, M.; Mason, A.; Dubois, P.;
Hedrick, J.; Waymouth, R. Angew. Chem., Int. Ed. 2007, 46, 4719.
(39) Bantu, B.; Pawar, G. M.; Decker, U.; Wurst, K.; Schmidt, A. M.;
Buchmeiser, M. R. Chem. Eur. J. 2009, 15, 3103.
(40) Bantu, B.; Pawar, G. M.; Wurst, K.; Decker, U.; Schmidt, A. M.;
Buchmeiser, M. R. Eur. J. Inorg. Chem. 2009, 2009, 1970.
(41) Naik, P. U.; Refes, K.; Sadaka, F.; Brachais, C.-H.; Boni, G.;
Couvercelle, J.-P.; Picquet, M.; Plasseraud, L. Polym. Chem. 2012, 3,
1475.
(75) Nyce, G. W.; Lamboy, J. A.; Connor, E. F.; Waymouth, R. M.;
Hedrick, J. L. Org. Lett. 2002, 4, 3587.
(76) Grasa, G. A.; Guveli, T.; Singh, R.; Nolan, S. P. J. Org. Chem.
̈
2003, 68, 2812.
(77) Starikova, O. V.; Dolgushin, G. V.; Larina, L. I.; Ushakov, P. E.;
Komarova, T. N.; Lopyrev, V. A. Russ. J. Org. Chem. 2003, 39, 1467.
(42) Holbrey, J. D.; Reichert, W. M.; Tkatchenko, I.; Bouajila, E.;
Walter, O.; Tommasi, I.; Rogers, R. D. Chem. Commun. 2003, 28.
(43) Tommasi, I.; Sorrentino, F. Tetrahedron Lett. 2006, 47, 6453.
(44) Duong, H. A.; Tekavec, T. N.; Arif, A. M.; Louie, J. Chem.
Commun. 2004, 112.
(45) Van Ausdall, B. R.; Glass, J. L.; Wiggins, K. M.; Aarif, A. M.;
Louie, J. J. Org. Chem. 2009, 74, 7935.
(46) Sauvage, X.; Demonceau, A.; Delaude, L. Adv. Synth. Catal.
2009, 351, 2031.
(47) Voutchkova, A. M.; Appelhans, L. N.; Chianese, A. R.; Crabtree,
R. H. J. Am. Chem. Soc. 2005, 127, 17624.
(48) Voutchkova, A. M.; Feliz, M.; Clot, E.; Eisenstein, O.; Crabtree,
R. H. J. Am. Chem. Soc. 2007, 129, 12834.
(49) Zhou, H.; Zhang, W.-Z.; Wang, Y.-M.; Qu, J.-P.; Lu, X.-B.
Macromolecules 2009, 42, 5419.
(50) Rauchfuss, T. B. In Inorganic Syntheses; Wiley: Hoboken, NJ,
2010; p 78.
10144
dx.doi.org/10.1021/jo301597h | J. Org. Chem. 2012, 77, 10135−10144