Oxidative esterification of aldehydes using mesoionic 1,2,3-triazolyl carbene organocatalysts

Article abstract of DOI:10.1021/ol501458p

The synthesis and catalytic activity of a new class of 1,2,3-triazolyl N-heterocyclic carbene organocatalysts is described. These new catalysts chemoselectively facilitate the oxidative esterification of aldehydes. NMR acidity studies show an inverse correlation between triazolium acidity and reactivity. Kinetic studies show that the resting state of the catalyst involves a NHC-aldehyde adduct. A catalytically active intermediate was synthesized and characterized by X-ray diffraction as the initial carbene-aldehyde adduct.

Full text of DOI:10.1021/ol501458p

Letter
pubs.acs.org/OrgLett
Oxidative Esterification of Aldehydes Using Mesoionic 1,2,3-Triazolyl
Carbene Organocatalysts
Matthew T. Berry, Disnay Castrejon, and Jason E. Hein*
Department of Chemistry and Chemical Biology, University of California, 5200 North Lake Road, Merced, California 95343, United
States
S
* Supporting Information
ABSTRACT: The synthesis and catalytic activity of a new
class of 1,2,3-triazolyl N-heterocyclic carbene organocatalysts
is described. These new catalysts chemoselectively facilitate the
oxidative esterification of aldehydes. NMR acidity studies show
an inverse correlation between triazolium acidity and reactivity.
Kinetic studies show that the resting state of the catalyst
involves a NHC−aldehyde adduct. A catalytically active
intermediate was synthesized and characterized by X-ray
diffraction as the initial carbene−aldehyde adduct.
he development of N-heterocyclic carbene catalysis in
Scheme 1. Synthesis of Triazolium Salt Precatalysts
T
the past 50 years has led to the discovery of a wide
library of synthetic transformations.1 Recent work in this field
by Enders,2,3 Bode,4,5 Rovis,6,7 and Scheidt8,9 has led to the
development of new carbene catalysts that feature the
thiazolyl, imidazolyl, and 1,2,4-triazolyl architectures. Despite
the wide structural diversity explored within these systems, the
feature common to all is the singlet carbene center flanked by
two heteroatomic neighbors, resulting in a stabilized neutral
carbene. Recently, Bertrand reported the synthesis, isolation,
and characterization of a new class of mesoionic carbene
based on the 1,2,3-triazolium scaffold.10,11 These unique
carbenes, sometimes termed remote or abnormal N-
heterocyclic carbenes, bear only a single flanking heteroatom
and exist as a zwitterionic structure with no neutral canonical
resonance form. Because these carbenes possess a unique
electronic structure, they are desirable as potential nucleo-
philic organocatalysts.
exchange with LiNTf2 to give triazolium salts 4a−j.13
Similarly, the 1,2,4-trisubstituted-1,2,3-triazolium salt 8 was
formed via coupling azide 5 with alkyne 6, followed by
cyclization of triazole 7 with (Tf)2O to yield 8. These
protocols allowed for a diverse set of triazolium salts to be
rapidly prepared on multigram scale (Figure 1).
To test the catalytic activity of the triazolium salts, we
screened different common NHC-catalyzed reactions (benzoin
condensation, Stetter reaction,14 and redox esterification15)
with 4c as a trial catalyst. While these trials did not yield any
of the expected products, we did find that our catalyst
facilitated the oxidative esterification of cinnamaldehyde with
methanol in trace yields. By introducing quinone 12 as
oxidant instead of ambient molecular oxygen, our yields
improved (Scheme 2). To optimize reaction conditions and
gain mechanistic insight, we used reaction monitoring
techniques (HPLC, NMR, React-IR) to understand the
individual contributions of each reaction component and to
identify the cause of catalyst deactivation.
In addition to these unique electronic properties, the 1,2,3-
triazolium core represents a highly modifiable catalyst
architecture, making the 1,2,3-triazolyl carbene an attractive
catalyst candidate. Via a simple two-step synthetic sequence
(Scheme 1), a variety of precatalysts with differing electronic
and steric properties were obtained quickly and cleanly. Two
possible regioisomers are accessible through this synthetic
strategy: the 1,3,4-substituted and 1,2,4-substituted 1,2,3-
triazolium salts. By varying the substituents of the azide and
alkyne precursors, both of these regioisomers can be obtained.
Herein we report the first demonstration of these compounds
as organocatalysts that facilitate the oxidative esterification of
aldehydes.
The candidate triazolium salts were synthesized via a simple
two-step protocol (Scheme 1). To obtain the 1,3,4-substituted
triazolium salts 4a−j, azides 1a−e were coupled to alkyne 2a
or 2b via the Cu-catalyzed azide−alkyne cycloaddition.12 The
resulting triazoles 3a−j were then cyclized and underwent ion
Received: May 20, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ol501458p | Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Figure 1. Scope of 1,2,3-triazolium salt precatalysts. Counterion for
−
all salts is bis(trifluoromethane)sulfonamide (NTf2 ).
Scheme 2. Initial Trial Reaction Conditions
Initially, we considered the possibility that the overall
moderate catalytic performance of these species was due to a
low population of the free carbene catalyst, stemming from a
lack of acidity at the C5 proton. To probe this possibility, we
used in situ NMR to monitor the rate of deuterium
incorporation at the C5 position starting from the protonated
azolium precatalyst in the presence of DIPEA (Figure 2a).
This experiment demonstrates that both the N-substituent
and exocyclic ring size impact the relative acidity of the
azolium C−H. Precatalysts with strongly electron-poor
aromatic substituents (4e and 4f) showed the fastest rate of
exchange, whereas species with the more electron-rich mesityl
substitution (4c and 4d) display the slowest exchange. While
it appears that the N-substituent plays the largest role in
modifying the triazolium acidity, the size of the saturated ring
also has an impact. In general, triazoliums bearing a five-
membered ring display faster rates of deuterium incorporation
than their corresponding six-membered analogues (Figure 2a,
4a cf. 4b; 4c cf. 4d).
Figure 2. (a) Deuterium exchange curves for each 1,2,3-triazolium
salt. (b) Reaction progress curves for each 1,2,3-triazolium salt.
these species. In part, this hypothesis is supported by the
detection of a decomposition product 4e′ by HPLC−MS,
believed to result from a SNAr-type addition of methanol to
the C6F5 ring (Scheme 3). While we were unable to isolate
Scheme 3. Possible Decomposition Pathway of C6F5-
Substituted Triazolium Catalysts
Following this survey, we attempted to correlate the
precatalyst’s relative acidity (as implied from the observed
deuterium incorporation rates) with their catalytic efficiency.
Individually, triazoliums 4a−f were applied to the oxidative
esterification of cinnamaldehyde, and product formation was
monitored by HPLC (Figure 2b). This experiment revealed
two key trends. First, the size of the saturated ring plays a
larger role in catalytic efficiency than the nature of the N-aryl
substituent, with the most effective catalysts 4a and 4c both
containing the 5-membered subunit. This result may be due
to unfavorable steric encumbrance adjacent to the carbene
center leading to hindered nucleophilicity. Second, the
catalytic activity appears to be inversely proportional to the
relative acidity of the triazolium species. Electron-poor C6F5-
substituted triazoliums (4e and 4f) performed very poorly,
while mesityl-substituted 4c displayed the highest efficiency.
Moreover, electron-poor 4e and 4f display no further
conversion beyond 5 h, leading us to conclude that rapid
deactivation is the root cause of the poor performance of
4e′ for full characterization, we have confirmed that 4e and 4f
are both stable in solutions of MeOH but rapidly form 4e′
and 4f′ on addition of base. This result suggests that the
pendant azolium group is an effective electrophilic activator of
the perfluoroaromatic group.
Ruling out carbene population as the reason for low yields,
we turned to investigating the individual contribution of each
component to the rate of product formation. Multiple
reactions with differing initial concentrations of either
cinnamaldehyde, TMG, MeOH, or oxidant 12 were
performed and monitored by HPLC sampling, giving reaction
progress curves (see the Supporting Information). Plotting the
initial rate of the oxidative esterification as a function of the
initial concentration for each component being varied (Figure
B
dx.doi.org/10.1021/ol501458p | Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
3) allows us to identify which reagents are involved in the
rate-limiting step and suggest a resting state for the catalyst.
isomerization to Breslow-like intermediate III all have similar
magnitude under these reaction conditions.
In addition to informing us of the possible mechanism, our
studies provided us with optimized conditions to carry out the
esterification. Running the reaction with 2 equiv of alcohol
and increasing catalyst loading to 10 mol % improved yields
and reaction times. Screening a variety of organic and
inorganic bases revealed that 0.2 equiv of KOtBu was optimal,
allowing reaction time to be reduced from 48 to 3 h. A brief
substrate scope of this reaction was performed using these
newly optimized conditions (Table 1). Electron-withdrawing
a
Table 1. Substrate Scope
Figure 3. Plot of initial rate vs initial concentration for each reaction
component. The zero slope of the MeOH and 12 curves indicate the
reaction is zero-order in both components. The positive slope in the
aldehyde and TMG curves shows the reaction to be positive order in
those components.
Our study revealed that changes in the initial concentration
of both MeOH and oxidant 12 do not impact the observed
rate, and thus, the reaction is zero-order with respect to these
reagents (Figure 3). By contrast, increasing concentrations of
both cinnamaldehyde and TMG lead to an increase in
reaction rate; however, the rate enhancement due to
additional TMG does reach a saturation point beyond 0.8
M. These results can be explained in the context of the
catalytic cycle (Scheme 4). Since both MeOH and 12 do not
Scheme 4. Proposed Catalytic Cycle
a
Standard reaction conditions: aldehyde (0.5 mmol, 1 equiv), alcohol
(1 mmol, 2 equiv), 12 (0.5 mmol, 1 equiv), and 4c (0.05 mmol, 0.1
equiv) were dissolved in anhydrous THF and heated to 40 °C. KOtBu
(0.1 mmol, 0.2 equiv) was added and the mixture heated overnight.
b
Isolated yield after column chromatography.
groups (entry 5) and heterocyclic substituents (entry 7) are
tolerated, while electron-donating aryl substrates (entry 6) do
not react. Primary alcohol nucleophiles are the most effective,
while secondary alcohols are far less reactive. These catalysts
also facilitate intramolecular cyclizations (entry 9), however,
reactions with t-BuOH or phenols give little to no product.
During the course of our reaction progress investigations,
we observed an intermediate by HPLC/MS with a mass that
corresponds to a carbene-aldehyde adduct. This compound
could be one of three tautomeric forms arising from the
nucleophilic addition of free catalyst I with the aldehyde
(Figure 4a, 14, 15, and 16).16,17
In hopes of isolating this intermediate for further
identification, we carried out a series of reactions where
oxidant 12 was omitted. The catalyst, aldehyde, base, and
solvent were varied and the intermediate formation and
decomposition was monitored by HPLC/MS (see the
Supporting Information for progress curves). Reacting
triazolium 4c with 4-chlorobenzaldehyde and NaOH appeared
to produce the desired intermediate in highest concentration
(Figure 4a). Using these conditions, we successfully isolated
and unambiguously characterized this compound by NMR
affect initial reaction rate, we can conclude that both the
oxidation and nucleophilic substitution are not rate limiting
(Scheme 4, III → IV and IV → I, respectively). However,
both TMG and cinnamaldehyde must be involved in
kinetically important steps, suggesting that the resting state
of the catalyst is not clearly defined as a single intermediate.
Instead, it is possible that the rate constants for the formation
of the free carbene I, addition of aldehyde to generate II and
C
dx.doi.org/10.1021/ol501458p | Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental procedures, characterization data, reaction
progress curves, and X-ray crystal data. This material is
available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
■
*E-mail: jhein2@ucmerced.edu.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Funding from UCM (Graduate and Research Council Faculty
Award). Further support was provided by NSF-COINS EEC-
0832819. We thank Dr. Arnold Rheingold and Dr. Curtis
Moore at the UCSD Crystallography Laboratory and Dr.
Alfred D. Bacher at UCLA for generous use of FTIR.
REFERENCES
■
(1) For reviews on NHC organocatalysis, see: (a) Enders, D.;
Niemeier, O.; Henseler, A. Chem. Rev. 2007, 107, 5606−5655.
(b) Grossmann, A.; Enders, D. Angew. Chem., Int. Ed. 2012, 51,
314−325. (c) Chauhan, P.; Enders, D. Angew. Chem., Int. Ed. 2014,
53, 1485−1487.
(2) Enders, D.; Breuer, K.; Teles, J. H. Helv. Chim. Acta 1996, 79,
1217−1221.
(3) Enders, D.; Kallfass, U. Angew. Chem., Int. Ed. 2002, 41, 1743−
1745.
Figure 4. (a) Possible tautomers of the catalytic intermediate formed
from 4-chlorobenzaldehyde and 4c. (b) X-ray crystal structure of
isolated catalytic intermediate 15.
and single-crystal X-ray diffraction, revealing that it is in fact
the benzylic alcohol carbene-aldehyde adduct 15.
Isolation and identification of 15 allows us to further refine
our mechanistic interpretation. This intermediate is stable in
solid form but can be reverted to free carbene and aldehyde
on treatment with base. Treating 15 solely with oxidant 12 or
with DDQ to allow direct benzylic oxidation were all
unsuccessful. However, subjecting 15 to the reaction
conditions including KOtBu, MeOH, and 12 did result in
ester formation (Scheme 5a). In addition, 15 can be used to
convert 4-nitrobenzaldehyde to the corresponding ester,
demonstrating its competence as a precatalyst for further
reaction (Scheme 5b).
(4) Sohn, S. S.; Bode, J. W. Org. Lett. 2005, 7, 3873−3876.
(5) Sohn, S. S.; Bode, J. W. Angew. Chem., Int. Ed. 2006, 45, 6021−
6024.
(6) Reynolds, N. T.; Rovis, T. J. Am. Chem. Soc. 2005, 127, 16406−
16407.
(7) Kerr, M. S.; Alaniz, J. R. De; Rovis, T. J. Org. Chem. 2005, 70,
5725−5728.
(8) Mattson, A. E.; Bharadwaj, A. R.; Zuhl, A. M.; Scheidt, K. A. J.
Org. Chem. 2006, 71, 5715−5724.
(9) Chan, A.; Scheidt, K. A. J. Am. Chem. Soc. 2006, 128, 4558−
4559.
(10) Guisado-Barrios, G.; Bouffard, J.; Donnadieu, B.; Bertrand, G.
Scheme 5. Catalytic Activity of Isolated Intermediate 15
Angew. Chem. 2010, 122, 4869−4872.
(11) Bouffard, J.; Keitz, B. K.; Tonner, R.; Lavallo, V.; Guisado-
Barrios, G.; Frenking, G.; Grubbs, R. H.; Bertrand, G. Organo-
metallics 2011, 30, 2617−2627.
(12) Hein, J. E.; Fokin, V. V. Chem. Soc. Rev. 2010, 39, 1302−1315.
(13) Tseng, M.; Cheng, H.; Shen, M.; Chu, Y. Org. Lett. 2011, 13,
4434−4437.
(14) Stetter, H. Angew. Chem., Int. Ed. Engl. 1976, 15, 639−712.
(15) Mahatthananchai, J.; Bode, J. W. Chem. Sci. 2012, 3, 192−197.
(16) Alwarsh, S.; Ayinuola, K.; Dormi, S. S.; McIntosh, M. C. Org.
Lett. 2013, 15, 3−5.
(17) Berkessel, A.; Elfert, S.; Etzenbach-Effers, K.; Teles, J. H.
Angew. Chem., Int. Ed. 2010, 49, 7120−7124.
Together these results confirm that the esterification likely
proceeds first via nucleophilic capture of the aldehyde by the
mesoionic carbene to give II, followed by proton transfer to
give the Breslow-like intermediate III and subsequent
oxidation, instead of a direct oxidation pathway from II to IV.
In conclusion, we have demonstrated the first report of
organocatalytic activity for mesoionic 1,2,3-triazolyl carbenes.
We have identified the likely resting state and have isolated a
catalytic intermediate, proving that these carbenes act as
nucleophilic organocatalysts.
D
dx.doi.org/10.1021/ol501458p | Org. Lett. XXXX, XXX, XXX−XXX