Expanding the catalytic activity of nucleophilic N-heterocyclic carbenes for transesterification reactions

Article abstract of DOI:10.1021/ol0267228

(graph presented) Currently, there is a renewed interest in reactions that are catalyzed by organic compounds. Typical organic catalysts for acylation or transesterification reactions are based on either nucleophilic tertiary amines or phosphines. This communication discusses the use of nucleophilic N-heterocyclic carbenes as efficient transesterification catalysts. These relatively unexplored and highly versatile organic catalysts were found to be mild, selective, and more active than traditional organic nucleophiles.

Full text of DOI:10.1021/ol0267228

ORGANIC
LETTERS
2002
Vol. 4, No. 21
3587-3590
Expanding the Catalytic Activity of
Nucleophilic N-Heterocyclic Carbenes
for Transesterification Reactions
,†
Gregory W. Nyce, Jorge A. Lamboy, Eric F. Connor, Robert M. Waymouth,* and
James L. Hedrick*
Center for Polymer Interfaces and Macromolecular Assemblies,
IBM Almaden Research, 650 Harry Rd., San Jose, California 95120,
and Department of Chemistry, Stanford UniVersity, Stanford, California 94305
hedrick@almaden.ibm.com
Received August 12, 2002
ABSTRACT
Currently, there is a renewed interest in reactions that are catalyzed by organic compounds. Typical organic catalysts for acylation or
transesterification reactions are based on either nucleophilic tertiary amines or phosphines. This communication discusses the use of nucleophilic
N-heterocyclic carbenes as efficient transesterification catalysts. These relatively unexplored and highly versatile organic catalysts were found
to be mild, selective, and more active than traditional organic nucleophiles.
Growing trends in asymmetric synthesis for classic reactions
that use organic molecules as promoters have provided an
alternative to traditional organometallic reagents. Simple
organic molecules in enantiomerically pure form have
demonstrated high reactivity and stereospecificity for a
number of useful organic transformations.1-5 MacMillan has
reported the first highly enantioselective organocatalytic
inter- and intramolecular Diels-Alder reactions, 1,3-dipolar
cycloadditions, and 1,4-conjugate Friedel-Crafts additions
involving pyrroles, as well as the first enantioselective
organocatalytic alkylation of indoles catalyzed by imidazo-
lidinone.1 Several groups have reported effective nonenzy-
matic catalysts for the kinetic resolution of secondary
alcohols using chiral phosphines2 or amine catalysts.3 These
nucleophilic catalysts, particularly the “planar-chiral” het-
erocycles based on the tertiary phosphine and amine frame-
works, provide good levels of enantiomeric excess. To some
degree this strategy mimics that carried out by enzymes.4
These trends toward environmentally sound catalysts have
been reviewed in a recent issue of Chemical and Engineering
News: “Improving classics: organocatalysts inspire “greener”
asymmetric versions of classic synthetic reactions”.5
† Stanford University. E-mail: waymouth@stanford.edu.
(2) For example: (a) Vedejs, E.; Daugulis, O.; Diver, S. T. J. Org. Chem.
1996, 61, 430. (b) Vedejs, E.; Rozners, E. J. Am. Chem. Soc. 2001, 123,
2428. (c) Qiao, S.; Fu, G. C. J. Org. Chem. 1998, 63, 4168.
(3) For example: (a) Fu, G. C. Acc. Chem. Res. 2000, 33, 412-420. (b)
Arai, S.; Bellemin-Laponnaz, S.; Fu, G. C. Angew. Chem., Int. Ed. 2001,
40, 234. (c) Tao, B.; Lo, M. C.; Fu, G. C. J. Am. Chem. Soc. 2001, 123,
353-354. (d) Fu, G. C. AdV. Organomet. Chem. 2001, 47, 101. (e) Fu, G.
C. Pure Appl. Chem. 2001, 73, 347. (f) Hodous, B. L.; Fu, G. C. J. Am.
Chem. Soc. 2002, 124, 1578-1579.
(1) For example: (a) Ahrendt, K. A.; Borths, C. J.; MacMillan, D. W.
C. J. Am. Chem. Soc. 2000, 122, 4243-4244. (b) Jen, W. S.; Wiener, J. J.
M.; MacMillan, D. W. C. J. Am. Chem. Soc. 2000, 122, 9874-9875. (c)
Dong, V. M.; MacMillan, D. W. C. J. Am. Chem. Soc. 2001, 123, 2448-
2449. (d) Fu, G. C. Acc. Chem. Res. 2000, 33, 412-420. (e) Arai, S.;
Bellemin-Laponnaz, S.; Fu, G. C. Angew. Chem., Int. Ed. 2001, 40, 234.
(f) Tao, B.; Lo, M. C.; Fu, G. C. J. Am. Chem. Soc. 2001, 123, 353-354.
(g) Cordova, A.; Notz, W.; Zhong, G.; Betancort, J. M.; Barbas, C. F. J.
Am. Chem. Soc. 2002, 124, 1842. (h) Cordova, A.; Notz, W.; Zhong, G.;
Betancort, J. M.; Barbas, C. F. J. Am. Chem. Soc. 2002, 124, 1866.
(4) Liu, L.; Breslow, R. J. Am. Chem. Soc. 2002, 124, 4978-4979.
(5) Borman, S. Chem. Eng. News 2002, Feb. 25, p33.
10.1021/ol0267228 CCC: $22.00 © 2002 American Chemical Society
Published on Web 09/25/2002

Recently, we reported the first example of N-heterocyclic
carbenes (NHC) as catalysts for the living ring-opening
polymerization (ROP) of cyclic esters. We believed that the
extension of organic catalysis to controlled polymerization
procedures would be a highly desirable alternative to
traditional organometallic approaches, and ongoing work in
this area shows promise. During our initial survey of a variety
of potential nucleophilic catalysts including tertiary amines
and phosphines for ROP, it was discovered that the N-
heterocyclic carbenes were by far the most active. It was
postulated that the NHC activates the substrate toward attack
from the initiating/propagating alcohol.6 Breslow, Setter, and
others demonstrated that thiazolium-based carbenes promoted
organic transformations such as the benzoin condensation
reaction, where Breslow proposed acyl activation by a
carbene intermediate.7
Scheme 1. Electronic and Steric Versatility of Nucleophilic
N-Heterocyclic Carbenes
mesityl and 2,6-diisopropylphenyl groups with smaller
substituents at the 1,3-positions. Alkylation of 1-methyl
imidazole with methyl iodide provides a convenient synthesis
of 1,3-dimethylimidazolium iodide salt precursor that can
subsequently be deprotonated to yield 1,3-dimethylimidazol-
2-ylidene, 3. In addition, 1-ethyl-3-methylimidazol-2-ylidene,
4, can be synthesized from commercially available 1-ethyl-
3-methyl imidazolium chloride. To demonstrate the feasibil-
ity of imidazolin-2-ylidene nucleophilic catalysts, we were
prompted to explore the utility of “Wanzlick” carbenes as
potential polymerization catalysts (5).10 Although the imi-
dazolin-2-ylidene carbenes are prone to dimerize in the
absence of bulky groups at the 1,3-positions, Wanzlick’s
original work demonstrated that tetraaminoethylene com-
plexes have reactivity characteristic of nucleophilic carbenes.
One can regard the conversion of a cyclic ester to a
polymer as successive transesterification reactions that
selectively give a ring-opened product where the length of
polymer chains is controlled by the amount of initiator added.
With this in mind, we decided to extend this general
transesterification principle to the synthesis of organic
molecules. Moreover, as a general transesterification catalyst,
we wish to extend the NHC catalysts platform to effect
polycondensation polymerization reactions as a practical
route to engineering polyesters.
The isolation of NHC is complicated by their extreme air
and moisture sensitivity and, in some cases, their tendency
to dimerize. Recent investigations by Nolan and Grubbs
demonstrated that free carbenes can be generated in situ and
directly used to form N-heterocyclic carbene-coordinated
catalysts in a greatly simplified process.11 The in situ
generation of the N-heterocyclic carbene catalysts directly
from their respective salts allowed the rapid screening of
these catalyst libraries and reaction conditions for both the
organic transformations and polycondensation reactions
(Scheme 2).12 Importantly, isolated carbene 1 and in situ
generated carbene 1 gave nearly identical results.
Since the initial description of the synthesis, isolation and
characterization of stable carbenes by Arduengo,8 the ex-
ploration of their versatility and chemical reactivity has
become a major area of research.9 Carbenes can be synthe-
sized with considerable diversity by the sterics and electron-
ics of the groups attached to the imidazole ring (R1-2) and
the nitrogen(s) (R3-4) and the ethylene backbone (e.g.,
saturated vs unsaturated), (Scheme 1). The high reactivity
of the 1,3-bis(2,4,6-trimethylphenyl)imidazol-2-ylidene, 1,
for the ring-opening polymerization of lactide inspired us to
investigate 1 as a general transesterification catalyst together
with 1,3-bis(2,6-diisopropyl)imidazol-2-ylidene, 2. We were
also interested in exploring the effect of replacing bulky
Scheme 2. Procedure for in Situ Generation of 1
(6) Olmstead W. N.; Margolin Z.; Bordwell F. G. J. Org. Chem. 1980,
45, 3295.
(7) Breslow, R. J. Am. Chem. Soc. 1958, 80, 3719. (b) Stetter, H. Angew.
Chem., Int. Ed. Engl. 1976, 15, 639. (c) Enders, D.; Breuer, J. H. HelV.
Chim. Acta 1996, 79, 1217. (d) Teles, J.; Melder, J.; Ebel, K.; Enders, D.;
Breuer, J. H. HelV. Chim. Acta 1996, 79, 61. (e) Kerr, M. S.; De Alaniz, J.
R.; Roves, T. J. Am. Chem. Soc. 2002, ASAP. (f) Enders, D.; Kallfass, U.
Angew. Chem., Int. Ed. 2002, 41, 1743.
(8) a) Arduengo, A. J.; Dais, R. L.; Harlow, R.; Kline, M. J. Am. Chem.
Soc. 1991, 113, 361. (b) Arduengo, A. J.; Dias, R. L.; Harlow, R.; Kline,
M. J. Am. Chem. Soc. 1992, 114, 5530.
(9) a) Arduengo, A. J. Acc. Chem. Res. 1999, 32, 913. (b) Arduengo, A.
J.; Krfczyk, R. Chem. Z. 1998, 32, 6. (c) Bourissou, D.; Guerret, O.; Gabbai,
F. P.; Bertrand, G. Chem. ReV. 2000, 100, 39. (d) Herrmann, W.; Kocher,
C. Angew. Chem., Int. Ed. Engl. 1997, 36, 2162. (e) Scholl, M.; Trnka, T.
M.; Morgan, J. P.; Grubbs, R. H. Tetrahedron Lett. 1999, 40, 2247. (f)
Huang, J.; Stevens, E. D.; Nolan, S. P. Petersen, J. L. J. Am. Chem. Soc.
1999, 121, 2674. (g) Weskamp, T.; Schattenmann, W. C.; Spiegler, M.;
Herrmann, W. A. Angew. Chem., Int. Ed. Engl. 1998, 37, 2490. (h)
Herrmann, W. A.; Elison, M. Fisher, J.; Kocher, C. Artus, G. J. Angew.
Chem., Int. Ed. Engl. 1995, 34, 2371. (i) Herrmann, W. A.; Goossen, L.
Kocher, C.; Artus, G. Angew. Chem., Int. Ed. Engl. 1996, 35, 2805. (j)
Lappert, M. J. J. Organomet. Chem. 1988, 358, 185. (k) Enders, D.; Breuer,
K.; Raabe, G.; Runsink, J.; Henrique, T.; Melder, J.-P.; Ebel, K.; Brode, S.
Angew. Chem., Int. Ed. Eng. 1995, 34, 1021.
To demonstrate the feasibility of NHC-catalyzed transes-
terification, methyl benzoate was chosen because it mimics
the central building block for important polyesters. Methyl
benzoate reacted with a 20-fold excess of either ethanol,
2-propanol, or tert-butyl alcohol (Table 1) to drive the
(10) Wanzlick, H. W.; Kleiner, H. J. Angew. Chem., Int. Ed. Engl. 1962,
1, 75.
(11) For example: (a) Zhang, C.; Huang, J.; Trudell, M. L.; Nolan, S.
P. J. Org. Chem. 1999, 64, 3804. (b) Morgan, J. P.; Grubbs, R. H. Org.
Lett. 2000, 2, 3153. (c) Morgan, J. P.; Morrill, C.; Grubbs, R. H. Org. Lett.
2002, 4, 67-70. (d) Jafarpour, L.; Nolan, S. P. Organometallics 2000, 19,
2055. (e) Grasa, G. A.; Nolan, S. P. Organometallics 2001, 20, 3607.
(12) To ensure complete consumption of tert-butoxide in the in situ
formation of carbenes 1-5, a 20% excess of imidazolium salt precursor
was used. See Supporting Information for details.
3588
Org. Lett., Vol. 4, No. 21, 2002

Scheme 4. Polymerization via Transesterificaton Reactions
Table 1. Model Transesterification Reactions with Subsitituted
Alcoholsa
with N-Heterocyclic Carbenes
a Typical procedure: inert atmosphere, 6.6 × 10-5 mol of N-heterocyclic
carbene in 2 mL of THF (generated in situ; 0.8 equiv of KO-tBu/1 equiv
of carbene salt, 25 °C, 1 h and then filtered) was added to 20 mol of EtOH
and 0.0015 mol of methylbenzoate, 20 h, 25 °C. Yield from GC/MS.
what analogous to the procedure by Frechet using DMAP,
and will be elaborated in a subsequent publication.
It is of great interest to extend this catalyst’s platform to
show polymerizations via typical transesterification/poly-
condensation reactions (Table 2). Examples of polyesters
were directed toward those traditionally obtained by classical
reaction to completion. For 1, transesterification with primary
alcohols was extremely effective and considerably better than
secondary or tertiary alcohols, consistent with the nucleo-
philic character of the alcohol and perhaps the sterics.
However, transesterifications in the presence of 2-propanol
with smaller catalysts 3 and 4 were much more efficient.
Transesterification reactions with tertiary alcohols, such as
tert-butyl alcohol, were ineffective irrespective of the catalyst
used.
Table 2. Transesterification Reactions That Generate
Biodegradable Polysetersa
Scheme 3. Acylation of Disubstituted Alcohol through
N-heterocyclic Carbene Catalysis
In addition, we examined the transesterification of 6 with
7 in the presence of either 2 or 3 carbene catalysts (Scheme
4). In this case, 7 was used in 20% excess and the reaction
was accomplished in bulk under vacuum to remove the
condensation byproduct. Only a 60% yield of 8 was obtained
with catalyst 2, whereas quantitative conversion to 8 was
achieved for the less sterically hindered catalyst, 3. In this
example, catalyst structure clearly plays an important role.
Removal of excess 7 was accomplished by distillation at 75
°C. This general procedure may provide an extremely simple
and accelerated route to dendrons, dendrimers, etc.,13 some-
a Typical procedure: reduced pressure (0.1 mm/Hg), 6.6 × 10-5 mol of
N-heterocyclic carbene (generated in situ; 0.8 equiv of KO-tBu/1 equiv of
carbene salt, 2 mL of THF, 25 °C, 1 h and then filtered) added to 20 mol
of AB monomer, 20 h, 25 °C. Mn, molecular weight; PDI, (Mn/Mw)
polydispersity index, measured by gel permeation chromatography (THF).
b Insoluble in THF. c Isolated yield.
Org. Lett., Vol. 4, No. 21, 2002
3589

ring-opening methods including poly(ꢀ-caprolactone), poly-
(glycolide), etc. Toward this goal, ethyl 6-hydroxyhexanoate
and ethyl glycolate were self-condensed in bulk at 60 °C
under vacuum for 24 h (Table 2). High molecular weight
poly(ꢀ-caprolactone) was obtained and characterized, but the
poly(glycolide) formed was highly crystalline and insoluble
in our GPC solvents. Thermal analysis of the poly(glycolide)
was comparable to that reported for high molecular weight
poly(glycolide). To facilitate solubility, a copolymer was
obtained simply by the condensation of the two monomer
types (Table 2). In addition, dimethyl adipate and ethylene
glycol were condensed in bulk at 65 °C under vacuum (24
h). This AA/BB monomer system also produced high
molecular weight polymer.
THF or precipitation. In each case the 1H NMR spectra was
identical to that of the commercial BHET. The melt
condensation of BHET was performed in the presence of
catalyst 5 using a slow heating ramp to 280 °C under
vacuum, generating PET polymer. Importantly, the spectral
and thermal (mp of 280 °C) characteristics of the polymer
made from 5 are identical to commercial PET polymer (see
Supporting Information).
In conclusion, NHCs are efficient transesterification
catalysts. Transesterification reactions with primary alcohols
were the most successful, and little to no reaction was
observed in the presence of tertiary alcohols. Transesterifi-
cation reactions with secondary alcohols were difficult with
the sterically demanding catalyst 1 but more facile when
smaller NHC catalysts (3 and 4) were used. We have also
demonstrated the utility of NHC catalysts for polymer-
forming reactions. NHC-mediated poly-condensation po-
lymerization reactions provide a convenient route to biode-
gradable and commodity polymers as well as rapid entry into
dendrimer precursors. Applications of NHC catalysts as a
general platform for synthesis are ongoing.
Another important commercial polyester, poly(ethylene
terephthalate) (PET), is generally prepared by a two-step
process: condensation of dimethyl terephthalate (DMT) with
excess ethylene glycol (EG) to generate bis(2-hydroxyethyl)
terephthalate (BHET), followed by the self-condensation of
BHET at high temperatures (270-290 °C) using mixed
organometallic catalysts optimized for their reactivity and
selectivity for each process. Catalysts 3 and 5 were inves-
tigated to condense DMT with excess ethylene glycol (EG)
in THF at room temperature. Quantitative conversion of
DMT to BHET was realized in 1 h, irrespective of the
catalysts employed (Scheme 4). The success of this strategy
prompted us to investigate the use of the corresponding ionic
liquid directly as a reactive media for the condensation of
DMT and EG. The room-temperature ionic liquid 1-pentyl-
3-methylimidazolium iodo salt proved an outstanding me-
dium for the formation of BHET.14 The ionic liquid was
activated with potassium tert-butoxide, dissolved in a drop
of THF, and the BHET could be retrieved by extraction using
Acknowledgment. The authors gratefully acknowledge
the NSF Center for Polymeric Interfaces and Macromolec-
ular Assemblies (CPIMA). The authors also acknowledge
partial support from NIST/ATP cooperative agreement
70NANB8H4013.
Supporting Information Available: References to syn-
theses of known compounds and experimental procedures
are provided. This material is available free of charge via
the Internet at http://pubs.acs.org.
OL0267228
(14) 1-Pentyl-3-methylimidazolium iodide was synthesized by heating
1-methy imidazole and 1-iodopentane in isopropyl alcohol at 60 °C for 6
h. For a recent review on ionic liquids, see: Welton, T. Chem. ReV. 1999,
99, 2071-2083 and references therein.
(13) Ihre, H.; Padilla, O. L.; De Jesus, P.; M. J.; Frechet, M. J. J. Am.
Chem. Soc. 2001, 123, 5908-5917.
3590
Org. Lett., Vol. 4, No. 21, 2002