Thiourea-based bifunctional organocatalysis: Supramolecular recognition for living polymerization

Article abstract of DOI:10.1021/ja0543346

A versatile, metal-free, organocatalytic approach to the living ring-opening polymerization of lactide using a bifunctional thiourea-tertiary amine catalyst is described. Mild and highly selective polymerization conditions produced poly(lactides) with predictable molecular weights and extremely narrow polydispersities (～ 1.05), characteristic of a living polymerization. The extraordinary selectivity of this catalyst system for polymerization relative to transesterification is remarkably unusual. The low polydispersities and exceptional control observed are a consequence of selective transesterification of lactide relative to the open chain esters. Presumably, the ring strain of lactide provides both a driving force for the polymerization and a kinetic preference for polymerization relative to transesterification with catalyst. We postulate that the initiating/propagating alcohol is activated by acid-base interaction with the tertiary amine moiety and the carbonyl of the lactide monomer is simultaneously activated by hydrogen bonding to the thiourea moiety of the catalyst. Copyright

Full text of DOI:10.1021/ja0543346

Published on Web 09/21/2005
Thiourea-Based Bifunctional Organocatalysis: Supramolecular Recognition
for Living Polymerization
†
†
†
,‡
Andrew P. Dove, Russell C. Pratt, Bas G. G. Lohmeijer, Robert M. Waymouth,* and
,
†
James L. Hedrick*
IBM Almaden Research Center, 650 Harry Road, San Jose, California 95120, and Department of Chemistry,
Stanford UniVersity, Stanford, California 94305
Received June 30, 2005; Revised Manuscript Received September 2, 2005; E-mail: waymouth@stanford.edu; hedrick@almaden.ibm.com
The development of biodegradable polymers as resorbable
biomaterials and, more recently, as commodity thermoplastics from
renewable resources has received significant attention. Synthetic
routes to these polyesters generally employ transition metal
compounds to effect the ring-opening polymerization (ROP) of a
Scheme 1. Thiourea-Amine Catalysts
1
cyclic ester monomer. There has been a recent resurgence of
interest in organocatalytic methods as an alternative to organome-
2
tallic reagents. These methods are proving particularly useful in
Table 1. Properties of Poly(lactide)s
situations where residual metals may compromise the purity of the
product, an important consideration for microelectronic and bio-
medical applications. During our initial survey of a variety of
organic catalysts for the ROP of cyclic esters, we showed that
nucleophilic catalysts, such as tertiary amines, phosphines, and, in
particular, stabilized singlet carbenes, were effective polymerization
catalysts for strained cyclic esters.3
timea (h)
conv %b
DPb,c
GPCd
M
n
PDId
[
M]/[I]
2
5
0
0
24
32
48
105
144
97
98
97
98
95
21
53
103
215
5200
12300
23000
42000
1.08
1.05
1.05
1.05
e
100
200
5
00
e
e
a
With 5 mol % of 1; [LA] ) 1 M in CH2Cl2. b Determined by H NMR.
1
Catalysis employing hydrogen bonding for substrate activation
has been shown to be an effective and versatile strategy in a wide
c Degree of polymerization. d Determined by gel permeation chromatogra-
phy. e Not soluble in THF.
4
variety of transformations, analogous to many enzymatic pathways.
experimental molecular weights, and polydispersities below 1.07
throughout the reaction (Figure 1).
Jacobsen5 has exploited a variety of urea and thiourea-based
catalysts for Strecker, Mannich, Pictet-Spengler, and hydrophos-
phonylation reactions. These catalysts were proposed to activate a
variety of carbonyl and sulfoxide substrates for stereoselective C-C
Variation of the [M]/[I] ratio led to narrowly dispersed poly-
(lactide)s with molecular weights matching those predicted for DP’s
from 20 to 200 (Table 1). The polydispersities are extremely low
and remained invariant at high monomer conversions (95-100%).
For example, when poly(rac-lactide) (M ) 21 300, M /M ) 1.06,
bond forming reactions.5 Takemoto demonstrated that chiral
bifunctional organic catalysts containing thiourea and tertiary amino
groups effectively activated nitro compounds for enantioselective
-7
8
n
w
n
measured by GPC) was left to react with the catalyst for a further
4 days after complete monomer conversion, analysis of the
9
aza-Henry and Michael reactions. Similarly, Berkessel demon-
strated the dynamic kinetic resolution of azalactones using urea-
based chiral bifunctional catalysts. This catalyst motif is versatile
in design as the hydrogen bonding group, and its strength can be
tailored, as well as steric congestion, stereochemistry, etc. In our
initial experiments, we used the bifunctional catalyst 1 developed
polylactide revealed a M
results strongly imply that even after extended reaction periods
minimal transesterification is observed. This was confirmed by
NMR spectroscopy (Figure S1). Moreover, analysis of isotactic
poly(L-lactide) by 13C NMR spectroscopy and differential scanning
n
) 20 900 g/mol and PDI ) 1.07; these
13
C
8
by Takemoto (Scheme 1). We anticipated that the thiourea and
m
calorimetry (T ) 170 °C) clearly shows that racemization does
amine functional groups of these bifunctional catalysts could
activate both the monomer and alcohol to catalyze the ROP of cyclic
esters.
not occur (Figures S1 and S4).
The extraordinary selectivity of this catalyst system for polym-
erization relative to transesterification is unusual. To test this, the
reaction between methyl benzoate and either ethanol or 2-propanol
The catalytic behavior of 1 in the polymerization of lactide was
2 2
in the presence of 1 (5 mol %, 1 M CH Cl ) gave no evidence for
2 2
studied in CH Cl (1 M) at 25 °C using pyrenebutanol as the
transesterification (48 h) and resulted in the quantitative recovery
of methyl benzoate. These data suggest that the low polydispersities
and exceptional control observed are a consequence of selective
transesterification of lactide relative to the open chain esters.
Presumably, the ring strain of lactide provides both a driving force
for the polymerization and a kinetic preference for polymerization
relative to transesterification using catalyst 1.
To further corroborate the living nature of the polymerization, a
chain extension experiment was carried out. Polymerization of
lactide using pyrene butanol as the initiator and 5 mol % of 1 with
initiator in the presence of 5 mol % of 1. At a monomer-to-initiator
ratio of 100 ([M]/[I] ) 100), lactide was converted to polylactide
with 97% conversion after 48 h. The polylactide generated exhibited
a number average molecular weight M
of polymerization DP ) 100) with a very narrow polydispersity
/M ) 1.05; Table 1). The polymerization exhibited charac-
n
) 23 000 g/mol (degree
(M
w
n
teristics of a living polymerization as evidenced by the linear
1
correlation between M
n
(measured by H NMR) and monomer
conversion, the close correlation between the theoretical and
1
00 equiv of lactide [M]/[I] ) 100 yielded polylactide with a DP
†
IBM Almaden Research Center.
Stanford University.
‡
1
of 103 ( H NMR) and a PDI of 1.05 after 48 h. An additional 100
13798
9
J. AM. CHEM. SOC. 2005, 127, 13798-13799
10.1021/ja0543346 CCC: $30.25 © 2005 American Chemical Society

C O M M U N I C A T I O N S
Scheme 2. Polymerization of Lactide by 1
dimethylformamide (Table S1). As these solvents are likely to bind
avidly to the thiourea, these results imply that the ability of the
catalyst to activate the monomer through hydrogen bonding is
essential for polymerization of lactide, consistent with the mech-
anism proposed in Scheme 2.
Figure 1. Mn (diamonds) and PDI (squares) versus % monomer conversion
for polymerization of lactide with 1. Theoretical Mn (dashed line). A typical
GPC trace is shown together with a polystyrene standard to facilitate
comparison.
In summary, the bifunctional catalyst 1 exhibits exceptional
selectivity in the ring-opening polymerization of lactide for the
synthesis of well-defined poly(lactide)s of narrow polydispersity.
While these catalysts exhibit lower activities than some of the more
equiv of lactide was then added. After 48 h, the polylactide exhibited
1
a DP of 215 ( H NMR), and the polydispersity remained unchanged
1
3
(
1.05). This catalyst system is also effective for the synthesis of
block copolymers; a hydroxyl-end-capped poly(N,N-dimethylacryl-
amide) macroinitiator (M ) 4100 g/mol, PDI ) 1.07) was used
to initiate the polymerization of lactide using catalyst 1 or 2 in
CH Cl to generate poly(N,N-dimethylacrylamide)-b-poly(lactide)
active transition metal or organic catalysts, the high selectivity
for polymerization relative to transesterification provides exciting
opportunities for the synthesis of well-defined polylactide archi-
tectures with end-group fidelity.
n
2
2
Acknowledgment. We gratefully acknowledge support from the
NSF Center on (Polymeric Interfaces and Macromolecular As-
semblies (CPIMA) (NSF-DMR-0213618), and a NSF-GOALI Grant
block copolymer. The resulting polymer had M
PDI ) 1.07 after 72 h.
n
) 16 900 g/mol,
The initiation efficiency was investigated by analysis of the end-
(NSF-CHE-0313993).
groups of a DP 20 polymer initiated from 4-pyrene-1-butanol by
1
Supporting Information Available: Figures S1-S4, Table S1, and
the experimental procedures. This material is available free of charge
via the Internet at http://pubs.acs.org.
H NMR spectroscopy. For this sample, the only end-groups
observed were the R-ester from the initiating alcohol and the
â-hydroxyl chain-ends, which is indicative of one initiator per
polymer chain (Figure S2). In addition, analysis of the gel
permeation chromatography (GPC) traces of the polymer using both
refractive index and UV detectors (410 and 350 nm, respectively)
clearly shows distribution of pyrene throughout the sample (Figure
References
(
1) Dechy-Cabaret, O.; Martin-Vaca, B.; Bourissou, D. Chem. ReV. 2004, 104,
147.
(2) For a recent review, see: Seayad, J.; List, B. Org. Biomol. Chem. 2005,
, 719.
6
3
1
S3). This corroborates the H NMR analysis and establishes the
(
3) (a) Nyce, G. W.; Glauser, T.; Connor, E. F.; M o¨ ck, A.; Waymouth, R.
W.; Hedrick, J. L. J. Am. Chem. Soc. 2003, 125, 3046. (b) Csihony, S.;
Culking, D. A.; Sentman, A. C.; Dove, A. P.; Waymouth, R. W.; Hedrick,
J. L. J. Am. Chem. Soc. 2005, 127, 9078.
fidelity of initiation from the alcohol initiator. Moreover, a single
turnover reaction using catalyst 1 (5% loading), lactide, and excess
2 2
benzyl alcohol (1 M CH Cl ) yielded the expected benzyl dilactate.
(4) For example, see: (a) Pihko, P. M. Angew. Chem., Int. Ed. 2004, 43,
062. (b) Schreiner, P. R. Chem. Soc. ReV. 2003, 32, 289. (c) Tanaka,
2
We postulate that catalysis proceeds by bifunctional activation
of the carbonyl of a lactide monomer via hydrogen bonding to the
thiourea group and of the initiating/propagating alcohol by the
Bronsted basic (tertiary amino) group of the catalyst, similar to
K.; Inoue, S. J. J. Org. Chem. 1990, 55, 181. (d) Grogan, M. J.; Corey,
E. J. Org. Lett. 1999, 1, 157. (e) Sigman, M. S.; Jacobsen, E. N. J. Am.
Chem. Soc. 1998, 120, 4901. (f) Huang, Y.; Unni, A. K.; Thadani, A. N.;
Rawal, V. H. Nature 2003, 424, 146. (g) Unni, A. K.; Takenaka, N.;
Yamamoto, H.; Rawal, V. H. J. Am. Chem. Soc. 2005, 127, 1336. (h)
McDougal, N. T.; Schaus, S. E. J. Am. Chem. Soc. 2003, 125, 12094. (i)
Nugent, B. M.; Yoder, R. A. Johnston, J. N. J. Am. Chem. Soc. 2004,
126, 3418.
9
the mechanism proposed by Berkessel. Nucleophilic ring-opening
of the lactide leads to propagation, whereby the ring-opened lactide
forms the propagating alcohol for the subsequent addition of
monomer (Scheme 2).
(
5) (a) Sigman, M. S.; Vachal, P.; Jacobsen, E. N. Angew. Chem., Int. Ed.
2
000, 39, 1279. (b) Vachal, P.; Jacobsen, E. N. Org. Lett 2000, 2, 867.
(
c) Vachal, P.; Jacobsen, E. N. J. Am. Chem. Soc. 2002, 124, 10012. (d)
Wenzel, A. G.; Jacobsen, E. N. J. Am. Chem. Soc. 2002, 124, 12964. (e)
Taylor, M. S.; Jacobsen, E. N. J. Am. Chem. Soc. 2004, 126, 10558. (f)
Yoon, T. P.; Jacobsen, E. N. Angew. Chem., Int. Ed. 2005, 44, 466. (g)
Joly, G. D.; Jacobsen, E. J. Am. Chem. Soc. 2004, 126, 4102. (h) Fuerst,
D. E.; Jacobsen, E. N. J. Am. Chem. Soc. 2005, 127, 8964-8965.
6) Curran, D. P.; Kuo, L. J. J. Org. Chem. 1994, 59, 3259.
The bifunctional nature of the catalysis was probed by attempts
to polymerize lactide in the presence of (i) a thiourea, 2; (ii) N,N-
dimethylcyclohexylamine, 3; and (iii) both 2 and 3. Polymer was
only obtained when 2 and 3 were present at the same time (M
n
)
(
12 500, PDI ) 1.1), indicating that the bifunctional nature of the
(7) Wittopp, A.; Schreiner, P. R. Chem. Eur. J. 2003, 32, 289.
(
8) (a) Okino, T.; Hoashi, Y.; Takemoto, Y. J. Am. Chem. Soc. 2003, 125,
catalyst is critical, but that both activating units are not required in
a single catalyst. The implication that hydrogen bonding is critical
for catalysis was studied by performing polymerizations in different
solvents. Polymerization was observed in non-hydrogen bonding
solvents, such as chloroform, methylene chloride, and toluene. In
contrast, no polymerization was observed in tetrahydrofuran and
1
2672. (b) Okino, T.; Nakamura, S.; Furukawa, T.; Takemoto, Y. Org.
Lett. 2004, 6, 625. (c) Hoashi, Y.; Okino, T.; Takemoto, Y. Angew. Chem.,
Int. Ed. 2005, 44, 4100.
9) (a) Berkessel, A.; Mukherjee, S.; Cleemann, F.; Muller, T. N.; Lex J.
Chem. Commun. 2005, 1898. (b) Berkessel, A.; Mukherjee, S.; Cleemann,
F.; Muller, T. N.; Lex, J. Angew. Chem., Int. Ed. 2005, 44, 807.
(
JA0543346
J. AM. CHEM. SOC.
9
VOL. 127, NO. 40, 2005 13799