Polymerization of racemic β-butyrolactone using gold catalysts: A simple access to biodegradable polymers

Article abstract of DOI:10.1021/om200271q

Novel gold(I) complexes containing N-heterocyclic carbene ligands have been synthesized and fully characterized. The resulting mononuclear gold complexes act as active initiators in the polymerization of rac-β-butyrolactone under solvent-free conditions to provide the corresponding biodegradable poly(3-hydroxybutyrate).

Full text of DOI:10.1021/om200271q

COMMUNICATION
pubs.acs.org/Organometallics
Polymerization of Racemic β-Butyrolactone Using Gold Catalysts: A
Simple Access to Biodegradable Polymers
Emilie Brulꢀe,†,‡ Sylvain Gaillard,§ Marie-No€elle Rager,^ Thierry Roisnel,|| Vincent Guꢀerineau,#
Steven P. Nolan,*,§ and Christophe M. Thomas*,†
†Chimie ParisTech, UMR CNRS 7223, 11 Rue Pierre et Marie Curie, 75005 Paris, France
‡Universitꢀe Pierre et Marie Curie, 4 Place Jussieu, 75005 Paris, France
§EaStCHEM School of Chemistry, University of St Andrews, St Andrews, KY16 9ST, U.K.
^NMR Facility, Chimie ParisTech, 11 Rue Pierre et Marie Curie, 75005 Paris, France
Centre de Diffractomꢀetrie X, Universitꢀe de Rennes 1, 35042 Rennes, France
#Institut de Chimie des Substances Naturelles (ICSN), Avenue de la Terrasse, 91198 Gif/Yvette, France
S Supporting Information
b
ABSTRACT: Novel gold(I) complexes containing N-heterocyclic carbene ligands
have been synthesized and fully characterized. The resulting mononuclear gold
complexes act as active initiators in the polymerization of rac-β-butyrolactone
under solvent-free conditions to provide the corresponding biodegradable poly(3-
hydroxybutyrate).
Scheme 1. Ring-Opening Polymerization of β-Butyrolactone
iodegradable aliphatic polyesters such as poly(hydroxy-
alkanoate)s (PHAs) represent an attractive class of materials
B
for a variety of applications.1 These polymers have been shown
to be nontoxic and biocompatible both as polymers and as
their degradation products, which in most cases are ultimately
metabolized.1 The most common PHA is poly(3-hydro-
xybutyrate) (PHB), which is an isotactic, highly crystalline
thermoplastic polyester produced by various bacteria and
algae.1,2
A range of catalytic species have been employed to mediate the
polymerization of BBL.4 However, in spite of great efforts in this
field, there is still demand for highly active, nontoxic, and stable
complexes that can be easily handled. For instance, ROP catalysts
have not been widely investigated in bulk polymerizations,
although there is significant interest in using metal-based cata-
lysts that are active in neat monomer. Indeed, bulk polymeriza-
tions are industrially attractive and compliant with the principles
of green chemistry.5 These goals are what prompted our efforts
to search for novel and simpler initiating systems that would
prove capable of efficiently ring-opening racemic BBL under
solvent-free conditions.
One of the most convenient and promising synthetic routes to
PHB uses organic, enzymatic, or metal-based initiators to per-
form the ring-opening polymerization (ROP) of β-butyrolactone
(BBL), where the relief of ring strain represents the driving force
for the polymerization (Scheme 1).2,3 However, despite the high
internal strain of β-lactones, BBL appears as a rather reluctant
monomer, significantly less reactive than related lactones such as
β-propiolactone or even lactide and ε-caprolactone.2 For in-
stance, BBL is not polymerized by common anionic initiators
(e.g., metal alkoxides, alkali metal carboxylate salts).3 Moreover, the
ROP of BBL is also challenging in that ring-opening can proceed by
bond cleavage either between the carbonyl carbon and oxygen
atom of the β-lactone ring, resulting in retention of stereochemis-
try, or between the β-carbon and oxygen atom, leading to inversion
of configuration and loss of end-group fidelity.2,3
Received: March 26, 2011
Published: April 29, 2011
r
2011 American Chemical Society
2650
dx.doi.org/10.1021/om200271q Organometallics 2011, 30, 2650–2653
|

Organometallics
COMMUNICATION
Scheme 2. Synthesis of Complexes 2ꢀ4
Scheme 3. Formation of Two Isomers, a and b, Depending
on the CoordinationꢀInsertion (i) or Anionic (ii)
Mechanism
The gold(I) complex [Au(IPr)(OH)] (1) (IPr = 1,3-bis(2,6-
diisopropylphenyl)imidazol-2-ylidene) was selected as a starting
initiator, as it possesses a strongly basic character (pKa 30.4).12
Complex 1 was also employed as a synthon to isolate, in
quantitative reactions, novel isopropoxide (2) and silylamide
(3, 4) gold derivatives by simple protonolysis with an appropriate
alcohol or disilazane (Scheme 2).13,14
Initial polymerizations were performed using 1 mol % of
[Au(IPr)(OH)] (1) with commercially available rac-BBL at
100 ꢀC without solvent. Gratifyingly, 1 proved an active initiator
for the ROP of rac-BBL under these conditions (Table 1, entry
1). To test the limits of reactivity of 1, the temperature was
lowered to 50 ꢀC and a disappointing 19% polymer was formed
after prolonged reaction times (Table 1, entry 2).
As isopropoxide and to a lesser extent amido ligands have been
shown to act as good initiating groups in ROP,2,4a,4b,10 the
reactivity of 2 and 3 was tested at 50 ꢀC. Both complexes
displayed improved activity compared to 1. Surprisingly, 3 was
much more active than 2, as it converted 83% BBL in 2 h vs 46%
in 24 h for 2 (entries 3 and 4). Complete monomer conversions
proved difficult, as polymerization reactions were conducted in
the absence of solvent, where mass transport issues can arise.
Since 3 proved effective, the more sterically hindered bis-
(dimethylphenyl)amide-containing 4 was then tested and also
displayed important activity, allowing a conversion of 75% of
BBL after 2 h (entry 5). However, the molecular weight
distribution was slightly higher (PDI = 1.57) compared to that
obtained with 3 (PDI = 1.27).
Table 1. Polymerization of rac-β-Butyrolactonea
entry
initiator
t (h)
yield (%)c
Mn (g molꢀ1 d
)
PDId
3
1b
2
1
67
46
24
2
85
19
46
83
75
66
31
72
26
7150
nd
1.19
nd
1
3
2
5060
1.14
1.27
1.57
1.14
1.19
1.52
1.15
4
3
14 890
10 510
9010
5
4
2
6e
7e
8f
9g
3 (THF)
4.5
24
7
3 (AN)
6870
3
3
14 500
10 750
7
a [BBL]/[Au] = 100 at 50 ꢀC in neat monomer. b Reaction at 100 ꢀC.
c As determined by the integration of methine 1H NMR resonances of
rac-BBL and PHB. d Mn and PDI (Mw/Mn) of polymer determined by
size-exclusion chromatography in THF at room temperature using
polystyrene standards. PDI = polydispersity index. e [BBL] = 2 M. AN =
CH3CN f [BBL]/[Au] = 150. g [BBL]/[Au] = 200.
Gold catalysis has emerged as an area of intense research
activity over the past decade, and this has resulted in significant
advances in synthetic chemistry.6 Some examples, although few
in number, have been reported where gold complexes have
served as initiators in polymerization reactions. Gold(III) N-het-
erocyclic carbene (NHC) compounds have been shown to
induce olefin polymerization,7 and gold(I)ꢀNHC complexes
display activity in the ROP of lactide,8 although these best
operate at elevated temperatures. The potential use of gold
complexes as initiators in β-butyrolactone polymerization is
largely motivated by advantageous features displayed by these
gold complexes such as lack of color, low toxicity, and high
activity.6 Therefore the use of NHC as supporting ligand
appeared judicious to generate simple and robust catalytic
systems. This ligand class has recently generated considerable
interest in metal coordination chemistry for its ability to stabilize
well-defined monomeric complexes for a wide range of reactive
metal centers. As with other systems, the NHC ligand properties
impart great stability to Au complexes, which generally are easy
to synthesize and convenient to handle.9
A significant solvent effect was also observed (entries 6, 7). For
[BBL]/[3] = 100, reactions performed in polar solvents lead to
some polymerization, THF providing the highest yield of PHB
(66% after 4.5 h) with narrow PDI (1.14). BBL was converted to
PHB in only 30% yield in acetonitrile (AN). No trace of
polymerization was observed in nonpolar solvents (toluene
and CH2Cl2). The increase of [BBL]/[Au] ratio was then
considered in experiments using 3 at 50 ꢀC without solvent.
With [BBL]/[Au] = 150, a conversion of 72% was reached in 7 h
with Mn 14 500 g molꢀ1 (entry 8). In contrast, the reaction with
3
[BBL]/[Au] = 200 was not as productive, and a modest 26%
conversion was reached after 7 h, though a narrow polydispersity
(1.15) was observed (entry 9). This reaction was allowed to
proceed at room temperature, and a similar yield of 27% was
obtained after 144 h. To summarize, the initiator activity appears
to follow the trend OHꢀ < OiPrꢀ < N(SiR3)2ꢀ (1 < 2 < 3 ≈ 4).15
As a control experiment, as imidazolium salts can be a
byproduct of the degradation of metalꢀNHC complexes,16 the
Juxtaposing recent developments in our laboratories, namely,
the living polymerization of β-lactones10 and the development of
well-defined [Au(NHC)]-containing complexes for catalytic
applications,11 the synthesis of biodegradable poly(3-hydro-
xybutyrate) using gold catalysis in neat monomer was envisaged.
Herein, we report the use of AuꢀNHC complexes as initiators in
the polymerization of rac-BBL.
imidazolium salt IPr HCl was tested in the polymerization of
3
BBL under the same conditions as 3 to verify if any background
reaction led to polymer formation. IPr HCl displayed some
3
activity but proved much less efficient than 3, as 75% BBL was
2651
dx.doi.org/10.1021/om200271q |Organometallics 2011, 30, 2650–2653

Organometallics
COMMUNICATION
observed species. Finally, heteronuclear multiple bond correla-
tion (HMBC) was needed to confirm that complex 5b was
3
formed. Indeed, HMBC analysis showed a JCH correlation
between Hb and Cc and Hc and Cb, which is diagnostic of 5b.
Thus, although it is difficult to compare the results obtained by
NMR in benzene to the catalytic data obtained in neat monomer,
the formation of compound 5b suggests that 2 might behave as
an anionic initiator under these conditions.
It is worth noting that the organocatalytic NHC-based ROP of
BBL operates by a combination of both mechanisms. Indeed
Dubois demonstrated the formation of adducts from both
hydroxyl- and carboxylic acid-terminated structures with free
carbene in the presence of alcohol initiators.4d Structural and
kinetics studies highlighted an anionic polymerization mecha-
nism where the protonated carbene played the role of associated
counterion. In our case after the initiation step with complex 3
the alkoxide chain end should behave like the isopropoxide
catalyst 2. However alkoxide and amido complexes show differ-
ent catalytic activities, suggesting that once the propagating
active species is formed from 3, the anionic mechanism is not
operative. By contrast, the low activity of 2 might be due to
coexistence of both alkoxide and carboxylate growing species (at
least in the early stage of the polymerization).20
In conclusion gold amido complexes exhibit high activity
under mild conditions for the ROP of rac-BBL. These robust
systems operate under solvent-free conditions to yield PHB with
controlled molecular weight and a narrow molecular weight
distribution. Depending on the nature of the gold initiator,
different mechanistic pathways can be involved during the
polymerization of BBL. These results point to a number of
new avenues to explore in the synthesis of biodegradable
polymers.
Figure 1. Alkyl region of 1H NMR (400 MHz, C6D6) spectrum of 5b in
the presence of unreacted BBL (m).
converted after 24 h (vs 2 h for 3) and crotonate product was also
observed. Finally, the polymerization of rac-BBL with sodium
bis(trimethylsilyl) amide was carried out, and the absence of
polymer formation even after 24 h at 50 ꢀC confirmed the crucial
role of the goldꢀNHC moiety on the polymerization activity.17
As significant differences in terms of catalytic activity were
observed, we envisaged that different mechanistic pathways
could be involved using 2 or 3 as an initiator. Possible mecha-
nisms of enchainment are depicted in Scheme 3. It is well
established that the ring-opening of BBL by metal-based com-
plexes can proceed by a coordinationꢀinsertion mechanism (i.e.,
acyl cleavage) or an anionic mechanism (i.e., alkyl cleavage).2,3
Depending on whether one or the other of two possible
mechanisms is at play in the present transformation, two isomers,
a or b, can be formed in this initial step (Scheme 3). Indeed, in
the coordinationꢀinsertion mechanism the acylꢀoxygen bond
cleavage would produce a, while the alkylꢀoxygen bond cleavage
occurring via an anionic mechanism would lead to the formation
of b. To determine the route involved in the ROP of BBL with 3,
reaction was conducted with a [BBL]/[Au] ratio of 25. The
resulting low molecular weight PHB sample was then analyzed by
MALDI-TOF mass spectrometry. A single mass distribution was
observed with BBL as the repeating unit and [Au(NHC)] and
OH (from hydrolysis of silylamide) as the chain-end groups.18
The observation of these end-groups suggests that the ROP
proceeds via a classical coordinationꢀinsertion mechanism, i.e.,
initial acylꢀoxygen bond cleavage by the transfer of the nucleo-
philic group of the metal complex to the monomer with the
eventual formation of a metalꢀalkoxide propagating species.
Furthermore, enantiomerically pure (R)-BBL was polymerized
using 3. As expected, the resulting polymer was shown to be
highly isotactic by 13C NMR. Polarimetry at 365 nm revealed that
the polymerization proceeds with retention of the R-stereocen-
ter, consistent with a coordinationꢀinsertion mechanism.19
The possible mechanism involved in the polymerization was
also explored by following the polymerization of 10 equiv of BBL
in C6D6 at 50 ꢀC by NMR spectroscopy using 2 as a mechanistic
probe. Interestingly, as this polymerization is considerably slow
in C6D6, the complex resulting from the opening of 1 equiv of
’ ASSOCIATED CONTENT
S
Supporting Information. Experimental section, NMR
b
and MALDI-TOF data, crystallographic data for 6, and calcula-
tions of hydrodynamic radii are available free of charge via the
Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: christophe-thomas@ens.chimie-paristech.fr, snolan@
st-andrews.ac.uk.
’ ACKNOWLEDGMENT
We are grateful to the ENSCP, the CNRS, and the French
Ministry of Higher Education and Research for financial support.
For work conducted at St Andrews, the ERC (Advanced
Investigator Award-FUNCAT) and the EPSRC are acknowl-
edged for financial support. S.P.N. is a Royal Society-Wolfson
Research Merit Award holder.
’ REFERENCES
(1) (a) Mecking, S. Angew. Chem., Int. Ed. 2004, 43, 1078–1085. (b)
Ulrich, K. E.; Cannizzaro, S. M.; Langer, R. S.; Shakesheff, K. M. Chem.
Rev. 1999, 99, 3181–3198. (c) M€uller, H.-M.; Seebach, D. Angew. Chem.,
Int. Ed. 1993, 32, 477–502. (d) Sudesh, K.; Abe, H.; Doi, Y. Prog. Polym.
Sci. 2000, 25, 1503–1555.
1
BBL by 2 could be spectroscopically observed. H (Figure 1),
13C, correlation protonꢀproton (COSY), and protonꢀcarbon
(HMQC) NMR analyses were performed to unambiguously
assign NMR resonances for the protons and carbons of the
2652
dx.doi.org/10.1021/om200271q |Organometallics 2011, 30, 2650–2653

Organometallics
COMMUNICATION
(2) (a) Carpentier, J.-F. Macromol. Rapid Commun. 2010, 31,
1696–1705. (b) Thomas, C. M. Chem. Soc. Rev. 2010, 39, 165–173.
(c) Kamber, N. E.; Jeong, W.; Waymouth, R. M.; Pratt, R. C.; Lohmeijer,
B. G. G.; Hedrick, J. L. Chem. Rev. 2007, 107, 5813–5840. (d) Lutz, J.-F.
Nat. Chem. 2010, 2, 84–85.
similar rH values to the ligand IPr HCl (around 5.5 Å), suggesting
3
mononuclear structures for the [Au(NHC)]-containing catalysts.
(16) Bacciu, D.; Cavell, K. J.; Fallis, I. A.; Ooi, L.-l. Angew. Chem., Int.
Ed. 2005, 44, 5282–5284.
(17) Under this reaction manifold no stereoselectivity was observed
and atactic PHBs were produced (13C NMR spectrum, see SI1).
(18) See for instance: Ma, H.; Okuda, J. Macromolecules 2005,
38, 2665–2673.
(3) (a) Jedlinski, Z.; Kurcok, P.; Lenz, R. W. Macromolecules 1998,
31, 6718–6720, and references therein.(b) Coulembier, O.; Dubois, P.
Handbook of Ring-Opening Polymerization; Dubois, P.; Coulembier, O.;
Raquez, J.-M., Eds.; Wiley: Weinheim, 2009; pp 227ꢀ254. (c) Okada,
M. Prog. Polym. Sci. 2002, 27, 87–133. (d) Dechy-Cabaret, O.; Martin-
Vaca, B.; Bourissou, D. Chem. Rev. 2004, 104, 6147–6176. (e) Wu, J.; Yu,
T.-L.; Chen, C.-T.; Lin, C.-C. Coord. Chem. Rev. 2006, 250, 602–626.
(f) Williams, C. K. Chem. Soc. Rev. 2007, 36, 1573–1580. (g) Stanford, M. J.;
Dove, A. P. Chem. Soc. Rev. 2010, 39, 486–494. (h) Billingham, N. C.;
Proctor, M. G.; Smith, J. D. J. Organomet. Chem. 1988, 341, 83–93.
(19) The specific rotation at 25 ꢀC in CHCl3 was [R]25365 = þ6.8
(c 0.010 g/mL).
(20) As a control experiment, the acetate derivative, [Au(IPr)(AcO)],
was tested in the polymerization of BBL at 100 ꢀC and displayed very low
catalytic activity.
ꢁ
(i) Moeller, M.; Kange, R.; Hedrick, J. L. J. Polym. Sci. Part A 2000,
38, 2067–2074. (j) Wu, B.; Lenz, R. W. Macromolecules 1998,
31, 3473–3477. (k) Bloembergen, S.; Holden, D. A.; Bluhm, T. L.;
Hamer, G. K.; Marchessault, R. H. Macromolecules 1989, 22, 1656–1663.
(l) Hori, Y.; Hagiwara, T. Int. J. Biol. Macromol. 1999, 25, 237–245.
(m) Kricheldorf, H. R.; Eggerstedt, S. Macromolecules 1997, 30, 5693–5697.
(4) (a) Rieth, L. R.; Moore, D. R.; Lobkovsky, E. B.; Coates, G. W.
J. Am. Chem. Soc. 2002, 124, 15239–15248. (b) Amgoune, A.; Thomas,
C. M.; Ilinca, S.; Roisnel, T.; Carpentier, J.-F. Angew. Chem., Int. Ed.
2006, 45, 2782–2784. (c) Connor, E. F.; Nyce, G. W.; Myers, M.; M€ock,
A.; Hedrick, J. L. J. Am. Chem. Soc. 2002, 124, 914–915. (d) Coulembier,
O.; Delva, X.; Hedrick, J. L.; Waymouth, R. M.; Dubois, P. Macro-
molecules 2007, 40, 8560–8567. (e) Ajellal, N.; Durieux, G.; Delevoye,
L.; Tricot, G.; Dujardin, C.; Thomas, C. M.; Gauvin, R. M. Chem.
Commun. 2010, 46, 1032–1034. (f) Zintl, M.; Molnar, F.; Urban, T.;
Bernhart, V.; Preishuber-Pfl€ugl, P.; Rieger, B. Angew. Chem., Int. Ed.
2008, 47, 3458–3460.
(5) (a)Horvath, I. T.; Anastas, P. T.Chem. Rev. 2007, 107, 2169–2173.
(b) Gupta, P.; Kumar, V. Eur. Polym. J. 2007, 43, 4053–4074.
(6) (a) Hashmi, A. S. K.; Hutchings, G. J. Angew. Chem., Int. Ed.
2006, 45, 7896–7936. (b) Li, Z.; Brouwer, C.; He, C. Chem. Rev. 2008,
108, 3239–3265. (c) Arcadi, A. Chem. Rev. 2008, 108, 3266–3325.
(d) Jimꢀenez-Nꢀu~nez, E.; Echavarren, A. M. Chem. Rev. 2008, 108,
3326–3350. (e) Marion, N.; Nolan, S. P. Chem. Soc. Rev. 2008,
37, 1776–1782. (f) Nolan, S. P. Acc. Chem. Res. 2011, 44, 91–100.
(7) Urbano, J.; Hormigo, A. J.; de Frꢀemont, P.; Nolan, S. P.; Diaz-
Requejo, M. M.; Pꢀerez, P. J. Chem. Commun. 2008, 759–761.
(8) (a) Ray, L.; Katiyar, V.; Raihan, M. J.; Nanavati, H.; Shaikh,
M. M.; Ghosh, P. Eur. J. Inorg. Chem. 2006, 3724–3730. (b) Ray, L.;
Katiyar, V.; Barman, S.; Raihan, M. J.; Nanavati, H.; Shaikh, M. M.;
Ghosh, P. J. Organomet. Chem. 2007, 692, 4259–4269.
(9) For the use of NHCs in late transition metal catalysis see: Díez-
Gonzꢀalez, S.; Marion, N.; Nolan, S. P. Chem. Rev. 2009, 109, 3612–3676.
(10) Kramer, J. W.; Treitler, D. S.; Dunn, E. W.; Castro, P. M.;
Roisnel, T.; Thomas, C. M.; Coates, G. W. J. Am. Chem. Soc. 2009,
131, 16042–16044.
(11) For recent examples, see: (a) Nun, P.; Gaillard, S.; Poater, A.;
Cavallo, L.; Nolan, S. P. Org. Biomol. Chem. 2011, 9, 101–104.
(b) Ramon, R. S.; Gaillard, S.; Poater, A.; Cavallo, L.; Slawin,
A. M. Z.; Nolan, S. P. Chem.—Eur. J. 2011, 17, 1238–1246.
(12) (a) Gaillard, S.; Slawin, A. M. Z.; Nolan, S. P. Chem. Commun.
2010, 46, 2742–2744. (b) Boogaerts, I. I. F.; Nolan, S. P. J. Am. Chem.
Soc. 2010, 132, 8858–8859.
(13) See SI1 for experimental synthetic and characterization details.
(14) Attempts of crystallization of 3 and 4 revealed that these
complexes are quite sensitive to prolonged exposure to moisture that
is present in recrystallization solvents, and reaction with water was
observed to produce the Au-O-SiR3 complex (6) (see SI1).
(15) To investigate a possible aggregation of goldꢀNHC catalysts in
solution, a pulse gradient spin echo (PGSE) NMR study was conducted
under polymerization conditions. The experimental translational diffu-
sion (Dt) of 1, 3, and IPr HCl in THF was used in the StokesꢀEinstein
3
equation to determine their hydrodynamic radii (rH). 1 and 3 exhibit
2653
dx.doi.org/10.1021/om200271q |Organometallics 2011, 30, 2650–2653