Epimerization-suppressed organocatalytic synthesis of poly-L-lactide in supercritical carbon dioxide under plasticizing conditions

Article abstract of DOI:10.1016/j.tetlet.2019.150987

Herein, an efficient (>95% yield, >99.0% ee) Br?nsted acid-catalyzed synthetic method of poly-L-lactide (PLLA) in supercritical carbon dioxide (scCO₂) under plasticizing conditions is presented. High-performance liquid chromatography analysis of the PLLA hydrolysis products indicated that, as opposed to the case of organic solvents, the use of a nucleophilic catalyst in scCO₂ suppressed the epimerization. The highly stereochemically pure PLLA prepared by the developed method under metal-free conditions meets the criteria of medicinal/engineering applications.

Full text of DOI:10.1016/j.tetlet.2019.150987

Tetrahedron Letters 60 (2019) 150987
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Epimerization-suppressed organocatalytic synthesis of poly-L-lactide in
supercritical carbon dioxide under plasticizing conditions
a
a
a
Nobuyuki Mase a,b,c, , Moniruzzaman , Shoji Yamamoto , Kohei Sato , Tetsuo Narumi a,b,c, Hikaru Yanai d
⇑
a Department of Engineering, Graduate School of Integrated Science and Technology, Shizuoka University, 3-5-1 Johoku, Hamamatsu, Shizuoka 432-8561, Japan
b Graduate School of Science and Technology, Shizuoka University, 3-5-1 Johoku, Hamamatsu, Shizuoka 432-8561, Japan
c Research Institute of Green Science and Technology, Shizuoka University, 3-5-1 Johoku, Hamamatsu, Shizuoka 432-8561, Japan
d School of Pharmacy, Tokyo University of Pharmacy and Life Sciences, 1432-1 Horinouchi, Hachioji, Tokyo 192-0392, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Herein, an efficient (>95% yield, >99.0% ee) Brønsted acid-catalyzed synthetic method of poly-L-lactide
Received 4 June 2019
Revised 22 July 2019
Accepted 26 July 2019
Available online 26 July 2019
(PLLA) in supercritical carbon dioxide (scCO2) under plasticizing conditions is presented. High-perfor-
mance liquid chromatography analysis of the PLLA hydrolysis products indicated that, as opposed to
the case of organic solvents, the use of a nucleophilic catalyst in scCO2 suppressed the epimerization.
The highly stereochemically pure PLLA prepared by the developed method under metal-free conditions
meets the criteria of medicinal/engineering applications.
Keywords:
Ó 2019 Elsevier Ltd. All rights reserved.
CO2 plasticizing polymerization
Organocatalysis
Polylactide
Supercritical fluid
Introduction
Hedrick et al. developed a ring-opening PLA synthesis promoted
by metal-free organocatalysts such as 4-(dimethylamino)pyridine
Polylactide (PLA), a biodegradable plastic that has attracted
considerable attention as a carbon–neutral material [1], contains
numerous chiral centers in its polymer main chain and exhibits
physical properties that are greatly influenced by its stereochemi-
cal purity [2]. Generally, mixing highly stereochemically pure poly-
(DMAP) in 2001 [7], and ever since, the organocatalytic approach
has progressed considerably [8]. Hedrick et al. also reported that
no epimerization was observed during the DMAP-catalyzed PLA
synthesis at low temperature (35 °C). However, in the abovemen-
tioned study, the stereochemical purity of PLA was determined
by 13C NMR analysis, which lacks quantitative accuracy (Fig. 1,
Eqs. (1) and (2)) [7].
Recently, we reported the organocatalytic synthesis of metal-,
organic solvent-, and residual monomer-free PLA in supercritical
carbon dioxide (scCO2) under CO2 plasticizing polymerization
(CPP) conditions using low temperature (60 °C) and a short reac-
tion time (1 h) [9]. In the present work, this CPP method was
applied to the synthesis of PLA with high stereochemical purity
and a quantitative method was established to determine the extent
of epimerization (Fig. 1).
L-lactide (PLLA) and poly-D-lactide (PDLA) results in the formation
of a stereocomplex with a higher melting point [3]. Although PLA is
susceptible to enzymatic hydrolysis, common PLA-degrading
enzymes can specifically degrade PLLA. Therefore, the incorpora-
tion of a D-isomer inhibits hydrolysis and decreases biodegradabil-
ity [4]. Considering these advantages, methods for the synthesis of
PLA with high stereochemical purity are much sought after.
Previously, metal catalyst-promoted ring-opening polymeriza-
tion (ROP) of lactide has been reported as an efficient way to
prepare PLA [5], and the selective synthesis of L- or D-PLA from
racemic lactide has been achieved by adjusting the chiral ligands
of the abovementioned metal catalysts [6]. However, this approach
raises certain safety concerns when thus-prepared PLA is used in
medicinal and engineering applications, because the metal content
of the product cannot be reliably reduced to zero.
Results and discussion
Usually, the stereochemical purity of PLA is determined based
on its stereoregularity using 13C and 1H NMR spectroscopy [10].
However, the inherently poor quantitativeness of NMR analysis
complicates the accurate determination of the degree of epimer-
ization. To mitigate this problem, chiral HPLC was used in this
study to determine the enantiopurity of lactic acid (3) produced
⇑
Corresponding author.
E-mail address: mase.nobuyuki@shizuoka.ac.jp (N. Mase).
https://doi.org/10.1016/j.tetlet.2019.150987
0040-4039/Ó 2019 Elsevier Ltd. All rights reserved.

2
N. Mase et al. / Tetrahedron Letters 60 (2019) 150987
Based on these reports, highly enantiopure (>99.0% ee) lactide 1
and dimer 4 were hydrolyzed in 1 M aqueous NaOH at reflux. No
epimerization was observed under these conditions (Scheme 1),
thereby demonstrating that 1 and 4 could not be deprotonated at
the a-position even by a strong base such as NaOH. The utilization
of this simple hydrolysis technique allowed the evaluation of the
degree of epimerization of both the polymer and unreacted
monomer.
To confirm reproducibility, an organocatalytic ROP of 1 was
conducted according to a previously reported epimerization-avoid-
ing procedure [7] and found that this reaction was not complete
even after 30 h. HPLC analysis of the hydrolysis products indicated
that the occurrence of epimerization decreased the enantiopurity
of the product to 98.0% ee (Scheme 2, Table 1, entry 1). To increase
the product yield, the reaction was carried out at 60 °C in CHCl3.
However, after 20 h under these conditions, the product enantiop-
urity decreased to 88.0% ee (entry 2). In contrast, under CPP condi-
tions (60 °C, 10 MPa) in scCO2, the polymerization reaction was
complete within 1 h and the obtained product exhibited an
increased enantiopurity of 96.5% ee (entry 3). Similar to DMAP,
nucleophilic and basic organocatalysts (9-azajulolidine (9-AJ),
1,8-diazabicyclo[5.4.0]undec-7-ene (DBU), and 1,1,3,3-tetram-
ethylguanidine (TMG)) suppressed the epimerization under CPP
conditions (entries 4–9). Furthermore, using a conventional cata-
lyst, tin(II) 2-ethylhexanoate [5], did not lead to epimerization in
toluene after 5 h polymerization (entry 10), although it did pro-
gress to a significant extent under the high-temperature conditions
employed in commercial bulk polymerization (entry 11).
Fig. 1. Synthesis of highly stereochemically pure PLLA.
by the hydrolysis of PLA, since the abovementioned hydrolysis is
known to proceed without epimerization when performed using
NaOH solution at room temperature [11] or under hydrothermal
conditions [12].
O
O
O
[a]
[a]
O
O
OH
EtO
OH
O
OH
O
O
1
3
4
(>99.0% ee)
(>99.0% ee)
(>99.0% ee)
Epimerization in nucleophilic or basic organocatalyst-promoted
PLLA synthesis is presumed to be caused by deprotonation at the
methine carbon. To shed further light on the mechanism of this
[a] 1 M NaOH, reflux, 4 h
Scheme 1. Hydrolysis of
L
-lactide (1) and dimer 4.
Catalysts:
O
O
catalyst (6.6 mol%)
EtOH (3.3 mol%)
O
O
N
N
NH
1 M NaOHaq.
reflux, 4 h
N
N
O
H
O
OH
N
N
O
N
EtO
N
60 °C, 1 h
O
n
OH
3
DMAP
9-AJ
DBU
TMG
O
1
2
(>99.0% ee)
Scheme 2. Organocatalyst-promoted synthesis of PLA in organic solvents and scCO2.
Table 1
Nucleophilic and basic organocatalyst-promoted syntheses of PLA in organic solvents and scCO2.a
Entry
Catalyst
Solvent
Conversion (%)b
Mnc
PDId
ee of 3 (%)e
1f
2 g
3
4
5
DMAP
DMAP
DMAP
9-AJ
9-AJ
DBU
CH2Cl2
CHCl3
scCO2
CHCl3
scCO2
CHCl3
scCO2
CHCl3
scCO2
toluene
Neat
79
3800
3900
4100
4100
4900
4100
5900
4100
4800
4200
3100
1.18
1.23
1.10
1.24
1.27
1.90
2.88
1.90
1.63
1.36
1.79
98.0
88.0
96.5
89.5
94.0
59.0
81.0
89.5
89.5
> 99.0
88.0
>95
>95
>95
>95
>95
>95
>95
>95
>95
89
6
7
8
9
DBU
TMG
TMG
10 h
11i
Sn cat.j
Sn cat.j
a
Procedure: A 12 mL pressure-resistant vessel was charged with L-lactide (1, 864 mg, 6.0 mmol, 1.0 equiv), catalyst (0.40 mmol, 6.6 mol%), and ethanol (11.6 lL, 0.2 mmol,
3.3 mol%), and the mixture was warmed to 60 °C using a water bath. scCO2 (60 °C, 10 MPa) was introduced into the vessel and the mixture was stirred for 5 min. After 1 h
without stirring, the pressure was reduced to ambient values, and the produced PLA (white solid) was removed.
b
Determined by 1H NMR.
Determined by gel permeation chromatography (GPC) using polystyrene as a standard.
Polydispersity index (PDI, Mw/Mn).
Determined by HPLC after hydrolysis of PLA to lactic acid.
Polymerization was carried out at 35 °C for 30 h.
Polymerization was carried out for 20 h (for 5 h, 65% conv., Mn = 2900, PDI = 1.19, ee = 93.5%).
Polymerization was carried out for 5 h under reflux condition.
Polymerization was carried out at 200 °C.
c
d
e
f
g
h
i
j
Tin(II) 2-ethylhexanoate.

N. Mase et al. / Tetrahedron Letters 60 (2019) 150987
3
Scheme 3. Proposed epimerization mechanisms.
epimerization, DMAP, DBU, or TMG was added to a solution of PLA
in chloroform, and the resulting mixture was stirred at 60 °C for
10 h (DMAP) or 1 h (DBU and TMG). Notably, there was no sub-
stantive decrease in enantiopurity, which implied that the epimer-
ization was extremely slow (Scheme 3–1). Therefore, in order to
confirm the epimerization of lactide 1, it was polymerized in the
absence of an initiator, and the recovered unreacted monomer
was shown to contain meso-lactide 1 (Scheme 3–2). The nucle-
use of phosphoric acid catalyst 8 afforded non-epimerized PLLA,
albeit in low yield (Scheme 4, Table 2, entry 1) [18]. Since fluori-
nated compounds generally have a good affinity with scCO2, the
catalytic ability of fluorinated catalysts would be higher in scCO2
because of a high solubility and low dielectric constant than that
in conventional organic solvent. Therefore, we selected the fluori-
nated Brønsted acids 9–12 as a catalyst. CAHAacidic catalyst 9
exhibited low reactivity (entry 2) [19], which was greatly improved
in the case of NAHAacidic catalysts. Notably, the polymerization
reaction was complete within 5 h in scCO2, whereas a reaction time
of 192 h was required in CH2Cl2 [16]. Also, cyclic sulfonimide 12 as
well as linear sulfonimides 10 and 11 effectively catalyzed ROP and
furnished PLLA without any reduction in stereochemical purity
(entries 3–5). Although there is the difference of the Mn depending
on the catalysts, the reason is still unclear.
ophilic addition of the catalyst to
6, which has a considerably acidic
the corresponding carbon could be deprotonated by the zwitterion,
with the subsequent re-protonation affording -lactide and
L-lactide afforded intermediate
a-methine proton. Therefore,
L
epimerized meso-lactide (Scheme 3-3). In fact, the higher extent
of epimerization observed for DBU compared to that for DMAP
was ascribed to the higher nucleophilicity of the former [13]. How-
ever, although 9-AJ is also more nucleophilic than DMAP [13], the
presence of substituents at the 3- and 5-positions of the pyridine
ring in the former case presumably led to steric hindrance, which
inhibited the deprotonation and intramolecular cyclization pro-
cesses. Although the reason for why epimerization was suppressed
in scCO2 as compared to CHCl3 is unclear, this phenomenon could
be ascribed to the formation of polymerization-promoting highly
reactive anionic intermediates. These intermediates were not sta-
O
Organocatalyst
O
EtOH (3.3 mol%)
O
H
O
O
O
EtO
scCO2, 10 MPa
60 °C, 5 h
O
n
O
1
2
bilized by scCO2 in view of its low dielectric constant (
10 MPa, 40 °C) [14], which is similar to those of pentane
r = 1.84) and hexane ( r = 1.89), and therefore, the occurrence of
er = 1.37 at
(e
e
Organocatalysts:
deprotonation-induced epimerization was suppressed.
Epimerization promoted by deprotonation is unavoidable when
nucleophilic or basic organocatalysts are used. However, the Sn
(Oct)2-catalyzed polymerization of L-lactide via Lewis acid activa-
tion reportedly proceeded without epimerization at low tempera-
ture (Table 1, entry 10).
Additionally, Brønsted acid-catalyzed ROP featuring a monomer
activation mechanism similar to that of Lewis acid catalysis, has
recently been reported. For example, the following Brønsted acids
have been successfully used for PLLA synthesis in CH2Cl2 (although
they still require long reaction times): triflic acid (3.3 mol%, 24 h,
>95% conversion, Mn = 1800, PDI = 1.38) [15], triflimide (10 mol%,
192 h, 91% conversion, Mn = 4070, PDI = 1.15) [16], and diphenyl
phosphate (DPP)/DMAP (DPP 6 mol%, DMAP 18 mol%, 28 h, 93%
conversion, Mn = 6830, PDI = 1.24) [17]. Therefore, the use of acidic
organocatalysts is expected to suppress epimerization. Indeed, the
OO OO
O2N
S
S
F3C
F
CF3
F
O
O
O
P
OH
F
F
O2N
F
F
F
8
9
F
F
F
F
OO OO
OO OO
S
S
S
S
O S
S O
O
F3C
N
H
CF3 C4F9
N
H
C4F9
N
H
O
10
11
12
Scheme 4. Brønsted acid–promoted syntheses of PLLA in organic solvents and
scCO2.

4
N. Mase et al. / Tetrahedron Letters 60 (2019) 150987
Table 2
Scientific Research on Innovative Areas (No. JP24105511, JP
Brønsted acid-catalyzed PLLA synthesis in scCO2.a.
26105722) from the Ministry of Education, Culture, Sports, Science
and Technology (MEXT). We would like to thank Editage (www.ed-
itage.jp) for English language editing.
Entry
Catalyst(mol%)
Conversion(%)b
Mnc
PDId
ee(%)e
1
2
3
4
5
8 (6.6)
14
1100
< 1000
3000
2800
5200
1.27
2.46
1.33
1.36
1.34
> 99.0
98.5
> 99.0
> 99.0
> 99.0
9 (2.0)
29
10 (2.0)
11 (6.6)
12 (2.0)
> 95
> 95
> 95
Appendix A. Supplementary data
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.tetlet.2019.150987.
a
Procedure: A 12 mL pressure-resistant vessel was charged with
L
-lactide (1,
864 mg, 6.0 mmol, 1.0 equiv), catalyst (0.12 mmol, 2.0 mol% or 0.40 mmol, 6.6 mol
%), and ethanol (11.6 L, 0.2 mmol, 3.3 mol%), and the mixture was warmed to 60 °C
l
References
using a water bath. scCO2 (60 °C, 10 MPa) was introduced into the vessel, and the
mixture was stirred for 5 min. After 5 h without stirring, the pressure was reduced
to ambient values and the produced PLLA (white solid) was removed.
[1] R. Auras, L.-T. Lim, S.E.M. Selke, H. Tsuji, In Poly(Lactic Acid): Synthesis,
Structures, Properties, Processing And Applications, John Wiley & Sons Inc,
Hoboken, New Jersey, 2010.
[2] a) S. Farah, D.G. Anderson, R. Langer, Adv. Drug Delivery Rev. 107 (2016) 367–
392;
b) O. Martin, L. Avérous, Polymer 42 (2001) 6209–6219.
[3] a) H. Tsuji, Adv. Drug Delivery Rev. 107 (2016) 97–135;
b) H. Tsuji, Macromol. Biosci. 5 (2005) 569–597.
b
Determined by 1H NMR.
Determined by GPC with polystyrene standard.
Polydispersity index (Mw/Mn).
Determined by HPLC after hydrolysis of PLA to lactic acid.
c
d
e
[4] a) M.S. Reeve, S.P. McCarthy, M.J. Downey, R.A. Gross, Macromolecules 27
(1994) 825–831;
b) S.H. Lee, W.S. Song, Appl. Biochem. Biotechnol. 164 (2011) 89–102;
c) F. Kawai, K. Nakadai, E. Nishioka, H. Nakajima, H. Ohara, K. Masaki, H. Iefuji,
Polym. Degrad. Stab. 96 (2011) 1342–1348.
[5] a) O. Dechy-Cabaret, B. Martin-Vaca, D. Bourissou, Chem. Rev. 104 (2004)
6147–6176;
b) M.J. Stanford, A.P. Dove, Chem. Soc. Rev. 39 (2010) 486–494;
c) K. Madhavan Nampoothiri, N.R. Nair, R.P. John, Bioresour. Technol. 101
(2010) 8493–8501;
d) S. Corneillie, M. Smet, Polym. Chem. 6 (2015) 850–867.
[6] a) X. Pang, R. Duan, X. Li, C. Hu, X. Wang, X. Chen, Macromolecules 51 (2018)
906–913;
b) P. Horeglad, M. Cybularczyk, A. Litwinska, A.M. Dabrowska, M. Dranka, G.Z.
Zukowska, M. Urbanczyk, M. Michalak, Polym. Chem. 7 (2016) 2022–2036;
c) S. Tabthong, T. Nanok, P. Sumrit, P. Kongsaeree, S. Prabpai, P. Chuawong, P.
Hormnirun, Macromolecules 48 (2015) 6846–6861;
d) K. Press, I. Goldberg, M. Kol, Angew. Chem. Int. Ed. 54 (2015) 14858–14861;
e) X. Pang, R. Duan, X. Li, X. Chen, Polym. Chem. 5 (2014) 3894–3900;
f) C. Bakewell, A.J.P. White, N.J. Long, C.K. Williams, Angew. Chem. Int. Ed. 53
(2014) 9226–9230;
Scheme 5. Proposed mechanism of Brønsted acid-catalyzed ROP in scCO2.
g) S. Bian, S. Abbina, Z. Lu, E. Kolodka, G. Du, Organometallics 33 (2014) 2489–
2495;
h) S. Abbina, G. Du, ACS Macro Lett. 3 (2014) 689–692.
[7] F. Nederberg, E.F. Connor, M. Moller, T. Glauser, J.L. Hedrick, Angew. Chem. Int.
Ed. 40 (2001) 2712–2715.
[8] a) L. Mezzasalma, A.P. Dove, O. Coulembier, Eur. Polym. J. 95 (2017) 628–634;
b) W.N. Ottou, H. Sardon, D. Mecerreyes, J. Vignolle, D. Taton, Prog. Polym. Sci.
56 (2016) 64–115;
In view of the above results, electrophilic Brønsted acids were
concluded to activate the carbonyl group of the L-lactide monomer,
affording activated monomer 10 that is more electrophilic than 1
and can undergo nucleophilic ring-opening by the initiator or
propagating chain-end alcohol, which results in ROP and regener-
ation of the catalyst (Scheme 5). Although the acidic catalyst–pro-
c) C. Thomas, B. Bibal, Green Chem. 16 (2014) 1687–1699;
d) A.P. Dove, ACS Macro Lett. 1 (2012) 1409–1412;
e) M.K. Kiesewetter, E.J. Shin, J.L. Hedrick, R.M. Waymouth, Macromolecules
43 (2010) 2093–2107;
f) N.E. Kamber, W. Jeong, R.M. Waymouth, R.C. Pratt, B.G.G. Lohmeijer, J.L.
Hedrick, Chem. Rev. 107 (2007) 5813–5840.
moted ROP of L-lactide was slow in CH2Cl2 (vide supra), high
reactivity was observed in scCO2 because activated 10 was less sus-
ceptible to solvation in the latter low-dielectric-constant solvent.
[9] a) N. Mase, Moniruzzaman, S. Yamamoto, Y. Nakaya, K. Sato, T. Narumi,
Polymers 10 (2018) 713;
Conclusion
b) N. Mase, Moniruzzaman, S. Mori, J. Ishizuka, F. Kumazawa, S. Yamamoto, K.
Sato, T. Narumi, Tetrahedron Lett. 59 (2018) 4392–4396.
[10] M.T. Zell, B.E. Padden, A.J. Paterick, K.A.M. Thakur, R.T. Kean, M.A. Hillmyer, E.J.
Munson, Macromolecules 35 (2002) 7700–7707.
[11] H. Urayama, T. Kanamori, Y. Kimura, Macromol. Mater. Eng. 286 (2001) 705–
713.
[12] M. Yagihashi, T. Funazukuri, Ind. Eng. Chem. Res. 49 (2010) 1247–1251.
[13] N. De Rycke, F. Couty, O.R.P. David, Chem.-Eur. J. 17 (2011) 12852–12871.
[14] A. Wesch, N. Dahmen, K.H. Ebert, Ber. Bunsenges. Phys. Chem. 100 (1996)
1368–1371.
[15] Bas´ko M, Kubisa P. J. Polym. Sci., Part A: Polym. Chem. 2006; 44: 7071–7081.
Triflic acid is not suitable for use in batch scCO2 reactors because exposure to
moisture needs to be strictly avoided.
In summary, compared to conventional polymerization in
organic solvents, Brønsted acid catalyst-promoted synthesis of
PLLA in scCO2 under CO2 plasticizing polymerization conditions
featured the advantage of reduced epimerization degree, thereby
affording PLLA and PDLA of high stereochemical purity, which
could be potentially used as materials for medical and engineering
applications.
[16] K. Makiguchi, S. Kikuchi, T. Satoh, T. Kakuchi, J. Polym. Sci., Part A: Polym.
Chem. 51 (2013) 2455–2463.
Declaration of Competing Interest
[17] K. Makiguchi, S. Kikuchi, K. Yanai, Y. Ogasawara, S. Sato, T. Satoh, T. Kakuchi, J.
Polym. Sci., Part A: Polym. Chem. 52 (2014) 1047–1054.
[18] In low-polarity scCO2, phosphoric acid forms a dimer exhibiting decreased
reactivity.
[19] a) K. Ishihara, A. Hasegawa, H. Yamamoto, Angew. Chem., Int. Ed. 40 (2001)
4077–4079;
The authors declare no competing financial interest.
Acknowledgments
We are grateful to Ricoh Co., Ltd. for providing highly pure
b) H. Yanai, A. Takahashi, T. Taguchi, Chem. Commun. 46 (2010) 8728–8730;
c) H. Yanai, T. Taguchi, Chem. Commun. 48 (2012) 8967–8969.
L
-lactide 1. This work was supported in part by a Grant-in-Aid for