Revolutionary phosgene-free synthesis of α-amino acid N-carboxyanhydrides using diphenyl carbonate based on activation of α-amino acids by converting into imidazolium salts

Article abstract of DOI:10.1002/pola.24213

The phosgene-free synthesis of α-amino acid n-carboxyanhydrides using diphenyl carbonate based on activation of α-amino acids by converting into imidazolium salts was reported. 1-Ethyl-3-methylimidazolium bromide (4.77 g, 25.0 mmol) was dissolved in water (25 mL) and was passed through a column of Amberlite IRA 400 CL (50 cm³) using 250 mL of water as an eluent to synthesize amino acid imidazolium salt. A solution of L-phenylalanine imidazolium salt (550 mg, 2.0 mmol) in acetonitrile (15 mL) was added dropwise to a solution of diphenyl-carbonate (427 mg, 2.00 mmol) in acetonitrile (5 mL) at room temperature, and the reaction mixture was stirred at room temperature to synthesize urethane derivative. The results showed that all of the obtained imidazolium salts were soluble in chloroform, dichloromethane, 2-butanone acetonitrile, and not soluble in tetrahydrofuran.

Full text of DOI:10.1002/pola.24213

Revolutionary Phosgene-Free Synthesis of a-Amino Acid
N-Carboxyanhydrides Using Diphenyl Carbonate Based on Activation
of a-Amino Acids by Converting into Imidazolium Salts
KOICHI KOGA, ATSUSHI SUDO, TAKESHI ENDO
Molecular Engineering Institute, Kinki University, 11-6 Kayanomori, Iizuka, Fukuoka 820-8555, Japan
Received 7 April 2010; accepted 26 June 2010
DOI: 10.1002/pola.24213
Published online in Wiley Online Library (wileyonlinelibrary.com).
KEYWORDS: monomers; polypeptides; ring-opening polymerization
INTRODUCTION N-Carboxylic anhydrides of a-amino acids
(NCAs) are highly useful monomers, because their living ani-
onic ring-opening polymerizations permit precise construc-
tion of various polymer architectures having well-defined
polypeptide sequences.1–10 Currently, NCAs are synthesized
conveniently by treatment of a-amino acids with phosgene,
phosgene dimer, or phosgene trimer.11–14 These carbonyl
sources are highly reactive, but at the same time, they are
lethally toxic so that their uses are limited to only small-
scale operations. Another method established eariler than
the phosgene method is ‘‘Leuch’s method,’’ which is based on
the cyclization of N-alkoxycarbonyl amino acid chloride.15
This method allows a straightforward access to NCA; how-
ever, its application to industrial production of NCA would
be difficult, because of the requirement of the use of corro-
sive halogenating reagents (such as PBr3, PCl5, and SOCl2)
and the formation of carcinogenic alkyl halides as a result of
the cyclization.
which was so serious to be overcome by weak electrostatic
interaction between a-amino acids and DPC. To accomp-
lish this challenging task, that is, use of DPC as a safer
and less-expensive phosgene-alternative for NCA synthesis,
we designed a new route that consisted of three steps
(Scheme 1): The first step is derivation of a-amino acids in a
zwitter ionic form into their imidazolium salts 1, which were
reported by Ohno and co-workers.21,22 We considered that
this derivation can be regarded as a facile method to liberate
the amino group from the protonation, and at the same time,
as a convenient method to transform a-amino acids into
‘‘soluble’’ derivatives. The second step is a reaction of the
free amino group of 1 with DPC to obtain the corresponding
urethane derivative 2, and the third step is the cyclization of
2 to give NCA 3.
RESULTS AND DISCUSSION
The imidazolium salts of various a-amino acids were readily
prepared according to the reported method by Ohno and
co-workers.21,22 All of the obtained imidazolium salts herein
were soluble in chloroform, dichloromethane, 2-butanone,
acetonitrile, and not soluble in tetrahydrofuran. N-Carbamoy-
lation of the obtained imidazolium salts 1 with DPC was
successfully conducted in acetonitrile, which can dissolve
L-phenylalanine imidazolium salt 1a, to give the correspond-
ing 2a within 10 min without any side reactions. 1H- and
13C-NMR of 2a are shown in Figure S1 (Supporting Informa-
tion). To confirm that this process was free from racemi-
zation, 2a was treated with trimethylsilyl chloride in dry
methanol to obtain the corresponding methyl ester 4a in
Figure 1(a),23 and its optical purity was studied by high-per-
formance liquid chromatography (HPLC) equipped with a
column of chiral stationary phase. The resulting chromato-
gram, which is shown in Figure 1(b) along with that for the
For industrial production of NCAs, development of new
methods with using safer reagents without formation is
strongly desired. As a phosgene-alternative, we have focused
our attention on diphenyl carbonate (DPC): DPC is a much
less toxic and inexpensive chemical, which can be produced
by a completely phosgene-free process from ethylene oxide,
phenol, and carbon dioxide.16–18 We examined the reactivity
of a-amino acids with DPC; however, despite surveying
various conditions, they did not react with DPC at all. On the
other hand, amino acids were successfully converted to
NCAs by treatment with analogous but much more electro-
philic carbonates such as bis(2,4-dinitrophenyl)carbon-
ate.19,20 These results implied that the protonated amino
group of a-amino acids in a zwitter-ionic form was not
enough nucleophilic to react with less electrophilic DPC.
Another barrier was the insoluble nature of a-amino acids,
Additional Supporting Information may be found in the online version of this article. Correspondence to: T. Endo (E-mail: tendo@mol-eng.fuk.
kindai.ac.jp)
C
V
Journal of Polymer Science: Part A: Polymer Chemistry, Vol. 48, 4351–4355 (2010) 2010 Wiley Periodicals, Inc.
RAPID COMMUNICATION, KOGA, SUDO, AND ENDO
4351

JOURNAL OF POLYMER SCIENCE: PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
SCHEME 1 Synthesis of urethane deriva-
tives 2 by reaction of imidazolium salts of
a-amino acids 1 with diphenylcarbonate
and their cyclization reactions.
authentic racemic mixture of 2a synthesized from DL-phenyl-
alanine, clearly confirmed that racemization of the chiral
center during the N-carbamoylation step was negligible. In a
similar way, the other imidazolium salts 1b and 1c were
also successfully converted to the corresponding urethanes
2b–2c. The reactions completed within 30 min, and the
urethanes were isolated in approximately 80% yield.
cyclization very slowly. As a result, the corresponding NCA
3a was obtained in 9% yield at 40 h (Fig. 2). Encouraged by
this result, we performed the reaction at higher temperature
for prolonged reaction time; however, it was interfered by
serious oligomerization of the formed NCA. Addition of basic
compounds such as triethylamine and sodium bicarbonate
for the purpose of enhancing the nucleophilicity of the
carboxyl group was also not suitable, because they induced
further serious oligomerization of the formed NCA. Eventu-
ally, we examined addition of acidic compounds and discov-
ered that acetic acid (AA) was effective additive to accelerate
the cyclization reaction of 2a into the corresponding NCA
3a. In Figure 2, the time-dependence of the yield of 3a,
which was measured by 1H-NMR, is shown. A series of the
corresponding spectra are shown in Figure 3. As Figure 2
shows, the addition of a 3.0 equivalent amount of AA
resulted in complete transformation of 2a into the
The last highlight of this work was the selective cyclization
of the urethane derivatives 2 into NCA. First, 2a was heated
in acetonitrile at 80 ꢀC to find whether it underwent the
FIGURE 1 (a) Derivation of urethane 2a into the corresponding
methyl ester 4a. (b) HPLC chromatogram of L-phenylalanine-
derived methyl ester 4a and that of DL-phenylalanine-derived
methyl ester rac-4a (column: Daicel Chiralpak OD-H, eluent:
hexane/2-propanol ¼ 5/2 (v/v)).
FIGURE 2 Time-conversion relationships for the cyclization of
the carbamate 2a into NCA 3a at 60 ꢀC.
4352
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

RAPID COMMUNICATION
FIGURE 3 1H-NMR spectra of the reaction
mixture (a) after 6 h, (b) after 15 h, and (c)
after 29 h in the intramolecular cyclization
of 2a into NCA 3a.
corresponding NCA 3a within 30 h. This acceleration effect
by AA was quite unexpected, and its mechanism is not clear,
but it may involve enhancement of electrophilicity of the
urethane moiety by the added carboxylic acid. In a similar
way, the cyclization reactions of L-methionine-derived urethane
2b and c-t-butyl-L-glutamate-derived urethane 2c were per-
formed. As shown in Figure 2, 2b and 2c were smoothly con-
verted to the corresponding NCAs 3b and 3c, respectively.
increased to enhance hydrophobic interaction between
phenol and charcoal. Then, activated charcoal (1250 wt % to
the theoretical weight of phenol released by the cyclization
reaction) was added to the mixture. The mixture was filtered
to obtain a solution of NCA 3a free from phenol, from which
the volatiles (acetonitrile and AA) were removed under
reduced pressure to obtain NCA 3a in 78% yield. Its 1H-
NMR spectrum is shown in Figure S2 (Supporting Informa-
tion). NCA 3a was stable even in the presence of water, if
the medium was enough acidified. The melting point of the
The cyclization of urethanes 2 was accompanied by forma-
tion of an equimolar amount of phenol, and, thus, NCAs 3
were obtained as a mixture with phenol. We found that
phenol was efficiently removed by adsorption on activated
charcoal to permit facile isolation of NCA: The cyclization of
2a was performed in acetonitrile with using AA as a pro-
moter as was described above. To the resulting reaction
mixture containing NCA 3a and phenol, 1 M hydrochloric
acid (1/10 volume of the solution) was added. By addition
of hydrochloric acid, hydrophilicity of the medium was
ꢀ
obtained phenylalanine NCA was 89.1–90.4 C, which was in
24
ꢀ
good agreement with the reported one (88–90 C) to con-
firm the high purity of the obtained NCA.
Summary
In summary, we developed a new ‘‘phosgene-free’’ synthetic
route from a-amino acids to NCAs with using DPC, a much
less toxic and more easily available carbonyl source than
phosgene. All the steps are free from toxic and corrosive
RAPID COMMUNICATION, KOGA, SUDO, AND ENDO
4353

JOURNAL OF POLYMER SCIENCE: PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
reagents, free from special caution to moisture, and free
from toxic by-products. This ‘‘safety’’ ensured in the present
phosgene-free synthesis will open up a new vista of the
future industrial production of NCAs and its contribution to
the prospect of polypeptide-related science and technology.
mixture was concentrated under reduced pressure, and
the resulting residue was transferred into a separating fun-
nel with diluting with water (25 mL) and ethyl acetate
(50 mL). The ethyl acetate layer was separated, and the
aqueous layer was extracted with 50 mL of ethyl acetate
twice. The combined organic layers were washed with water
and brine, dried over anhydrous sodium sulfate, filtered, and
concentrated under reduced pressure. The residue was
charged on a silica gel column (4 cm ꢁ 15 cm) and eluted
with ethyl acetate/n-hexane ¼ 1/3 to remove phenol first;
then, 2a was eluted with ethyl acetate. The ethyl acetate
solution was concentrated under reduced pressure to
obtained 2a (502 mg, 1.76 mmol, 88%) as a white solid:
[a2D0] þ 91.2 (in chloroform, c ¼ 1.0); 1H-NMR (CDCl3, d in
ppm): 3.00–3.35 (m, 2H), 4.76 (m, 1H), 5.45 (d, J ¼ 8.2 Hz,
1H), 7.09 (d, J ¼ 7.7 Hz, 2H), 7.24 (m, 3H), 7.34 (m, 5H);
13C-NMR (CDCl3, d in ppm) 37.60, 54.66, 121.5, 125.6, 127.4,
128.8, 129.30, 129.33, 135.3, 150.7, 154.2, 176.2.
EXPERIMENTAL
Materials and Measurements
L-Phenylalanine, L-methionine, 1-ethyl-3-methylimidazolium
bromide, AA, D,L-phenylalanine, and activated charcoal were
purchased from Wako Pure Chemical Industries (Wako). c-t-
Butyl-L-glutamic acid was purchased from Watanabe Chemi-
cal. Amberlite IRA 400 CL (anion exchange resin) was
purchased from Aldrich. Diphenylcarbonate was purchased
from Tokyo Chemical Industry Co. These reagents were used
as received. Acetonitrile was purchased from Wako Pure
Chemical Industries and dried over calcium hydride and then
distilled before use. 1H-NMR (400 MHz; dtetramethylsilane
¼
0.00 ppm, dchloroform ¼ 7.26 ppm) and 13C-NMR (100.6 MHz;
dchloroform ¼ 77.00 ppm) were recorded on a Varian NMR
spectrometer model Unity INOVA. HPLC analysis was per-
formed on JASCO Intelligent LC1500 with detection at
258 nm at 20 ꢀC, with using a chiral column (Daicel Chiral-
pak OD-H). The flow rate was 1.0 mL/min. Optical rotation
was measured with JASCO DIP-1000 at 21.6 ꢀC, using a
quartz cell of 100 mm length.
From L-methionine imidazolium salt 1b (519 mg, 2.00 mmol),
the corresponding urethane 2b (439 mg, 1.62 mmol, 82%)
was obtained as a white solid: 1H-NMR (CDCl3, d in ppm):
2.02–2.34 (m, 5H), 2.65 (m, 2H), 4.58 (m, 1H), 5.45 (d, J ¼
8.3 Hz, 1H), 7.10–7.42 (m, 5H).
From c-t-butyl-L-glutamic acid imidazolium salt 1c (627 mg,
2.00 mmol), the corresponding urethane 2c (470.1 mg,
2.00 mmol, 76%) was obtained as a white solid: 1H-NMR
(CDCl3, d in ppm): 1.45 (s, 9H), 1.98–2.32 (m, 2H), 2.45 (m,
2H), 4.45 (m, 1H), 5.89 (d, J ¼ 8.3 Hz, 1H), 7.15–7.42 (m, 5H).
Synthesis of Amino Acid Imidazolium Salt 1
Typical procedure is as follows: 1-Ethyl-3-methylimidazo-
lium bromide (4.77 g, 25.0 mmol) was dissolved in water
(25 mL) and was passed through a column of Amberlite IRA
400 CL (50 cm3) using 250 mL of water as an eluent. The
eluted solution of 1-ethyl-3-methylimidazolium hydroxide
(25.0 mmol) was added dropwise to a solution of L-phenyl-
Derivation of Urethane 2a into the
Corresponding Methyl Ester 4a
To a solution of the alanine-derived urethane 2a (285 mg,
1.00 mmol) in dry methanol (10 mL), trimethylsilyl chloride
(0.297 mL, 2.33 mmol) was added at room temperature, and
the resulting solution was stirred at room temperature. After
18 h, the solution was concentrated under reduced pressure
to obtain the corresponding methyl ester 4a (294 mg, 98%)
as a colorless oil: [a2D5] þ 83.4 (in chloroform, c ¼ 1.0); 1H-
NMR (CDCl3, d in ppm) 3.00–3.35 (m, 2H), 3.77 (s, 3H), 4.76
(m, 1H), 5.45 (d, J ¼ 8.2 Hz, 2H), 7.09 (d, J ¼ 7.7 Hz, 2H),
7.24 (m, 3H), 7.34 (m, 5H). Optical purity of the resulting
methyl ester 4a was 99% e.e., which was determined by HPLC
analysis with chiral column (Daicel Chiralpak OD-H). A mixture
of hexane and 2-propanol (5/2 in v/v) was used as the eluent.
ꢀ
alanine (4.96 g, 30.0 mmol) in water at 0 C, and the result-
ꢀ
ing solution was stirred at 0 C. After 12 h, the solution was
concentrated under reduced pressure with heating at 40–
50 ꢀC. To the resulting residue, acetonitrile (400 mL) and
methanol (50 mL) were added, and the resulting mixture
ꢀ
was stirred vigorously at 0 C. By this process, unconsumed
L-phenylalanine precipitated out of the solution. The mixture
was filtered, and the filtrate was concentrated under reduced
pressure and then dried in vacuo for 5 h at 60 ꢀC to obtain
L-phenylalanine imidazolium salt 1a (5.78 g, 21.2 mmol, 84%).
L-Methionine imidazolium salt 1b (4.99 g, 19.3 mmol, 77%)
was obtained in a manner similar to that of L-phenylalanine
imidazolium salt 1a. Similarly, from c-t-butyl-L-glutamic acid
(2.03 g, 12.0 mmol) and 1-ethyl-3-methylimidazolium bro-
mide (1.91 g, 10.0 mmol), c-t-butyl-L-glutamic acid imidazo-
lium salt 1c (2.97 g, 9.47 mmol, 95%) was obtained.
Intramolecular Cyclization of 2 into NCA 3
and Its Isolation
Typical procedure is as follows: In an oven-dried glassware,
2a (570 mg, 2.00 mmol) and AA (360 mg, 6.00 mmol) 2a
were dissolved in acetonitrile (20 mL), and the resulting
solution was stirred at 80 ꢀC under nitrogen. During
the reaction, several portions of the reaction mixture were
taken out at prescribed times, concentrated under reduced
pressure, and dissolved in CDCl3 for 1H-NMR analysis. The
characteristic multiplet signal for the methine proton of NCA
was observed at 4.4 ppm, and this signal was used for calcu-
lating NMR yields of NCA. The reaction mixture was cooled
to room temperature, and then 1 M HCl solution (2.0 mL)
Synthesis of Urethane Derivative 2
Typical procedure is as follows: To a solution of diphenyl-
carbonate (427 mg, 2.00 mmol) in acetonitrile (5 mL), a
solution of 1a (550 mg, 2.0 mmol) in acetonitrile (15 mL)
was added dropwise at room temperature, and the reaction
mixture was stirred at room temperature. After 15 min, the
reaction mixture was poured into 1 M hydrochloric acid. The
4354
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

RAPID COMMUNICATION
J Chem Soc 1950, 3218–
11 Coleman, D.; Farthing, A. C.
and activated charcoal (2.4 g) were added to the reaction
mixture. After stirring for 1 h at room temperature, the mix-
ture was filtered, and the filtrate was concentrated under
reduced pressure to obtain NCA 3a (302 mg, 1.56 mmol,
3222.
12 Wilder, R.; Mobashery, S. J Org Chem 1992, 57, 2755–
2756.
ꢀ
78%) as a white solid: m.p. ¼ 89.1–90.4 C.
13 Katakai, R.; Iizuka, Y. J Org Chem 1985, 50, 715–716.
14 Daly, W. H.; Poche, D. Tetrahedron Lett 1988, 29, 5859–5862.
15 Leuchs, H. Ber Dtsch Chem Ges 1906, 39, 857–859.
REFERENCES AND NOTES
1 Hattori, I.; Hirao, A.; Yamaguchi, K.; Nakahama, S.; Yamazaki,
16 Fukuoka, S.; Hachiya, H.; Matsuzaki, K.; Miyaji, H. PCT Int.
N. Makromol Chem 1983, 184, 1355–1362.
Appl. WO2006022294, 2006.
2 Chong, S. C.; Byung, W. J.; Joong, K. K.; Komoto, T. Macro-
17 Fukuoka, S.; Miyaji, H.; Hachiya, H.; Matsuzaki, K. PCT Int.
mol Chem Phys 1994, 195, 2195–2206.
Appl. WO2006043491, 2006.
3 Hirokawa, Y.; Hayashi, T. Eur Polym J 1994, 30, 1397–1401.
4 Deming, T. J. Nature (London) 1997, 390, 386–389.
18 Fukuoka, S.; Miyaji, H.; Hachiya, H.; Matsuzaki, K. PCT Int.
Appl. WO2006041075, 2006.
5 Brezezinska, K. R.; Deming, T. J. Macromolecules 2001, 34,
19 Fujita, Y.; Koga, K.; Kim, H.-K.; Wang, X.-S.; Sudo, A.; Nish-
ida, H.; Endo, T. J Polym Sci Part A: Polym Chem 2007, 45,
5365–5369.
4348–4354.
6 Fan, Y.; Chen, G.; Tanaka, J.; Tateishi, T. Biomacromolecules
2005, 6, 3051–3056.
20 Fujita, Y.; Sudo, A.; Nishida, H.; Endo, T. U.S. Patent 2007/
0015932, January 18, 2007.
7 Higuchi, M.; Inoue, T.; Miyoshi, H.; Kawaguchi, M. Langmuir
2005, 21, 11462–11467.
21 Fukumoto, K.; Yoshizawa, M.; Ohno, H. J Am Chem Soc
2005, 127, 2398–2399.
8 Jurgen, M. S.; Peter, S.; Matthijs, B. J. O.; Sylvia, S.; Hendrik,
G. S.; Jan, C. V. H. Macromolecules 2006, 39, 2989–2997.
22 Ohno, H.; Fukumoto, K. Acc Chem Res 2007, 40, 1122–1129.
9 Aoi, K.; Katsube, M.; Tsutsumuchi, K.; Okada, M. Kobunshi
23 Edmonds, M. K.; Abell, A. D. J Org Chem 2001, 66,
Ronbunshu 2002, 59, 742–752.
3747–3752.
10 Appoh, F. E.; Thomas, D. S.; Kraatz, H. B. Macromolecules
24 Kricheldorf, H. R.; von Lossow, C.; Schwarz, G. Macromol
2006, 39, 5629–5638.
Chem Phys 2005, 206, 282–290.
RAPID COMMUNICATION, KOGA, SUDO, AND ENDO
4355