Fluorenylmethoxycarbonyl-N-methylamino acids synthesized in a flow tube-in-tube reactor with a liquid-liquid semipermeable membrane

Article abstract of DOI:10.1002/ejoc.201300705

Both steps of the N-methylation of 9-fluorenylmethoxycarbonyl (Fmoc) amino acids were carried out in a microstructured tube-in-tube reactor equipped with a semipermeable Teflon AF 2400 membrane as the inner tubing. In the first step, gaseous formaldehyde

Full text of DOI:10.1002/ejoc.201300705

SHORT COMMUNICATION
DOI: 10.1002/ejoc.201300705
Fluorenylmethoxycarbonyl-N-methylamino Acids Synthesized in a Flow Tube-
in-Tube Reactor with a Liquid–Liquid Semipermeable Membrane
Annette E. Buba,[a] Stefan Koch,*[a] Horst Kunz,*[a] and Holger Löwe[a]
Keywords: Amino acids / Flow chemistry / Heterocycles / Microreactors / Oxazolidinones
Both steps of the N-methylation of 9-fluorenylmethoxycarb-
onyl (Fmoc) amino acids were carried out in a microstruc-
tured tube-in-tube reactor equipped with a semipermeable
Teflon® AF 2400 membrane as the inner tubing. In the first
step, gaseous formaldehyde was passed through the inner
membrane to effect the acid-catalyzed conversion of the
Fmoc amino acids into the corresponding N-Fmoc oxazolid-
inones. In the second step, liquid–liquid transfer of trifluoro-
acetic acid was used for the first time in such a reactor for
the reductive opening of these oxazolidinones to give Fmoc
N-methylamino acids in high yields.
Introduction
N-Methylated amino acids (NMAs) are known constitu-
ents of important biologically active peptides, as for exam-
ple the immunosuppressant cyclosporine (1)[1] and dolastat-
ine D10 (2),[2] which is a promising anticancer drug (Fig-
ure 1). N-Methylamino acid residues lack the ability to be
hydrogen donors and significantly change the physical
properties of peptides. They influence the conformation of
NMA-containing peptides[3] and cause an increased pro-
teolytic resistance that is useful for medical applications.[4]
N-Methylamino acids have been synthesized by nucleo-
philic substitution of α-bromo acids[5] and α-hydroxy acid
derivatives,[6] by methylation of sulfonamides,[7] carbamates
or amides,[8] as well as through reductive amination.[9,10] Se-
lective access to NMAs by avoiding hazardous chemicals,
racemization, and harsh reaction conditions was introduced
by Freidinger et al.[11] in a two-step process via oxazolidin-
ones 4 (Scheme 1). However, these syntheses suffer from in-
convenient up-scaling and require careful purification pro-
cedures. In particular, long reaction times make them less
attractive for industrial applications. Flow chemistry[12]
using special mixers that are more efficient than stirrers
with regard to concentration and temperature gradient[13]
can eliminate most of these drawbacks. In addition, up-scal-
ing can easily be facilitated in a microreactor by continuous
processing, and the purification can be performed during
the process, for example, with integrated columns.
Figure 1. N-Methylamino acids in natural peptide drugs.
Scheme 1. N-Methylation of amino acids in two steps according
to Freidinger et al. Fmoc = 9-fluorenylmethoxycarbonyl, TFA =
trifluoroacetic acid.
[a] Institut für Organische Chemie, Johannes Gutenberg-
Universität Mainz,
Results and Discussion
Duesbergweg 10–14, 55128 Mainz, Deutschland
Fax: +49-6131-39-24786
To demonstrate these advantages, we developed a con-
tinuous flow version of the second step of Freidinger’s[11a]
two-step synthesis by using a tube-in-tube reactor. A semi-
permeable membrane, such as Teflon® AF 2400, was ap-
E-mail: schtefkoch@hotmail.com
hokunz@uni-mainz.de
Homepage: http://www.ak-kunz.chemie.uni-mainz.de/
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201300705.
Eur. J. Org. Chem. 2013, 4509–4513
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4509

A. E. Buba, S. Koch, H. Kunz, H. Löwe
SHORT COMMUNICATION
Figure 2. Flow reactor used for the second step of the NMA synthesis.
plied for the first time as a mixing system with optimal li- temperatures, acetonitrile and chlorobenzene proved to be
quid–liquid contact over a large exchange surface, which the most effective solvents, and conversions up to 100%
would prevent early undesired side reactions, for example, within 2 h were achieved (Table 1, entries 4–6). A further
Mannich-type reactions, and the need for a Lewis acid sup- increase in the temperature allowed a flow rate up to
port.[11f] Teflon® AF is a stable amorphous copolymer con- 8 mLh–1, a reaction time of 1 h (Table 1, entries 7–9), and a
sisting of 2,2-bis(trifluoromethyl)-4,5-difluoro-1,3-dioxol- reduction in the amount of triethylsilane (Table 1, entry 10).
ane and tetrafluoroethylene.[14] It is highly permeable to
gases[15] and, to a limited extent, to solutes.[16] AF mem-
Table 1. Optimization of the reaction conditions for NMA synthe-
sis.
branes were successfully applied in pioneering work by
Ley[17] in gas–liquid transfers in syntheses.
To demonstrate the liquid–liquid transfer through a Tef-
lon® AF 2400 membrane in synthesis, oxazolidinones 4 as
substrates were prepared by heating Fmoc amino acids 3
with paraformaldehyde under acid catalysis (Scheme 1).[11a]
The reductive ring opening of the oxazolidinones was per-
formed in a flow reactor with a tube-in-tube configura-
tion[17] (Figure 2), which consisted of an inner Tef-
lon® AF 2400 membrane (60 cm) and an outer polytetra-
fluoroethylene (PTFE) tubing. Connection through a T-
piece ensured independent liquid flow into the reactor. Tri-
fluoroacetic acid (TFA) was chosen as the liquid acid cata-
lyst that was injected into the outer tubing of the tube-in-
tube system. It penetrated the inner membrane at room
temperature without affecting the material, as demonstrated
by repeated use of the reactor.
Entry Solvent
T
[°C]
Flow rate Et3SiH Conversion[a]
[mLh–1] [equiv.]
[%]
1
2
3
4
5
6
7
8
9
10
chloroform
chloroform
dioxane
50
50
70
70
70
70
70
75
75
75
6
3
3
3
3
3
6
6
8
8
3
3
3
3
3
3
3
3
3
2
33
40
2
toluene
71
chlorobenzene
acetonitrile
acetonitrile
acetonitrile
acetonitrile
acetonitrile
100
100
Ͼ95
100
100
100
The dissolved substrates, Fmoc amino acid oxazolid-
inone 4, and triethylsilane were placed into the sample loop
and the solvent was pumped by using a syringe pump
through the reactor. During solvent flow, the level of TFA
in the outer tubing significantly declined. Transfer of the
substrates into the outer tube was never observed. No
amino acid derivatives were detectable in the trapping hy-
[a] Fmoc-l-alanine oxazolidinone (97.0 mg, 0.30 mmol), solvent
(1.5 mL), triethylsilane, trifluoroacetic acid. All reactions were per-
formed in a solvent-free apparatus. Conversion was determined by
1H NMR spectroscopy.
After optimization of the conditions, the general applica-
drogen carbonate solution. The high yields of the products bility of this flow-through synthesis to Fmoc α-amino acids
also give evidence that substrates and products do not pass 3a–g was studied. The results for amino acids with aliphatic
the inner tube. In a subsequent coil, the temperature was (i.e., 5a,b), aromatic (i.e., 5c,d), and heterofunctional side
elevated to accelerate the reaction. For optimization of the chains (i.e., 5e–g) are given in Table 2. Although each
conditions, Fmoc-protected alanine oxazolidinone 4a was amino acid required an optimized flow rate, as determined
chosen as a model substrate (Table 1). By using chloroform by monitoring by 1H NMR spectroscopy, all N-meth-
as the solvent according to Freidinger’s protocol at 50 °C ylamino acids except for methionine (i.e., product 5e) were
with a flow rate of 6 mLh–1, a conversion of 33% was ob- obtained in high yields. With methionine, side products
served within 1 h (Table 1, entry 1). A lower flow rate re- were formed, which resulted in difficult purification. Thus,
sulted in a higher conversion (Table 1, entry 2). At higher the yield was determined by NMR spectroscopy.
4510
www.eurjoc.org
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 4509–4513

Fmoc-N-methylamino Acids from a Flow Tube-in-Tube Reactor
Table 2. Fmoc N-methylamino acids from Fmoc amino acids.
was used, which was generated by heating paraformalde-
hyde in a separate pressure-resistant vessel. The formalde-
hyde permeability of the Teflon® AF 2400 membrane was
tested by using a Schiff reagent.[18] Gaseous formaldehyde
is stable at 80–100 °C. Below 75 °C, it polymerizes and pa-
raformaldehyde precipitates on the membrane. In processes
with gaseous reactants, outgassing is often observed.[17d]
Performance of these conversions in a closed reactor under
pressure avoids this problem and enables a reaction in a
homogeneous liquid–gas solution. To maintain a constant
and high pressure (1–2 bar), an HPLC pump was used in-
stead of a syringe pump, as shown in Figure 3.
Fmoc-alanine (3a) and catalytic p-toluenesulfonic acid
were dissolved in acetonitrile and injected into the sample
loop. The reactor was flushed with gaseous formaldehyde
through the membrane. Subsequently, the solvent was
pumped into the reactor with a flow rate of 3 mLh–1. A
pressure of 0.5–1.0 bar was sufficient to avoid outgassing of
formaldehyde.
The model reaction yielded 91% of N-Fmoc-l-4-methyl-
oxazolidin-5-one (4a, 100% conversion; Table 3, entry 1)
within 1 h. No azeotropic distillation of water was neces-
sary. Trifluoroacetic acid could be used as a substitute for
p-toluenesulfonic acid as the catalyst, and its use also led
to complete conversion. However, some side products were
observed under these conditions (Table 3, entry 2).
Table 3. Optimization of the reaction conditions for oxazolidinone
syntheses.
Entry
Acid
Conversion[a] [%]
1
2
pTsOH
TFA
100
100
[a] Fmoc-l-alanine (96.5 mg, 0.31 mmol), acid in catalytic
amounts, gaseous formaldehyde from paraformaldehyde, acetoni-
trile (1.5 mL). Conversion was determined by 1H NMR spec-
troscopy.
Notably, performing both steps, that is, the formation of
the oxazolidinone and reductive ring opening, in a single
combined two tube-in-tube flow reactor at a flow rate of
3 mLh–1 and an elevated reaction temperature (80 °C) was
not successful, because the excess amount of formaldehyde
from the first step preferably reacted with triethylsilane,
which is required for the reductive opening of the oxazolid-
inone ring.
[a] First step, batch reaction: Fmoc amino acid, p-toluenesulfonic
acid, paraformaldehyde, toluene, reflux, 2 h; isolated yield. [b] Sec-
ond step: Fmoc-oxazolidinone (0.30 mmol), triethylsilane
(0.10 mL, 0.63 mmol, 2.0 equiv.), trifluoroacetic acid (1.5 mL,
19.5 mmol, 65 equiv.), acetonitrile (1.5 mL), flow rate = 6 mLh–1,
reaction time = 1 h; isolated yield. [c] Flow rate = 3 mLh–1; reac-
tion time = 2 h. [d] 0.15 mL (0.94 mmol, 3.0 equiv.) of triethylsil-
1
Conclusions
ane. [e] Yield was determined by H NMR spectroscopy.
Teflon® AF 2400 membranes in tube-in-tube flow reac-
Further investigations showed that the first step of Frei- tors were demonstrated to be permeable to gaseous reac-
dinger’s synthesis, the formation of oxazolidinone 4a, can tants and also to liquid organic acids. The liquid–liquid
also be performed in a tube-in-tube reactor by exploiting transfer of trifluoroacetic acid in a flow reactor enabled the
the high gas permeability of Teflon® AF 2400 membranes. highly efficient synthesis of Fmoc N-methylamino acids,
Instead of solid paraformaldehyde, gaseous formaldehyde which are interesting components used in the syntheses of
Eur. J. Org. Chem. 2013, 4509–4513
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4511

A. E. Buba, S. Koch, H. Kunz, H. Löwe
SHORT COMMUNICATION
Figure 3. Flow tube-in-tube reactor used for the synthesis of Fmoc-oxazolidinones.
natural products and drugs. Liquid transfer through semi-
Acknowledgments
permeable membranes, in general, should open up attract-
ive perspectives for processes performed in flow reactors,
for example, if trapping of a desired but unstable product
with a liquid reactant is demanded.
This work was supported by the Stiftung Rheinland-Pfalz für Inno-
vation and by the Fonds der Chemischen Industrie. S. K. is grateful
for a Ph.D. grant from the Studienstiftung des Deutschen Volkes.
[1] T. A. Starzl, G. B. G. Klintmalm, K. A. Porter, S. Iwatsuki,
G. P. J. Schröter, N. Engl. J. Med. 1981, 305, 266–269.
[2] G. R. Pettit, D. L. Herald, S. B. Singh, T. J. Thornton, J. T.
Mullaney, J. Am. Chem. Soc. 1991, 113, 6692–6693.
[3] A. E. Tonelli, J. Am. Chem. Soc. 1971, 93, 7153–7155.
[4] D. P. Fairlie, G. Abbenante, D. R. March, Curr. Med. Chem.
1995, 2, 654–686.
[5] E. Fischer, L. V. Mechel, Ber. Dtsch. Chem. Ges. 1915, 48, 360.
[6] F. Effenberger, U. Burkard, J. Willfahrt, Liebigs Ann. Chem.
1986, 314–333.
Experimental Section
General Procedure for the Flow-Through Synthesis of 9H-Fluoren-
9-ylmethyl-(S)-4-methyl-5-oxo-1,3-oxazolidine-3-carboxylate (4a) in
a Tube-in-Tube Reactor: A solution of Fmoc-l-alanine (3a; 100 mg,
0.32 mmol) and p-toluenesulfonic acid (23.7 mg, 138 μmol) in
acetonitrile (1.5 mL) was injected into the sample loop. The tube-
in-tube reactor was subsequently flooded with gaseous formal-
dehyde generated by heating paraformaldehyde in a separate vessel.
Acetonitrile was passed through the reactor at a flow rate of
3 mLh–1 and 0.5 bar. The temperature of the coils was set to 75–
80 °C. The product was collected at the output stream. The solu-
tion was concentrated in vacuo. Codistillation with dichlorometh-
ane gave the product as a colorless solid, yield 95.2 mg (294 μmol,
91%).
[7] a) E. Fischer, W. Lipschitz, Ber. Dtsch. Chem. Ges. 1915, 48,
ˇ
360; b) J. Hlavácˇek, K. Podusˇka, F. Sorm, K. Sláma, Collect.
Czech. Commun. 1976, 41, 2079–2087; c) D. Papaioannou, C.
Athanassopoulos, V. Magafa, N. Karamanos, G. Stavropoulos,
A. Napoli, G. Sindona, D. W. Aksnes, G. W. Francis, Acta
Chem. Scand. 1994, 48, 324–333.
[8] a) B. C. Das, S. D. Gero, E. Lederer, Biochem. Biophys. Res.
Commun. 1967, 29, 211–215; b) N. L. Benoiton, K. Kuroda,
S. T. Cheung, F. M. F. Chen, Can. J. Biochem. 1979, 57, 776–
781; c) J. R. McDermott, N. L. Benoiton, Can. J. Chem. 1973,
51, 2555–2561; d) J. R. McDermott, N. L. Benoiton, Can. J.
Chem. 1973, 51, 2562–2570; e) J. R. McDermott, N. L. Beno-
iton, Can. J. Chem. 1973, 51, 1915–1919; f) R. K. Olsen, J. Org.
Chem. 1970, 35, 1912–1915; g) S. Coulton, G. A. Moore, R.
Ramage, Tetrahedron Lett. 1976, 17, 4005–4008; h) S. L. Bela-
gali, T. Mathew, M. Himaja, P. Kocienski, Indian J. Chem. Sect.
B 1995, 34, 45–47.
[9] a) P. Quitt, Proceedings of the Fifth European Peptide Sympo-
sium, Oxford, Pergamon, Oxford, 1963, p. 165–169; b) P. Quitt,
J. Hellerbach, K. Vogler, Helv. Chim. Acta 1963, 46, 327–333;
c) R. E. Bowman, H. H. Stroud, J. Chem. Soc. 1950, 1342–
1345; d) R. E. Bowman, J. Chem. Soc. 1950, 1346–1348; e)
R. E. Bowman, J. Chem. Soc. 1950, 1349.
General Procedure for Flow-Through Syntheses of N-Methyl Amino
Acids 5a–g: The reactor set up is shown in Figure 2. The Fmoc-
protected α-amino acid derived oxazolidinone was dissolved in
acetonitrile (1.50 mL) and triethylsilane was added. The tempera-
ture of the coils was set to 75 °C, and the outer tube of the tube-
in-tube reactor was filled with trifluoroacetic acid (1.50 mL,
19.5 mmol). Acetonitrile was passed through the reactor by a sy-
ringe pump with a set flow rate. The product was collected at the
output stream. To record the NMR spectra, the obtained solution
was concentrated in vacuo and codistilled with dichloromethane
twice. For isolation, the product solution was diluted with a satu-
rated aqueous solution of NaHCO3 (15 mL). The aqueous layer
was washed with dichloromethane and then acidified with 1 m HCl
(pH ≈ 2). The acidic layer was then extracted with dichloromethane
(2ϫ 15 mL), and the combined organic layer was dried with
MgSO4. The solvent was removed under reduced pressure. The ob-
tained solids were recrystallized from diethyl ether. Oils were puri-
fied by flash chromatography on silica (cyclohexane/ethyl acetate,
3:1).
[10]
L. Aurellio, R. T. C. Brownlee, A. B. Hughes, Chem. Rev. 2004,
104, 5823–5846.
[11]
a) R. M. Freidinger, J. S. Hinkle, D. S. Perlow, B. H. Arison, J.
Org. Chem. 1983, 48, 77–81; b) D. Ben-Ishai, J. Am. Chem.
Soc. 1957, 79, 5736–5738; c) J. Auerbach, M. Zamore, S. M.
Weinreb, J. Org. Chem. 1976, 41, 725–726; d) L. Aurelio, J. S.
Box, R. T. C. Brownlee, A. B. Hughes, M. M. Sleebs, J. Org.
Chem. 2003, 68, 2652–2667; e) B. E. Sleebs, A. B. Hughes, J.
Org. Chem. 2007, 72, 3340–3352; f) S. Zhang, T. Govender, T.
Norström, P. I. Arvidsson, J. Org. Chem. 2005, 70, 6918–6920;
Supporting Information (see footnote on the first page of this arti-
cle): Experimental description of the batch and flow-through syn-
1
theses, characterization data, and H NMR and 13C NMR spectra
for all key intermediates and products.
4512
www.eurjoc.org
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 4509–4513

Fmoc-N-methylamino Acids from a Flow Tube-in-Tube Reactor
g) T. Govender, P. I. Arvidsson, Tetrahedron Lett. 2006, 47,
1691–1694.
[12] For a recent review, see: J. Wegner, S. Ceylan, A. Kirschning,
Adv. Synth. Catal. 2012, 354, 17.
[13] a) K. Jähnisch, V. Hessel, H. Löwe, M. Baerns, Angew. Chem.
2004, 116, 410–451; Angew. Chem. Int. Ed. 2004, 43, 406–446;
b) . J. Kobayashi, Y. Mori, S. Kobayashi, Chem. Asian J. 2006,
1, 22–35; c) T. Fukuyama, Md. T. Rahman, M. Sato, I. Ryu,
Synlett 2008, 2, 151–163.
[14] P. R. Resnick, W. H. Buck, in: Modern Fluoropolymers: High
Performance Polymers for Diverse Applications (Ed.: J. Scheirs),
Wiley, Chichester, UK, 1997, p. 397.
[15] a) I. Pinnau, L. G. Toy, J. Membr. Sci. 1996, 109, 125–133;
b) T. C. Merkel, V. I. Bondar, K. Nagai, B. D. Freeman, Y. P.
Yampolskii, Macromolecules 1999, 32, 8427–8440; c) A. Y.
Alentiev, V. P. Shantarovich, T. C. Merkel, V. I. Bondar, B. D.
Freeman, Y. P. Yampolskii, Macromolecules 2002, 35, 9513–
9522; d) Y. P. Yampolskii, Russ. J. Gen. Chem. 2009, 79, 657–
665; e) T. C. Merkel, V. I. Bondar, B. D. Freeman, Macromole-
cules 1999, 32, 370–374.
[16] a) H. Zhao, K. Ismail, S. G. Weber, J. Am. Chem. Soc. 2004,
126, 13184–13185; b) H. Zhao, J. Zhang, N. Wu, X. Zhang, K.
Crowley, S. G. Weber, J. Am. Chem. Soc. 2005, 127, 15112–
15119; c) H. Zhang, A. Hussam, S. G. Weber, J. Am. Chem.
Soc. 2010, 132, 17867–17879.
[17] a) M. O’Brien, I. R. Baxendale, S. V. Ley, Org. Lett. 2010, 12,
1596–1598; b) A. Polyzos, M. O’Brien, T. P. Petersen, I. R.
Baxendale, S. V. Ley, Angew. Chem. 2011, 123, 1222–1225; An-
gew. Chem. Int. Ed. 2011, 50, 1190–1193; c) P. Koos, U. Gross,
A. Polyzos, M. O’Brien, I. Baxendale, S. V. Ley, Org. Biomol.
Chem. 2011, 9, 6903–6908; d) M. O’Brien, N. Taylor, A. Po-
lyzos, I. R. Baxendale, S. V. Ley, Chem. Sci. 2011, 2, 1250–
1257; e) S. Newton, S. V. Ley, E. Casas Arcé, D. M. Grainger,
Adv. Synth. Catal. 2012, 354, 1805–1812; f) P. B. Cranwell, M.
O’Brien, D. L. Browne, P. Koos, A. Polyzos, M. Peña-Lopéz,
S. V. Ley, Org. Biomol. Chem. 2012, 10, 5774–5779.
[18] H. Wieland, G. Scheuing, Ber. Dtsch. Chem. Ges. 1921, 54,
2527–2555.
Received: May 14, 2013
Published Online: June 17, 2013
Eur. J. Org. Chem. 2013, 4509–4513
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4513