Process development on the preparation of trans-(+)-2-methylaminocyclohexanol: A fascinating resolution example

Full text of DOI:10.1021/op0000428

Organic Process Research & Development 2001, 5, 184−185
Process Development on the Preparation of
trans-(+)-2-Methylaminocyclohexanol: A Fascinating Resolution Example
Xinbo Lu,* Zunle Xu, and Guangming Tang
Beijing HSD Medicinal Research, Zhong Guan Cun Chuang Ye Building 1008, Shang-Di, Beijing, P.R. China, and
Chemistry Department, Zhong Shan UniVersity, Guang Zhou, 135 Xingang Road, P.R. China
Scheme 1
A new simple process is described to produce the highly
optically pure single isomer trans-(+)-2-methylaminocyclo-
hexanol [(+)-MAC] by mixing (+)-di-p-toluoyltartaric acid
(DTTA) with (()-MAC in an appropriate ratio, agitating at
60 °C for 30 min. After filtration and recovery by basifica-
tion, a 99% ee (+)-MAC is obtained.
Table 1. Solubility of salt (+)-MAC2-(+)-DTTA versus
(+)-MAC-(-)-DTTA
Introduction:
Single isomers of trans-(()-2-methylaminocyclohexanol
[(+)-MAC and (-)-MAC] are important intermediates for
the preparation of nitrogen-containing cycloalkylaminoaryl
derivatives for CNS disorder,1b analgesia agents,1c and anti-
arteriostenotic agents.1a They are also important chiral
intermediates useful as ligands, chiral auxiliaries, or in other
asymmetric synthetic applications.2 The (()-MAC was first
prepared by reacting an aqueous methylamine with cyclo-
hexene oxide in the Mousseron3 method. Another method
was reported in 1983 by Japanese chemists.4,5 The major
problems faced with this reaction were the high-pressure
involved and that the excessive primary amine released into
air would cause particular difficulty for scale up.
mp
(°C)
solubility in H2O
crystalline
(mg/mL)
(+)-MAC2(+)-DTTA.
(+)-MAC(-)-DTTA
190-192
180-182
9
>2000
less than 3%. The trace amount of 1 and methylamine residue
could be easily removed during the distillation since they
both have low boiling points. The (() MAC was prepared
in over 90% yield and purity above 95%.
In our recent attempts on the (()-MAC resolution, we
obtained one crystalline solid with a structure of (+)-MAC2-
(+)-DTTA (by NMR) containing 100% pure (+)-MAC. The
crystalline solid had very limited solubility in most solvents.
Subsequently, we used 2 equiv of pure (+)-MAC mixed with
1 equiv of (+)-DTTA and 1 equiv of (-)-DTTA, respec-
tively. The solubility test data on these solids is listed in
Table 1.
The resolution of (()-MAC with (+)-tartaric acid required
5-7 recrystallizations6 with a yield of approximately 8%.
The current (+)-MAC development is a practical method to
make (()-MAC suitable for large-scale preparation
Results and Discussion
From Table 1, the solubility of (+)-MAC-(-)-DTTA was
200 times more than that of (+)-MAC2-(+)-DTTA, which
provides a guideline for the efficient resolution.
(()-MAC Preparation. The test results in our lab
indicated that 40% aqueous methylamine solution reacted
with cyclohexene oxide quite readily under reflux condition
as described in Scheme 1. No high pressure or special
equipment was required as in previous examples.4,5
In our tests, 2.5 equiv of aqueous methylamine solution
was added to 1 equiv of 1 in 95% ethanol and refluxed for
3 h, the conversion of 1 being over 97%. The dimer impurity
caused by further reaction of MAC with the epoxide was
When we reacted >4 equiv of (()-MAC with 1 equiv of
(+)-DTTA, most of the (+)-MAC2-(+)-DTTA precipitated
out owing to the very low solubility and left most of the
other isomer with much higher solubility in the mother liquor.
Once the reaction conditions favored the formation of the
(+)-MAC2-(+)-DTTA (a “eutectic” type solid) a good
resolution could be obtained. The solvents were screened,
and the best was ethanol, and actually the resolution
efficiency was independent of the solvents used, when the
other isomer salt of (-)-MAC-(+)-DTTA was with relatively
high solubility and easily removed by adjusting the solvent
volume. The resolution efficiency was even more easily
achieved by running the resolution at 60 °C for 20 min,
proving that the resolution efficiency was primarily depend-
ent on the eutectic composition.
(1) (a) Koga, Y.; et al. Bioorg. Med. Chem. Lett. 1998, 8, 1471-1476. (b) De
Costa, et al. PCT Int. Appl. WO 9112247 A1 22. August, 1991. (c) Bain,
A. I. PCT Int. Appl. WO 9916431 A1 8. April, 1999. (d) White, A. C.;
Bradley, G. Eur. Pat. Appl. EP 61900 A16. October, 1982.
(2) (a) Pracejus, H. J. Prakt. Chim. 1987, 329, 235-245. (b) Beaton, M.; Gani,
D. Tetrahedron Lett. 1998, 39, 8549-8552. (c) Rosling, A.; Fulop, F.;
Askolin, C.-P.; Mattinen, J. J. Chem. Res., Synop. 1998, 9, 492, 2237-
2272. (d) Pecunioso, A.; Maffeis, M.; Marchioro, C.; Rossi, L.; Tamburini,
B. Tetrahedron: Asymmetry 1997, 8, 775-778. (e) Takagi, Y.; Kawashima,
O.; Tsutomu, ; Sano, H.; Umezawa, S. Bull. Chem. Soc. Jpn. 1976, 49
(11), 3108-3112. (f) Faber, K.; Hoenig, H.; Seufer-Wasserthal, P.
Tetrahedron Lett. 1988, 29 (16), 1903-1904.
Table 2 shows the resolution examples with various molar
(3) Mousseron, M.; Granger, R. Bull. Soc. Chim. Fr. 1947, 14, 850.
(4) Moikawa, T. Chim. Pharm. Bull. 1983, 31 (5), 1646-1651.
(5) Spessard, G. O. Tetrahedron Lett. 1983, 24 (7), 655-656, 1983.
(6) Kay, J. B. J. Chem. Soc. C 1969, 248-252.
ratios of (+)-DTTA over (()-MAC.
The data listed in table were results from mixing various
molar ratios of (()-MAC to (+)-DTTA. The (()-MAC and
184
•
Vol. 5, No. 2, 2001 / Organic Process Research & Development
10.1021/op0000428 CCC: $20.00 © 2001 American Chemical Society and The Royal Society of Chemistry
Published on Web 03/01/2001

Table 2. Resolution examples with various molar ratios of
(+)-DTTA over (()-MAC
analytical column: Chiralpak AD, 25 cm × 4.6 mm (Daicel),
mobile phase: hexane/2-propanol (85:15); column T °C )
rt; flow rate: 1.0 mL/min; wavelength: 254 nm; run time:
20 min, sample preparation: precolumn derivation with
m-toluoyl chloride.
(+)-DTTA (()-MAC
ee
T
(+)-MAC
entry
(equiv)
(equiv)
mp (%) (°C) yield (%)
1
2
3
4
5
6
1
1
1
1
1
1
4
5
6
6
4
6
184 92.9 25
186 97.5 25
187 98.5 25
190 99.3 25
190 99.5 60
190 99.9 60
40.7
33.4
27.3
30.6
45.8
30.2
trans-(()-2-Methylaminocyclohexanol (2). A mixture of
1 (500.0 g, 5094 mmol), methylamine (40% aqueous, 988.9
g, 12736 mmol), and ethanol (2500 mL) was mixed and
heated to reflux for about 90 min. When the 1 residue was
<1% by GLC, vacuum distillation was started at 60 °C. After
removal of all the solvent and reactants, 650 g of a light
yellow oil was obtained (yield > 95%, purity 95% by GC),
bp 107-109/17 mm, FT-IR 3150 (bonded NH), 3320
Scheme 2
1
(bonded OH) cm-1; HNMR (300 MHz) δ 0.93-1.02 (dd,
1H, J ) 3.1,16.2), 1.18-1.32 (m, 3H), 1.68-1.72 (d, 2H,
) 10) 1.95-2.22 (m, 3H), 2.39 (s, 3H), 3.18-3.26 (m, 1H),
3.4 (s, 1H). 13C NMR (60 MHz) δ 24.5, 24.9, 29.5, 33.2,
33.8, 65.0, 73.5.
(+)-trans-2-Methylaminocyclohexanol (+)-Di-p-toluoyl-
tartrate [(+)-MAC]2-[(+)-DTTA] (4). In a three necked
mL round-bottom flask equipped with a mechanic stirrer and
a thermometer were charged 200 g of (+)-DTTA (517.7
mmol) and 600 mL of EtOH; the mixture then was heated
to 45 °C, and then (()-MAC (267.1 g, 2070.5 mmol) in
400 mL of ethanol was added dropwise in 15 min. The
mixture was further heated to 60 °C and agitated for 30 min
and then cooled gradually in about 30 min to 23 °C and
agitated for another 20 min. The mixture was filtered, and a
white solid was obtained and rinsed with 100 mL of ethanol
twice. The white solid after being dried over vacuum at 45
°C for about 5 h was weighed: about 316.8 g (95%); purity
>99%. [R]25 +69.5° (c 5, in H2O); mp ) 190-193 °C;
D
(+)-DTTA were mixed in the correct molar ratio of 4:1,
suspended in ethanol, agitated at 60 °C for 20 min, and then
gradually cooled to room temperature and filtered. The crude
product with optical purity >99%, chemical purity >99%
was obtained in ∼45% yield.
FT-IR 3150 (bonded NH), 3320, 3000 (bonded OH), 1700
1
(CdO) cm-1; HNMR (250 MHz) δ 1.17-1.27 (m, 8H),
1.65-1.71 (m, 4H), 1.93-1.97 (m, 2H), 2.03-2.07 (m, 2H),
2.33 (s, 6H), 2.60 (s, 6H), 2.79-2.83 (m, 2H), 3.42-3.50
(m, m, 2H), 5.62 (s, 2H), 7.28-7.31 (d, 4H, J ) 7.95),7.89-
7.92 (d, 4H, J ) 7.63); 13C NMR (60 MHz) δ 23.6, 26.1,
28.4, 32.3, 36.2, 66.1, 72.9, 77.8, 128.8, 132.1, 132.7, 148.2,
170.1, 175.8.
The free base of (+)-MAC was recovered by agitating
the (+)-MAC2-(+)-DTTA salt in a mixture of aqueous 20%
NaOH and TBME and then extracted back with TBME. After
removal of TBME, the single isomer of (+)-MAC was
obtained.
trans-(()-2-Methylaminocyclohexanol (5). 4 (300 g,
465.5 mmol) was mixed with TBME (1200 mL, 50% (wt/
wt)), NaOH (82.0 mL, 1025 mmol), and water (1000 mL).
The mixture was stirred at rt for 3 h to get a clear
heterogeneous solution when standing still. After phase
separation, the aqueous phase was extracted with TBME
(1000 mL) three times. The combined organic phase was
washed with brine and then concentrated into a white solid:
117.6 g (yield: 98%); purity > 99%. [R]25D +89.5° (c 5, in
H2O); FT-IR 3150 (bonded NH), 3320 (bonded OH) cm-1;
1HNMR (300 MHz) δ 0.93-0.99 (dd, 1H, J ) 2.7, 11.9),
1.18-1.32 (m, 3H), 1.67-1.72 (d, 2H, J ) 12.5), 1.92-
2.21 (m, 3H), 2.39 (s, 3H), 3.18-3.26 (m, 1H), 3.67 (s, 1H);
13C NMR (60 MHz) δ 24.5, 24.9, 25.7, 29.5, 33.2, 33.8,
65.0, 73.5
Conclusions
From the economic point of view, the (()-MAC is much
cheaper than (+)-DTTA and (-)-DTTA, without considering
the recovery of (+)- or (-)-DTTA, and thus using 4 equiv
of (()-MAC with 1 equiv of (+)-DTTA at 60 °C to get an
initial precipitate with ee > 99% gives the most efficient
resolution with the best usage of the relatively expensive
chiral reagent (Scheme 2).
Experimental Section
General Methods. Melting points are uncorrected.
IR spectra were recorded on an FT-IR spectrometer as
1
thin films or KBr disks. H and 13C spectra were recorded
at 250 and 63 MHz, using CDCl3, CD3OD, or D2O as internal
reference. Enantiopurities were determined by running a
chiral HPLC test method. Chromatographic conditions:
Received for review April 20, 2000.
OP0000428
Vol. 5, No. 2, 2001 / Organic Process Research & Development
•
185