An efficient asymmetric synthesis of key intermediates in the synthesis of aphanorphine and eptazocine

Article abstract of DOI:10.1016/S0957-4166(03)00117-4

Efficient, formal syntheses of aphanorphine and eptazocine are reported that involve epoxide cyclizations. The necessary chiral epoxides were prepared following treatment of diols prepared by Sharpless asymmetric osmylations of an alkene. The cyclizations formed a ring compound with the same enantiomeric purity as the starting epoxide.

Full text of DOI:10.1016/S0957-4166(03)00117-4

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 14 (2003) 743–747
An efficient asymmetric synthesis of key intermediates in the
synthesis of aphanorphine and eptazocine
Stephen K. Taylor,* Milica Ivanovic, Lloyd J. Simons and Matthew M. Davis
Department of Chemistry, Hope College, Holland, MI 49422-9000, USA
Received 5 December 2002; accepted 6 January 2003
Abstract—Efficient, formal syntheses of aphanorphine and eptazocine are reported that involve epoxide cyclizations. The
necessary chiral epoxides were prepared following treatment of diols prepared by Sharpless asymmetric osmylations of an alkene.
The cyclizations formed a ring compound with the same enantiomeric purity as the starting epoxide. © 2003 Elsevier Science Ltd.
All rights reserved.
1. Introduction
yield to the corresponding aldehyde 4a. Shiotani11,12
and co-workers synthesized (−)-aphanorphine from 3a,
and have accomplished a formal syntheses of (−)-epta-
zocine from the enantiomeric alcohol 3b. Oxidation of
the alcohols to the corresponding aldehydes (4a and 4b
in 90 and 88%, respectively) produced the compounds
used to introduce the nitrogen functionality (e.g. via a
Schiff’s base) in the total synthesis.
The narcotic aphanorphine1–7 1 and the analgesic
eptazocine7–12 2 have received significant recent atten-
tion. The interest in these morphine analogs arises from
their significant bioactivity,1–12 and also from the syn-
thetic challenge of constructing their quaternary ben-
zylic carbon centers.13–15 Both the natural3–7,11,12 and
unnatural2,8,11,12 enantiomers have been prepared,
although with 2 the unnatural isomer has been reported
to be the more active and less toxic9,10 isomer.
2. Results
Shiotani’s syntheses of alcohols 3a and b required
eleven steps for each of the enantiomers.12 We wish to
report the synthesis of both enantiomers by cyclizations
that require only four steps each, and give very good
ees (see the reaction envisaged for 5a below). The
epoxides used in the syntheses were prepared by Sharp-
less asymmetric osmylation of alkenes,16 and transfor-
mation of the resulting diols into epoxides by Sharpless’
method.17,18 Isomer 3a was prepared by cyclization of
5a, and 3b was obtained analogously from 5b. The
Fadel and Arzel7 accomplished the formal syntheses of
(−)-aphanorphine and (+)-eptazocine via a synthesis of
alcohol 3a (see Scheme 1), which was oxidized in 92%
Scheme 1. Retrosynthetic analysis of enantiomeric products.
* Corresponding author. Fax: 616-395-7118; e-mail: staylor@hope.edu
0957-4166/03/$ - see front matter © 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0957-4166(03)00117-4

744
S. K. Taylor et al. / Tetrahedron: Asymmetry 14 (2003) 743–747
Scheme 2. Synthesis and cyclization of epoxides.
asymmetric syntheses show the generality and useful-
ness of the Sharpless methods (Scheme 2).16,17
not result in this duality of isomers, although some
aldehyde was formed during the cyclization reaction
(21% with MeAlCl2, see Section 3).
We have carried out many epoxy-arene cyclizations
before,19–23 but never ones that involve optically active
epoxides. The ees obtained suggest that this is a very
good approach to asymmetric synthesis.
Corey and Sodeoka25 reported that Me2AlCl was a
good catalyst to minimize rearrangements to carbonyl
compounds in cyclizations of some epoxides at tertiary
positions. In our earlier studies with 10a, SnCl4 did
cause rearrangement to significant amounts of aldehyde
(88%).20 MeAlCl2 did result in less aldehyde (a 3:9 ratio
of 75:25) in our work with the enantiomers of 3, but
BF3·OEt2 (25–29%) did too. Unfortunately our ees
using BF3·OEt2 for the cyclization reaction were lower
(e.g. 78% ee) than those from MeAlCl2-promoted
reactions.
It is interesting that recrystallization of the diol (8a or
b) was required for high enantiomeric purity. When the
diol was used directly as it crystallized from the osmyla-
tion reaction (in 92% yield), the ee of cyclization
product 3a was 75%. However, when the diol was
carefully recrystallized from 1:1 benzene/hexane, the
yield was 66–76%, but the enantiomeric purity of the
final alcohol product was much higher (94–95% ee—as
determined by chiral GC on a b-cyclodextrin column).
This also demonstrates that the enantiomeric purity of
the epoxide is the key factor in determining the ee of
the product. We were unable to determine the ee of the
diol or the epoxide by any means attempted.
3. Experimental
The equipment used has been described earlier.21 All
the NMR spectra but those of 3a (for which a Varian
400 MHz spectrometer was used) were obtained on a
General Electric 300 MHz spectrometer (in CDCl3).
Perkin–Elmer 241 or a Jasco DIP 370 polarimeters
were used for optical rotation measurements. 13C NMR
spectra are available to prove product identities and
purities.
Earlier, we successfully completed a ring closure reac-
tion at a racemic quaternary epoxide position (10a,
R=H).20 Julia and Labia24 reported a cyclization of the
racemic meta-substituted analog of
3 (10b), and
observed that the reaction occured in 63% yield (with
21% aldehyde), but afforded equal amounts of ortho
and para cyclization products. The para isomer does
3.1. 2-Methyl-5-(p-methoxyphenyl)-1-pentene, 7
Using the methods of Julia and Labia24 and Taylor, et
al.,20 a Grignard reagent was combined with 3-chloro-
2-methylpropene. A sample of p-methoxyphenethyl
bromide (4.4 g, 20.6 mmol) in dry THF (27 mL) was
added with stirring to pure Mg (0.54 g, 22 mmol)
turnings covered with THF (3 mL). After heating under
reflux for 0.5 h, the mixture was cooled, and a solution

S. K. Taylor et al. / Tetrahedron: Asymmetry 14 (2003) 743–747
745
of 3-chloro-2-methylpropene (1.9 g, 21 mmol) was
added dropwise. The resulting mixture was stirred for
16 h. After dropwise addition of H2O (30 mL), ether
(30 mL) and sat. NaCl (10 mL), the organic layer was
collected and the aqueous layer was extracted with
ether. The combined organic layers were extracted with
sat. aqueous NaCl, dried (MgSO4) and evaporated.
Distillation of the product with a 1 inch Vigreux
column gave four fractions, with the last two affording
7 (2.8 g, 14.7 mmol, 71%), bp 80–81°C (0.25 mmHg).
Julia and Labia24 obtained 63% yield in the preparation
of the meta analog. The MS and NMR data matched
those of the reported compound:26 IR 3073 (m), 1613
the sample (not rigorously dried) showed [h]2D1 +3.6 (c
2.7, CHCl3). A 1.04 g sample of 7 gave 8b (0.956 g,
92%). The compound gave the same spectral data as 8a.
3.4. Racemic 1,2-epoxy-2-methyl-5-(p-methoxyphenyl)-
pentane, 5
Prepared by MCPBA epoxidation of 7 (as described for
other epoxide preparations)19–21 by adding 7 (1.38 g,
7.26 mmol) in CH2Cl2 (12 mL) to an ice-cooled solu-
tion of MCPBA (1.86 g, 70%, 7.5 mmol)) in CH2Cl2 (20
mL), and stirring for 50 min. Standard workup19 (dilu-
tion with petroleum ether, filtration of MCBA, and
washes with 20% NaHSO3, 5% sodium bicarbonate and
10% NaCl)21 left product that distilled at 72–78°C
(0.01–0.02 mmHg), or rotary TLC (3.5:1 hex-
anes:EtOAc) gave racemic 5 (0.821 g, 55%), 98% pure;
1
(m), 1512, 1246, 1039, 887 (m) and 827 (m) cm−1; H
NMR (CDCl3), l 1.71 (s, 3H), 1.7–1.8 (m, 2H), 2.04 (t,
J=7.8 Hz, 2H), 2.54 (t, J=7.8 Hz, 2H), 3.77 (s, 3H),
4.7, (2 vinyl H, pseudo doublet), 7.1 and 6.8 (AB
pattern, wAB=82 Hz, JAB=8.8 Hz, 4H); 13C NMR
(CDCl3) l 22.3, 29.5, 34.5, 37.3, 55.2, 109.9, 113.7 (2C),
129.2 (2C), 134.6, 145.7 and 157.7; MS 190 (5, M+), 134
(100), 121 (49). Anal calcd for C13H18O: C, 82.06; H,
9.53. Found: C, 81.55; H, 9.51%.
1
IR 1613 (m), 1512, 1244, 831 (m) and 812 (m) cm−1: H
NMR (CDCl3), l 1.3 (s, 3H), 1.51–1.7 (m, 4H), 2.56
(2m, 4H), 3.78 (s, 3H), 6.95 (AB pattern, wAB=79.4 Hz,
J
AB=8.7 Hz, 4H); 13C NMR (CDCl3) l 20.9, 27.1, 34.9,
36.2, 53.7, 55.2, 56.7, 113.8 (2C), 129.2 (2C), 134.2,
157.9. Anal calcd for C13H18O2: C, 75.69; H, 8.79.
Found: C, 75.52; H, 8.89%. The product was used to
verify the identity and purity of the R- and S-enan-
tiomers below.
3.2. (2S)-1,2-Dihydroxy-2-methyl-5-(p-methoxyphenyl)-
pentane, 8a
Prepared by asymmetric osmylation with AD-mix-b: a
mixture of t-butanol (18.5 mL), H2O (22.5 mL), and
AD-mix-b (6.77 g) was cooled in an ice water bath, and
7 (0.910 g, 4.8 mmol) was added dropwise to the stirred
mixture. After stirring for 5.5 h, TLC (3:1 hex-
ane:EtOAc) showed no alkene remained, and sodium
sulfite (7.5 g) was slowly added to the cooled mixture.
After stirring for 45 min, the product was extracted
twice with CH2Cl2 and once with ether. The combined
organic extracts were washed with sat. NaCl and dried
(MgSO4). The evaporated mixture was dissolved in
benzene (3 mL), and hexane (3 mL) was added. On
cooling overnight, crystals formed, which were isolated
by vacuum filtration. After drying under high vacuum
at room temperature, the product was isolated as a
solid (1.04 g, 93%); mp 55.2–57.4°C [h]2D7 −5.0 (c 1.05,
CHCl3); IR 3400 (br), 1613 (m), 1512, 1246,1038, 831
(m) and 810 (m) cm−1; 1H NMR (with 400 MHz NMR)
(CDCl3), l 1.14 (s, 3H), 1.4–1.5 (m, 2H), 1.5–1.7 (m,
2H), 2.0 (s, 2OH), 2.56 (t, J=7.5 Hz, 2H), 3.4 (AB
pattern, wAB=20 Hz, JAB=10.8 Hz, 2H), 3.78 (s, 3H),
6.95 (AB pattern, wAB=108 Hz, JAB=8 Hz, 4H); 13C
NMR (CDCl3) l 23.6, 26.2, 35.7, 38.5, 55.6, 70, 73.2,
114 (2C), 129.4 (2C), 134.5, 157.9.4. A purified sample
showed a mp of 59.1–60.4°C, but the sample was
hygroscopic, and so C,H analysis was not attempted.
Crude diol afforded a final product (3a or b) that was
typically 75% ee (rather than 90–95% ee). When the
diol was carefully recrystallized from 1:1 ben-
zene:hexane, the 3a or b ultimately produced was 94–
95% ee.
3.5. (2S)-1,2-Epoxy-2-methyl-5-(p-methoxyphenyl)-
pentane, 5a
Was prepared from 8a and purified by the procedures
detailed below, [h]2D1 −6.8 (c 0.58, CHCl3), mass spec-
trum, exact mass m/z calcd for C13H18O2 206.1307,
found 206.1312.
3.6. (2R)-1,2-Epoxy-2-methyl–5-(p-methoxyphenyl)-
pentane, 5b
The diol 8b (0.66 g, 2.95 mmol) was dissolved in dry
pyridine (3.5 mL), and p-toluenesulfonyl chloride (0.57
g, 3 mmol) was added in three portions over 15 min to
the ice/water-cooled solution. The reaction vial was
tightly sealed and kept in the refrigerator for 5.5 h. The
pyridine hydrochloride crystals were removed by filtra-
tion, and the filtrate was added to H2O (75 mL) and the
resulting mixture was extracted twice with ether. The
combined ethereal extract was washed twice with 10%
HCl, once with 5% NaHCO3 and sat. NaCl and dried
(MgSO4). The product was dissolved in MeOH (6 mL),
and K2CO3 (1.04 g, 7.5 mmol) and the mixture was
stirred for 5.5 h. Filtration left material that was dis-
solved in ether (30 mL), and the aqueous layer was
extracted with ether, and the combined ethereal extract
was washed with sat. NaCl and evaporated. The residue
was purified by rotary chromatography using 3:1 (or
3.5:1) hexane:EtOAc, giving 0.3769 g of 5b (62%), [h]D20
+6.3 (c 0.66, CHCl3), mass spectrum, exact mass m/z
calcd for C13H18O2 206.1307, found 206.1307. Chiral
GC did not separate the enantiomers.
3.3. (2R)-1,2-Dihydroxy-2-methyl-5-(p-methoxyphenyl)-
pentane, 8b
3.7. Cyclizations of epoxides
Prepared as above using AD-mix-a. The product was
recrystallized from benzene/hexane, mp 54–57.5°C, and
3.7.1. Racemic alcohols. Racemic 5 (0.253 g, 1.2 mmol)
was dissolved in dry CH2Cl2 (21 mL-distilled under N2

746
S. K. Taylor et al. / Tetrahedron: Asymmetry 14 (2003) 743–747
from calcium hydride). The mixture was cooled under
N2 with a dry ice/acetone bath, and MeAlCl2 (1 M
solution in hexane, 0.52 mL, 0.52 mmol) was added
dropwise via syringe. After 30 min at −78°C, 5% HCl
(70 mL) was added dropwise to the mixture over 15
min, and ether (50 mL) was added. The aqueous layer
was washed with ether, and the combined ethereal
extract was washed with 5% NaHCO3 and brine and
dried (MgSO4). Rotary silica chromatography with 3:1
hexane:EtOAc produced racemic alcohol 3 (0.158 g,
62%): IR 3400 (br), 1611 (m), 1503, 1497, 1240, 1038,
3H), 1.3–1.8 (m, 4H) 2.4 (q of d, J=7 and 2 Hz, 1H),
2.57 (t, J=7.3 Hz, 2H), 3.78 (s, 3H), 6.84 and 7.07 (AB
pattern, wAB=77.6 Hz, JAB=9 Hz, 4H), 9.6 (d, J=2
Hz, 1H); 13C NMR (CDCl3) l 13, 29, 30, 35, 46, 55,
114 (2C), 129 (2C), 134, 158, 205: mass spec. 206 (15),
134 (12), 121 (100), 128 (13). Anal calcd for C13H18O2:
C, 75.69; H, 8.79. Found: C, 75.22; H, 8.74%: mass
spectrum, exact mass m/z calcd for C13H18O2 206.1307,
found 206.1303.
1
806 (m), 736 (w) and 702 (w) cm−1: H NMR (CDCl3),
Acknowledgements
l 1.23 (s, 3H), 1.4 (br, OH), 1.45–1.6 (m, 1H), 1.65–1.9
(m, 2H), 1.9–2.1 (m, 1H), 2.7 (t, J=6.4 Hz, 2H), 3.64
(AB pattern, wAB=86 Hz, JAB=10.7 Hz, 2H), 3.78 (s,
3H), 6.71 (dd, J=8.3, 2.9 Hz, 1H), 6.83 (d, J=2.9 Hz,
1H), 7.01 (d, J=8.3 Hz, 1H); 13C NMR (CDCl3) l
19.6, 26.6, 29.8, 33.4, 39.5, 55.2, 71.7, 111.5, 112.2,
130.2, 130.4, 142.3, 157.9: mass spec. 206 (12), 175
(100), 129 (15), 128 (14), 125 (15), 115 (15).
This work was supported by a grant from Research
Corporation. L.J.S. was supported by an award from
Pfizer, and the Howard Hughes Medical Institute pro-
vided funds. Charles P. Kulier of Pfizer performed a
valuable literature search in this area.27
References
3.8. (R)-(C1)-1,2,3,4-Tetrahydro-7-methoxy-1-methyl-1-
naphthalenemethanol, 3a
1. Gulavita, N.; Hori, A.; Shimizu, Y.; Laszlo, P.; Clardy, J.
Tetrahedron Lett. 1988, 29, 4381–4384.
2. Takano, S.; Inomata, K.; Sato, T.; Takahashi, M.;
Ogasawara, K. J. Chem. Soc., Chem. Commun. 1990,
290–292.
3. Takano, S.; Inomata, K.; Sato, T.; Takahashi, M.;
Ogasawara, K. J. Chem. Soc., Chem. Commun. 1989,
1591–1592.
4. ElAzab, A. S.; Taniguchi, T.; Ogasawara, K. Heterocy-
cles 2002, 56, 39–43.
5. Tamura, O.; Yanagimachi, T.; Kobayashi, T.; Ishibashi,
H. Org. Lett. 2001, 3, 2427–2429.
6. Fadel, A.; Arzel, P. Tetrahedron: Asymmetry 1995, 6,
893–900.
7. Fadel, A.; Arzel, P. Tetrahedron: Asymmetry 1997, 8,
371–373.
Was made by combining 5a (29 mg) in dry CH2Cl2 (2
mL) at −78°C under N2 with MeAlCl2 (1 M in hexane,
70 mL). The mixture was stirred 30 min, and then 5%
HCl (8 mL) and ether (30 mL) were added. The result-
ing aqueous layer was extracted with ether (20 mL),
and the combined ethereal extract was extracted with
5% NaHCO3 and dried (MgSO4). Column chromatog-
raphy (4.5:1 hexane:EtOAc) gave 19.2 mg (66%) of 3a,
89.5% ee [h]D22 −18.3 (c 1.1, CHCl3), lit.12 [h]2D6 −16.5 (c
0.36, CHCl3) and lit.7 [h]D20 −20 (c 1, CHCl3). When
recrystallized (from 1:1 benzene:hexane) 8a was used to
make epoxide 5a, the ee of the cyclization product 3a
was 94–95%.
8. Hulme, A. N.; Henry, S. S.; Meyers, A. I. J. Org. Chem.
1995, 60, 1265–1270.
9. Takemoto, T.; Sodeoka, M.; Sasai, H.; Shibasaki, M. J.
Am. Chem. Soc. 1993, 115, 8477–8478 and see correction
ibid 1994, 116, 11207.
10. Shiotani, T.; Kometani, K.; Nozawa, T.; Kurobe, A.;
Futsukaichi, O. J. Med. Chem. 1979, 22, 1558–1560 and
references cited therein.
11. Shiotani, S.; Okada, H.; Yamamoto, T.; Nakamata, K.;
Adachi, J.; Nakamoto, H. Heterocycles 1996, 43, 113–
126.
12. Shiotani, S.; Okada, H.; Nakamata, K.; Yamamoto, T.;
Sekino, F. Heterocycles 1996, 43, 1031–1047.
13. Fuji, K. Chem. Rev. 1993, 93, 2037–2066.
14. Corey, E. J.; Guzman-Perez, A. Angew. Chem., Int. Ed.
1998, 37, 388–401.
3.9. (S)-(C1)-1,2,3,4-Tetrahydro-7-methoxy-1-methyl-1-
naphthalenemethanol, 3b
A sample of 5b (123.4 mg) was dissolved in dry CH2Cl2
(8 mL), and the solution was cooled to −78°C. MeAlCl2
(1 M in hexane, 230 mL) was added and the cooled
mixture was stirred for 42 min. TLC showed the reac-
tion was over, and 5% HCl (25 mL) was added over 15
min, and then at room temperature, ether (30 mL) was
added. The aqueous layer was extracted with ether, and
the combined ethereal extract was washed with 5%
NaHCO3 and sat. NaCl and dried (MgSO4). Rotary
TLC gave 9.3 mg of aldehyde 9 (below) and 81.6 mg
(66%) of 3b, 88% ee, [h]2D0 +18.1 (c 0.85, CHCl3), lit.12
[h]2D4 +16.6 (c 3.4, CHCl3). Capilliary GC showed the
fractions collected were pure, and this was the preferred
method of purification. The crude product was
analysed by GC and found to be 84% alcohol and 14%
aldehyde (2% total impurity). Recrystallized diol led to
3b with a higher ee (90–94%).
15. Carpenter, N. E.; Kucera, D. J.; Overman, L. E. J. Org.
Chem. 1989, 54, 5846–5848.
16. Sharpless, K. B.; Amberg, W.; Bennani, Y. L.; Crispino,
G. A.; Hartung, J.; Jeong, K.-S.; Kwong, H.-L.;
Morikawa, K.; Wang, Z.-M.; Xu, D.; Zhang, X.-L. J.
Org. Chem. 1992, 57, 2768–2771.
3.10. 2-Methyl-5-(p-methoxyphenyl)pentanal, 9
17. Kolb, H. C.; Sharpless, K. B. Tetrahedron 1992, 48,
10515–10530.
IR 2772 (w), 1724 (s), 1613, 1514, 1246, 831 (m), and
18. Cink, R. D.; Forsyth, C. J. J. Org. Chem. 1995, 60,
8122–8123.
1
812 (m) cm−1: H NMR (CDCl3), l 1.08 (d, J=7.3 Hz,

S. K. Taylor et al. / Tetrahedron: Asymmetry 14 (2003) 743–747
747
19. Taylor, S. K.; Davisson, M. E.; Hissom, B., Jr.; Brown,
S. L.; Pristach, H. A.; Schramm, S. B.; Harvey, S. M. J.
Org. Chem. 1987, 52, 425–429.
20. Taylor, S. K.; Blankespoor, C. L.; Harvey, S. M.;
Richardson, L. J. J. Org. Chem. 1988, 53, 3309–3312.
21. Taylor, S. K.; Hockerman, G. H.; Karrick, G. L.; Lyle,
S. B.; Schramm, S. B. J. Org. Chem. 1983, 48, 2449–2452.
22. Taylor, S. K. Org. Prep. Proc. Int. 1992, 24, 245–
284.
23. Taylor, S. K.; Dickinson, M. G.; May, S. A.; Pickering,
D. A.; Sadek, P. C. Synthesis 1998, 1133–1136.
24. Julia, M.; Labia, R. Bull. Soc. Chim. Fr. 1972, 4151–4157.
25. Corey, E. J.; Sodeoka, M. Tetrahedron Lett. 1991, 32,
7005–7008.
26. Lehmkuhl, H.; McLane, R. Liebigs Ann. Chem. 1980,
736–743.
27. For a recent synthesis of ( )-aphanorphine, see: Fuchs, J.
R.; Funk, R. L. Org. Lett. 2001, 3, 3923–3925.