Enantioselective synthesis of (2R,3R)- and (2S,3S)-2-[(3-chlorophenyl)-(2- methoxyphenoxy)methyl]morpholine

Article abstract of DOI:10.1016/j.tetasy.2005.06.012

The enantioselective synthesis of the (R,R)- and (S,S)-enantiomers of 1 from commercially available 3-chlorocinnamic acid is reported. The Sharpless asymmetric epoxidation was used to establish the stereocenters in the synthesis of both enantiomers of 1.

Full text of DOI:10.1016/j.tetasy.2005.06.012

Tetrahedron:
Asymmetry
Tetrahedron: Asymmetry 16 (2005) 2249–2256
Enantioselective synthesis of (2R,3R)- and (2S,3S)-2-
[(3-chlorophenyl)-(2-methoxyphenoxy)methyl]morpholine
Wayne W. Harding,a Matthis Hodge,a Zhixia Wang,b William L. Woolverton,b
Damon Parrish,c Jeffrey R. Deschampsc and Thomas E. Prisinzanoa,*
aDivision of Medicinal & Natural Products Chemistry, College of Pharmacy, The University of Iowa, Iowa City, IA 52242, USA
bDepartment of Psychiatry, University of Mississippi Medical Center, Jackson, MS 39216, USA
cLaboratory for the Structure of Matter, Code 6030, Naval Research Laboratory, Washington, DC 20735, USA
Received 22 April 2005; accepted 8 June 2005
Abstract—The enantioselective synthesis of the (R,R)- and (S,S)-enantiomers of 1 from commercially available 3-chlorocinnamic
acid is reported. The Sharpless asymmetric epoxidation was used to establish the stereocenters in the synthesis of both enantiomers
of 1.
ꢀ 2005 Elsevier Ltd. All rights reserved.
1.Introduction
ing the aryloxy-propanolamine or aryloxy-methyl-mor-
pholino moiety, such as nisoxetine, tomoxetine, and
reboxetine (Fig. 1), have been shown to possess high
affinities and selectivities for NET.13 Compounds of
these classes have been evaluated as potential treatments
for depression and attention-deficit/hyperactivity disor-
der (ADHD).14–16 In fact, reboxetine (EdronaxTM) is cur-
rently marketed in Europe as an antidepressant while
tomoxetine (StratteraTM), has been approved by the
FDA as a treatment for ADHD.17,18 The enantiomers
of these compounds often display significantly different
activities at NET. For example, reboxetine is sold com-
mercially as an enantiomeric mixture of (R,R)- and
(S,S)-isomers. However, the (S,S)-enantiomer of
reboxetine is approximately 24 times more potent than
the (R,R)-enantiomer.19 In the case of tomoxetine,
almost all activity resides in the (R)-enantiomer.13
Current interest in the development of pharmacothera-
pies for the treatment of depression and ADHD has
Stimulant abuse is a serious problem in the US with 37%
of state and local law enforcement agencies identifying
cocaine as their greatest drug threat.1 It has been postu-
lated that the addictive properties of cocaine are due to
the blockade of dopamine transporters (DAT) in the
brain by cocaine and subsequent increase in extracellu-
lar dopamine levels (the dopamine hypothesis).2 It is
known that cocaine displays similar in vitro affinity for
the DAT, serotonin transporter (SERT), and norepi-
nephrine transporter (NET) in nervous tissue.3 Further-
more, a growing body of evidence indicates that the
norepinephrine transporter (NET) and serotonin trans-
porter (SERT) also play a role in mediating cocaineꢀs
pharmacological effects.4–12 It has been suggested that
blockade of DATby cocaine contributes positively to
cocaineꢀs reinforcing effects, while blockade of SERT
may have a negative influence.11,12 The influence of
NETblockade is, as yet, unclear though it has been sug-
gested that the blockade of NETmay mediate the aver-
sive effects of cocaine.7,10
MeO
O
H3C
O
EtO
O
To probe the role of NET in mediating the pharmaco-
logical effects of cocaine, high-affinity ligands are
required as pharmacological tools. Compounds possess-
NHCH3
NHCH3
NH
O
(R)-tomoxetine
(R)-nisoxetine
(2S,3S)-reboxetine
*
Corresponding author. Tel.: +1 319 335 6920; fax: +1 319 335 8766;
e-mail: thomas-prisinzano@uiowa.edu
Figure 1. Structures of reboxetine, nisoxetine, and tomoxetine.
0957-4166/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2005.06.012

2250
W. W. Harding et al. / Tetrahedron: Asymmetry 16 (2005) 2249–2256
necessitated the development of enantioselective routes
to these classes of compounds.
with guaiacol to provide diol 5. Epoxide 6 was prepared
via a sequence of reactions from 5 involving the selective
protection of the primary and secondary alcohol func-
tionalities as TMS ether and mesylate, respectively,
and subsequent cleavage of the primary alcohol fol-
lowed by ring closure by elimination of the secondary
mesyloxy group. The overall sequence of events was
conducted without any intermediate purification and
the overall transformation was completed in 57% yield
from 5. The reaction of 6 with ethanolamine in isopro-
panol gave amino alcohol 7. Treatment of 7 with tosyl
chloride gave 8a, which was subsequently cyclized to
9a under phase transfer catalysis conditions. However,
N-detosylation of 9a to give racemic 1 as described in
the literature for similar morpholino substrates,25 gave
unsatisfactory yields of product. Attempted deprotec-
tion of 9a with sodium naphthalide resulted in recovery
of starting material. Since the tosyl group proved so dif-
ficult to remove we decided to investigate use of the
more labile nosyl group. Thus, compound 9b was pre-
pared from 7 according to Scheme 1. However, removal
of the nosyl protecting group could not be achieved with
benzenethiol and some decomposition of starting mate-
rial occurred. After this disappointing result, we con-
Racemic 1 was previously reported to inhibit norepi-
nephrine uptake in rat hypothalamic synaptosomes with
20
high selectivity versus DATand SERT.
However, the
chiral isomers of 1 have never been evaluated at these
monoamine transporters. This, together with the possi-
bility for the development of 1 as a NET-selective PET
radiotracer for brain imaging studies, makes it a partic-
ularly attractive target for synthesis. The route described
herein provides an efficient enantioselective synthesis of
(R,R)- and (S,S)-enantiomers of 1.
2.Results and discussion
Initially, we embarked on the racemic synthesis of 1
(Scheme 1) to validate the reaction sequence and to
provide a racemic material for biological testing. Thus,
DIBAL-H reduction of ethyl 3-chlorocinnamate 221,22
afforded cinnamyl alcohol 323 in 93% yield. Racemic
epoxy alcohol 424 was prepared from 3 by reaction with
peracetic acid. This epoxide was subsequently cleaved
MeO
CO2Et
OH
OH
O
O
b
c
a
OH
Cl
Cl
Cl
OH
d-g
4
3
2
Cl
5
MeO
O
MeO
MeO
O
MeO
O
O
R
R
j
i
h
N
N
NH
OH
O
O
OH
Cl
OR1
Cl
Cl
OH
Cl
8a: R, R1 = Ts
8b: R, R1 = Ns
9a: R = Ts
7
6
9b: R = Ns
m
MeO
O
MeO
O
R
n
N
NBoc
O
OH
Cl
Cl
OH
10
11: R = Boc
1: R = H
o
k or l
Scheme 1. Reagents and conditions: (a) DIBALH, THF, 93%; (b) CH3CO3H, CH2Cl2, 80%; (c) guaiacol, NaOH, CH2Cl2, 39%; (d) TMSCl, Et3N,
EtOAc; (e) MsCl, Et3N, EtOAc; (f) 2 M HCl, EtOAc; (g) NaOH, toluene, methyltributylammonium chloride, 57% from 5; (h) ethanolamine, i-PrOH,
reflux, 48%; (i) tosyl chloride or nosyl chloride, Et3N, CH2Cl2, 80–85%; (j) NaOH, toluene, methyltributylammonium chloride, 65%; (k) Na-
naphthalide for 9a; (l) Benzenethiol for 9b; (m) Boc2O, CH2Cl2, 92%; (n) NaH, TsIm, 60%; (o) TFA, CH2Cl2, 90%.

W. W. Harding et al. / Tetrahedron: Asymmetry 16 (2005) 2249–2256
2251
templated the use of other nitrogen protecting groups
for compound 7. The Boc group seemed particularly
appealing in terms of ease of cleavage. Thus compound
10 was prepared. It was envisaged that cyclization to 1
could be effected by tosylation of the primary alcohol
function in 10 to give the corresponding tosylate fol-
lowed by ring closure under PTC conditions and Boc
deprotection. However, tosylation of 10 led to decompo-
sition. At this juncture, with 10 in hand, we sought to
effect cyclization to 11 in one pot by treatment of 10
with NaH and tosylimidazole in THF.25 The Boc group
was then removed from 11 with trifluoroacetic acid to
give racemic 1. The HCl salt of 1 was prepared and
found to have a melting point identical to the previously
reported value.20
3-chlorocinnamate (2)21,22 and the ester subsequently
reduced with DIBAL-H to give the cinnamyl alcohol 3
in 98% yield.23 The key step in our sequence to (S,S)-1
(Scheme 2) is the Sharpless asymmetric epoxidation
(SAE) of 3. We envisaged that SAE of 3 with D-DET
should provide (R,R)-4 and thus the (S,S)-enantiomer
of 1, while in a similar fashion, L-DETshould provide
(R,R)-1. Attempts to effect the SAE of 3 under condi-
tions reported for similar substrates31 [Ti(i-OPr)4, tert-
butylhydroperoxide, D-DET, À20 ꢁC] resulted in poor
and variable enantiomeric excesses (ees). It is known
that the SAE can suffer from poor ees due to the pres-
ence of adventitious water in the reaction.31 However,
even when rigorous attempts were made at keeping the
reaction anhydrous the maximum ee obtained was
45%. Significant improvements in the absolute value
and variability of ees was achieved by substitution of
tert-butylhydroperoxide with cumene hydroperoxide.
Thus, both (R,R)-4 and its enantiomer were prepared
by SAE in greater than 84% ee when cumene hydroper-
oxide was used as oxidant. The epoxide (R,R)-4 was
cleaved with guaiacol to give diol (S,R)-5. Differences
in the reactivities of the primary and secondary alcohol
functions of (S,R)-5 were exploited in the selective pro-
tection of the primary alcohol functionality as the TMS
ether 14a. Subsequently, the secondary hydroxyl group
was mesylated and the primary hydroxyl function
exposed by treatment of the silyl ether with acid. The
intermediate mesylate was converted to the epoxide
(S,S)-6 under phase transfer catalysis conditions in basic
medium. Mechanistically, this reaction proceeds via an
SN2 displacement of the mesyloxy group thus providing
a predictable stereochemical outcome. The smooth
conversion of intermediates from (S,R)-5 to (S,S)-6
obviated the need for any purification and a 60% overall
yield was realized for this transformation.
Racemic 1 was then reevaluated for its ability to inhibit
uptake of [3H]NE.20 The racemate displayed high
potency in inhibiting uptake of norepinephrine (Ki =
0.46 nM) with good selectivity versus dopamine and
serotonin uptake in vitro (Ki = 1732 and 39.4 nM,
respectively).
Having established a route to racemic 1 and bolstered by
its high NETpotency, we next focused our attentions on
preparing chiral material. Racemic syntheses of com-
pounds of the reboxetine class have been previously de-
scribed in the literature.20,26,27 Chiral products have
been accessed by optical resolution20,26,27 or by asym-
metric synthesis.28–30 The attractiveness of the present
enantioselective route lies in its amenability to the prep-
aration of all four possible stereoisomers of 1.
Our enantioselective route to (R,R)- and (S,S)-enantio-
mers of 1 started with commercially available 3-chloro-
cinnamic acid. This was esterified to provide ethyl
MeO
O
MeO
(R)
OH
O
OH
(R)
O
a
b
c,d,e,f
(S)
OH
O
(S)
Cl
OH
Cl
3
(R,R)-4
Cl
Cl
(S,R)-5
(S,S)-6
g
MeO
MeO
O
MeO
O
O
(S)
h
R
i
(S)
(S)
(S)
(S)
NH
N
(S)
NBoc
OH
O
OH
Cl
OH
Cl
Cl
OH
(S,S)-10
(S,S)-7
(S,S)-11: R = Boc
(S,S)-1: R = H
j
˚
Scheme 2. Reagents and conditions: (a) Ti(i-OPr)4, cumene hydroperoxide, D-DET, CH2Cl2, 4 A mol. sieves, 72%, 85% ee; (b) Guaiacol, NaOH,
CH2Cl2, 99%; (c) TMSCl, Et3N, EtOAc; (d) MsCl, Et3N, EtOAc; (e) 2 N HCl; (f) NaOH, toluene, methyltributylammonium chloride, 60% from 5;
(g) ethanolamine, i-PrOH, reflux, 70%; (h) Boc2O, CH2Cl2, 83%; (i) NaH, TsIm, THF, 61%; (j) TFA, CH2Cl2, 93%.

2252
W. W. Harding et al. / Tetrahedron: Asymmetry 16 (2005) 2249–2256
Compound (S,S)-7 was prepared by reaction of (S,S)-6
with ethanolamine at reflux in toluene. It is envisioned
that treatment of (S,R)-5 with 1 equiv of p-toluenesulfo-
nyl chloride followed by reaction with ethanolamine
under basic conditions will afford (S,R)-7.32 Boc-protec-
tion of the secondary amine in (S,S)-6 gave (S,S)-10. As
in the racemic synthesis, ring closure to (S,S)-11 was
achieved in one pot by reaction of the N-Boc diol
(S,S)-10 with NaH and tosylimidazole.25 Finally, N-
Boc deprotection of (S,S)-11 in trifluoroacetic acid fur-
nished (S,S)-1. An X-ray diffraction analysis of (S,S)-1
(Fig. 2) confirmed the absolute configuration to be
(2S,3S). In similar fashion, (R,R)-1 was also prepared
from 3 utilizing an SAE reaction using L-DET.
Figure 3. Diagnostic 1H NMR spectral region for ureas prepared from
reaction of (R)-(+)-1-phenylethyl isocyanate and compound 1 for (a)
racemic 1; (b) (S,S)-1; (c) (R,R)-1; and (d) (S,S)-1 + 4% (R,R)-1.
3.Conclusions
Figure 2. Displacement ellipsoid of (S,S)-1 drawn at 20% probability
levels.
We have demonstrated the utility of the Sharpless asym-
metric epoxidation in the enantioselective synthesis of 1.
Higher ees of the key intermediate epoxide 4 were
obtained by using cumene hydroperoxide as the oxidant
component than with tert-butylhydroperoxide. The
enantiomeric excess of (S,S)-1 was determined by
NMR analysis of the urea formed from (R)-(+)-1-phen-
ylethyl isocyanate and the stereostructure confirmed by
single crystal X-ray crystallographic analysis of the di-
benzoyl-D-tartarate salt. The NMR experiment de-
scribed here is a rapid method for determining ees of 1
and should also prove useful for other similar morpho-
lino substrates.
Enantiomeric purities of (R,R)- and (S,S)-1 were deter-
mined by a H NMR experiment in which the corre-
1
sponding ureas were prepared from enantiomerically
pure (R)-(+)-1-phenylethyl isocyanate and the spectra
analyzed.33 In this experiment the urea prepared from
racemic 1, exhibited a quartet of peaks at d 5.20, 5.20,
5.19, and 5.18, which proved to be diagnostic in assess-
ing purity. The 1H NMR spectrum of the individual iso-
meric ureas each showed a doublet in this region with
the (S,S)-urea derivative having peaks at d 5.20 and
5.18 and the (R,R)-derivative with peaks at d 5.20 and
5.19. To determine the limits of detection, in a separate
experiment, increasing amounts (2% increments by
weight) of purified (R,R)-derivative were added to a
known amount of purified (S,S)-urea derivative and
the d 5.2–5.1 region of the 1H NMR spectrum examined.
Addition of up to 2% by weight of (R,R) derivative did
not cause any detectable change in the appearance of the
spectrum. However, upon addition of 4% by weight of
the (R,R)-derivative a noticeable difference in the base-
line appearance of the diagnostic region of the spectrum
was observed (Fig. 3). Similarly, a 4% detection limit
was observed for the (R,R)-derivative. Thus, we can
conclude that the isomers prepared by this route are at
least 96% enantiomerically pure (ee 92%).
4.Experimental
4.1. General
Unless otherwise indicated, all reagents were purchased
from commercial suppliers and used without further
purification. All melting points were determined on a
Thomas–Hoover capillary melting apparatus and are
uncorrected. NMR spectra were recorded at 300 MHz
on a Bruker Avance-300 spectrometer using CDCl3 as
solvent, d values in parts per million (TMS as internal
standard), and J (Hz) assignments of 1H resonance cou-
pling unless otherwise stated. Thin-layer chromatogra-

W. W. Harding et al. / Tetrahedron: Asymmetry 16 (2005) 2249–2256
2253
phy (TLC) was performed on 0.25 mm plates Analtech
GHLF silica gel plates. Spots on TLC were visualized
with vanillin/H2SO4 in ethanol. Column chromatogra-
phy was performed with silica gel (32–63 l particle size)
from Bodman Industries (Atlanta, GA). Optical rota-
tions were performed on a Jasco P-1020 polarimeter at
room temperature. Elemental analyses were performed
by Atlantic Microlabs, Norcross, GA and were within
0.4% of the theoretical values.
oxide (12 mL) were added sequentially to a mixture of
˚
4 A molecular sieves (3 g) and CH2Cl2 (150 mL) at
À35 ꢁC under a nitrogen atmosphere. The mixture was
then stirred at À35 ꢁC for 2 h. A solution of 3 (5.3 g,
32.0 mmol) in CH2Cl2 (20 mL) was added dropwise over
1 h and the resulting mixture stirred for an additional
8 h at À35 ꢁC. The solution was filtered and washed
with 10% NaOH (2 · 80 mL) and brine (100 mL). The
organic layer was dried over Na2SO4 and the solvent
removed under reduced pressure to afford a crude resi-
due. The residue was purified by flash column chroma-
tography (eluent: 2% ethyl acetate–CH2Cl2) to give
4.78 g (81%, 82% ee) of (S,S)-434 as a colorless oil.
The ee was determined by the preparation of the mosher
ester and examination of the terminal methylene proton
signals at d 4.00–4.35 of the 1H NMR spectrum in
C6D6.31,33 [a]D = À24.1 (c 1.66, CHCl3); 1H NMR
(CDCl3): d 2.30 (br s, 1H), 3.20 (dt, 1H, J = 2.4,
3.6 Hz), 3.81 (br d, 1H, J = 12.6 Hz), 3.93 (d, 1H,
J = 2.4 Hz), 4.06 (br d, 1H, 12.6 Hz), 7.18 (m, 1H),
7.28 (m, 3H); 13C NMR (CDCl3): d 55.0, 61.2, 62.8,
124.2, 125.9, 128.7, 130.0, 134.8, 139.1.
4.2. Preparation of 3-chlorocinnamyl alcohol, 3
A
solution of ethyl 3-chlorocinnamate 2 (6.5 g,
30.9 mmol) in THF (30 mL) was cooled to 0 ꢁC and
DIBAL-H (73 mL, 1 M in THF) added dropwise over
30 min. The mixture was then stirred at 0 ꢁC for 5 h.
H2O (3.6 mL), 10% NaOH (3.6 mL) and Celite were
added to the solution and the mixture filtered. The solid
material was washed with THF and the filtrate was dried
over Na2SO4 and concentrated in vacuo to a pale yellow
syrup. The crude oil was purified by flash column chro-
matography (eluent: 30% ethyl acetate/hexanes) to give
5.1 g (98%) of 323,24 as a colorless oil: 1H NMR
(CDCl3): d 1.93 (br s, 1H), 4.33 (d, 2H, J = 5.1 Hz),
6.34 (dt, 1H, J = 5.1, 15.9 Hz), 6.53 (dt, 1H, J = 1.5,
15.9 Hz), 7.2 (m, 1H), 7.36 (s, 1H); 13C NMR (CDCl3):
63.3, 124.9, 126.6, 127.8, 129.6, 130.0, 130.3, 134.7,
138.8; HRMS calcd 168.0342, found 168.0333.
4.5. (2R,3S)-3-(3-Chlorophenyl)-3-(2-methoxyphenoxy)-
propane-1,2-diol, (R,S)-5
A mixture of (R,R)-4 (4.7 g, 25.7 mmol), CH2Cl2
(150 mL), H2O (18 mL), 50% NaOH (1.9 mL), methyl-
tributyl ammonium chloride (6.6 mL), and guaiacol
(5.9 mL, 53.7 mmol) was stirred at room temperature
for 2 h. CH2Cl2 was distilled over 1 h and the mixture
heated to 65 ꢁC for 3 h. The mixture was cooled to room
temperature, toluene (25 mL) added and the mixture
stirred for 5 min. The layers were separated and the
organic layer washed with 1 M NaOH (2 · 20 mL) and
water (2 · 30 mL), dried over Na2SO4, and the solvent
removed under reduced pressure to afford a viscous
oil. The crude product was purified by flash column
chromatography (eluent: 50% ethyl acetate/hexanes) to
4.3. (2R,3R)-2,3-Epoxy-3-(3-chlorophenyl)-1-propanol,
(R,R)-4
Diethyl-D-tartarate (856 lL, 4.99 mmol), titanium iso-
propoxide (1.22 mL, 4.17 mmol), and cumene hydroper-
oxide (12 mL) were added sequentially to a mixture of
˚
4 A molecular sieves (4 g) and CH2Cl2 (180 mL) at
À35 ꢁC under a nitrogen atmosphere. The mixture was
then stirred at À35 ꢁC for 2 h. A solution of 3 (7.0 g,
41.7 mmol) in CH2Cl2 (20 mL) was added dropwise over
1 h and the resulting mixture stirred for an additional
8 h at À35 ꢁC. The solution was filtered and washed
with 10% NaOH (2 · 80 mL) and brine (100 mL). The
organic layer was dried over Na2SO4 and the solvent
removed under reduced pressure to afford a crude resi-
due. The residue was purified by flash column chroma-
tography (eluent: 2% ethyl acetate–CH2Cl2) to give
5.5 g (72%, 85% ee) of (R,R)-4 as a colorless oil. The
ee was determined by the preparation of the mosher es-
ter and examination of the terminal methylene proton
signals at d 4.00–4.35 of the 1H NMR spectrum in
C6D6.16,18 [a]D = +21.3 (c 1.93, CHCl3); 1H NMR
(CDCl3): d 2.30 (br s, 1H), 3.20 (dt, 1H, J = 2.4,
3.6 Hz), 3.81 (br d, 1H, J = 12.6 Hz), 3.93 (d, 1H,
J = 2.4 Hz), 4.06 (br d, 1H, 12.6 Hz), 7.18 (m, 1H),
7.28 (m, 3H); 13C NMR (CDCl3): d 55.0, 61.2, 62.8,
124.2, 125.9, 128.7, 130.0, 134.8, 139.1; HRMS calcd
184.0291, found 184.0292.
give 7.8 g (99%) of (R,S)-5 as
a colorless oil:
[a]D = +11.4 (c 5.45, CHCl3); 1H NMR (CDCl3): d
3.29 (br s, 1H), 3.72 (dd, 1H, J = 4.5, 12.3 Hz), 3.92 (s,
3H), 3.96 (m obscured, 2H), 5.20 (d, 1H, J = 4.5 Hz),
6.65 (dd, 1H, J = 1.5, 8.1 Hz), 6.76 (ddd, 1H, J = 2.1,
6.9, 8.1 Hz), 6.94 (m, 2H), 7.31 (m, 3H), 7.44 (s, 1H);
13C NMR (CDCl3): d 56.0, 62.5, 74.5, 85.0, 112.0,
116.6, 121.2, 123.0, 125.0, 127.0, 128.6, 130.2, 135.0,
140.4, 147.0, 150.0; HRMS calcd 308.015, found
308.0816.
4.6. (2S,3S)-1,2-Epoxy-3-[(3-chlorophenyl)-(2-methoxy-
phenoxy)]propane, (S,S)-6
Chlorotrimethylsilane (3.2 mL, 25.4 mmol) was added
to a solution of (R,S)-5 (7.83 g, 25.4 mmol) and NEt3
(3.53 mL, 88.9 mmol) in ethyl acetate (85 mL) at
À20 ꢁC. After stirring for 15 min, methanesulfonyl chlo-
ride (2.3 mL, 29.6 mmol) and triethylamine (5.69 mL,
40.8 mmol) were added and the mixture stirred for
15 min. HCl (2 M, 50 mL) was added and the mixture
warmed to room temperature and stirred for 45 min.
The phases were separated and the organic phase
washed with NaHCO3 (2 · 70 mL) and brine (100 mL)
4.4. (2S,3S)-2,3-Epoxy-3-(3-chlorophenyl)-1-propanol,
(S,S)-4
Diethyl-L-tartarate (658 lL, 3.83 mmol), titanium iso-
propoxide (0.94 mL, 3.20 mmol), and cumene hydroper-

2254
W. W. Harding et al. / Tetrahedron: Asymmetry 16 (2005) 2249–2256
and dried over Na2SO4. The solvent was removed under
reduced pressure to give 7.5 g (76%) of a crude oil, which
was used without further purification. The oil was dis-
solved in toluene (160 mL) and NaOH (50%, 20 mL)
and a catalytic amount of methyltributylammonium
chloride were added and the resulting mixture stirred
at room temperature overnight. The layers were sepa-
rated and the aqueous layer extracted with toluene
(2 · 40 mL). The combined organic extract was washed
with brine (100 mL), dried over Na2SO4, and the solvent
removed under reduced pressure to afford a crude resi-
due. The residue was purified by flash column chroma-
tography (eluent: 40% ethyl acetate/hexanes) to give
4.4 g (60%) of (S,S)-6 as a colorless oil: [a]D = +14.5 (c
0.57, CHCl3); 1H NMR (CDCl3): d 2.73 (dd, 1H,
J = 2.6, 4.8 Hz), 2.85 (dd, 1H, J = 4.0, 4.8 Hz), 3.47
(ddd, 1H, 2.6, 4.0, 5.7 Hz), 4.92 (d, 1H, J = 5.7 Hz),
6.82 (m, 2H), 6.94 (m, 2H), 7.32 (m, 3H), 7.49 (d, 1H,
J = 1.5 Hz); 13C NMR (CDCl3): d 44.8, 54.9, 56.2,
82.5, 112.7, 117.6, 121.0, 123.0, 125.1, 127.1, 128.8,
130.1, 134.8, 139.8, 147.2, 150.6; HRMS calcd
290.0710, found 290.0715.
4.9. (2S,3S)-2-[(3-Chlorophenyl)-(2-methoxyphenoxy)-
methyl]morpholine-4-carboxylic acid tert-butyl ester,
(S,S)-11
A solution of (S,S)-10 (2.3 g, 5.1 mmol) in THF (60 mL)
was added in a dropwise manner to a suspension of
NaH (60% dispersion in oil, 0.5 g, 21.3 mmol) in THF
(250 mL) at 0 ꢁC. The resulting suspension was stirred
for 5 min and then allowed to reach room temperature
over 1 h. The mixture was cooled to 0 ꢁC and 1-(p-tolu-
enesulfonyl)imidazole (1.1 g, 5.1 mmol) was added in
one portion. The mixture was stirred for 15 min at
0 ꢁC and then was allowed to warm to room tempera-
ture and stirred for 19 h. The reaction mixture was then
cooled to 0 ꢁC and then saturated NH4Cl solution
(80 mL) was added cautiously. The mixture was warmed
to room temperature and then extracted with EtOAc
(2 · 200 mL). The combined EtOAc portion was washed
with
a 1:1 mixture of brine/saturated NaHCO3
(2 · 200 mL), dried over Na2SO4, and the solvent was
removed under reduced pressure to afford a crude oil.
The crude product was purified by flash column chro-
matography (eluent: 50% EtOAc/hexanes) to give 1.4 g
(61%) of 11 as a colorless oil: 1H NMR (CDCl3): d
1.45 (s, 9H), 2.86 (t, 1H, J = 11.7 Hz), 2.97 (t, 1H,
J = 11.7 Hz), 3.55 (dt, 1H, 2.7, 11.7 Hz), 3.80–3.85 (m,
3H), 3.87 (s, 3H), 3.97 (dd, 1H, J = 2.1, 11.4 Hz), 5.14
(d, 1H, J = 5.1 Hz), 6.74–6.77 (m, 2H), 6.86–6.95 (m,
2H), 7.26–7.30 (m, 3H), 7.45 (d, 1H, J = 1.5 Hz); 13C
NMR (CDCl3): d 28.6, 56.1, 66.9, 76.7, 78.0, 78.0,
80.3, 112.6, 117.8, 120.9, 122.8, 125.7, 127.6, 128.6,
129.8, 134.5, 139.9, 147.2, 150.8, 155.0; HRMS calcd
433.1656, found 433.1656.
4.7. (2S,3S)-[3-(3-Chlorophenyl)-2-hydroxy-3-(2-meth-
oxyphenoxy)]-(1-hydroxyethylamino)propane, (S,S)-7
Ethanolamine (3.7 mL, 6.1 mmol) was added to a solu-
tion of (S,S)-6 (4.4 g, 15.3 mmol) in isopropanol
(175 mL) and the mixture heated at reflux overnight.
The solvent was removed under reduced pressure and
EtOAc (120 mL) added to the residue. The EtOAc solu-
tion was washed with saturated NaHCO3 (2 · 50 mL),
brine (70 mL), dried over Na2SO4, and concentrated in
vacuo. The crude product was purified by flash column
chromatography (eluent: 10% MeOH/CH2Cl2) to give
3.78 g (70%) of (S,S)-7 as a colorless oil: 1H NMR
(CDCl3): d 2.57 (d, 2H, J = 5.7 Hz), 2.67 (dd, 2H,
J = 5.4, 11.1 Hz), 3.60 (t, 2H, J = 5.4 Hz), 3.88 (s, 3H),
4.11 (dt, 1H, J = 5.7, 7.2 Hz), 4.93 (d, 1H, J = 7.2 Hz),
6.66 (dd, 1H, J = 1.7, 8.3 Hz), 6.74 (ddd, 1H, J = 1.7,
7.1, 7.1 Hz), 6.88 (dd, 1H, J = 1.8, 8.1 Hz), 6.95 (ddd,
1H, J = 1.3, 7.1, 7.1 Hz), 7.29 (m, 3H), 7.45 (s, 1H);
13C NMR (CDCl3): d 50.5, 51.4, 55.8, 60.5, 73.8, 85.2,
112.1, 117.8, 121.0, 122.9, 125.5, 127.4, 128.5, 129.9,
134.6, 140.7, 147.4, 150.2; HRMS calcd 352.1316, found
352.1329.
4.10. (2S,3S)-2-[(3-Chlorophenyl)-(2-methoxyphenoxy)-
methyl]morpholine, (2S,3S)-1
A solution of (S,S)-11 (1.3 g, 3.0 mmol) in CH2Cl2
(100 mL) was treated with trifluoroacetic acid (7.4 mL,
96.0 mmol) and stirred for 2 h at room temperature.
The mixture was washed with 2 M NaOH (2 · 50 mL)
and H2O (2 · 50 mL), dried over Na2SO4, filtered, and
the solvent removed under reduced pressure to afford
a crude residue. The crude product was purified by flash
column chromatography (eluent: 5% MeOH/CH2Cl2) to
give 0.9 g (93%) of (S,S)-1 as a colorless oil: [a]D = +5.9
1
(c 2.44, CHCl3); H NMR (CDCl3): d 1.83 (br s, 1H),
4.8. (2S,3S)-[3-(3-Chlorophenyl)-2-hydroxy-3-
(2-methoxyphenoxy)propyl]-(2-hydroxyethyl)carbamic
acid tert-butyl ester, (S,S)-10
2.60–2.85 (m, 3H), 2.87 (dt, 1H, J = 3.3, 11.6 Hz), 3.65
(dt, 1H, J = 3.3, 10.8 Hz), 3.85 (s, 3H), 3.91 (ddd, 1H,
J = 3.6, 5.7, 9.0 Hz), 3.97 (ddd, 1H, J = 1.5, 3.0,
11.4 Hz), 5.10 (d, 1H, J = 6.0 Hz), 6.70–6.80 (m, 2H),
6.83–6.92 (m, 2H), 7.23–7.30 (m, 2H), 7.44 (d, 1H,
J = 1.2 Hz); 13C NMR (CDCl3): d 44.9, 46.5, 56.1,
67.1, 77.4, 78.0, 82.0, 112.5, 117.8, 120.9, 122.9, 125.7,
127.7, 128.6, 129.8, 134.5, 139.8, 147.2, 150.7; HRMS
calcd 333.1132, found 333.1126.
A solution of (S,S)-7 (5.4 g, 15.2 mmol) and di-tert-bu-
tyl dicarbonate (9.9 g, 45.7 mmol) in CH2Cl2 (170 mL)
was stirred at rt for 1 h. The solvent was removed under
reduced pressure and the crude product subjected to
flash column chromatography (eluent: 50% EtOAc/hex-
anes) to afford 5.7 g (83%) of (S,S)-10 as a colorless oil:
1H NMR (CDCl3): d 1.29 (s, 9H), 3.20 (m, 1H), 3.27–
3.42 (m, 3H), 3.60–3.85 (m, 2H), 3.92 (s, 3H), 4.20–
4.40 (m, 1H), 4.72 (d, 1H, J = 5.4 Hz), 6.63 (dd, 1H,
J = 8.4, 8.4 Hz), 6.75 (ddd, 1H, J = 1.8, 7.2, 7.2 Hz),
6.85–7.00 (m, 2H), 7.28–7.31 (m, 3H), 7.42 (s, 1H).
HRMS calcd 452.1840, found 452.1842.
4.11. (2S,3S)-2-[(3-Chlorophenyl)-(2-methoxyphenoxy)-
methyl]morpholine dibenzoyl-D-tartarate, (S,S)-1
dibenzoyl-D-tartarate
(2S,3S)-1 (0.8 g, 1.9 mmol) was dissolved in ethanol
(20 mL) and a solution of dibenzoyl-D-tartaric acid

W. W. Harding et al. / Tetrahedron: Asymmetry 16 (2005) 2249–2256
2255
(753 mg, 2.10 mmol) in ethanol (10 mL) added. The salt
precipitated on standing overnight and was collected,
washed with ethanol (100 mL) and dried to afford
1.2 g (89%) of (S,S)-1 dibenzoyl-D-tartarate: mp 160–
162 ꢁC; [a]D = +30.0 (c 0.1, acetic acid); Anal. Calcd
for (C54H54Cl2N2O14ÆC2H6O): C, 62.74; H, 5.64; N,
2.61. Found: C. 62.43; H, 5.68; N, 2.57.
Data collection was performed and the unit cell was ini-
tially refined using SMART [v5.625].35 Data Reduction
was performed using SAINT [v6.36A]36 and XPREP
[v6.12].37 Corrections were applied for Lorentz, polari-
zation, and absorption effects using SADABS [v2.03].38
The structure was solved and refined with the aid of
the programs in the SHELXTL-plus [v6.10] system of
programs.39
4.12. (2R,3R)-2-[(3-Chlorophenyl)-(2-methoxyphenoxy)-
methyl]morpholine, (2R,3R)-1
The crystal was a merohedral twin, consisting of two
components. It was described as such with the appropri-
ate twin law and refined to a relative component ratio of
approximately 64:36. A solvent methanol molecule was
located in the asymmetric unit, however, it was disor-
dered over two positions. The combined population of
these positions equated to the methanol molecule being
there half of the time. This is most likely a result of the
lack of hydrogen bonding between the solvent and the
other molecules. The methanol molecule was refined
using a global thermal parameter. For additional stabil-
A solution of (R,R)-11 (830 mg, 1.91 mmol) in CH2Cl2
(60 mL) was treated with trifluoroacetic acid (3.2 mL,
41.7 mmol) and stirred for 2 h at room temperature.
The mixture was washed with 2 M NaOH (2 · 50 mL)
and H2O (2 · 50 mL), dried over Na2SO4, filtered, and
the solvent removed under reduced pressure to afford
a crude residue. The crude product was purified by flash
column chromatography (eluent: 5% MeOH/CH2Cl2) to
give 0.55 g (86%) of (R,R)-1 as a colorless oil:
[a]D = À6.8 (c 1.56, CHCl3); 1H NMR (300 MHz,
CDCl3): d 1.83 (br s, 1H), 2.60–2.85 (m, 3H), 2.87 (dt,
1H, J = 3.3, 11.6 Hz), 3.65 (dt, 1H, J = 3.3, 10.8 Hz),
3.85 (s, 3H), 3.91 (ddd, 1H, J = 3.6, 5.7, 9.0 Hz), 3.97
(ddd, 1H, J = 1.5, 3.0, 11.4 Hz), 5.10 (d, 1H,
J = 6.0 Hz), 6.70–6.80 (m, 2H), 6.83–6.92 (m, 2H),
7.23–7.30 (m, 2H), 7.44 (d, 1H, J = 1.2 Hz); 13C NMR
(CDCl3): d 44.9, 46.5, 56.1, 67.1, 77.4, 78.0, 82.0,
112.5, 117.8, 120.9, 122.9, 125.7, 127.7, 128.6, 129.8,
134.5, 139.8, 147.2, 150.7.
˚
ity, the C1S–O1S bond length was fixed at 1.43 A. T he
remaining difference peaks were located in the area of
the methanol molecule, indicating that there may be
additional positions. The variables relating to the sol-
vent molecule were fixed in the final least squares
refinements.
For the rest of the model, the full-matrix least-squares
refinement on F2 included atomic coordinates and
anisotropic thermal parameters for all non-H atoms.
The H atoms were included using a riding model for
the entire system. The absolute configuration was estab-
lished by the structure determination of a compound
containing a chiral reference molecule of known abso-
lute configuration and confirmed by anomalous disper-
sion effects in diffraction measurements on the crystal,
with a resulting Flack parameter of 0.02(5).40 Coordi-
nates have been deposited with the Cambridge Crystal-
lographic Data Centre (Cambridge University
Chemical Laboratory, Cambridge CB2 1EW, UK).
CDCC number 268586.
4.13. (2R,3R)-2-[(3-Chlorophenyl)-(2-methoxyphenoxy)-
methyl]morpholine dibenzoyl-L-tartarate, (R,R)-1
dibenzoyl-L-tartarate
(2R,3R)-1 (0.55 g, 1.65 mmol) was dissolved in ethanol
(20 mL) and a solution of dibenzoyl-L-tartaric acid
(650 mg, 1.82 mmol) in ethanol (10 mL) added. The salt
precipitated on standing for 2 h and was collected,
washed with ethanol (100 mL) and dried to afford
1.1 g (94%) of (R,R)-1 dibenzoyl-L-tartarate: mp 160–
162 ꢁC; [a]D = À30.0 (c 0.1, acetic acid); Anal. Calcd
for (C54H54Cl2N2O14ÆC2H6O): C, 62.74; H, 5.64; N,
2.61. Found: C. 62.41; H, 5.69; N, 2.64.
Acknowledgements
The authors (W.W.H., T.E.P.) thank the Biological Sci-
ences Funding Program of the University of Iowa and
(W.L.W.) the National Institute on Drug Abuse for
financial support. W.L.W. is a recipient of National
Institute on Drug Abuse grant K05-DA15343. The
X-ray crystallographic work was supported in part by
the National Institute on Drug Abuse, NIH, DHHS,
and the Office of Naval Research. The authors also
thank Vic Parcell and Lynn Teesch of the High Resolu-
tion Mass Spectrometry Facility, University of Iowa for
mass spectral analysis.
4.14. Single crystal X-ray diffraction analysis of (S,S)-1
dibenzoyl-D-tartarate
C
a = 8.70440(10) A, b = 8.70440(10) A, c = 73.005(2) A, V =
54.50H56Cl2N2O14.50, FW = 1041.91, Tetragonal, P43,
˚
˚
˚
3
5531.4(2) A , Z = 4,
q
calcd = 1.251 mg/m3, l = 1.603
˚
mmÀ1, F(000) = 2188, R1 = 0.0972 for 6732 observed
(I > 2rI) reflections and 0.1148 for all 8700 reflections,
Goodness-of-fit = 1.077, 651 parameters.
A thin, colorless needle of dimensions 0.01 · 0.04 ·
0.28 mm2 was mounted on glass fiber using a small
amount of Epoxy. Data were collected on a Bruker
three-circle platform diffractometer equipped with a
SMART6000 CCD detector. The crystals were irradi-
References
1. National, D. I. C. National Drug Threat Assessment 2004;
US Department of Justice, Report 2004-Q0317-002, 2004.
2. Kuhar, M. J.; Ritz, M. C.; Boja, J. W. Trends Neurosci.
1991, 14, 299–302.
ated using a rotating anode Cu Ka source (k =
1.54178) with incident beam Go¨bel mirrors.

2256
W. W. Harding et al. / Tetrahedron: Asymmetry 16 (2005) 2249–2256
3. Rothman, R. B.; Baumann, M. H. Eur. J. Pharmacol.
2003, 479, 23–40.
4. Johanson, C. E.; Barrett, J. E. J. Pharmacol. Exp. Ther.
1993, 267, 1–8.
5. Rowlett, J. K.; Platt, D. M.; Spealman, R. D. J.
Pharmacol. Exp. Ther. 2004, 310, 342–348.
6. Gainetdinov, R. R.; Caron, M. G. Annu. Rev. Pharmacol.
Toxicol. 2003, 43, 261–284.
22. Hagiwara, H.; Shimizu, Y.; Hoshi, T.; Suzuki, T.; Ando,
M.; Ohkubo, K.; Yokoyama, C. Tetrahedron Lett. 2001,
42, 4349–4351.
23. Crilley, M. M. L.; Golding, B. T.; Pierpoint, C. J. Chem.
Soc., Perkin Trans. 1 1988, 2061–2067.
24. Von Sprecher, A.; Beck, A. Eur. Patent 89-105427, 1989.
25. Lanman, B. A.; Myers, A. G. Org. Lett. 2004, 6, 1045–
1047.
7. Uhl, G. R.; Hall, F. S.; Sora, I. Mol. Psychiat. 2002, 7, 21–
26.
8. Stephens, D. N.; Mead, A. N.; Ripley, T. L. Behav.
Pharmacol. 2002, 13, 327–345.
9. Rocha, B. A. Eur. J. Pharmacol. 2003, 479, 107–115.
10. Hall, F. S.; Li, X. F.; Sora, I.; Xu, F.; Caron, M.; Lesch,
K. P.; Murphy, D. L.; Uhl, G. R. Neuroscience 2002, 115,
153–161.
11. Ritz, M. C.; Kuhar, M. J. J. Pharmacol. Exp. Ther. 1989,
248, 1010–1017.
26. Orjales, A.; Mosquera, R.; Toledo, A.; Pumar, M. C.;
Garcia, N.; Cortizo, L.; Labeaga, L.; Innerarity, A.
J. Med. Chem. 2003, 46, 5512–5532.
27. Melloni, P.; Della Torre, A.; Lazzari, E.; Mazzini, G.;
Meroni, M. Tetrahedron 1985, 41, 1393–1399.
28. Boot, J.; Cases, M.; Clark, B. P.; Findlay, J.; Gallagher, P.
T.; Hayhurst, L.; Man, T.; Montalbetti, C.; Rathmell, R.
E.; Rudyk, H.; Walter, M. W.; Whatton, M.; Wood, V.
Bioorg. Med. Chem. Lett. 2005, 15, 699–703.
29. Prabhakaran, J.; Majo, V. J.; Mann, J. J.; Kumar, J. S.
Chirality 2004, 16, 168–173.
12. Roberts, D. C.; Phelan, R.; Hodges, L. M.; Hodges, M.
M.; Bennett, B.; Childers, S.; Davies, H. Psychopharmacol.
(Berl.) 1999, 144, 389–397.
30. Brenner, E.; Baldwin, R. M.; Tamagnan, G. Org. Lett.
2005, 7, 937–939.
13. Olivier, B.; Soudjin, W.; Wijngaarden, I. Prog. Drug Res.
2000, 54, 60–119.
14. Tejani-Butt, S. M. J. Pharmacol. Exp. Ther. 1992, 260,
427–436.
15. Schatzberg, A. F. J. Clin. Psychiat. 2000, 61(Suppl 10),
31–38.
16. Szabadi, E.; Bradshaw, C. M.; Boston, P. F.; Langley, R.
W. Hum. Psychopharmacol. 1998, 13(Suppl 1), S3–S12.
17. Scates, A. C.; Doraiswamy, P. M. Ann. Pharmacother.
2000, 34, 1302–1312.
31. Gao, Y.; Hanson, R. M.; Klunder, J. M.; Ko, S. Y.;
Masamune, H.; Sharpless, K. B. J. Am. Chem. Soc. 1987,
109, 5765–5780.
32. Prisinzano, T.; Hsin, L. W.; Folk, J. E.; Flippen-Ander-
son, J. L.; George, C.; Jacobson, A. E.; Rice, K. C.
Tetrahedron: Asymmetry 2003, 14, 3285–3289.
33. Greiner, E.; Folk, J. E.; Jacobson, A. E.; Rice, K. C.
Bioorg. Med. Chem. 2004, 12, 233–238.
34. Reddy, K. R.; Erion, M. D.; Matelich, M. C.; Kopcho, J.
J. US Patent 2003229225 A1, 2003.
18. Bymaster, F. P.; Katner, J. S.; Nelson, D. L.; Hemrick-
Luecke, S. K.; Threlkeld, P. G.; Heiligenstein, J. H.;
Morin, S. M.; Gehlert, D. R.; Perry, K. W. Neuropsycho-
pharmacology 2002, 27, 699–711.
35. Bruker. SMART v5.625. Bruker AXS Inc., Madison, WI,
USA, 2001.
36. Bruker. SAINT v6.36A. Bruker AXS Inc., Madison, WI,
USA, 2002.
19. Strolin Benedetti, M.; Frigerio, E.; Tocchetti, P.; Briance-
schi, G.; Castelli, M. G.; Pellizzoni, C.; Dostert, P.
Chirality 1995, 7, 285–289.
20. Melloni, P.; Carniel, G.; Torre, A. D.; Bonignori, A.;
Buonamici, M.; Pozzi, V.; Ricciardi, S.; Rossi, A. C. Eur.
J. Med. Chem. 1984, 19, 235–242.
37. Bruker. XPREP v6.12. Bruker AXS Inc., Madison, WI,
USA, 2001.
38. Bruker. SADABS v2.03, Bruker AXS Inc., Madison, WI,
USA, 2000.
39. Bruker. SHELXTL v6.10. Bruker AXS Inc., Madison, WI,
USA, 2000.
40. Flack, H. D. Acta Cryst. A 1983, 39, 876–881.
21. Huang, Z. Z.; Tang, Y. J. Org. Chem. 2002, 67, 5320–5326.