Asymmetric hydrogenation of a 4,4-diaryl-3-butenoate; a novel approach to sertraline.

Article abstract of DOI:10.1039/b301175p

The asymmetric hydrogenation of a selectively crystallised (E)-4,4-diaryl-3-butenoate with a rhodium-PhanePhos catalyst is described, providing an intermediate to the antidepressant sertraline.

Full text of DOI:10.1039/b301175p

View Article Online / Journal Homepage / Table of Contents for this issue
Asymmetric hydrogenation of a 4,4-diaryl-3-butenoate; a novel
approach to sertraline
Lee T. Boulton,* Ian C. Lennon and Raymond McCague
Dowpharma, Chirotech Technology Limited, A Subsidiary of The Dow Chemical Company,
Cambridge Science Park, Unit 321, Milton Road, Cambridge, UK CB4 0WG.
E-mail: lboulton@dow.com; Fax: ϩ44 (0)1223 506701; Tel: ϩ44 (0)1223 728010
Received 29th January 2003, Accepted 20th February 2003
First published as an Advance Article on the web 11th March 2003
The asymmetric hydrogenation of a selectively crystallised
(E)-4,4-diaryl-3-butenoate with
a rhodium-PhanePhos
catalyst is described, providing an intermediate to the anti-
depressant sertraline.
Asymmetric hydrogenation of an olefin functionality with
rhodium-diphosphine catalysts constitutes a powerful method
to construct stereogenic centres. For high enantioselectivity
and catalyst reactivity, a suitable substrate contains a frag-
ment of the form C᎐C–X–C᎐O where X is any single spacer
᎐
᎐
atom (often carbon or nitrogen) and the carbonyl provides
secondary co-ordination to the metal of the catalyst.1 Fre-
quently the stereocentre is generated at the position of the
olefin closest to the carbonyl function; an example from our
laboratories is the practical access to 2-alkylsuccinates by the
asymmetric hydrogenation of Stobbe derived itaconates
(Scheme 1, Y = CO2Me).2 Alternatively, a stereocentre can be
created at the more distant end of the olefin from the co-
ordinating functionality.3 Then, attainment of
a single
product enantiomer requires the substrate olefin to be a single
geometric isomer, because (Z)- and (E)-olefin isomers are
expected to supply opposite enantiomers with a given catalyst
(Scheme 1, Y = H). 4,4-Diarylbutanoates are useful inter-
mediates for pharmaceutical agents, particularly in single
enantiomer form, but only a (Z)/(E) isomeric mixture of
olefins is readily available. Sertraline (Zoloft®) 5, an import-
ant commercial pharmaceutical agent for the treatment of
depression, can be derived from a 4,4-diarylbutanoate.4,5 The
current published route for the commercial production relies
on a racemic synthesis and resolution of the final compound
as the mandelic acid salt.4 A means to establish the chirality
much earlier in the synthesis would be preferred economic-
ally.6–8
Scheme 2 Synthesis of sertraline 5. Reagents and conditions: i)
potassium tert-butoxide, diethyl succinate, tert-butanol; ii) 48% hydro-
bromic acid, acetic acid, followed by recrystallisation from MTBE, 32%
for two steps (17% 19 : 1 ratio, 15% 30 : 1 ratio); iii) tert-butylamine,
ethyl acetate, 99%; iv) H2, pre-catalyst, MeOH (see Table 1); v) 2 M
sulfuric acid, ethyl acetate; vi) chlorosulfonic acid, dichloromethane,
91% (two steps).
Following the known route, Stobbe condensation of 3,4-
dichlorobenzophenone and diethyl succinate and then acidic
treatment of the crude reaction residue afforded allylic acid 1 as
a 1 : 1 mixture of geometric isomers.5a We discovered that sub-
sequent recrystallisation from methyl tert-butyl ether (MTBE)
provided isomer enrichment. Two crops of crystals [19 : 1 and
30 : 1 (E) : (Z) isomeric ratio respectively] were obtained in a
combined yield of 32% overall from the starting dichloro-
benzophenone (Scheme 2). In an attempt to obtain additional
isomeric enrichment, various amine salts of the crude material
were screened. However, no significant improvement was
observed.
For the hydrogenation reactions we chose to use tert-butyl
ammonium salt 2 due to its favourable physical form and
because a carboxylate salt was expected to give better second-
ary co-ordination to the catalyst than the free acid. Thus hydro-
genation of salt 2 (19 : 1 (E) : (Z) isomeric ratio) was screened
with a variety of rhodium and ruthenium pre-catalysts (Table 1
and Fig. 1).† Initially the achiral pre-catalyst [DiPFc Rh
COD]BF4 was used to prepare a racemic sample, on which a
supercritical fluid chromatography (SFC) analytical method for
enantiomeric excess (ee) was developed for the saturated free
acid.9d
Scheme 1 Orientation of asymmetric hydrogenation for mixed-olefin
stereoisomers.
Herein we report that an enantiomerically enriched 4,4-
diarylbutanoate intermediate for sertraline 5 can be obtained
by separating out an olefin geometric isomer by crystallisation,
then subjecting it to asymmetric hydrogenation with an
appropriate rhodium-diphosphine catalyst (Scheme 2).
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 1 0 9 4 – 1 0 9 6
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 3
1094

View Article Online
Table 1 Summary of results for the catalytic asymmetric hydrogenation of tert-butyl ammonium salt 2
Entry
Pre-catalyst
Temp./ЊC
Pressure/psi
Time/ha
Conversion (%)
ee (%)bc
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16e
17
18
19
20
21
22
23
24f
[DiPFc Rh COD]BF4
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
0
120
120
120
120
120
120
120
120
120
120
120
120
120
120
120
120
120
120
180
60
1
16
16
16
4
100
77
86
rac
ndd
ndd
Ϫ1
20
[(R,R)-Me-DuPHOS Rh COD]BF4
[(R,R)-Et-DuPHOS Rh COD]BF4
[(R,R)-iPr-DuPHOS Rh COD]BF4
[(S,S)-Me-BPE Rh COD]BF4
100
100
100
100
100
100
100
100
100
100
93
100
100
100
100
100
100
100
100
100
100
[(R,R)-Me-BPE Rh COD]BF4
16
16
16
16
16
16
16
16
16
16
16
16
16
64
64
16
16
16
16
9
[(S,S)-iPr-BPE Rh COD]BF4
Ϫ13
Ϫ30
46
[(R,R)-Me-FerroTANE Rh COD]BF4
[(S,S)-Et-FerroTANE Rh COD]BF4
[(S,S)-nPr-FerroTANE Rh COD]BF4
[(S,S)-iPr-FerroTANE Rh COD]BF4
[(R,R)-Me-5-Fc Rh COD]BF4
33
Ϫ53
Ϫ65
82
[(R,R)-iPr-5-Fc Rh COD]BF4
[(R,R)-tBu-5-Fc Rh COD]BF4
ndd
90
[(R)-PhanePhos Rh COD]BF4
[(R)-PhanePhos Rh COD]BF4
[(S)-Xyl-PhanePhos Rh COD]BF4
[(S)-Xyl-PhanePhos Rh COD]BF4
[(S)-Xyl-PhanePhos Rh COD]BF4
[(S)-Xyl-PhanePhos Rh COD]BF4
[(R)-Cl-MeO-BIPHEP Rh COD]BF4
[(R)-Cl-MeO-BIPHEP Rh COD]BF4
[(R,R)-iPr-DuPHOS Ru (OCOCF3)2]
[(R,R)-iPr-BPE Ru (2-methylallyl)2]
84
Ϫ77
Ϫ72
Ϫ89
Ϫ89
3
36
Ϫ44
Ϫ50
0
22
40
22
22
22
120
120
120
120
a Time is not necessarily for complete reduction. b Negative ee corresponds to R enantiomer of product (by comparison of optical rotation). c rac =
racemic. d Not determined, starting olefin interferes in analytical assay if conversion is incomplete. e Dibenzylammonium salt used. f Free acid used.
Fig. 1 Structure of chiral diphosphine ligands used during the hydrogenation screen.9,10
The majority of reactions performed gave full conversion to
acid salt 3 over 16 hours at 22 ЊC and 120 psi hydrogen pressure
with 1 mol% catalyst. Unlike our experience with the hydrogen-
ation of itaconates,2 DuPHOS, BPE and FerroTANE based
rhodium catalysts showed little enantioselectivity for this sub-
strate (entries 2–11).9a–c The ferrocenylphospholane ligands
used for entries 12, 13 and 14 showed some increased selectivity.
Unexpectedly, rhodium-PhanePhos, a catalyst that typically
shows lower selectivity than other diphosphine complexes,
proved the most useful for this particular substrate, with 90% ee
being achieved (entry 15).10 The parent diphenylphosphinyl
ligands provided slightly better selectivity than the corre-
sponding xylyl analogue. Lowering the temperature and
pressure slightly increased the enantioselectivity for the xylyl
catalyst albeit with a lowering of the reaction rate (entries 18,
19 and 20). Hydrogenation of free acid 1 rather than the salt 2
gave consistently worse results under rhodium catalysis. A
variety of ruthenium pre-catalysts were examined against salt 2
and free acid 1, the best of these results are presented in the
table (entries 23 and 24) but selectivities were moderate.
In order to verify the enantioenriched diarylbutanoate salt 3
as an intermediate for sertraline, the hydrogenation product was
cyclised by treatment firstly with 2 M sulfuric acid to liberate
the free carboxylic acid and then with chlorosulfonic acid to
afford tetralone 4 in 91% yield, with no loss of enantiopurity
as judged by HPLC analysis (Scheme 2).6 It was found that
catalysts with the (S)-PhanePhos ligand provided the required
(R)-enantiomer of the diarylbutanoate and from that the
(S)-tetralone 4, which has been used for the synthesis of
sertraline 5 by reductive amination with methylamine.7f The
stereochemistry of the major component of the enriched olefin
mixtures used for the hydrogenation screen was determined to
1
be the (E)-isomer by X-ray crystallographic analysis and H
NMR (Fig. 2).‡
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 1 0 9 4 – 1 0 9 6
1095

View Article Online
1
reduced pressure to afford a crude residue, which was analysed by H
NMR spectroscopy for conversion and SFC for enantiomeric excess.
δH(400 MHz; d4-MeOD; ref. MeOH) 7.40–7.14 (8H, m, Ar), 3.96 (1H,
t, J 8, CHCH2), 2.34–2.28 (2H, m, CHCH2), 2.09 (2H, t, J 7,
CHCH CH2) and 1.33 (9H, s, 3 × CH3); Free acid [α]2D2 = ϩ12.4 [c = 2.2
in benz2ene (product from run 15, 90% ee)]; δH(400 MHz; CDCl3; ref.
Me4Si) 7.36–7.18 (7H, m, Ar), 7.07 (1H, d, J 9, Ar), 3.91 (1H, br t, J 8,
CHCH2) and 2.40–2.27 (4H, m, CH2CH2). Determination of enantio-
meric excess was performed using a Gilson SFC system [Diacel Chiral-
cel OD column, 250 × 4.6 mm, 10 µm particle size, using 95% CO2, 5%
methanol (with 1% v/v TFA modifier), 30 ml minϪ1 flow rate, 3000 psi
pressure, 35 ЊC column temperature, 220 nm UV detection]. Retention
times: first enantiomer (S) 12.0 minutes, second enantiomer (R) 14.1
minutes.
‡ Crystal data for (E)-allylic acid 1, C16H12Cl2O2, M = 307.16: colour-
less, monoclinic, P21/C, a = 7.474(4) Å, b = 12.744(4) Å, c = 15.237(3) Å,
β = 91.49(3)Њ, V = 1450.7(10) Å3, Z = 4, dcalc = 1.406 g cmϪ3, F(000) =
632, µ(Mo-Kα) = 0.455 mmϪ1, θ = 6.9–8.2Њ (Mo-Kα), 2777 measured
reflections on a Rigaku AFC-5R diffractometer, 2567 independent
reflections [R(int) = 0.0346], structure determination using direct
methods (SHELXS86), R(F) = 0.0575, Rw = 0.1135 for 1321 reflections
with I > 2σ(I ), GOF = 1.022. CCDC reference numbers 196474. See
http://www.rsc.org/suppdata/ob/b3/b301175p/ for crystallographic data
in .cif or other electronic format.
Fig. 2 ORTEP diagram of the molecular structure of allylic acid 1.
Thermal ellipsoids shown at the 50% probability level.
The rhodium-PhanePhos catalysed hydrogenation of a 1 : 1
isomeric mixture of tert-butyl ammonium salt 2 using our best
conditions (cf. run 15) gave racemic product. The successful
synthesis of an intermediate for candoxatril,11 where the stereo-
centre is created at the more distant end of the olefin from the
coordinating carboxylate functionality, is presumably due to
the substrate being available as a single geometric isomer.
In summary we have demonstrated the principle of a route to
an enantioenriched 4,4-diarylbutanoate using asymmetric
hydrogenation, in which unpredictably, a catalyst based on the
PhanePhos ligand gave the best results. These findings illustrate
the importance of screening a wide range of ligands against
any new substrate.12 The route may have practical utility for
sertraline and stereochemically related compounds, particularly
when an efficient crystallisation based separation procedure for
the olefin isomers can be demonstrated. It may ultimately be
possible to recycle the unwanted olefin isomer, or hydrogenate it
with the opposite enantiomer of catalyst to converge all the
material into one enantiomer of product.
1 S. K. Armstrong, J. M. Brown and M. J. Burk, Tetrahedron Lett.,
1993, 34, 879–882.
2 M. J. Burk, F. Bienewald, M. Harris and A. Zanotti-Gerosa,
Angew. Chem., Int., Ed., 1998, 37, 1931–1933.
3 M. J. Burk, M. F. Gross and J. P. Martinez, J. Am. Chem. Soc., 1995,
117, 9375–9376.
4 A. Kleemann and J. Engel, Pharmaceutical Substances, Thieme,
Stuttgart, 1999, 3rd edn.
5 (a) W. M. Welch, A. R. Kraska, R. Sarges and K. B. Coe, J. Med.
Chem., 1984, 27, 1508–1515; (b) J. A. Lowe III, D. L. Hageman,
S. E. Drozda, S. McLean, D. K. Bryce, R. T. Crawford, S. Zorn,
J. Morrone and J. Bordner, J. Med. Chem., 1994, 37, 3789–3811.
6 E. J. Corey and T. G. Gant, Tetrahedron Lett., 1994, 35, 5373–
5376.
7 (a) K. Maeda and J. M. Brown, Chem. Commun., 2002, 310–311;
(b) S. Chandreskar and M. Venkat Reddy, Tetrahedron, 2000, 56,
1111–1114; (c) C.-Y. Chen and R. A. Reamer, Org. Lett., 1999, 1,
293–294; (d ) H. M. L. Davies, D. Stafford and T. Hansen, Org. Lett.,
1999, 1, 233–236 (Corrigendum: H. M. L. Davies, D. Stafford and
T. Hansen, Org. Lett., 2000, 2, 417); (e) C. Barbieri, E. Caruso,
P. D’Arrigo, G. P. Fanton and S. Servi, Tetrahedron: Asymmetry,
1999, 10, 3931–3937; ( f ) M. Williams and G. Quallich, Chem. Ind.
(London), 1990, 10, 315–319.
8 Alternative syntheses: (a) K. Vukics, T. Fodor, J. Fischer,
I. Fellegvári and S. Lévai, Org. Process Res. Dev., 2002, 6, 82–85;
(b) J. Yun and S. L. Buchwald, J. Org. Chem., 2000, 65, 767–774;
(c) M. Lautens and T. Rovis, Tetrahedron, 1999, 55, 8967–8976;
(d ) M. Lautens and T. Rovis, J. Org. Chem., 1997, 62, 5246–5247;
(e) G. J. Quallich and T. M. Woodall, Tetrahedron, 1992, 48, 10239–
10248; ( f ) G. J. Quallich, M. T. Williams and R. C. Friedmann,
J. Org. Chem., 1990, 55, 4971–4973.
9 (a) M. J. Burk and M. F. Gross, Tetrahedron Lett., 1994, 35, 9363–
9366; (b) U. Berens, M. J. Burk, A. Gerlach and W. Hems,
Angew. Chem., Int. Ed., 2000, 39, 1981–1984; (c) M. J. Burk, Acc.
Chem. Res., 2000, 33, 363; (d ) M. J. Burk, T. G. P. Gregory,
P. Harper, J. R. Lee and C. Kalberg, Tetrahedron Lett., 1994, 35,
4963–4966.
Acknowledgements
We thank Dr Madeleine Helliwell, Department of Chemistry,
University of Manchester for determination of the X-ray crys-
tal structure. We are grateful to Natasha Cheeseman and
Catherine Rippé for development of analytical methods.
Notes and References
† General hydrogenation procedure for the preparation of 4-(3,4-
dichlorophenyl)-4-phenylbutanoic acid tert-butyl ammonium salt
3. Geometrically enriched (E)-4-(3,4-dichlorophenyl)-4-phenylbut-3-
enoic acid tert-butyl ammonium salt 2 and the hydrogenation pre-
catalyst (1 mol%, see Table 1) were placed in a glass liner within a 50 ml
stainless steel autoclave and the vessel was assembled. The vessel was
pressurised to 120 psi with nitrogen and then the gas was released, this
process was repeated a further three times. After the final vent,
deoxygenated methanol (5 ml) was introduced into the bomb. The ves-
sel was then charged and vented three times with hydrogen to 120 psi.
The vessel was equilibrated to the appropriate temperature and charged
with the appropriate pressure of hydrogen. After stirring overnight, the
vessel was allowed to cool, the hydrogen pressure was released and the
vessel was disassembled. The reaction mixture was concentrated under
10 P. J. Pye, K. Rossen, R. A. Reamer, N. N. Tsou, R. P. Volante and
P. J. Reider, J. Am. Chem. Soc., 1997, 119, 6207–6208.
11 M. J. Burk, F. Bienewald, S. Challenger, A. Derrick and J. A.
Ramsden, J. Org. Chem., 1999, 64, 3290–3298.
12 T.-Y. Yue and W. A. Nugent, J. Am. Chem. Soc., 2002, 124, 13692–
13693.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 1 0 9 4 – 1 0 9 6
1096