SCHEME 3a
In conclusion we have demonstrated a six-step syn-
thesis of (S)-(+)-3-aminomethyl-5-methylhexanoic acid
(1) that delivers the product in high yield and excellent
enantiopurity. The synthetic sequence described is as
short as the previously preferred route,3 but potentially
provides significant improvements in cost of goods, waste
reduction, and throughput.
a
Reagents and conditions: (a) [(R,R)-(Me-DuPHOS)Rh(COD)]-
BF4, H2 (45 psi), MeOH, 55 °C, 100% conversion, 97.7% ee; (b) (i)
Sponge Ni, KOH, H2 (50 psi), H2O, EtOH; (ii) AcOH, 61% (two
steps), 99.8% ee.
Exp er im en ta l Section
ter t-Bu t yla m m on iu m 3-Cya n o-5-m et h yl-h ex-3-en oa t e
(2b). Ethyl ester 2a (20.0 g, 110 mmol, see Supporting Informa-
tion) and lithium hydroxide hydrate (13.0 g, 310 mmol) were
suspended in a mixture of tetrahydrofuran (75 mL) and water
(25 mL). The slurry was vigorously stirred for 4 h at room
temperature. The mixture was acidified to pH 2 (HCl, 3 N) and
extracted into ethyl acetate (3 × 150 mL). The combined organic
layers were dried (MgSO4) and concentrated to give crude
3-cyano-5-methylhex-3-enoic acid: IR (film) νmax 2222, 1714
conditions. Thus, the favored process is to prepare the
tert-butylammonium salt 2b from the purified ethyl ester
2a , followed by asymmetric hydrogenation to tert-butyl-
ammonium salt 3b (Scheme 3). This process has been
scaled up to multi-killogram quantities without signifi-
cant difficulties (see Supporting Information).
The final step in the synthesis of (S)-(+)-3-amino-
methyl-5-methylhexanoic acid (1), the reduction of the
nitrile group, was accomplished via a heterogeneous
hydrogenation of the tert-butylammonium salt 3b over
sponge nickel. The crude product was crystallized from
a mixture of ethanol, water, and acetic acid to give
Pregabalin 1 in 61% yield and 99.8% ee (Scheme 3).
The much higher reactivity and selectivity observed
for the asymmetric hydrogenation of salts 2b and 2c
compared to the ethyl ester 2a is due to enhanced
coordination between the substrate and the catalyst. It
is well established that the bisphosphine rhodium cata-
lysts of this type are most effective when the substrate
is able to behave as a bidentate ligand 7 (Scheme 4).10
In the presence of the strongly coordinating nitrile ligand
this chelating binding mode is disrupted. The 31P{1H}
NMR spectrum of the complex formed between the
catalytic intermediate [(R,R)-(Me-DuPHOS)Rh(CD3OD)2]-
BF4 (6) and tert-butylammonium salt 2b shows a dynamic
mixture of species, characterized by complex and broad-
ened signals. Within this is a pair of double doublets at
δ 79.3 and 88.6 ppm (J PP ) 34 Hz, J PRh ) 149 Hz) which
can be assigned to the rhodium chelate 7. The spectro-
scopic data for this complex are similar to those previ-
ously reported for an analogous vinyl acetate complex.10
Upon treatment with hydrogen, the olefin is reduced and
the 31P{1H} NMR spectrum collapses to a doublet at δ
95.4 ppm (J PRh ) 171 Hz), assigned to the bis-(S)-3-cyano-
5-methylhexanoate complex 8 as the 31P{1H} NMR
spectrum is almost identical with that of [(R,R)-(Me-
DuPHOS)Rh(NCCH3)2]BF4 [δ 95.3 (d, J PRh ) 175 Hz)].
It is proposed that the small standing concentration of 7
in the reaction mixture (ca. 10%) serves as a conduit
through which all the hydrogenation substrate is con-
verted to product. The relatively low reactivity observed
in this reaction compared to, for example, itaconic salts4c
or amido itaconates11 is attributed to the low levels of
the reactive intermediate 7 in the reaction mixture.
Similar observations, where a minor component of a
mixture gives rise to the major product, are well estab-
lished in the asymmetric hydrogenation of prochiral
olefins by chiral bisphosphine rhodium complexes.12
1
cm-1; H NMR (CDCl3, 400 MHz) δ (major isomer) 1.09 (3H, d,
J ) 7.0 Hz), 2.91 (1H, dheptet, J ) 10.0, 7.0 Hz), 3.25 (2H, br),
6.16 (1H, d, J ) 10.0 Hz), (minor isomer) 1.05 (3H, d, J ) 6.7
Hz), 2.63 (1H, dheptet, J ) 10.0, 6.7 Hz), 3.31 (2H, s), 6.40 (1H,
d, J ) 10.0 Hz); 13C NMR (CDCl3, 100 MHz) δ (major isomer)
22.1, 31.9, 39.0, 104.6, 116.9, 159.7, 175.4 (minor isomer) 21.9,
28.9, 34.2, 104.8, 119.5, 159.1, 175.0. m/z 152 (M - H), 305
(2M - H). The acid was dissolved in ethyl acetate (400 mL) and
a solution of tert-butylamine in ethyl acetate (20 mL) was added.
The temperature of the solution rose by approximately 10 °C as
the salt 2b precipitated as a white crystalline solid. The product
was collected by filtration and dried in vacuo (22.15 g, 89%):
mp 161 °C; IR (KBr) νmax 2216, 1557 cm-1 1H NMR (CD3OD,
;
400 MHz) δ (major isomer) 1.09 (6H, d, J ) 6.5 Hz), 1.37 (9H,
s), 2.81 (1H, dheptet, J ) 10.0, 6.5 Hz), 3.04 (2H, d, J ) 1 Hz),
6.13 (1H, d, J ) 10.0 Hz), (minor isomer) 1.05 (6H, d, J ) 6.5
Hz), 1.37 (9H, s), 2.74 (1H, dheptet, J ) 10.1, 6.5 Hz), 3.11 (2H,
s), 6.25 (1H, d, J ) 10.1 Hz); 13C NMR (CD3OD, 100 MHz) δ
(major isomer) 22.7, 28.3, 33.0, 44.1, 52.9, 110.3, 119.2, 157.3,
177.1, (minor isomer) 22.1, 28.3, 29.7, 38.8, 52.9, 110.8, 122.1,
157.0, 176.5; m/z 74 (tBuNH3+), 305 (2M + H).
Rep r esen ta tive P r oced u r e for Hyd r ogen a tion Scr een -
in g Rea ction s. A solution of ethyl ester 2a (0.19 mL, 1.0 mmol)
in methanol (4 mL) was placed in a glass-lined 50-mL PARR
microreactor modified with an injection septum and valve. The
vessel was heated to an internal temperature of 55 °C. A
hydrogen atmosphere was established and a solution of [(R,R)-
(iPr-DuPHOS)Rh(COD)]BF4 (7.2 mg, 10 µmol) in methanol (1
mL) was added via syringe. The vessel was pressurized with
hydrogen to 100 psi and stirred overnight. The pressure was
then released and the solvent was removed in vacuo. 1H NMR
analysis showed approximately 80% conversion to 3a , GC
analysis showed 86.4% conversion, 43.8% ee (S): IR (film) νmax
2242, 1738 cm-1; 1H NMR (CDCl3, 400 MHz) δ 0.96 (3H, d, J )
6.8 Hz), 0.98 (3H, d, J ) 6.5 Hz), 1.29 (3H, t, J ) 7.1 Hz), 1.34
(1H, ddd, J ) 13.4, 9.4, 5.0 Hz), 1.64 (1H, ddd, J ) 13.8, 10.9,
4.7 Hz), 1.87 (1H, m), 2.53 (1H, dd, J ) 16.6, 6.9 Hz), 2.69 (1H,
dd, J ) 16.3, 7.3 Hz), 3.06 (1H, m), 4.20 (2H, q, J ) 7.1 Hz); 13
C
NMR (CDCl3, 100 MHz) δ 14.5, 21.6, 23.2, 25.7, 26.2, 26.5, 37.5,
41.1, 61.6, 121.5, 170.1. Screening reactions at room temperature
were carried out via a modified procedure. The liner was charged
with a stir bar, the substrate, and catalyst. The vessel was
assembled and a hydrogen atmosphere established as described
above. Methanol was added via the septum, the vessel was again
purged before pressurizing to the reaction pressure and stirring
was then initiated.
ter t-Bu t yla m m on iu m (S)-3-Cya n o-5-m et h ylh exa n oa t e
(3b). A pressure reactor was charged with a solution of tert-
butylammonium salt 2b (125.8 g, 0.56 mol) in methanol (1 L).
A hydrogen atmosphere was established and the vessel was
heated to 45 °C. A solution of [(R,R)-(Me-DuPHOS)Rh(COD)]-
BF4 (0.125 g, 0.206 mmol) in methanol (15 mL) was added via
syringe. The vessel was charged with hydrogen to 65 psi and
the reaction was stirred at 45 °C until hydrogen uptake ceased
(10) Burk, M. J .; Bienwald, F.; Challenger, S.; Derrick, A.; Ramsden,
J . A. J . Org. Chem. 1999, 64, 3290.
(11) Berens, U.; Burk, M. J .; Gerlach, A.; Hems, W. Angew. Chem.,
Int. Ed. 2000, 39, 1981.
(12) Landis, C. R.; Halpern, J . J . Am. Chem. Soc. 1987, 109, 1746.
J . Org. Chem, Vol. 68, No. 14, 2003 5733