Efficient preparation of an N-aryl β-amino acid via asymmetric hydrogenation and direct asymmetric reductive amination en route to Ezetimibe

Article abstract of DOI:10.1016/j.tetasy.2010.04.013

Two routes for the preparation of an N-aryl β-amino acid, an important precursor for the cholesterol-lowering drug Ezetimibe, were investigated. The first pathway proceeds via an Rh- or Ir-catalyzed asymmetric hydrogenation of N-aryl enamine giving the desired product with up to 82% ee. The other pathway involves a direct asymmetric reductive amination (DARA) of the β-keto ester which yielded the β-amino ester in high yield and 97% ee. Subsequent copper-catalyzed N-arylation gave the target compound.

Full text of DOI:10.1016/j.tetasy.2010.04.013

Tetrahedron: Asymmetry 21 (2010) 1709–1714
Contents lists available at ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
Efficient preparation of an N-aryl b-amino acid via asymmetric
hydrogenation and direct asymmetric reductive amination en route
to Ezetimibe
a
a
a
a
Guuske F. Busscher a, , Laurent Lefort , Jozef G. O. Cremers , Marco Mottinelli , Roel W. Wiertz ,
*
Ben de Lange a, Yutaka Okamura b, Yukinori Yusa b, Kazuhiko Matsumura b, Hideo Shimizu b,
Johannes G. de Vries a, André H. M. de Vries a,
*
a DSM Innovative Synthesis B.V. A Unit of DSM Pharma Chemicals, PO Box 18, 6160 MD Geleen, The Netherlands
b Takasago International Corporation, Corporate Research & Development Division, 1-4-11 Nishi-Yawata, Hiratsuka City, Kanagawa 254-0073, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Two routes for the preparation of an N-aryl b-amino acid, an important precursor for the cholesterol-low-
ering drug Ezetimibe, were investigated. The first pathway proceeds via an Rh- or Ir-catalyzed asymmet-
ric hydrogenation of N-aryl enamine giving the desired product with up to 82% ee. The other pathway
involves a direct asymmetric reductive amination (DARA) of the b-keto ester which yielded the b-amino
ester in high yield and 97% ee. Subsequent copper-catalyzed N-arylation gave the target compound.
Ó 2010 Elsevier Ltd. All rights reserved.
Received 24 March 2010
Accepted 13 April 2010
Available online 16 June 2010
This article is dedicated to Professor Henri
Kagan on the occasion of his 80th birthday
1. Introduction
Asymmetric hydrogenation of prochiral N-aryl enamines can be
consideredas acost-efficientandatom-economicmethodtosynthe-
Enantiomerically pure b-amino acid derivatives are recurring
structural motives in many important biologically active com-
pounds such as b-peptides and b-lactams. The N-aryl b-amino acid
derivative 1 described here is an attractive building block en route
to Ezetimibe (Scheme 1). Ezetimibe is a member of a class of lipid-
altering agents known as cholesterol absorption inhibitors.1,2 In
contrast to other drugs that reduce blood cholesterol levels by
inhibiting its biosynthesis, Ezetimibe exerts its activity by limiting
the intestinal absorption of both dietary and biliary cholesterol.3
size N-aryl b-amino acids. Although many catalysts are known for
hydrogenation of the structurally related N-acyl enamide ana-
logues,5 the N-substituted and unsubstituted enamines have been
much less studied.6–8,13b This is probably related to the conviction
that the presence of the amido group is a prerequisite for bidentate
binding of the substrate with the metal, leading to high reactivity
and selectivity.9,10 Very few reports exist on the asymmetric hydro-
genation of N-aryl b-enamino esters.11 Bruneau et al. also described
the preparation of a wide range of similar substrates but reported
only their non-enantioselective hydrogenation.12 Herein we report
on our screening efforts leading to the discovery of a hydrogenation
catalyst for the conversion of 3 into 1 (Scheme 2, pathway A).
Another approach to obtain b-amino acids is via direct asym-
metric reductive amination of the corresponding b-keto esters. In
spite of its attractiveness there are only a limited number of re-
ports on this method.13 Recently, Takasago and Merck jointly ap-
plied this technology to obtain Sitagliptin.14 In this article we
also explored the use of this route to obtain the chiral primary
amine 4 starting from keto ester 2 and its further conversion into
the N-aryl b-amino acid 1a by Cu-catalyzed N-arylation (Scheme 2,
pathway B).15
F
OH
OH
NH
O
R
N
O
F
O
1a R = H
1b R = Et
Ezetimibe
BnO
F
Scheme 1.
Many methods have been reported for the preparation of enan-
tiopure b-amino acids.4 In this article we describe the preparation
of N-aryl b-amino acid 1a via two short and efficient pathways,
that is, asymmetric hydrogenation of the N-aryl enamine 3 and di-
rect asymmetric reductive amination (DARA) of the b-keto ester 2
followed by N-arylation (Scheme 2).
2. Results and discussion
2.1. Asymmetric hydrogenation, pathway A
* Corresponding authors. Tel.: +31 464 761 572; fax: +31 46 4767604 (G.F.B.).
E-mail addresses: Guuske.busscher@dsm.com (G.F. Busscher), Andre.Vries-
de@dsm.com (A.H.M. de Vries).
The synthesis started from the inexpensive and readily avail-
able p-hydroxy-acetophenone 5. O-Benzylation followed by a
0957-4166/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2010.04.013

1710
G. F. Busscher et al. / Tetrahedron: Asymmetry 21 (2010) 1709–1714
F
Pathway A
asymmetric
NH
O
O
hydrogenation
F
O
3
O
O
NH
O
BnO
BnO
O
R
O
2
NH2
BnO
1a R = H
1b R = Et
direct asymmetric
red. amination
Cu-catalyzed
BnO
N-arylation
O
4
Pathway B
Scheme 2.
Claisen condensation16 afforded the b-keto ester 2 in high yield
(Scheme 3). Subsequent reaction with p-fluoro-aniline in the pres-
ence of molecular sieves and p-toluenesulfonic acid yielded the
hydrogenation substrate, that is, enamine 3.
O
O
O
i)
O
5
2
HO
BnO
(78%)
We started our screening by applying the Rh/TangPhos catalyst
developed by Zhang et al. (L-1, Scheme 4).11b This led to formation
of the desired product 1b with an acceptable enantiomeric access
(81%, Table 1, entry 1). The hydrogenation proceeds in various sol-
vents (CH2Cl2, 2,2,2-trifluoroethanol (TFE), MeOH, entries 1–3) but
in all cases the activity is rather low, possibly due to product inhi-
bition.17 An extensive screening of solvents and additives (Brön-
sted acids (CH3COOH), Lewis acids (La, Cu, Sc, Mg, Fe, Ti), NH4Cl,
phenol, iodine) did not improve the activity of the Rh/TangPhos
catalyst. We then tested some ligands from the Josiphos family,
L-2 to L-4 (Scheme 4), as this class of ligands had been used
successfully in the rhodium-catalyzed hydrogenation of an unpro-
tected enamine.8 Higher yields were obtained (Table 1, entries
F
(60%) ii) H2N
F
NH
O
O
3
BnO
Scheme 3. Synthesis of the enamine 3. Reagents: (i) BnBr, K2CO3, acetone, (95%)
then (EtO)2CO, NaH, THF, (83%); (ii) p-TsOH, molecular sieves (60%).
Table 1
Asymmetric hydrogenation of enamine 3
F3C
H
H
F
PtBu2
P
P
P
Fe
Fe
tBu tBu
H2
NH
O
3
L-1
L-2
O
CF3
P
BnO
1b
Entry
Catalyst
S/C
50
Solvent
P (bar)
T (°C)
1ba
eeb (%)
PtBu2
PtBu2
P
Fe
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
L-1c
L-1c
L-1c
L-2c
L-3c
L-4c
L-5c
L-6c
L-5c
L-5c
L-5c
L-5c
L-6c
L-6c
L-6c
L-7d
L-8e
L-8d
L-9e
L-9d
TFE
CH2Cl2
MeOH
TFE
TFE
TFE
TFE
TFE
TFE
TFE
TFE
TFE
TFE
TFE
TFE
TFE
TFE
TFE
TFE
TFE
25
25
25
25
25
25
25
25
25
25
5
50
50
50
50
50
50
50
50
30
40
50
70
50
40
30
50
50
50
50
50
45
32
49
62
99
93
80
82
69
75
88
89
75
91
81
100
94
100
74
93
81
87
67
38
74
39
82
80
73
74
81
53
67
78
80
45
42
42
41
36
50
50
50
50
50
50
50
50
50
50
50
50
50
50
L-3
L-4
P
P
P
P
Fe
Fe
5
5
N
25
25
25
25
25
25
25
L-5
L-6
1000
80
80
80
80
O
O
Ph2P N
P
N
R
2
a
b
c
HPLC area%.
Determined by chiral HPLC using a Chiralcel AD-H column.
Rh-precursor = [Rh(COD)2]BF4.
Ir-precursor = [Ir(COD)2]PF6.
L-7 R = i-Pr
L-8 R = t-Bu
L-9
d
e
Ir-precursor = [Ir(COD)Cl]2. Reaction time 18 h.
Scheme 4. Ligands used for the asymmetric hydrogenation of 3.

G. F. Busscher et al. / Tetrahedron: Asymmetry 21 (2010) 1709–1714
1711
Table 2
4–6) although the ee’s were lower than those achieved with Tang-
Screening of nitrogen sources in the direct reductive amination of b-keto ester 2
Phos. Stimulated by these results, we decided to screen a wide
range of commercially available ligands using our high throughput
screening equipment (HTE).18 Among the 32 ligands tested, two
new ligands that induced a high enantiomeric excess were identi-
fied, that is, L-5 Taniaphos 001-1 (82% ee, Table 1, entry 7) and L-6
Josiphos 404-1 (80% ee, Table 1, entry 8). Next, we examined the
effect of the hydrogen pressure and the temperature using these
two ligands. Unfortunately, the variations of these two parameters
did not result in any improvement, of the activity or of the enanti-
oselectivity (entries 9–15).
Since the activity of all rhodium systems we tested so far re-
mained low (incomplete conversions at S/C = 50–100), we decided
to turn toward another metal, that is, iridium. Inspired by the work
of Pfaltz et al.,6g we included chiral P,N-ligands in the large library
of ligands (monodentate and bidentate phosphines) tested with
this metal. Again we used our HTE set-up in order to screen many
ligands and conditions at the same time. Gratifyingly, we found out
that the use of Ir combined with P,N-ligand L-7 led to full conver-
sion at S/C = 1000 (entry 16). Encouraged by this result, we per-
formed an extensive screening of P,N-ligands in combination
with iridium as a metal. The best results were obtained with phos-
phine–oxazoline ligands for which we made the following observa-
tions: changes in the hydrogen pressure (5–25 bar) or the
temperature (25–50 °C) (not shown in Table 1) had little effect.
Good conversions were only obtained in TFE; much lower conver-
sions were found in most common solvents such as CH2Cl2, MeOH,
EtOH, i-PrOH, toluene, TFE/CH2Cl2 mixture, and TFE/EtOH mixture
even at S/C = 40. We found no influence of the counter-ions having
tested PF6, Cl (entry 17), and BArF (not shown). A summary of the
best results obtained from the screening of the P,N-ligands is
shown in Table 1 (entries 16–20). Although the Ir/phosphine–oxaz-
oline catalysts were highly active, the ee remained too low.
Entry S/C T (°C) NH4X
Yielda (%)
HPLC area %
(equiv)
4
6
2
7
8a,b Xb
1
2
3
4
5
6
7
8
9
200 80
500 95
200 80
500 95
500 90
200 80
200 90
200 90
200 90
500 95
200 90
AcONH4 (1)
AcONH4 (2)
AcONH4 (3)
12 1.5 30
39
0.9 13
9
12 0.1 4.3 68
11 0.1 4.3 74
18 2.8 1.4 49
27 0.2 4.3 42
2.3 10
0.6 5.8
11.0 10
HCO2NH4 (2) 14
PhCO2NH4 (2) 22
NH4-SAc (1)
6
18
55 4.2 27
1.8 0.9 6.8
NH4-SAc (1)
74 5.5 9.2 0.6 3.1 4.9
85 3.4 3.9 0.4 3.6 1.1
88 1.7 0.0 0.1 6.5 0.0
88 0.9 0.0 0.3 7.7 0.3
91 0.8 0.0 0.0 5.4 0.0
NH4-SAc (1.5)
NH4-SAc (2)
10
11
NH4-SAc (2)
NH4-SAc (3)
90
a
b
c
Assay yield of 4.
Unidentified side product.
SA = salicylate. Reaction conditions: Ru(OAc)2((R)-dm-segphos), H2 (30 bar),
EtOH (10 mL/g).
·HX
NH2
O
Ru-Cat H2
O
2
NH4X, additive
EtOH
BnO
4
Identified side-products
OH
NH2
CO2Et
CO2Et
CO2Et
BnO
BnO
BnO
BnO
6
7
2.2. Direct asymmetric reductive amination, pathway B
Next, the direct asymmetric reductive amination of b-keto ester
2 into chiral primary amine 4 was investigated (pathway B,
Scheme 2). This approach in which ammonium salts are used as
a nitrogen source was pioneered by Takasago13a using ruthe-
nium-based catalysts.
By reacting b-keto ester 2 with ammonium acetate in the pres-
ence of Ru(OAc)2((R)-DM-SEGPHOS) (see Scheme 5 for the ligand
structures) under 30 bar hydrogen pressure, only 12% of the chiral
amine 4 was formed (Table 2, entry 1). We also found several by-
products such as the hydroxyl-ester 6, the enamine ester 7, and the
elimination products 8a,b (Scheme 6).
CO2Et
8b
8a
Scheme 6. Direct reductive amination of 2 including various side products.
showed that the highest yield was obtained with DM-SEGPHOS
(not depicted). Variation of the hydrogen pressure from 10 to
50 bar revealed that the reaction accelerates as the hydrogen pres-
sure increases. Varying the substrate concentration also influenced
the product yield. The optimum substrate concentration of 0.2 g/
mL in EtOH yielded the product in 90% yield. Further fine-tuning
of the ruthenium catalyst was performed by testing two neutral
ruthenium acetate complexes with SEGPHOS (Table 3, entry 1)
and DM-SEGPHOS (entry 2) as ligand, one cationic arene ruthe-
nium DM-SEGPHOS complex (entry 3) and one anionic dimeric
complex (entry 4). The best results were obtained with the anionic
O
segphos Ar = Ph
dm-segphos Ar = 3,5-di-Me-C6H3
O
PAr2
PAr2
O
O
catalyst [NH2Me2][{RuCl((R)-DM-SEGPHOS)}2(l-Cl)3]. In all cases
Scheme 5. SEGPHOS ligands.
the enantiomeric excess was high (>97%). An additive screening
(not shown) revealed that further addition of salicylic acid acceler-
ates the reaction.
A number of other nitrogen sources were tested such as ammo-
nium iodide, ammonium citrate, and ammonium tartrate, but this
led to low yields of the desired product (<9%, not depicted). With
HCO2NH4 or PhCO2NH4, the yields increased slightly up to 14%
and 22%, respectively (entries 4 and 5). Much better results were
obtained with ammonium salicylate (entries 6–11). The amount
of ammonium salicylate had a strong impact on the yield of the
reaction. Addition of 2 equiv of this nitrogen source led to 90% as-
say yield (entry 10). After having established the best nitrogen
source, a ligand screening of SEGPHOS, DM-SEGPHOS, and BINAP
Having thus established the best nitrogen source, additive, and
Ru-catalyst we optimized the reaction conditions to further in-
crease the substrate to catalyst ratio. So far, all of the experiments
were performed at a maximum S/C of 750. Upon careful evaluation
of the effect of hydrogen pressure, temperature, and substrate con-
centration, we succeeded to find conditions where we could isolate
compound 4 in 80% yield at an S/C ratio of 5000. This is the highest
value ever obtained in a DARA. These conditions, relying on higher

1712
G. F. Busscher et al. / Tetrahedron: Asymmetry 21 (2010) 1709–1714
Table 3
Screening of Ru catalysts
O
O
Xyl2
Xyl2
Cl
Cl
Ru Cl Ru
Cl
O
O
O
O
P
P
P
P
NH2Me2
Cl
Xyl2
Xyl2
O
O
[NH2Me2][{RuCl((R)-dm-segphos}2(µ-Cl)3]
Entry
Ru-catalyst
Yielda (%)
ee (%)b
HPLC area %
4
6
2
7
8
Xc
1.0
4.7
12
1.1
1
2
3
4
Ru(OAc)2((R)-segphos)
Ru(OAc)2((R)-dm-segphos)
[RuCl(p-cymene)((R)-dm-segphos)]Cl
[NH2Me2][{RuCl((R)-dm-segphos)}2(
53
78
38
88
98
98
97
98
56
80
42
89
0.1
0.7
0.4
0.8
14
5.2
22
15
4.5
22
0.8
1.9
0.3
5.6
l
-Cl)3]
0.0
0.8
a
b
c
Assay yield of 4.
Determined by chiral HPLC using chiralcel AS-H column.
Unidentified side product. Reaction conditions: S/C = 750, ammonium salicylate (2 equiv), H2 (30 bar), EtOH (10 mL/g), 90 °C, 15 h.
pressure (50 bar H2 instead of 30 bar) and lower temperature
(85 °C instead of 90 °C), made this step sufficiently economic for
scale-up.
aryl halides.20 However, most of the arylations described in this
article were preformed with aryl iodides. Since the bromo-com-
pound is 20 times cheaper than the related iodine compound we
decided to use 4-fluoro-bromobenzene.21 The reaction of chiral
b-amino ester 4 with 4-fluoro-bromobenzene was investigated
using several different copper-sources and bases with acetylace-
tone (acac) as ligand (Table 4).22 In addition to the desired product
1b, several side products such as acid 1a and the elimination prod-
uct, cinnamic ester 8a, were formed. Although the choice of copper
source had an effect on the yield, the influence of the base was
much more important. Unacceptable amounts (21–44%) of unde-
sired elimination product were obtained when K2CO3 was used
as a base (entries 1–5). Other carbonates and amines such as
2.3. Copper-catalyzed arylation of 4
To achieve the final step, we investigated the copper-catalyzed
N-arylation of 4 to 1. Formation of the carbon–nitrogen bond by
CuI-catalyzed coupling of aryl halides with amines is one of the
typical Ullmann coupling reactions.19 In our group we have devel-
oped the use of diketone ligands for this reaction.15 A few years ago
Ma et al. published an efficient method for the synthesis of N-ary-
lated b-amino esters via a CuI-catalyzed coupling reaction with
Table 4
Variation copper source
F
NH
O
O
F
Br
+
OR
OEt
4
Cu-catalyst
Base
BnO
BnO
8a
1a R = H
1b R = Et
Entry
Copper source
Base
HPLC area %
4
1b
1a
9
1
2
3
4
5
6
7
8
9
10
11
12
13
14
CuCl
CuBr
CuI
Cu(OAc)2
Cu(acac)2
CuCl
CuCl
CuCl
CuCl
CuCl
K2CO3
K2CO3
K2CO3
K2CO3
7.3
13
25
15
48
37
40
45
62
16
3.7
—
—
7.3
13
14
12
3.8
—
37
37
21
28
33
27
17
22
20
61
82
21
—
K2CO3
0.4
Li2CO3
Na2CO3
Cs2CO3
K3PO4
57
79
7
14
36
18
33
3
—
71
65
—
—
—
Et3N
2.5
CuCl
CuCl
CuCl
CuCl
Imidazole
K-malonate
KOtBu
KOH
—
45
—
91
97
—
—
—
Reaction conditions: DMF, base 2 equiv, 15 mol % copper source, 20 mol % acac, 120 °C.

G. F. Busscher et al. / Tetrahedron: Asymmetry 21 (2010) 1709–1714
1713
Et3N and imidazole also led to the formation of considerable
amounts of the elimination product (entries 6–12). Much more
rewarding results were obtained using KOtBu and KOH, giving
exclusively the hydrolyzed product 1a (entries 13 and 14). We
were able to exclude a benzyne-type mechanism which has been
observed by us and others when using strong bases at high tem-
peratures,15,23 since in that case a mixture of 3- and 4-fluoro-aryl
products would be formed. The b-amino acid 1a is easily converted
to the corresponding ester using thionyl chloride in EtOH (87%
yield of 1b). Chiral HPLC determination of the obtained amino ester
showed no erosion of ee (97%).
lowed to cool to room temperature and subsequently added to
25.4 mL of an acetic acid/ water mixture (1/3, V/V) over 30 min.
The phases were separated and analyzed with HPLC. The THF phase
was washed with satd aq NaHCO3. The THF was partially removed
under reduced pressure. To the yellowish reaction mixture (14.3 g),
toluene (3.6 mL) and cyclohexane (62 mL) were added and the sus-
pension was heated to 80 °C. The resulting orange solution was al-
lowed to cool down to room temperature and seeded with a small
amount of the b-keto ester (0.17 g). The suspension was stirred for
an additional 2 h and filtrated. An aliquot was taken from the fil-
trate and analyzed with HPLC. The yellow residue was dried at
45 °C under reduced pressure for two hours to yield the product
(10.9 g, 36.6 mmol, 83% yield). 1H NMR (CDCl3, 300 MHz, ppm) d:
7.92–7.95 (m, 2H, arom.), 7.37–7.47 (m, 5H, arom.), 7.02–7.05
(m, 2H, arom.), 5.15 (s, 2H), 4.18–4.26 (q, 2H), 3.94 (s, 2H), 1.24–
1.27 (t, 3H). 13C NMR (CDCl3, 75 MHz, ppm) d: 191.3, 168.1,
163.5, 136.4, 131.3, 129.8, 129.1, 128.7, 127.9, 115.2, 70.6, 61.8,
46.2, 14.5.
3. Conclusions
We have demonstrated that the desired product 1b, an interme-
diate en route to Ezetimibe, can be prepared by two different
asymmetric pathways.24 For pathway A, relying on the hydrogena-
tion of enamine ester 3, good ee’s were obtained with Rh in com-
bination with chiral bisphosphines but the activity remained
relatively low. Better turnover numbers were obtained with Ir
and P,N-ligands but the enantioselectivity remained moderate
(ee of 45%). Pathway B, relying on the direct asymmetric reductive
amination (DARA), was found to be most efficient. Using Ru in
combination with DM-SEGPHOS, ammonium salicylate as ammo-
nium source, and salicylic acid as an additive, the DARA was highly
enantioselective, even at an unprecedented S/C of 5000. The b-ami-
no ester 4 was isolated in 80% yield with 97% ee. This compound
could be converted into 1a via a copper-catalyzed arylation with
4-bromo-fluorobenzene. By careful selection of the base, this reac-
tion was fine-tuned to yield 97% of the acid 1a which was con-
verted in good yield into the ester 1b upon treatment with EtOH/
SOCl2.
4.1.2. Ethyl (Z)-3-(4-fluorophenylamino)-3-(4-(benzyloxy)
phenyl) acrylate 3
At first, ethyl 3-(-4-benzyloxy)phenyl)-3-oxopropanoate (10.0 g
33.5 mmol) was dissolved in 4-fluoro-aniline (6.36 mL, 67.2 mol).
p-TsOHꢀH2O (1.29 g, 6.68 mmol) and molecular sieves 5 Å (10 g)
were added. The reaction mixture was stirred for 7 h at 100 °C until
the starting material was totally consumed. Subsequently the reac-
tion mixture was filtered, rinsed with CH2Cl2, washed with satd aq
NaHCO3, and dried with Na2SO4. 4-Fluoro-aniline was removed by
vacuum distillation (3 mm Hg, 67 °C). The residue was dissolved in
CH2Cl2, stirred with charcoal for a few minutes, and filtered off. The
solvent was evaporated and the residue was crystallized from
EtOH to give off white crystals (7.82 g, 60%). Rf 0.19 (EtOAc/n-hep-
tane, 1/5). 1H NMR (CDCl3, 300 MHz, ppm): d 10.12 (br s, 1H), 7.32–
7.14 (m, 5H, arom.), 7.13–7.09 (m, 2H, arom.), 6.78–6.61 (m, 4H,
arom.), 6.58–6.52 (m, 2H, arom.), 4.89 (br s, 2H), 4.87 (br s, 1H),
4.09 (q, J = 7.3 Hz, 2H), 1.19 (dt, J = 7.3 Hz, 3H). 13C NMR (CDCl3,
75 MHz, ppm) d: 170.2, 160.5, 159.8, 159.0, 157.2, 136.4, 129.7,
128.6, 128.1, 127.5, 124.0, 123.9, 115.5, 115.2, 114.7, 90.3, 70.0,
59.2, 14.5.
4. Experimental section
4.1. General information
Commercially available reagents and solvents were used as
delivered (reagent grade typically). Ligands were obtained from
Strem or Sigma–Aldrich. All reactions were carried out under inert
atmosphere of dry nitrogen. Standard syringe techniques were ap-
plied to the transfer of dry solvents and air- or moisture-sensitive
reagents.
Silica Gel 60 F254 pre-coated glass plates were used for analyt-
ical thin layer chromatography (Merck KGaA). Visualization was
accomplished by UV light and/or dipping the plate into an aq. solu-
tion of KMnO4 and K2CO3 followed by heating.
4.2. Preparation of the catalyst
Rhodium(I) tetrafluoroborate bis-1,5-cyclooctadiene complex25
(0.255 mmol, 104 mg) and (R,R)-1-[1-(di-tert-butylphosphi-
no)ethyl]-2-(diphenylphosphino)ferrocene (Josiphos SL-J002-1, L-
3)26 (0.255 mmol, 139 mg) were dissolved in degassed CH2Cl2
(10 mL). The reaction mixture was degassed three times and stirred
for 1 h at 40 °C. After stirring, the solvent was removed in vacuo.
Flash chromatography was performed on Silica Gel 60 (40–
63 lm, Merck KGaA). Melting points were determined with a Büchi
4.2.1. Ethyl-3-(4-fluorophenylamino)-3-(4-(benzyloxy)phenyl)
propanoate 1b
B-535 melting point apparatus. Optical rotations were determined
with a Perkin–Elmer 241 polarimeter (1 dm cell). NMR spectra
were recorded with a Bruker AMX-300 spectrometer. 1H NMR:
300 MHz, 21 °C, residual non-deuterated solvent as internal stan-
dard (CHCl3: dH 7.27 ppm, DMSO-d5: dH 2.50 ppm). 13C NMR:
75.5 MHz, 27 °C, deuterated solvent as internal standard (CDCl3:
dC 77.2 ppm, DMSO-d6: dC 39.5 ppm), DEPT experiments performed
additionally. Enantiomeric purities were determined on an Agilent
HPLC, with indicated column and solvent mixture.
Ethyl (Z)-3-(4-fluorophenylamino)-3-(4-(benzyloxy) phenyl)
acrylate (2 g, 5.1 mmol) was added to the autoclave which was
closed firmly and placed under an inert atmosphere. Next, de-
gassed TFE (50 mL) was added via the injection port. The pre-
formed catalyst (Rh/L-3) dissolved in 5 mL of degassed TFE was
added via the injection port to the reaction mixture. The reaction
mixture was allowed to react under 50 bar of H2 at 50 °C overnight.
The solvent was removed in vacuo and the crude compound was
purified by column chromatography (EtOAc/n-heptane, 1/5) to
yield the product (1) (1.69 g, 84%,) as a white solid. Rf 0.21
(EtOAc/n-heptane, 1/5). 1H NMR (CDCl3, 300 MHz, ppm) d: 7.42–
7.13 (m, 7H, arom.), 6.82 (d, J = 8.4 Hz, 2H, arom.), 6.73 (t,
J = 8.8 Hz, 2H, arom.), 6.49–6.34 (m, 2H, arom.), 4.95 (s, 2H), 4.62
(t, J = 6.9 Hz, 1H), 4.06 (q, J = 6.9, 7.3 Hz, 2H), 7.72 (d, J = 6.5 Hz,
2H), 1.12 (t, J = 6.9 Hz, 3H). 13C NMR (CDCl3, 75 MHz, ppm) d:
4.1.1. Ethyl 3-(4-(benzyloxy)phenyl)-3-oxopropanoate 2
A suspension of diethyl carbonate (12.5 g, 0.105 mol, 2.4 equiv),
sodium hydride (2.46 g, 0.0616 mol, 1.4 equiv, 60 wt % in mineral
oil), and THF (45 mL) was heated to 60 °C. To this suspension
was added p-BnO-acetophenone (10 g, 0.044 mol, dissolved in
THF (56 mL) in 1 h. Aliquots were taken from the reaction mixture
and analyzed with HPLC over time. The reaction mixture was al-

1714
G. F. Busscher et al. / Tetrahedron: Asymmetry 21 (2010) 1709–1714
3. Charlton-Menys, V.; Durrington, P. N. Exp. Physiol. 2007, 93, 27–42.
4. (a) Juaristi, E. Enantioselectivie Synthesis of b-Amino Acids; Wiley-VCH: New
York, 1997; (b) Kobayashi, S.; Ishitani, H. Chem. Rev. 1999, 99, 1069–1094; (c)
Ma, J.-A. Angew. Chem., Int. Ed. 2003, 42, 4290–4299.
171.5, 158.6, 158.1, 155.0, 143.17, 137.3, 134.4, 128.9, 128.4, 127.9,
116.12, 115.8, 115.5, 77.8, 77.6, 77.4, 77.0, 70.5, 61.2, 55.8, 43.1,
14.5. HPLC enantiomeric purity of 74% ee (AD-H column, 1.0 mL/
5. For a review, see: Drexler, H. J.; You, J.; Zhang, S.; Fischer, C.; Baumann, W.;
Spannenburg, A.; Heller, D. Org. Process Res. Dev. 2003, 7, 355.
min, 2-propanol/n-heptane, (1/9), 5 lL): retention times (min):
18.4 (major) and 23.6 (minor).
6. Tertiary enamines: (a) Lee, N. E.; Buchwald, S. L. J. Am. Chem. Soc. 1994, 116,
5985–5986; (b) Tararov, V. I.; Kadyrov, R.; Riermeier, T. H.; Holz, J.; Börner, A.
Tetrahedron Lett. 2000, 41, 2351–2355; (c) Hou, G.-H.; Xie, J.-H.; Wang, L.-X.;
Zhou, Q.-L. J. Am. Chem. Soc. 2006, 128, 11774–11775; (d) Cheruku, P.; Church,
T. L.; Trifonova, A.; Wartmann, T.; Andersson, P. G. Tetrahedron Lett. 2008, 49,
7290–7293; (e) Hou, G.-H.; Xie, J. H.; Yan, P.-C.; Zhou, Q.-L. J. Am. Chem. Soc.
2009, 131, 1366–1367; (f) Yan, P.-C.; Xie, J.-H.; Hou, G.-H.; Wang, L.-X.; Zhou,
Q.-L. Adv. Synth. Catal. 2009, 351, 3243–3250; (g) Baeza, A.; Pfaltz, A. Chem. Eur.
2009, 15, 2266–2269.
7. Secondary enamines: Wang, X.-B.; Wang, D.-W.; Lu, S.-M.; Yu, C.-B.; Zhou, Y.-G.
Tetrahedron: Asymmetry 2009, 20, 1040–1045.
8. (a) Enamino esters and amides: (a) Matsumura, K.; Zhang, X.; Saito, T. U.S.
Patent 7,015,348 B2, 2004.; (b) Matsumura, K.; Hori, K.; Kakizawa, T.; Saito, T.
In Proceedings of the Summer Symposium, Tokyo, July 28–29, 2005, The Japanese
4.3. Preparation of ethyl 3-amino-3-(4-(benzyloxy)phenyl)
propanoate 4 via D.A.R.A. of 2 (S/C = 5000)
A
100 mL Hastelloy autoclave was filled with [Me2NH2]
{RuCl(R)-dm-segphos)}2(
l
-Cl)3]27 (6.2 mg, 0.0033 mmol), ammo-
nium salicylate (10.4 g, 67.0 mmol), and salicylic acid (0.463 g,
3.35 mmol). The atmosphere was replaced with nitrogen (purging:
5 cycles of filling to 3 bar N2/emptying) followed by addition of
ethyl
3-(4-benzyloxy)phenyl)-3-oxopropanoate
(10.0 g,
33.5 mmol) in EtOH (50 mL). The reaction mixture was stirred for
10 h at 85 °C, with 50 bar hydrogen pressure. After cooling down
to ambient temperature and filtration, EtOH was evaporated and
the crude reaction mixture was stirred with satd aq NaHCO3 for
30 min. Next, the product was extracted with toluene
(3 ꢁ 250 mL). The toluene layer was washed with satd aq Na2CO3.
Evaporation of the toluene yielded the product as a brownish solid
(8.0 g, 80%). 1H NMR (CDCl3, 300 MHz, ppm) d: 7.3–7.10 (m, 7H),
6.84 (d, J = 8.4 Hz, 2H), 4.94 (s, 2H), 4.27 (t, J = 6.9 Hz, 1H), 4.02
(q, J = 6.9, 7.3 Hz, 2H), 2.53 (d, J = 7.3 Hz, 2H), 1.13 (t, J = 6.9 Hz,
3H). HPLC: enantiopurity of 96.6% ee (AS-H column, 0.5 mL/min,
Society for Process Chemistry: Tokyo;
p 146.; (c) Saito, T.; Zhang, X.;
Matsumura, K.; Yokozawa, T.; Shimizu, H. In 19th North American Meeting,
Philadelphia, PA, May 22–27, 2005, North American Catalysis Society Home
Page. http://www.nacatsoc.org/19nam/abstracts/O_244.pdf.; (d) Hsiao, Y.;
Rivera, N. R.; Rosner, T.; Krska, S. W.; Njolito, E.; Wang, F.; Sun, Y.-K.;
Armstrong, J. D.; Grabowski, E. J. J.; Tillyer, R. D.; Spindler, F.; Malan, C. J. Am.
Chem. Soc. 2004, 126, 9918–9919; (e) Clausen, A. M.; Dziadul, B.; Cappuccio, K.
L.; Kaba, M.; Starbuck, C.; Hsiao, Y.; Dowling, T. M. Org. Process Res. Dev. 2006,
10, 723–726; (f) Kubryk, M.; Hansen, K. B. Tetrahedron: Asymmetry 2006, 17,
205–209; (g) Hansen, K. B.; Hsiao, Y.; Xu, F.; Rivera, N.; Clausen, A.; Kubryk, M.;
Krska, S.; Rosner, T.; Simmons, B.; Balsells, J.; Ikemoto, N.; Sun, Y.-K.; Spindler,
F.; Malan, C.; Grabowski, E. J. J.; Armstrong, J. D., III J. Am. Chem. Soc. 2009, 131,
8798–8804.
9. Halpern, J. Science 1982, 217, 401–407.
10. (a) Lubell, W. D.; Kitamura, M.; Noyori, R. Tetrahedron: Asymmetry 1991, 2,
543–554; For a review, see: (b) Halpern, J. In Asymmetric Synthesis; Morrison, J.
D., Ed.; Academic Press: Orlando, FL, 1985; p 41.
2-propanol/2-hexane/diethylamine = 60/40/0.1,
times (min): 12.4 (minor) and 14.0 (major).
5 lL): retention
11. (a) Moroi, T.; Sotoguchi, T.; Matsumura, K.; Takenaka, M.; Kuriyama, W.;
Murayama, T.; Nara, H.; Yokozawa, T.; Yagi, K. WO 2004/074255.; (b) Dai, Q.;
Yang, W.; Zhang, X. Org. Lett. 2005, 7, 5343–5345.
12. Hebbache, H.; Hank, Z.; Bruneau, C.; Renaud, J. L. Synthesis 2009, 15, 2627–
2633.
4.3.1. Ethyl 3-(4-(benzyloxy)phenyl)-3-(4-fluorophenylamino)
propanoate 1b
A
round-bottomed flask was charged with CuCl (1.03 g,
13. (a) Matsumura, K.; Saito, T. WO 2005/028419.; (b) Bunlaksananusorn, T.;
Rampf, F. Synlett 2005, 2682–2684; (c) Shimizu, H.; Nagasaki, I.; Matsumura,
K.; Sayo, N.; Saito, T. Acc. Chem. Res. 2007, 40, 1385–1393; For a similar
approach using aryl ketones, see: Kadyrov, R.; Riermeier, T. H. Angew. Chem.,
Int. Ed. 2003, 42, 5472–5474.
14. Steinhuebel, D.; Sun, Y.-K.; Matsumura, K.; Sayo, N.; Saito, T. J. Am. Chem. Soc.
2009, 131, 11316–11317.
15. de Lange, B.; Lambers-Verstappen, M. H.; Schmieder-van de Vondervoort, L.;
Sereinig, N.; de Rijk, R.; de Vries, A. H. M.; de Vries, J. G. Synlett 2006, 3105–
3109.
10.4 mmol) and crushed KOH (17 g, 0.31 mol). Amino ester 4
(20.0 g, 66.8 mmol in 110 mL DMF), acetylacetone (1.65 mL,
16.1 mmol), and 4-fluoro-bromobenzene (12 mL, 0.11 mol) were
successively added. The resulting mixture was heated at 120 °C un-
der N2 for 20 h. The reaction was allowed to cool down to room
temperature and 100 mL of water was added. The suspension
was filtered over Celite and extracted with pentane (4 ꢁ 150 mL).
The aqueous layer was acidified to pH 2 with HCl (1 M) and ex-
tracted again with EtOAc (4 ꢁ 150 mL). The combined organic lay-
ers were extracted with water (10 ꢁ 175 mL). After drying over
Na2SO4 and concentration in vacuo, 1a was obtained as a brown
oil (25 g, 96%, 96.6% ee).
A round-bottomed flask was charged with the amino acid 1a
(56.8 g, 155 mmol) and EtOH (277 mL). The solution was cooled to
5 °C and stirred for 30 min. Subsequently, SOCl2 (22.2 g, 186 mmol)
was added in such a way that the temperature was kept between 10
and 15 °C. The mixture was heated to 60 °C for 1.5 h and then cooled
to room temperature. To the reaction mixture was added toluene
(283 mL). The reaction mixture was cooled with an ice bath while
NaHCO3 (362 g, 9 wt %, 388 mmol) was added and stirred for one
hour. After phase separation the toluene phase was concentrated
in vacuo to yield the b-amino ester 1b (53 g, 87%, 96.6% ee). Analyt-
ical data were in agreement with the data shown above.
16. Krapcho, A. P.; Diamanti, J.; Cayen, C.; Bingham, R. Org. Synth. Coll 1973, 5, 198–
201.
17. Heller, D.; de Vries, A. H. M.; de Vries, J. G. In Handbook of Homogeneous
Hydrogenation; de Vries, J. G., Elsevier, C. J., Eds.; Wiley: Weinheim, 2007.
18. (a) Lefort, L.; Boogers, J. A. F.; de Vries, A. H. M.; de Vries, J. G. Org. Lett. 2004, 6,
1733–1736; (b) Lefort, L.; Boogers, J. A. F.; de Vries, A. H. M.; de Vries, J. G. Top.
Catal. 2006, 40, 185–191; (c) de Vries, J. G.; Lefort, L. Chem. Eur. J. 2006, 12,
4722–4734.
19. (a) Monnier, F.; Taillefer, M. Angew. Chem., Int. Ed. 2009, 48, 6954–6971; (b)
Ley, S. V.; Thomas, A. W. Angew. Chem., Int. Ed. 2003, 42, 5400–5449.
20. (a) Ma, D.; Zhang, Y.; Yao, J.; Wu, S.; Tao, T. J. Am. Chem. Soc. 1998, 120, 12459–
12467; (b) Ma, D.; Xia, C. Org. Lett. 2001, 3, 2583–2586; (c) Ma, D. W.; Cai, Q. A.
Acc. Chem. Res. 2008, 41, 1450–1460.
21. Compare: 4-bromo-fluorobenzene 214, -€/kg and 4-iodo-fluorobenzene 4660, -
€/kg according to prices obtained from Sigma Aldrich.
22. No desired product (1a and 1b) is formed when acac is omitted.
23. Shi, L.; Wang, M.; Fan, C.-A.; Zhang, F.-M.; Tu, Y.-Q. Org. Lett. 2003, 5, 3515–
3517.
24. Busscher, G. F.; de Vries, J. G.; Lefort, L. WO2010006954 to DSM IP Assets BV.
25. CAS number: [35138-22-8].
26. CAS number: [155830-69-6].
27. MDL
number:
MFCD09753038,
dimethylammonium
dichlorotri(l-
References
chloro)bis[(R)-(+)-5,50-bis[di(3,5-xylyl)phosphino]-4,40-bi-1,3-
benzodioxole]diruthenate(II). Ligand obtained from Takasago. JP Registration
No. 3148136, 3549390, SEGPHOS is a registered trademark of Takasago.
1. Rosenblum, S. B.; Huynh, T.; Afonso, A.; Davis, H. R.; Yumibe, N.; Clader, J. W.;
Burnett, D. A. J. Med. Chem. 1998, 41, 973–980.
2. Wu, G.; Wong, Y.; Cheng, X.; Ding, Z. J. Org. Chem. 1999, 64, 3714–3718.