Palladium-catalysed enantioselective conjugate addition of aromatic amines to α,β-unsaturated N-imides. Effect of the chelating moiety

Article abstract of DOI:10.1002/adsc.200505126

Palladium-catalysed enantioselective additions of aromatic amines to α,β-unsaturated N-imides 4-6 are reported. The N-substituent of the Michael acceptor appears to have an unusual modulating effect on the amine activity. The effects of introducing differ

Full text of DOI:10.1002/adsc.200505126

FULL PAPERS
DOI: 10.1002/adsc.200505126
Palladium-Catalysed Enantioselective Conjugate Addition of
Aromatic Amines to a,b-Unsaturated N-Imides. Effect of the
Chelating Moiety
Pim Huat Phua,a Johannes G. de Vries,b King Kuok (Mimi) Hiia,*
a
Department of Chemistry, Imperial College London, Exhibition Road, South Kensington, London SW7 2AZ, U. K.
E-mail: mimi.hii@imperial.ac.uk
DSM Research-Life Sciences, Advanced Synthesis, Catalysis and Development, P. O. Box 18, 6160 MD Geleen, The
b
Netherlands
E-mail: Hans-JG.Vries-de@dsm.com
Received: March 24, 2005; Accepted: May3, 2005; Published online: September 26, 2005
Dedicated to John M. Brown, FRS on the occasion of his 65th birthday.
Supporting Information for this article is available on the WWW under http://asc.wiley-vch.de/home/.
Abstract: Palladium-catalysed enantioselective addi- and selectivityof the system appear to be less sensi-
tions of aromatic amines to a,b-unsaturated N-imides tive to environmental effects than previous systems.
4–6 are reported. The N-substituent of the Michael The importance of the counteranion is revealed.
acceptor appears to have an unusual modulating ef-
fect on the amine activity. The effects of introducing Keywords: asymmetric catalysis; conjugate addition;
different substituents are examined. Overall, the yield hydroamination; palladium; substituent effects
Introduction
Intrigued bythis dramatic change in reactivitybe-
tween the cyclic and acyclic Michael acceptors, we initi-
Previously, we reported the dicationic diphosphinepal- ated a studyto examine other olefin substrates contain-
ladium(II)
bis(trifluoromethanesulfonate)
[(BINAP)Pd(solvate)2]2þ[TfO]À2 (1) as an efficient cat-
alyst for the conjugate addition of aromatic amines to al-
kenoyl-N-oxazolidinones 2 and N-carbamate esters 3
with high enantioselectivity(Scheme 1). [1]
Both substrates 2 and 3 contain 1,3-dicarbonyl groups.
We postulated that theyare able to form templates with
the Lewis-acidic palladium metal centre, predisposing
the attack of the nucleophile stereoselectivelyat the b-
position (Fig. 1). In the absence of these chelating moi-
eties, the addition of aniline to methyl crotonate pro-
ceeded in just 24% ee under the same reaction condi-
tions.[1d]
Given that the reaction conditions are identical, the
disparitybetween the reactivityof the two substrates is
striking, and appears to be dependent on the nature of
the chelating moiety. For example, the addition of aro-
matic amines to alkenoyl-N-oxazolidinones 2a requires
a relativelyhigh catalytic loading (10 mol %) to achieve
93% yield in 18 h (93% ee). In contrast, amine addition
to the analogous substrate 3a is much more facile, only
2–5 mol % of the catalystwasrequired to afford compa-
rable yields.
Scheme 1. Enantioselective addition of aromatic amines to
a,b-unsaturated N-oxazolidinones 2 and N-carbamates 3 cat-
alysed by 1.
Adv. Synth. Catal. 2005, 347, 1775 – 1780
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1775

Pim Huat Phua et al.
FULL PAPERS
Figure 1. Chelation-assisted stereoselectivity.
Scheme 2. Preparation of N-alkenoyl N-imides 4a – c and 5a:
(i) heat; (ii) P(OEt)3, heat; (iii) R1CHO, DBU, THF.
Catalytic Conjugate Addition Reactions
Effect of Solvents (Table 1)
The reaction between aniline and crotonyl N-imide 4a
was performed initiallyin five different reaction media
(Table 1).[7] As was observed in previous systems, the
Figure 2. a,b-Unsaturated N-imides.
ing a structural motif containing non-cyclic 1,3-chelating non-coordinating toluene offered the highest yield of
carbonyls. Previously, a,b-unsaturated N-imides have the product with the best stereoselectivity(Table 1, en-
been employed as Michael acceptors in enantioselective try1), whereas polar, more coordinating, solvents such
addition of a remarkable range of nucleophiles, includ- as acetonitrile and methanol were detrimental to yield
ing hydrazoic acid,[2] as well as cyanide,[3] stabilised car- and selectivity(entries 4 and 5). However, unlike previ-
banions[4] and oximes,[5] catalysed by Salen-Al com- ous systems,[1c–d] dichloromethane and THF were well-
plexes. In this paper, the palladium-catalysed conjugate tolerated (entries 2 and 3), thus suggesting that the che-
addition of aromatic amines to alkenoyl N-imide sub- late ring obtained with the N-imide (Fig. 1) is more sta-
strates, 4–6 (Fig. 2), will be examined. The result will ble towards solvents of weak coordinating ability.
be compared with those obtained in the previous studies
(Scheme 1). Additionally, substituent (Y) at the para-
position of the benzoyl group was introduced, to probe Rates of Addition of Different Aromatic Amines (Ta-
the electronic effects exerted bythe terminal carbonyl
group.
ble 2)
Previously, the relative rates of amine addition to theN-
Boc substrate 3a were found to be in the order: 4-
chloroaniline>aniline>4-anisidine (unusuallyde-
creasing in the order of increasing nucleophilicity).[1c]
The same experiments were replicated for the addition
Results and Discussion
Synthesis of N-Imide Substrates
The N-imide substrates maybe prepared in multigram
quantities in three steps byadopting published proce-
dures (Scheme 2).[6] Following acyl substitution of chlor-
oacetyl chloride with the corresponding benzamides,
Table 1. Effect of solvents in the addition of aniline to 4a
(Scheme 3).[a]
[b]
EntrySolvent
Conversion
[%]
ee[c] [%]
À
Arbuzov, and Horner–Wadsworth Emmons (HWE)
1
2
3
4
5
toluene
84
82
86
49
47
89
81
84
39
28
reactions were performed to furnish imides 4a–c, and
5a. Due to the low nucleophilicityof 4-chlorobenza-
mide, a different procedure was adopted for the syn-
thesis of 6a, obtained from the reaction between croton-
yl chloride and deprotonated 4-chlorobenzamide.
dichloromethane
tetrahydrofuran
acetonitrile
methanol
[a]
Reaction conditions: catalyst/amine/N-imide 4a in a ratio
of 0.05/1.0/1.1, in the chosen solvent, at room temperature,
18 h.
Calculated by H NMR integration.
Determined bychiral HPLC.
[b]
[c]
1
1776
asc.wiley-vch.de
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2005, 347, 1775 – 1780

Addition of Aromatic Amines to a,b-Unsaturated N-Imides
FULL PAPERS
tion to substrate 3a–3c, ee values varied considerably
between 16 to>99% ee.[1c] In contrast, the enantioselec-
tivityof the present system appeared to be relativelyin-
sensitive to the electronic structure of the amine, rang-
ing between 73–89% for the addition to 4a (Table 3, en-
tries 1–4), 51–63% to 4b (entries 5–8) and 55–70% to
4c (entries 9–12). Also noticeable are the comparatively
low ees obtained from the addition of the electron-defi-
cient 4-chloroaniline. In each case, it is substantiallyin-
ferior compared to the addition of the electronically
ꢁneutralꢂ aniline and 4-toluidine (entries 1 vs. 2/3, 5 vs.
6/7 and 9 vs. 10/11). Nonetheless, the present system of-
fers an improvement for the addition of electron-rich ar-
omatic amines. In the previous system, the addition of 4-
anisidine to substrates 3 was particularlynon-selective;
when R1 ¼Me (a), Et (b) and n-Pr (c), ee values of 73,
16 and 17% were obtained, respectively. In comparison,
the addition to 4a–c proceeded with ees of 80, 51 and
55%, respectively(entries 4, 8 and 12).
Scheme 3. Conjugate addition of aromatic amines to 4a.
Table 2. Relative rates of addition of substituted anilines to
4a (Scheme 3).[a]
[b]
EntryTime [h] Conversion
[%]
4-ClC6H4NH2 PhNH2 4-MeOC6H4NH2
1
2
3
4
5
0.5
1
4
8
24
71
74
81
83
86
74
76
80
84
87
39
52
73
77
80
[a]
Reaction monitored by 1H NMR. Catalyst/amine/N-imide 4a
in a ratio of 0.05/1.0/1.1, in toluene-d8 at room temperature.
Calculated byrelative integrals of resonance signals of
substrate and product.
As expected, the R1 substituent has a dramatic effect
on the rate of catalytic turnover. Additions to pentenoyl
(4b) and hexenoyl (4c) substrates were slower, requiring
a longer reaction time than the equivalent addition to
[b]
ofthe aminesto 4a in toluene-d8, using1H NMR spectros- the 4a. Once again, the yields of these additions are far
copyto monitor the progress of the reaction (Table 2).
Interestingly, a different trend was obtained. Although a given Michael acceptor (entries 1–4, 5–8 and 9–12).
the addition of the most electron-rich anisidine re- Next, the electronic effect exerted bythe terminal car-
mained sluggish, the relative rates of addition of the oth- bonyl group was examined by studying the addition to
less sensitive to the electronic nature of the amine for
er two were found to be virtuallythe same! The nature of
the N-substituent evidentlyhas a modulating effect on
the reactivityof the amines, although the origin of this
is not clear at this stage.
butenoyl N-imides 5a and 6a, where electron-donating
(OMe) or withdrawing (Cl) groups were introduced at
the para-position of the benzoyl unit (Table 4).
Remarkably, neither the yield nor enantioselectivity
of the reactions are much affected bythe para-substitu-
ent of the benzoyl group. For the addition of 4-chloro-
aniline, decreasing the electron-densityof the N-benzo-
yl moiety led to a slight enhancement of the enantiomer-
Substituent Effects (Table 3)
The homologation of the b-substituent was examined in ic excess (entries 1–3). Apart from this, other additions
the next part of our investigation. In the analogous addi- of aniline, p-toluidine and p-anisidine appeared to be in-
Table 3. Addition of ArNH2 to 4a–c (Scheme 3).[a]
EntryY
Ar
Product
t [h]
Conversion[b] [%]
ee[c] [%]
1
2
3
4
5
6
7
8
9
Me
Me
Me
Me
Et
Et
Et
Et
Pr
4-ClC6H4
Ph
4-MeC6H4
4-MeOC6H4
4-ClC6H4
Ph
4-MeC6H4
4-MeOC6H4
4-ClC6H4
Ph
4-MeC6H4
4-MeOC6H4
9a
9b
9c
9d
9e
9f
9 g
9 h
9i
18
18
18
18
72
72
72
72
72
72
72
72
88 (82)
84 (76)
81 (80)
70 (64)
70 (68)
75 (74)
70 (54)
76 (70)
78 (75)
69 (65)
71 (69)
69 (62)
73
89
83
80
54
63
59
51
58
67
70
55
10
11
12
Pr
Pr
Pr
9j
9k
9 l
[a]
Reaction conditions: catalyst/amine/N-imide 4a in a ratio of 0.05/1.0/1.1, in toluene at room temperature (25 0C).
Calculated by H NMR integration. Value in parenthesis denotes isolated yield after column chromatography.
Determined bychiral HPLC.
[b]
[c]
1
Adv. Synth. Catal. 2005, 347, 1775 – 1780
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1777

Pim Huat Phua et al.
[%]
ee[c] [%]
FULL PAPERS
Table 4. Addition of ArNH2 to 4a, 5a and 6a (Scheme 3).[a]
[b]
EntryY
Ar
Product
Conversion
1
2
3
4
5
6
7
8
9
H (4a)
OMe (5a)
Cl (6a)
H (4a)
OMe (5a)
Cl (6a)
H (4a)
OMe (5a)
Cl (6a)
H (4a)
4-ClC6H4
4-ClC6H4
4-ClC6H4
Ph
Ph
Ph
4-MeC6H4
4-MeC6H4
4-MeC6H4
4-MeOC6H4
4-MeOC6H4
4-MeOC6H4
9a
9 m
9n
9b
9o
9p
9c
9q
9r
9d
9 s
9 t
88 (82)
89 (83)
85 (80)
81 (76)
89 (85)
83 (75)
81 (80)
86 (85)
87 (85)
70 (64)
82 (77)
85 (82)
73
67
78
89
89
89
83
78
83
80
83
80
10
11
12
OMe (5a)
Cl (6a)
[a]
As described in footnote[a] to Table 3.
Calculated by H NMR integration. Value in parenthesis denotes isolated yield after purification by column chromatogra-
phy.
[b]
[c]
1
Determined bychiral HPLC.
sensitive to the N-benzoyl substituent (entries 4–6, 7–9 The authors reasoned that this suppresses uncatalysed
and 10–12). Previously, we also observed little differ- addition of the aromatic amine, and prevents catalyst
ence between the addition of aniline to tert-butyl carba- deactivation byamine coordination. Thus, the reaction
mate 3a (where R¼t-Bu, Scheme 1) and its methyl ana- between 4a and the trifluoromethanesulfonate salt of
logue (R¼Me).[1c] Thus, the enantioselectivityappears
p-anisidine was carried out. In our case, the use of the
to be more dependent on the electronic nature of the amine salt led to a decrease in the yield and selectivity
amine substrate and the homologyof the olefin substitu-
ent (R1), with the N-substituent exerting a more global
effect.
(Scheme 5).[9]
Assignment of Absolute Configuration
The product 9a was converted to the corresponding N-
phenyl-substituted b-alanine 10 byhydrolysis with aque-
ous potassium hydroxide (Scheme 4). The (S)-(þ)-con-
figuration was thus established bycomparison of the op-
Scheme 5. Addition of trifluoromethanesulfonate salts.
In the concluding part of this investigation, we exam-
[1c]
tical rotation with previouslyestablished values.
In ined the effect of the counteranion. We attempted to
this instance, the sense of stereoinduction is unchanged prepare the tetrafluoroborate salt of complex 1 via the
bya change in the N-substituent.
commonlyadopted halide abstraction route. The addi-
tion of silver(I) tetrafluoroborate to a solution of [R-
(BINAP)PdCl2] led to the formation of a mixture of
compounds (31P NMR), including the expected mono-
meric complex 11, and a m-hydroxy dimeric palladiu-
m(II) complex 12 (Scheme 6), characterised bya distinct
1
singlet H resonance at ca. À3 ppm (corresponding to
the bridging OH groups). Treating the mixture of com-
pounds with molecular sieve 4 Š in dichloromethane
for 3 days shifts the equilibrium to the dimeric com-
pound 12.[10] Conversely, by employing acetone/water
during the halide abstraction reaction, the monomeric
palladium species 11 maybe isolated. [11]
Scheme 4. Establishing the absolute configuration of 10.
Additives and Counteranion
Following our initial work on the conjugate addition of
When the isolated complexes were used as catalysts in
aromatic amines to N-oxazolidinones,[1b] Sodeoka the addition of aniline to 4a, lower yields and ee values
et al. reported that the enantioselectivityof these addi-
tions maybe enhanced byemploying the trifluorome-
were obtained: 86% (75% ee) and 31% (36% ee) for
11 and 12, respectively. The dimeric species appears to
thanesulfonate salt of the amine as the Michael donor.[8] be much less catalytically active than the monomeric
1778
asc.wiley-vch.de
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2005, 347, 1775 – 1780

Addition of Aromatic Amines to a,b-Unsaturated N-Imides
FULL PAPERS
autoinjector with a 20 mL loop); detection was effected by
UV absorption at 254 nm. Elemental analyses were carried
out byElemental Analysis Service, Universityof North Lon-
don. Mass spectra (MS) were recorded on a Micromass Auto-
spec-Q Mass Spectrometer using EI techniques. All dried sol-
vents were purchased from Sigma Aldrich and stored under a
nitrogen atmosphere. All commercial reagents were used as re-
ceived without further purification.
Scheme 6. Equilibrium mixture between monomeric and di-
meric palladium(II) complexes.
Preparations of catalyst 1,[1a] 4a–c,[6] 11 and 12[10] have been
previouslydescribed.
complex. We speculate that the tetrafluoroborate salt of
monomeric complex 11 undergoes dynamic exchange
with the dimeric structure in solution – the small amount
of HBF4 released during this interchange catalyses a
competitive racemic process, which leads to the erosion
of the enantioselectivity. Thus, the trifluoromethanesul-
fonate counteranion plays an important role by stabilis-
ing the monomeric structure (possiblythrough hydro-
gen bonding), which leads to enhanced catalyst per-
formance.
4-Methoxyl-N-(but-2-enoyl)benzamide (5a)
Prepared bythe procedure described previously. [6] White solid;
yield: 1.53 g (70%); anal. found: C 65.66, H 5.95, N 6.26;
C12H13NO3 requires: C 65.74, H 5.98, N 6.39%; Rf ¼0.45
(EtOAc/hexanes, 4/1); mp 168.0–168.38C; IR (KBr): nmax
¼
1
3290 (NH), 1708 and 1673 cmÀ1 (C O); H NMR (270 MHz,
CDCl3): d¼8.88 (1H, s, NH), 7.88 (2H, d, J¼6.9 Hz, Ar),
7.25 (1H, dq, J¼15.3 and 6.9 Hz, CH3CH), 7.15 (2H, d, J¼
¼
¼
6.9 Hz, Ar), 6.95 (1H, d, J¼15.3 Hz, CH CH), 3.86 (3H, s,
OCH3), 1.97 (3H, d, J¼6.9 Hz, CH3); 13C NMR (67.5 MHz,
¼
¼
CDCl3): d¼167.7 (NC O), 165.2 (ArC O), 163.6 (Ar), 146.5
Conclusion
¼
¼
(CH3CH CH), 130.0 (Ar), 125.1 (Ar), 124.4 (CH3CH CH),
114.1 (Ar), 55.5 (OCH3), 18.5 (CH3); MS (EI): m/z¼220/219
(Mþ, 5/36%), 135 (100), 107 (8), 77 (15), 69 (28).
The enantioselective addition of primaryaromatic
amines to a,b-unsaturated N-imides, catalysed by the di-
cationic palladium trifluoromethanesulfonate complex
1, has been examined under a number of reaction condi-
tions. In certain cases, significant improvements were re-
ported. These studies reveal important factors govern-
ing reactivityand selectivityin these reactions. The im-
portant role of the counteranion was also revealed.
The reaction mechanisms of these reactions will be fur-
ther elucidated bykinetic and spectroscopic techniques,
and reported in due course.
N-(But-2-enoyl)-4-chlorobenzamide (6a)
To a stirred solution of 4-chlorobenzamide (1.5 g, 10 mmol) in
dryTHF (10 mL) at À208C was added EtMgBr (1 M in THF
solution, 10 mL, 10 mmol) slowly. After stirring for 1 h, trans-
crotonyl chloride (1.4 mL, 14 mmol) was added. The resulting
mixture was stirred for 2 h, then warmed up to room tempera-
ture. Stirring was continued for 1 h, before the reaction was
quenched bythe addition of saturated aqueous NH 4Cl
(10 mL). The layers were separated, and the aqueous layer
was extracted with ethyl acetate (3ꢀ10 mL). The combined or-
ganic layers were washed with water and brine, dried over Na2
SO4, and evaporated. The crude product was subject to column
Experimental Section
chromatography, then recrystallised from CHCl-hexane to
3
give 6a as an analytically pure white solid; yield: 0.6 g (27%);
mp 129.8–130.0 0C (CHCl3-hexane); Anal. found: C 59.05, H
4.50, N 6.26; C11H10ClNO2 requires: C 59.07, H 4.51, N
6.26%; Rf ¼0.55 (EtOAc/hexanes, 1/1); IR (KBr): nmax ¼3272
General Remarks
All manipulations involving air- and moisture-sensitive orga-
nometallic reagents were performed under a drynitrogen at-
mosphere using standard Schlenk techniques. Column chro-
matographywas performed on silica gel (Kieselgel 60, 63–
(NH), 1708 and 1670 cmÀ1 (C O); 1H NMR (270 MHz,
¼
CDCl3): d¼9.17 (1H, s, NH), 7.87 (2H, d, J¼6.7 Hz, Ar),
1
200 mm). H, 13C and 31P NMR spectra were obtained on a
7.46 (2H, d, J¼6.7 Hz, Ar), 7.39 (1H, dq, J¼8.6 and 5.2 Hz,
Jeol EX-270 instrument (1H at 270 MHz, 31P at 109.3 MHz
and 13C at 67.5 MHz). The chemical shifts are reported in d
(ppm), reference to residual protons and 13C signals of
CDCl3. 31P spectra were referenced to H3PO4. The coupling
constants are in Hertz (J Hz). Infrared spectra were recorded
on a Mattson Instrument Satellite FTIR spectrometer; the
samples were prepared either in Nujol mull or as liquid film be-
tween NaCl plates, or pressed into KBr discs. Optical rotation
values were measured on a Perkin Elmer 343 polarimeter using
a 10 cm solution cell. Concentration of the samples is given in
g/mL. Melting points (uncorrected) were determined on an
Electrothermal Gallenkamp apparatus. Chiral HPLC analyses
were performed on a Gilson HPLC system equipped with an
¼
CH3CH), 7.23 (1H, d, J¼8.6 Hz, CH CH), 1.99 (3H, d, J¼
5.2 Hz, CH3); 13C NMR (67.5 MHz, CDCl3): d¼167.7
¼
¼
¼
(NC O), 165.1 (PhC O), 147.4 (CH3CH CH), 139.6 (Ar),
131.4 (Ar), 129.4 (Ar), 124.1 (Ar), 18.6 (CH3); MS (EI):
m/z¼225/223 (Mþ, 13/36), 141 (34), 139 (100), 111 (35), 75
(25), 69 (100).
Catalytic Reactions Conducted in Parallel using a
Radleyꢀs 12-Position Reaction Carousel
For the typical catalytic experiment, each reaction tube was
loaded with the catalyst, a stirrer bar and fitted with a screw
Adv. Synth. Catal. 2005, 347, 1775 – 1780
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1779

Pim Huat Phua et al.
FULL PAPERS
cap (with a nitrogen inlet and rubber septum). The reaction
tubes were placed in the carousel and purged and filled with ni-
trogen. The solvent, N-imide and amine were then added suc-
cessivelyto each tube. The reaction mixtures were magnetical-
lystirred, while the temperature of the heating block was main-
tained at 25 0C bymeans of a thermostat. The progress of the
reaction was monitored byextracting the aliquots of reaction
mixture periodically via the rubber septum and recording the
corresponding 1H NMR spectra. All reactions were duplicated
to within ꢁ3% accuracy. After the prescribed period of time,
the solvent was evaporated and the residue subjected to col-
umn chromatography.
plc through a precious metal loan scheme. We thank Professor
Paul Pregosin (ETH Zurich, Switzerland) for valuable discus-
sions on counteranion effects.
References and Notes
[1] a) K. Li, P. N. Horton, M. B. Hursthouse, K. K. Hii, J. Or-
ganometallic Chem. 2003, 65, 250–258; b) K. Li, K. K.
Hii, Chem. Commun. 2000, 1132–1133; c) K. Li, X.
Cheng, K. K. Hii, Eur. J. Org. Chem. 2004, 959–964;
d) P. H. Phua, K. Li, K. K. Hii, Tetrahedron, 2005, in
press.
[2] J. K. Myers, E. N. Jacobsen, J. Am. Chem. Soc. 1999, 121,
8959–8960.
b-Amino N-Imides 9
[3] a) G. M. Sammis, E. N. Jacobsen, J. Am. Chem. Soc.
2003, 125, 4441–4443; b) G. M. Sammis, H. Danjo,
E. N. Jacobsen, J. Am. Chem. Soc. 2004, 126, 9928–9929.
[4] M. S. Taylor, E. N. Jacobsen, J. Am. Chem. Soc. 2003,
125, 11204–11205.
[5] C. D. Vanderwall, E. N. Jacobsen, J. Am. Chem. Soc.
2004, 126, 14724–14725.
[6] S. N. Goodman, E. N. Jacobsen, Adv. Synth. Catal. 2002,
344, 953–956.
[7] Although the solubilityof complex 1 in toluene is limit-
ed, the addition of aniline afforded a homogeneous yel-
low solution, which graduallyturned a darker orange-red
solution over the course of the reaction (18 h). The reac-
tion mixtures in all of the other solvents were homogene-
ous from the onset.
[8] Y. Hamashima, H. Somei, Y. Shimura, T. Tamura, M. So-
deoka, Org. Lett. 2004, 6, 1861–1864.
[9] No product formation was observed in the absence of the
catalyst.
The characterization data for 9a–9 t can be found in the Sup-
porting Information.
Hydrolysis of Compound 9b
Compound (À)-9b (0.10 g, 0.35 mmol, 89% ee) was dissolved
in 1 N KOH in MeOH (2.5 equivs.) at room temperature for
1 h, after which the solvents were evaporated. The residue
was dissolved in water, and washed with portions of Et2O.
The acidityof the aqueous phase was adjusted carefullyto
pH 4 bythe addition of 1 N HCl, before extracting with EtOAc
(4ꢀ10 mL). The combined organic layers were washed with
brine, dried over MgSO4 and evaporated to afford the b-amino
1
acid 10; yield: 64 mg (100%); [a]2D5: þ14.68 (c 1, CHCl3); H
NMR (270 MHz, CDCl3): d¼7.21 (2H, t, J¼7.4 Hz, Ph), 6.78
(1H, t, J¼7.4 Hz, Ph), 6.68 (2H, d, J¼7.4 Hz, Ph), 6.30 (2H,
br s, NH and COOH), 3.89–3.98 (1H, m, CH), 2.66 (1H, dd,
J¼15.4 and 5.9 Hz, CH2), 2.51 (1H, dd, J¼15.4, 6.4 Hz,
CH2), 1.29 (3H, d, J¼6.4 Hz, CH3); 13C NMR (67.5 MHz,
CDCl3): d¼176.6 (COOH), 146.1 (Ar), 129.4 (Ar), 118.7
(Ar), 114.5 (Ar), 46.5 (CH), 40.6 (CH2), 20.1 (CH3).
[10] A. Fujii, E. Hagiwara, M. Sodeoka, J. Am. Chem. Soc.
1999, 121, 5450–5458.
[11] In the absence of acetonitrile, the product is plagued
with the presence of the dimeric compound. The mono-
meric complex mayalso be obtained bythe addition of
HBF4 to the equilibrating mixture. However, the com-
plex prepared in this manner is not used in the catalytic
reactions as residual acid could catalyse the Michael ad-
dition.
Acknowledgements
The authors are grateful to the Committee of Vice-Chancellors
and Principals (CVCP) for an ORSA studentship (PHP), as
well as Imperial College London and DSM for additional finan-
cial support. Palladium salts were provided by Johnson Matthey
1780
asc.wiley-vch.de
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2005, 347, 1775 – 1780