Evidence That Protons Can Be the Active Catalysts in Lewis Acid Mediated Hetero-Michael Addition Reactions

Article abstract of DOI:10.1002/chem.200305407

The mechanism of Lewis acid catalysed hetero-Michael addition reactions of weakly basic nucleophiles to α,β-unsaturated ketones was investigated. Protons, rather than metal ions, were identified as the active catalysts. Other mechanisms have been ruled out by analyses of side products and of stoichiometric enone-catalyst mixtures and by the use of radical inhibitors. No evidence for the involvement of π-olefin-metal complexes or for carbonyl-metal-ion interactions was obtained. The reactions did not proceed in the presence of the non-coordinating base 2,6-di-tert-butylpyridine. An excellent correlation of catalytic activities with cation hydrolysis constants was obtained. Different reactivities of mono- and dicarbonyl substrates have been rationalised. A ¹H NMR probe for the assessment of proton generation was established and Lewis acids have been classified according to their propensity to hydrolyse in organic solvents. Bronsted acid-catalysed conjugate addition reactions of nitrogen, oxygen, sulfur and carbon nucleophiles are developed and implications for asymmetric Lewis acid catalysis are discussed.

Full text of DOI:10.1002/chem.200305407

FULL PAPER
Evidence That Protons Can Be the Active Catalysts in Lewis Acid Mediated
Hetero-Michael Addition Reactions
Tobias C. Wabnitz, Jin-Quan Yu, and Jonathan B. Spencer*[a]
Abstract: The mechanism of Lewis
acid catalysed hetero-Michael addition
reactions of weakly basic nucleophiles
to a,b-unsaturated ketones was investi-
gated. Protons, rather than metal ions,
were identified as the active catalysts.
Other mechanisms have been ruled out
by analyses of side products and of
stoichiometric enone catalyst mixtures
and by the use of radical inhibitors. No
evidence for the involvement of p-
olefin metal complexes or for carbon-
yl metal-ion interactions was obtained.
The reactions did not proceed in the
presence of the non-coordinating base
2,6-di-tert-butylpyridine. An excellent
correlation of catalytic activities with
cation hydrolysis constants was ob-
tained. Different reactivities of mono-
and dicarbonyl substrates have been
rationalised. A H NMR probe for the
1
assessment of proton generation was
established and Lewis acids have been
classified according to their propensity
to hydrolyse in organic solvents.
Br˘nsted acid-catalysed conjugate ad-
dition reactions of nitrogen, oxygen,
sulfur and carbon nucleophiles are de-
veloped and implications for asymmet-
ric Lewis acid catalysis are discussed.
Keywords: Br˘nsted
acids
¥
enones ¥ hydrolysis ¥ Lewis acids ¥
Michael addition
Introduction
allow hetero-Michael addition reactions to proceed under
much milder conditions.
Michael and hetero-Michael addition reactions are amongst
the most important bond forming strategies for both
carbon carbon and carbon heteroatom bonds in organic
chemistry. The first reports of hetero-Michael addition reac-
tions date back to 1874 and 1878, when Sokoloff and Lat-
schinoff observed the addition of ammonia to mesityl oxide
and Loydl described the reaction of sodium hydroxide with
fumaric acid;[1] even before the discovery of the actual Mi-
chael addition of carbon nucleophiles by Komnenos in 1883
and the studies of Michael in 1887.[2]
Normally, either the donor or the acceptor component
need to be activated in hetero-Michael addition reactions.
The classical method to achieve this has been deprotonation
of the nucleophile with strong bases. In order to perform
the reaction under conditions more compatible with other
functional groups, alternative methodologies have been de-
veloped. Important advances have been made with Lewis
acid catalysts, which activate the acceptor components and
Due to the importance of b-amino carbonyl moieties in
natural product chemistry and in drug development, synthe-
ses of these compounds by aza-Michael addition reactions
have received considerable attention.[3] Lewis acid catalysts
based on lanthanide triflates, FeCl3, InCl3, CeCl3/NaI, and
platinum group metals have been used and asymmetric ver-
sions with chiral complexes of main-group and transition-
metal cations have been reported.[4,5] Recently, significant
advances in metal-salt-catalysed, intermolecular aza-Michael
addition reactions of weakly basic nitrogen nucleophiles
such as carbamates, oxazolidinones and oximes to a,b-unsat-
urated ketones were reported by Kobayashi et al., Banik
and Srivastava and by our group. [Pd(CH3CN)2Cl2],
[Pd(CH3CN)4](BF4)2,[6] Cu(OTf)2,[7] Bi(NO3)3,[8] and noble
metal chlorides[9] such as RhCl3¥3H2O, ReCl5, PtCl4¥5H2O,
and AuCl were found to be efficient catalysts for these
transformations under very mild conditions. However, con-
clusive evidence for the reaction mechanism has not been
presented.
Lewis acid catalysed hetero-Michael addition reactions of
alcohols and thiols under nonbasic conditions have also
been reported. Achiral oxa-Michael addition reactions
mediated by PdII and ZnII have been described,[10,11] and ad-
dition reactions of sulfur nucleophiles have been achieved
[a] Dipl.-Chem. T. C. Wabnitz, Dr. J.-Q. Yu, Dr. J. B. Spencer
Cambridge University Chemical Laboratory
Lensfield Road, Cambridge CB2 1EW (UK)
Fax : (+44)1223-336362
[12]
[8]
both with achiral catalysts such as InBr3 and Bi(NO3)3
E-mail: jbs20@cam.ac.uk
and chiral transition-metal complexes.[13]
Supporting information for this article is available on the WWW
under http://www.chemeurj.org or from the author.
484
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/chem.200305407
Chem. Eur. J. 2004, 10, 484 493

484 493
Table 1. Aza-Michael addition reactions in the presence of radical inhibi-
tors.
In two very recent communications we described simple
protocols for the hetero-Michael addition reactions of nitro-
gen, oxygen and sulfur nucleophiles mediated by homoge-
neous[14] or heterogeneous[15] protic acid catalysts. These
methods were very general and a wide variety of nucleo-
philes and acceptors could be used.
We were intrigued that catalysts ranging from
[Pd(CH3CN)2Cl2] over noble metal chlorides and Bi(NO3)3
to strong Br˘nsted acids could be used for very similar reac-
tions and we suspected common mechanistic principles for
all transformations. Herein, we report the first comprehen-
sive investigation of the reaction mechanism of catalytic
hetero-Michael addition reactions with weakly basic nucleo-
philes.
Run
Catalyst
Solvent
Inhibitor
t[h]
Yield[%]
1
2
3
4
5
6
7
[Pd(CH3CN)4](BF4)2
[Pd(CH3CN)4](BF4)2
In(OTf)3
In(OTf)3
Cu(OTf)2
CH2Cl2
CH2Cl2
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
7
7
97
93
97
94
99
96
94
4a
4b
0.5
0.5
3
0.25
0.25
Results and Discussion
Cu(OTf)2
Cu(OTf)2
4a
4b
Four principal mechanisms can be envisaged for the mode
of catalyst action in conjugate addition reactions to enones
under nonbasic conditions. Lewis acid mediated activation
of the enone towards nucleophilic addition can occur by for-
mation of carbonyl metal-ion complexes such as 1a. Similar-
Scheme 1. p-Activation of enones by PdII.
The crucial step in this sequence is the protonolytic cleav-
age of the palladium carbon bond. This step is known to be
accelerated by phosphine ligands on palladium, the presence
of excess halide ions or weak acids.[19] However, the aza-Mi-
chael reaction of the carbamate 3 with the enone 2b, which
can be catalysed by [Pd(CH3CN)2Cl2], was completely shut
down when [Pd(PPh3)2Cl2] was used or LiCl was added, and
no increase in reaction rate was observed in the presence of
trifluoroacetic acid (TFA) (Table 2).[14,18]
ly, the carbonyl oxygen can be protonated by Br˘nsted
acids, leading to 1b. Direct interactions between olefinic
double bonds and transition-metal catalysts can furnish the
activated complexes 1c. Finally, coordination of metal ions
is known to facilitate attack of free radicals on enones, lead-
ing to the intermediates 1d.
Free radical conjugate addition reactions have been used
to create carbon carbon and carbon sulfur bonds,[16] but ap-
plications of nitrogen nucleophiles are not known. To deter-
mine if single-electron-transfer processes are involved, aza-
Michael addition reactions of benzyl carbamate (Cbz-NH2,
3) and the enone 2a were carried out in the presence of rad-
ical inhibitors such as 3,5-di-tert-butyl-4-hydroxytoluene
(BHT) (4a) and hydroquinone (4b) (Table 1). In agreement
with results obtained by Kobayashi et al. for PtCl4¥5H2O,[9]
reaction rates and yields were virtually unchanged when
[Pd(CH3CN)4](BF4)2 and In(OTf)3 were used in the pres-
ence or absence of the inhibitors (runs 1 4). Dramatic rate
accelerations observed with Cu(OTf)2 (runs 5 7) were due
to reduction of CuII to CuI by the phenolic inhibitors,[17] ac-
companied by the liberation of one equivalent of catalytical-
ly active triflic acid (TfOH).[14,18] The involvement of free
radicals in the catalytic cycle is therefore very unlikely.
A transition-metal-catalysed pathway proceeding via the
p-complexes and a-palladated intermediates 6 and 7, respec-
tively, was postulated for PdII-mediated oxa- and aza-Mi-
chael addition reactions (Scheme 1).[6,10]
Table 2. PdII-catalysed aza-Michael addition reactions under different
conditions.
Run
Catalyst
Additive
t[h]
Yield[%]
66
1
2
3
4
[Pd(CH3CN)2Cl2]
[Pd(PPh3)2Cl2]
[Pd(CH3CN)2Cl2]
[Pd(CH3CN)2Cl2]
24
48
48
24
LiCl (200%)
TFA (20%)
64
Careful analyses of the [Pd(CH3CN)4](BF4)2-catalysed
aza-Michael addition reactions to the cyclohexenones 2b,c
in CH3CN[20] revealed that small quantities of the anilines
8a,b and the tetracyclic compound 9 were formed as side
485
Chem. Eur. J. 2004, 10, 484 493
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

J. B. Spencer et al.
FULL PAPER
products (Scheme 2), supporting reaction pathways involv-
ing catalytic activation of the carbonyl group. The anilines
8a,b can be formed through 1,2-addition of benzyl carba-
mate to the acceptor and subsequent oxidation of the en-
amines by PdII, as recently demonstrated by Saito et al.[21]
tures of the intermediates could not be determined and che-
late complexes such as 10 are tentatively proposed based on
the data available. Decomplexation by reduction of PdII
with formic acid gave 11 as the only product.[23,24] None of
these results provided evidence for catalytic activation of
enones through coordination to the olefinic double bond.
Stoichiometric mixtures of other metal salts and enones
were also analysed by NMR spectroscopy. Sc(OTf)3, which
is a strong, oxophilic Lewis acid but a poor catalyst for the
aza-Michael addition reactions of carbamates, did form com-
plexes with the enone 2e and significant changes in 13C
chemical shifts of the enone were detected (Scheme 5). In
Scheme 2. Formation of side products in the PdII-catalysed aza-Michael
addition reactions.
Scheme 5. 13C NMR analysis of enone ScIII complexes.
The structure of the tetracyclic compound 9 was confirmed
by X-ray crystallography.[22] The formation of this product is
also likely to be initiated by 1,2-attack on the carbonyl
group, followed by Diels Alder (or stepwise conjugate addi-
tion reactions) and intramolecular Mannich reactions
(Scheme 3).
contrast, when Cu(OTf)2 or In(OTf)3 and the enone 2e
were mixed, no complexes were observed and slow, unselec-
tive condensation reactions occurred. Neither complex for-
mation nor decomposition was detected with PtCl4. When
ReCl5 was used, release of HCl led to the formation of the
b-chloroketone 12. The same reaction took place when the
catalytically inactive chlorides TiCl4 and AlCl3 were used
(Scheme 6).
Scheme 3. Postulated routes of side product formation in PdII-catalysed
aza-Michael addition reactions of cyclohexenones.
Scheme 6. Enone metal-salt interactions in CD3CN at 08C.
These NMR experiments did not reveal carbonyl/Lewis
acid interactions for active catalysts, but showed that
Br˘nsted acids can be generated. Therefore, investigations
aimed at separating Lewis and Br˘nsted acidity were carried
out with the weak base 2,6-di-tert-butylpyridine (13), which
only binds to protons and is unable to coordinate to metal
ions due to the bulky tert-butyl groups.[25] In the presence of
the pyridine 13, none of the active metal salt catalysts re-
Attempts to detect the postulated complexes 6 were also
made by NMR analysis of stoichiometric mixtures of
[Pd(CH3CN)4](BF4)2 and (E)-pent-3-en-2-one (2d). No sub-
strate catalyst complexes were detected in CD2Cl2. In
CD3CN, a slow Ritter reaction of the enone took place with-
out prior complex formation (Scheme 4). The precise struc-
9]
ported to date[6 were able to induce conversion in the aza-
Michael addition of benzyl carbamate (3) to the enone 2a
(Table 3, runs 1 11).
These results strongly suggest that the presence of Br˘n-
sted acids is crucial for the reaction. As the release of more
than one equivalent of protons per unit Lewis acid is possi-
ble through hydrolysis, increased amounts of base 13 were
required to suppress the reaction with the easily hydrolys-
able salts Fe(ClO4)3¥9H2O and In(OTf)3 (runs 10 and 11).
Slow conversion was observed when stoichiometric amounts
of dibasic phenanthroline (14) were present as a ligand to
Scheme 4. Stoichiometric enone PdII interactions.
486
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 484 493

Lewis Acid Mediated Hetero-Michael Addition Reactions
484 493
Table 3. Inhibition of the metal-salt-catalysed aza- and oxa-Michael addition reactions by nitrogen bases.
Poor conversion was observed
with cations in the borderline
region p*K1 =7.7 9.0 (ZnII,
PbII), while no reaction took
place with the weakly hydrolys-
ing cations LiI, NaI, MgII,
MnII, CoII, NiII, AgI, and CdII
(p*K1 >9.0). Rare earth triflates
induced slightly less conversion
than predicted from their hy-
drolysis constants. No reaction
was observed with LaIII and
SmIII and only slow conversion
took place with ScIII and YbIII.
Apart from these cases, an ex-
cellent correlation of p*K1 and
catalytic activity was obtained.
As p*K1 values are only
meaningful in aqueous solu-
tions, the presence of protons in
organic solutions was verified
spectroscopically for all Lewis
acids used. The wide separation
Run
[a]
Enone
RXH
Catalyst
Base(amount)
Product
Yield[%]
80 99
2a
2a
2a
2a
2a
2a
2a
2a
2a
2a
2a
2a
2a
2 f
2 f
3
3
3
3
3
3
3
3
3
3
3
3
3
15
15
MXn
5a
1[b]
2
3
4
5
6
7
8
9
10
11
12[c]
13[b]
14[b]
[Pd(CH3CN)2Cl2]
[Pd(CH3CN)4](BF4)2
Cu(OTf)2
RhCl3¥3H2O
ReCl5
PtCl4
AuCl
AuCl3
Bi(NO3)3¥5H2O
Fe(ClO4)3¥9H2O
13(11%)
13(11%)
13(11%)
13(11%)
13(11%)
13(11%)
13(11%)
13(11%)
13(11%)
13(22%)
13(22%)
14(10%)
In(OTf)3
[d]
5a
16a
45
85
Pd(CH3CN)2Cl2
Pd(CH3CN)2Cl2
1
of H chemical shifts of the pyr-
idine 13 and its conjugated acid
17 allowed the extent of proto-
13(11%)
[a] Yields with catalysts in runs 1 11 ranged from 80% (ReCl5) to 99% (Cu(OTf)2) in the absence of bases.
[b] In CH2Cl2. [c] In CH3NO2. [d] [Pd(phenanthroline)(CH3CN)2](BF4)2 was used (no further base added).
Table 4. Aza-Michael addition reactions with cationic catalysts.
PdII, confirming the Br˘nsted basic role of the pyridine 13
as an inhibitor (run 12). The identical effect of 13 was ob-
served in PdII-catalysed oxa-Michael addition reactions of
benzyl alcohol (15) to the enone 2 f (runs 13 and 14).
For metal salts with weakly coordinating anions, the gen-
eration of Br˘nsted acids through cation hydrolysis in aque-
ous solution is well-documented and hydrolysis constants
*K1 have been obtained for ions of almost all metals in the
periodic table (Scheme 7).[26,27] Surprisingly, no systematic
studies of catalyst hydrolysis are available for reactions in
organic solvents, despite the important implications for
Lewis acid mediated transformations.[28]
[a]
Run
Catalyst
t[h]
Yield
p*K1
1
Li(OTf)
7
13.7
14.6
11.4
5.3[b]
4.8[c]
2.4
2
3
NaClO4¥H2O
Mg(ClO4)2
7
7
4
5
6
7
8
9
Al(ClO4)3
Sc(OTf)3
5
7
5
5
7
7
1
7
7
7
1
7
2
1
7
7
0.5
1
7
7
7
2
7
0.5
93
42
79
76
V(ClO4)3¥nH2O
Cr(ClO4)3¥6H2O
Mn(ClO4)2¥6H2O
Fe(ClO4)2¥6H2O
Fe(ClO4)3¥9H2O
Co(BF4)2¥6H2O
Ni(ClO4)2¥6H2O
Cu(CH3CN)4PF6
Cu(OTf)2
Zn(BF4)2¥nH2O
ZrO(ClO4)2¥8H2O
Pd(CH3CN)4(BF4)2
AgClO4¥H2O
Cd(ClO4)2¥nH2O
In(OTf)3
4.0
10.6
7.1[d]
2.7[e]
9.7[d]
92
93
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
9.9[d]
[f]
81
18
94
76
7.3
9.0
5.1[g]
1.6
Scheme 7. Description of hydrolysis equilibria using the notation of Sillÿn
and Martell (see ref. [27]).
11.7
10.1
4.0
94
94
Sn(OTf)2
La(OTf)3¥nH2O
Sm(OTf)3
3.4[d]
8.5
The aza-Michael addition of benzyl carbamate (3) to the
enone 2e was carried out with Lewis acids containing highly
cationic metal parts and catalytic activities and hydrolysis
7.9
7.7
2.4
Yb(OTf)3
20
91
16
94
33]
constants were compared (Table 4).[26,29
Hg(ClO4)2¥3H2O
Pb(ClO4)2¥3H2O
Bi(OTf)3
7.9[h]
1.6
Cations with p*K1 ꢀ7.3 were found to release sufficient
amounts of acid to catalyse the reaction and good yields
were obtained with AlIII, VIII, CrIII, FeII/III, CuII, ZrO2+, PdII,
InIII, HgII and BiIII. In all cases, the reactions were complete-
ly suppressed in the presence of the pyridine 13 (22%).
[a] Unless otherwise noted, values were taken from ref. [29] and ref. [26]
[b] See ref. [30] [c] See ref. [31] [d] Reported *K1 for FeII, CoII, NiII, SnII
vary dramatically; ref. [29] [e] See ref. [32] [f] *K1 immeasurably low. See
+
ref. [26] [g] p*K3 given (for Zr(OH)3 + H+). [h] See ref. [33]
487
Chem. Eur. J. 2004, 10, 484 493
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

J. B. Spencer et al.
FULL PAPER
nation to be determined easily in the presence of metal salts
(Table 5).[34] It was confirmed that protons were liberated
from all good catalysts in the presence of one or more
equivalents of water (17:13>10:1). Without exception,
metal salts that did not hydrolyse (17:13 < 1:10) failed to
induce conversion in the aza-Michael addition reactions. No-
tably, protons were released from [Pd(CH3CN)2Cl2] in
be released from heteroatom nucleophiles and carbonyl
compounds through imine condensation (Scheme 3), or
acetal/thioacetal formation. As these reversible processes
occur easily under nonbasic conditions, it is virtually impos-
sible to perform the hetero-Michael addition reactions de-
scribed in the strict absence of water. In addition, in situ
drying agents such as molecular sieves led to irreproducible
results and could not be used for mechanistic studies as they
are known not only to retain water, but also to exchange
ions and neutralise acids.[42]
Table 5. Br˘nsted acid detection by assessing the protonation equilibri-
1
um of 13 and 17 using H NMR spectroscopy.
The important aspect of water content was therefore stud-
ied by kinetic experiments with Lewis acid catalysts and var-
iable amounts of water. The addition of up to two equiva-
lents of water with respect to the catalyst led to a significant
rate increase in the [Pd(CH3CN)4](BF4)2-catalysed aza-Mi-
chael addition of the carbamate 3 to the enone 2a in
CD3CN (Figure 1). Due to its Br˘nsted basic properties, in-
Extent of proto- Metal salts
nation
17:13 > 10:1
AlCl3, Al(ClO4)3, Sc(OTf)3, TiCl4, V(ClO4)3¥nH2O,
Cr(ClO4)3¥6H2O, Fe(ClO4)2¥6H2O, Fe(ClO4)3¥9H2O,
Cu(OTf)2, ZrO(ClO4)2¥8H2O, RhCl3¥3H2O,
[Pd(CH3CN)4](BF4)2, In(OTf)3, Sn(OTf)2, Sm(OTf)3,
Yb(OTf)3, ReCl5, PtCl4, AuCl[a], AuCl3, Hg(ClO4)2¥
3H2O, Pb(ClO4)2¥3H2O, Bi(OTf)3, Bi(NO3)3¥5H2O
Co(BF4)2¥6H2O, Zn(BF4)2¥nH2O, [Pd(CH3CN)2Cl2][b]
[Pd(phenanthroline)(CH3CN)2](BF4)2[c], La(OTf)3
equilibrium
,
17:13 < 1:10
Li(OTf), NaClO4¥H2O, Mg(ClO4)2, Mn(ClO4)2¥6H2O,
Ni(ClO4)2¥6H2O, [Cu(CH3CN)4]PF6, Pd(OAc)2,
[Pd(CH3CN)2Cl2], AgClO4¥H2O, Cd(ClO4)2¥nH2O
[a] 6.0 equiv used. [b] Suspension in CD2Cl2. [c] In CD3NO2.
CD2Cl2, but not in CD3CN, in which hydrolysis was hindered
due to solvation. This coincides with the observed catalytic
activity of [Pd(CH3CN)2Cl2].[35] These experiments also con-
firmed the ability of rare earth triflates to generate acid, but
due to their high affinity to carbonyl groups even in the
presence of water, hydrolysis is minimised in catalysis.[36] As
expected, AlCl3 and TiCl4 are also readily hydrolysed. How-
ever, these salts failed to catalyse the aza-Michael addition
reactions due to competitive conjugate addition of HCl
(Scheme 6).
A common feature of all active metal salt catalysts is the
generation of both protons and weakly nucleophilic counter
ions in solution. Catalytically active metal chlorides are able
to form complex acids with defined pKa values, for instance
trans-[PtCl4(H2O)2] (pKa =2.64),[37] [AuCl3(H2O)] (pKa =
2.7), [RhCl3(H2O)3] (pKa =4.8)[27] and [PdCl2(H2O)2] (pKa =
2.1).[29] ReV can form stable, weakly basic oxohalide com-
plexes such as [ReCl4O]À.[38] The involvement of these or
similar species in ReCl5-mediated aza-Michael addition re-
actions explains how active acid catalysts can be generated,
while yields are reduced due to competitive conjugate addi-
tion of HCl (Scheme 3 and Table 3).[39] Although electron-
donating ligands reduce the propensity of cations to hydro-
lyse, diamine PdII complexes can still generate acid and a
Figure 1. Initial rates (and linear best fits) of [Pd(CH3CN)4](BF4)2-cata-
lysed aza-Michael addition reactions of 3 to 2a in the presence of varying
amounts of water in CD3CN (15% catalyst).
creased amounts of water caused the reaction to slow down.
The same relationship between water content and reaction
rate was observed with PtCl4 and also with SbF5, which
forms strongly Br˘nsted acidic SbF5(H2O) upon hydroly-
sis.[18,43] The rate decrease caused by the action of water as a
Br˘nsted base was also detected with the acid catalyst
HBF4¥OMe2.[18]
In a similar manner, other molecules with oxygen func-
tionalities can act as Lewis or Br˘nsted bases. Earlier stud-
ies revealed that no or only reactions takes place when the
acid-catalysed aza-Michael addition is carried out in THF,
diethyl ether, acetone or ethyl acetate, whereas the reaction
is faster in acetonitrile, dichloromethane, and nitrome-
thane.[14]
Dicarbonyl acceptors, such as alkylidene malonates and
enoyl oxazolidinones, are known to coordinate more strong-
ly to oxophilic metal ions bearing multiple charges than to
protons,[44] forming conformationally rigid catalyst substrate
complexes. Both Br˘nsted and Lewis acids such as Tf2NH
and Yb(OTf)3 were suitable catalysts for the aza-Michael
addition of benzyl carbamate (3) to the acceptors 2 g and h
(Table 6, runs 1 4). However, Tf2NH and hydrolysable
pKa value of 4.7 has been reported for [Pd(bipyridine)-
[29]
(H2O)2]2+
.
AuCl is the only catalyst that cannot directly
liberate a Br˘nsted acid through hydrolysis. However, in the
absence of stabilising ligands, AuCl disproportionates in so-
lution, furnishing Au0 and hydrolysable AuCl3.[40,41]
The origin of water required for hydrolysis must also be
addressed. Water can be present as residual moisture or can
488
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 484 493

Lewis Acid Mediated Hetero-Michael Addition Reactions
484 493
Table 6. Aza-Michael addition of benzyl carbamate (3) to dicarbonyl
acceptors.
N-methylaniline (20) could be used successfully (runs 4 6).
Oxa-Michael addition reactions reactions of benzyl alcohol
(15) could also be catalysed by Br˘nsted acids (runs 7 and
8) and did not require PdII catalysts. The same method
could be applied to the thiols 21 and 22, which reacted rap-
idly in the presence of Tf2NH (entries 9 10). Similarly, the
weakly basic carbon nucleophile indole (23) underwent
acid-catalysed Michael addition reactions with the enones
2a, 2e and 2 f (runs 11 13).
Transformations with similar nucleophiles and acceptors
as used in runs 5, 6, 9 and 10 have also been carried out
with chiral Lewis acid catalysts.[5g,f,13] The existence of
Br˘nsted acid catalysed pathways has important implica-
tions for the development of enantioselective processes. Cat-
alytic enantioselective reactions which are sensitive to
proton-mediated background reactions cannot succeed
under nonbasic conditions when trace amounts of water
cannot be excluded, and when hydrolysable cationic cata-
lysts are used. The impact of hydrolysis can be reduced by
employing Lewis basic solvents such as THF or chelating di-
carbonyl substrates. Alternatively, the propensity of Lewis
acid catalysts to hydrolyse can be lowered by modification
of the metal centre with strongly electron-donating ligands
Run
Acceptor
Catalyst
Product
t[h]
Yield[%]
1[a]
2[a]
3[a]
4[a]
5
2g
2g
2h
2h
2i
2i
2i
2i
2i
Tf2NH[d]
Yb(OTf)3
Tf2NH
Yb(OTf)3
Tf2NH
Cu(OTf)2
In(OTf)3
Yb(OTf)3
Yb(OTf)3
Yb(OTf)3
5g
5g
5h
5h
72
48
12
12
48
48
48
12
12
48
49
21
91
87
6
7
8
5i
5i
5i
5i
5i
3
19
51
21
38
9[b]
10[c]
2i
[a] At À208C. [b] 2,6-Di-tert-butylpyridine (13, 11%) was added. [c] In
THF. [d] Tf2NH=bis(trifluoromethanesulfon)imide.
*
or by using cations with a small hydrolysis constants K1. To
metal salts such as Cu(OTf)2 or In(OTf)3 turned out to be
poor catalysts when benzylidene malonate 2i was used
(runs 5 7). Yb(OTf)3-mediated reactions with 2i, on the
other hand, proceeded even in the presence of the pyridine
13 and THF; this rules out background proton catalysis and
confirms YbIII as the active cat-
date, Lewis acid catalysed enantioselective aza- and thia-Mi-
chael addition reactions to monodentate enones under non-
basic conditions have only been carried out with weakly hy-
drolysing complexes of ScIII and CdII.[5h,13b 13c] High ees in ad-
dition reactions to chelating acceptors have been achieved
alyst (runs 8 10). Therefore,
Lewis acid catalysed processes
Table 7. Br˘nsted acid catalysed conjugate addition reactions of nitrogen, oxygen, sulfur and carbon
nucleophiles.[a]
involving
dicarbonyl
com-
pounds such as alkylidene mal-
onates can benefit from minimi-
sation of proton-mediated back-
ground reactions.
In combination, these results
unambiguously show that pro-
tons can be the active catalysts
in Lewis acid mediated hetero-
Michael reactions to a,b-unsat-
urated ketones and to some di-
carbonyl acceptors. Conse-
quently, Lewis acid catalysts re-
ported for aza- and oxa-Mi-
chael addition reactions of
Run
Acceptor
RXH
Solvent
Product
t[h]
Yield[%]
weakly
basic
nucleophi-
1
2
3
4
2a
2e
2j
3
3
3
CH3CN
CH3CN
CH3CN
CH3CN
CH2Cl2
CH3CN
CH3CN
CH3CN
CHCl3
CH2Cl2
CH3CN
CH3CN
CH3CN
5a
5e
5j
24
25
26
16a
16b
27
10 min
98
94
71
94
87
98
79
72
96
93
95
88
98
9,10a,c]
les[6
were outperformed
10 min
48
1
10
12
12
48
2
by homogeneous or heteroge-
neous Br˘nsted acid catalysts
such as Tf2NH or Nafion SAC-
2a
2g
2g
2 f
2a
2b
2g
2a
2e
2 f
18
19
20
15
15
21
22
23
23
23
5[b]
6
¾
13.[14,15] Examples and exten-
sions of this methodology are
listed in Table 7. In addition to
benzyl carbamate (3) (runs 1
3), other nitrogen nucleophiles
such as oxazolidin-2-one (18),
7
8
9
10
11[c]
12[c]
13
28
2
0.5
2
29a
29b
29c
12
benzaldehyde oxime (19) and [a] The dashed arrows indicate the reactive centres of 19 and 23. [b] Tf2NH (50%) was used. [c] At 508C.
489
Chem. Eur. J. 2004, 10, 484 493
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

J. B. Spencer et al.
FULL PAPER
with chiral catalysts based on MgII, NiII, ZnII, YIII,
YbIII,[5b,d,f,g,13a,13d] or with AlIII complexes[5c] bearing strongly
basic ligands.[45,46] All these catalysts have only weak tenden-
cies to hydrolyse, confirming the importance of the concept
described in this report.[47]
spectrometer at 258C. Acquisitions (2 scans) were made at regular inter-
vals. Conversions were determined from the relative intensities of the
methyl protons of the enone 2a (d=1.10 ppm in CD3CN) and the prod-
uct 5a (d=0.92 ppm in CD3CN).
General procedure for Br˘nsted acid detection with 2,6-di-tert-butylpyri-
dine: 2,6-Di-tert-butypyridine (13) (12 mL, 11 mg, 0.05 mmol), metal salt
(0.1 mmol) and–when anhydrous metal salts were used–water (0.9 mL,
0.9 mg, 0.05 mmol) were dissolved in CD3CN (0.75 mL). The relative
amounts of 13 and the conjugated acid 17 were determined by 1H NMR
spectroscopy. 13: 1H NMR (400 MHz, CD3CN): d=1.33 (s, 18H; CH3),
7.16 (d, J=7.8 Hz, 2H; m-Ar-H), 7.59 ppm (t, J=7.8 Hz, 1H; p-Ar-H).
17; 1H NMR (400 MHz, CD3CN): d=1.53 (s, 18H; CH3), 7.90 (d, J=
8.2 Hz, 2H; m-Ar-H), 8.47 (t, J=8.2 Hz, 1H; p-Ar-H), 11.21 (br, 1H;
NH).
Interaction between ScIII and (E)-hex-4-en-3-one (2e): Sc(OTf)3 (98 mg,
0.21 mmol) and 2e (22 mL, 20 mg, 0.20 mmol) were dissolved in CD3CN
(0.75 mL). 13C NMR spectra were recorded at 08C and stoichiometric
complex formation was observed. 2e: 13C NMR (100 MHz, CD3CN,
08C): d=7.4, 17.4 (CH3), 32.4 (CH2), 131.4, 142.4 (=CH), 200.5 ppm (C=
O). Sc(OTf)3+2e: 13C NMR (100 MHz, CD3CN, 08C): d=7.6, 17.8
(CH3), 32.4 (CH2), 130.4, 151.3 (=CH), 208.7 ppm (C=O).
Conclusion
The generation of Br˘nsted acids through hydrolyis of metal
complexes is a known, but often underestimated phenomen-
on in Lewis acid catalysis. In this report, Br˘nsted acids
have been identified as the active catalysts in hetero-Mi-
chael addition reactions to a,b-unsaturated ketones. Acid
mediated conjugate addition reactions of weakly basic nitro-
gen, oxygen, sulfur and carbon nucleophiles have been de-
veloped. Aza-Michael addition reactions of benzyl carba-
1
mate (3) and H NMR studies with 2,6-di-tert-butylpyridine
N-[(E)-Pent-2-enoyl]acetamide (2j): Sulfuric acid (110 mL, 205 mg,
2.10 mmol) was added to a mixture of (E)-pent-2-enoyl amide (2.1 g,
21 mmol) and acetic acid anhydride (20 mL, 210 mmol). The solution was
stirred at 808C for 30 min, volatile components were evaporated in vacuo
and the product was isolated as a colourless solid after column chroma-
tography (2.42 g, 82%); Rf =0.35 (PE/EtOAc 1:1); M.p.=84 858C;
1H NMR (400 MHz, CDCl3): d=1.03 (t, J=7.4 Hz, 3H; CH2CH3), 2.22
(dq, J=6.5, 7.4 Hz, 2H; CH2CH3), 2.42 (s, 3H; C(O)CH3), 6.19 (d, J=
15.4 Hz, 1H; C(O)CH), 7.10 (dt, J=6.5, 15.4 Hz, 1H; C=CH), 9.21 ppm
(br, 1H; NH); 13C NMR (100 MHz, CDCl3): d=12.1 25.2 (CH3), 25.5
(CH2), 122.0, 152.0 (=CH), 165.4, 173.6 ppm (C=O); IR (neat, film): n˜ =
3259, 3199, 2967, 1726, 1682, 1643, 1500, 1278 cmÀ1; HRMS (+EI, 70 eV)
calcd for: [C7H11NO2]+ : 141.0790; found: 141.0793 [M]+.
(13) have been used as probes to detect proton liberation
from Lewis acids in organic solvents. These investigations
provide very useful guidelines for the development of Lewis
acid catalysed reactions of mono- and dicarbonyl com-
pounds under nonbasic conditions and are not limited to
hetero-Michael addition reactions.
Experimental Section
General: Solvents were distilled from CaH2 prior to use. Thin-layer chro-
matography was performed with Merck 60 PF254 0.2 mm plates (glass
backed) and Merck 9385 Kieselgel 60 was used for column chromatogra-
phy. NMR spectra were recorded on Bruker Avance 400 and 500 ma-
chines, running at 400 and 500 MHz (1H) or 100 and 126 MHz (13C), re-
spectively. Chemical shifts (in ppm) are reported relative to residual sol-
vent peaks. CD2Cl2 (< 0.02% H2O) and CD3CN (< 0.05% H2O) were
used as prescored ampoules (0.75 mL). Assignments of 1H and 13C spec-
tra were verified by 1H,1H-COSY, DEPT-135 and 1H,13C-HMQC techni-
ques. Infrared spectra were recorded on a Perkin Elmer Spectrum One
FT-IR as neat solids or liquids. Mass spectra were recorded on a Kratos
890 spectrometer. Melting points were determined on a Reichert hot-
stage apparatus and are uncorrected. Al(ClO4)3,[48] Cu(CH3CN)4PF6,[49]
trans-3-Benzyloxycarbonylamino-4-methylcyclohexanone (5c): Com-
pound 5c was prepared according to GP1 ([Pd(CH3CN)4](BF4)2,
1.0 mmol scale, 24 h) and was isolated in dr>20:1 as a colourless oil
(86 mg, 33%). The assignment of trans-geometry was based upon the
coupling constant between CHCH3 and CHN, which was determined by
1H 1H homodecoupling experiments at 08C (3J=10.5 Hz). Rf 0.24 (PE/
diethyl ether 1:3); 1H NMR (500 MHz, CDCl3, 08C): d=1.04 (d, J=
6.6 Hz, 3H; CHCH3), 1.42 (dm, J=13.9 Hz, 1H; CHH), 1.78 (dm, J=
10.5 Hz, 1H; CHCH3), 2.03 (dm, J=13.9 Hz, 1H; CHH), 2.22 2.43 (m,
3H; C(O)CH2, C(O)CHH), 2.73 (m, 1H; C(O)CHH), 3.59 (dm, J=
10.5 Hz, 1H; NCH), 4.82 (br, 1H; NH), 5.10 (s, 2H; OCH2), 7.32
7.43 ppm (m, 5H; Ar-H); 13C NMR (100 MHz, CDCl3): d=17.8 (CH3),
30.7 (CH2), 36.6 (CH), 40.4, 47.5 (CH2), 54.4 (CH), 66.7 (CH2), 128.1,
128.2, 128.5 (Ar-C), 136.4 (ipso-Ar-C), 155.6, 208.7 ppm (C=O); IR (neat,
[50]
Bi(OTf)3 and compounds 2a,[51] 2g,[52] 2h[53] and 2i,[54] were prepared
according to literature procedures. V(ClO4)3¥nH2O was prepared in situ
from VCl3 and AgClO4¥H2O. Commercially available materials were used
as received, except enones, which were distilled prior to use. The prod-
ucts 5a,b,e,g,h and 16b have previously been characterised by us[6,14] and
full characterisation data for compounds 8a and b have already been re-
ported in the literature.[55]
film): n˜ =3324, 2957, 2928, 1717, 1680, 1531, 1275, 1251, 1020 cmÀ1
;
HRMS [+ESI] calcd for: [C15H19NO3Na]+ : 284.1263; found: 284.1256
[M+Na]+.
Diethyl ester of 2-(benzyloxycarbonylaminophenylmethyl)malonic acid
(5i): Compound 5i was prepared according to GP1 (Yb(OTf)3, 1.0 mmol
scale, 12 h) and was isolated as a colourless oil (202 mg, 51%). Rf 0.26
(PE/diethyl ether 1:1); 1H NMR (400 MHz, CDCl3): d=1.11 (t, J=
7.2 Hz, 3H; CH2CH3), 1.20 (t, J=7.2 Hz, 3H; CH2CH3), 3.88 (br, 1H;
C(O)CH), 4.02 4.19 (m, 4H; CH2CH3), 5.08 (d, J=12.3 Hz, 1H;
OCHH), 5.11 (d, J=12.3 Hz, 1H; OCHH), 5.54 (dd, J=4.2, 8.7 Hz, 1H;
NCH), 6.47 (br, 1H; NH), 7.21 7.33 ppm (m, 10H; Ar-H); 13C NMR
(100 MHz, CDCl3): d=14.2, 14.3 (CH3), 54.4, 57.2 (CH), 62.0, 62.4, 67.3
(CH2), 126.7, 128.1, 128.4, 128.4, 128.8, 129.0 (Ar-C), 136.9, 136.9 (ipso-
Ar-C), 156.1, 167.3, 168.4 ppm (C=O); IR (neat, film): n˜ =3344, 2982,
General procedure (GP1) for catalytic hetero-Michael addition reactions
(0.5 mmol scale): The nucleophile (0.75 mmol) and the catalyst
(0.05 mmol, 10%) were dissolved in CH3CN (1 mL). The a,b-unsaturated
carbonyl compound (0.5 mmol) was added and the reaction mixture was
stirred at ambient temperature and monitored by TLC. When maximum
conversion was reached,[56] an excess of silica gel containing 10%
Na2CO3 was added, the solvent was evaporated and the products were
isolated by column chromatography. The nature of the catalyst and devia-
tions from this procedure (such as different solvents, reaction tempera-
tures and reaction scales) are noted in brackets with the individual com-
pounds.
1725, 1498, 1217, 1020, 696 cmÀ1
;
HRMS [+ESI] calcd for:
[C22H25NO6Na]+ : 422.1580; found: 422.1602 [M+Na]+.
N-[3-(Benzyloxycarbonylamino)pentanoyl]acetamide (5j): Compound 5j
was prepared according to GP1 (Tf2NH, 48 h) and was obtained as a col-
ourless solid (103 mg, 71%). Rf 0.21 (PE/EtOAc 1:1); M.p. 123 1258C;
1H NMR (400 MHz, CDCl3): d=0.91 (t, J=7.4 Hz, 3H; CH2CH3), 1.40
1.52 (m, 2H; CH2CH3), 2.26 (s, 3H; C(O)CH3), 2.68 2.73 (m, 2H;
C(O)CH2), 3.95 (m, 1H; NCH), 5.05 (s, 2H; OCH2), 5.32 (br, 1H; NH),
General procedure (GP2) for kinetic experiments: NMR samples con-
taining equal amounts of benzyl carbamate (3) (78 mg, 0.52 mmol), cata-
lyst (3%-15%) and CD3CN (or CD2Cl2) (0.75 mL) were prepared from
stock solutions. If appropriate, water (or another additive) was added to
these samples. The enone 2a (80 mL, 80 mg, 0.50 mmol) was injected by
syringe, the mixtures were shaken and immediately placed into a NMR
490
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 484 493

Lewis Acid Mediated Hetero-Michael Addition Reactions
484 493
7.26 7.33 (m, 5H; Ar-H), 9.39 ppm (br, 1H; NH); 13C NMR (100 MHz,
CDCl3): d=10.8, 25.4 (CH3), 28.3, 42.3 (CH2), 50.2 (CH), 67.0 (CH2),
128.3, 128.5, 128.9 (Ar-C), 136.9 (ipso-Ar-C), 156.5, 172.7, 172.8 ppm (C=
O); IR (neat, film): n˜ =3317, 3144, 2976, 1734, 1709, 1686, 1534,
1245 cmÀ1; HRMS [+ESI]: calcd for [C15H20N2O4Na]+ : 315.1321, found:
315.1313 [M+Na]+.
(100 MHz, CD3CN): d=8.3, 26.2 (CH3), 37.6, 53.2 (CH2), 54.2 (CH),
208.3 ppm (C=O); IR (neat, film): n˜ =2961, 1726, 1393, 1250, 1066 cmÀ1
;
HRMS [EI, 70 eV] calcd for: [C6H11ClO]+ : 134.0498; found: 134.0494
[M]+.
1-(Benzyloxy)pentan-3-one (16a): Compound 16a was prepared accord-
ing to GP1 (Tf2NH, 1.0 mmol scale, 12 h) and was obtained as a colour-
less oil (151 mg, 79%). Rf 0.44 (CH2Cl2/acetone 30:1). 1H NMR
(500 MHz, CDCl3): d=1.08 (t, J=7.4 Hz, 3H; CH2CH3), 2.49 (q, J=
7.4 Hz, 2H; CH2CH3), 2.71 (t, J=6.3 Hz, 2H; C(O)CH2), 3.77 (t, J=
6.3 Hz, 2H; OCH2), 4.53 (s, 2H; OCH2), 7.30 7.41 ppm (m, 5H; Ar-H);
13C NMR (100 MHz, CDCl3): d=7.4 (CH3), 36.4, 42.3, 65.2, 73.0 (CH2),
127.4, 127.5, 128.2 (Ar-C), 138.0 (ipso-Ar-C), 209.6 ppm (C=O); IR (neat,
film): n˜ =2975, 2871, 1711, 1454, 1364, 1090, 735 cmÀ1; HRMS [EI, 70 eV]
calcd for: [C12H16O2]+ : 192.1150; found: 192.1155 [M]+.
Compound 9 was isolated as a side product in the synthesis of 5b accord-
ing to GP1 ([Pd(CH3CN)4(BF4)2], 1.0 mmol scale) as a colourless solid
(17 mg, 5%). Samples from several runs were combined and crystals suit-
able for X-ray crystallography were obtained from pentane/diethyl ether.
Rf 0.35 (PE/diethyl ether 1:3); M.p. 136 1378C; 1H NMR (400 MHz,
CDCl3): d=1.42 1.69 (m, 7H; CH2, CH), 1.80 1.86 (m, 1H; CHH), 1.90
2.01 (m, 4H; CH2, CH), 2.20 2.28 (m, 1H; CH), 2.34 (t, J=6.2 Hz, 1H;
C(O)CH), 2.71 (m, 1H; C(O)CH), 4.92 (br, 1H; NH), 5.05 (s, 2H;
OCH2), 7.28 7.36 ppm (m, 5H; Ar-H); 13C NMR (100 MHz, CDCl3): d=
20.9, 23.5, 23.5, 23.7 (CH2), 28.2, 35.8 (CH), 36.1 (CH2), 37.1, 51.5, 53.4
(CH), 56.7 (Cquart), 66.9 (CH2), 128.6, 128.7, 129.0 (Ar-C), 136.7 (ipso-Ar-
C), 155.8, 218.9 ppm (C=O); IR (neat, film): n˜ =3304, 3248, 2924, 2855,
3-(2-Oxo-oxazolidin-3-yl)-1-phenylpentan-1-one (24): Compound 24 was
prepared according to GP1 (Tf2NH, 1.0 mmol scale, 1 h) and was isolated
as a colourless solid (232 mg, 94%). Rf 0.17 (diethyl ether); m.p. 48
1
508C; H NMR (400 MHz, CDCl3): d=0.92 (t, J=7.4 Hz, 3H; CH2CH3),
1733, 1690, 1535, 1262 cmÀ1; HRMS [+ESI]: calcd for [C20H23NO3Na]+
348.1576; found: 348.1581 [M+Na]+).
:
1.61 1.78 (m, 2H; CH2CH3), 3.12 (dd, J=5.7, 16.2 Hz, 1H; C(O)CHH),
3.33 (dd, J=7.9, 16.2 Hz, 1H; C(O)CHH), 3.57 (dt, J=2.3, 7.9 Hz, 2H;
NCH2), 4.07 (m, 1H; NCH), 4.21 (t, J=7.9 Hz, 2H; OCH2), 7.42 (dd, J=
7.5, 7.9 Hz, 2H; m-Ar-H), 7.53 (t, J=7.5 Hz, 1H; p-Ar-H), 7.91 ppm (d,
J=7.9 Hz, 2H; o-Ar-H); 13C NMR (100 MHz, CDCl3): d=11.2 (CH3),
25.6, 41.4, 43.0 (CH2), 52.9 (CH), 62.4 (CH2), 128.4, 129.0, 133.7 (Ar-C),
136.9 (ipso-Ar-C), 158.1, 198.3 (C=O); IR (neat, film): n˜ =2963, 2910,
(Z)-4-(2,2,2-Trideuteroacetylamido)pent-2-en-2-olato palladium(ii) com-
plexes (10): The complexes were prepared by dissolving pent-3-en-2-one
(2d) (21 mL, 18 mg, 0.21 mmol) and [Pd(CH3CN)4(BF4)2] (100 mg,
0.210 mmol) in CD3CN (0.75 mL). The starting enone was completely
consumed within 2 h and the products were analysed in situ (70:30 mix-
ture). Major species: 1H NMR (400 MHz, CD3CN): d=1.66 (d, J=
6.7 Hz, 3H; CHCH3), 1.79 (s, 3H;=CCH3), 4.00 (m, 1H; NCH),
4.79 ppm (m, 1H; C=CH); 13C NMR (100 MHz, CD3CN): d=15.9, 23.0
(CH3), 53.7 (CH), 101.1 (=CH), 144.8 (=Cquart), 166.9 ppm (C=O);[57]
Minor species: 1H NMR (400 MHz, CD3CN): d=1.35 (d, J=6.6 Hz, 3H;
CHCH3), 1.89 (s, 3H;=CCH3), 4.28 (m, 1H; NCH), 5.20 ppm (m, 1H;
C=CH); 13C NMR (100 MHz, CD3CN): d=16.3, 21.5 (CH3), 43.0 (CH),
102.7 (=CH), 145.4 (=Cquart), 171.7 ppm (C=O);[57] IR (mixture, in
CD3CN): n˜ =1734, 1647, 1524, 1197 cmÀ1; HRMS [+ESI]: calcd for
[C7H9D3NO2Pd]+ : 251.0091; found: 251.0089 [M+H]+.[58]
1736, 1686, 1425, 1226, 1057, 756 cmÀ1
; HRMS [+ESI] calcd for
[C14H17NO3Na]+ : 270.1106; found: 270.1098 [M+Na]+.
3-[3-(Benzylideneamino)propionyl]-2-oxazolidinone N-oxide (25): Com-
pound 25 was prepared according to GP1 (Tf2NH (50%), in CH2Cl2,
10 h) and was obtained as a colourless solid (112 mg, 87%). Rf 0.05 (PE/
EtOAc 1:1); m.p. 169 1718C; 1H NMR (500 MHz, CD3CN): d=3.48 (t,
J=6.3 Hz, 2H; C(O)CH2), 3.90 (t, J=8.3 Hz, 2H; NCH2), 4.24 (t, J=
6.3 Hz, 2H; NCH2), 4.36 (t, J=8.3 Hz, 2H; OCH2), 7.42 7.48 (m, 3H; m-
Ar-H, p-Ar-H), 7.74 (s, 1H; N=CH), 8.22 8.24 ppm (m, 2H; o-Ar-H);
13C NMR (126 MHz, CDCl3): d=32.5, 42.4, 60.8, 62.8 (CH2), 128.5, 128.7,
130.6 (Ar-C), 130.6 (ipso-Ar-C), 136.3 (N=CH), 154.3, 170.9 ppm (C=
O); IR (neat, film): n˜ =2975, 1776, 1761, 1694, 1380, 1214, 1118,
755 cmÀ1; HRMS [+ESI] calcd for [C13H14N2O4Na]+ : 285.0851; found:
285.0858 [M+Na]+.
N-(1-Methyl-3-oxo-butyl)-2,2,2-trideuteroacetamide (11): Formic acid
(1.0 mL, 0.80 g, 18 mmol) was added to the solution described in the
preparation of the complexes 10. The mixture was stirred at 508C. After
2 h the initially dark orange solution became clear and Pd0 had precipi-
tated. The solid was filtered off, aqueous Na2CO3 (5 mL) was added and
the mixture was extracted with CH2Cl2 (3î10 mL). The combined organ-
ic phases were dried over Na2SO4 and filtered, and the solvent was re-
moved in vacuo. The product was obtained as a colourless oil (26 mg,
3-[3-(N-Methyl-N-phenylamino)propionyl]oxazolidin-2-one (26): Com-
pound 26 was prepared according to GP1 (Tf2NH, 12 h) and was ob-
tained as
a colourless oil (122 mg, 98%). Rf 0.29 (diethyl ether);
1H NMR (500 MHz, CDCl3) d=2.96 (s, 3H; NCH3), 3.20 (t, J=7.1 Hz,
2H; NCH2), 3.73 (t, J=7.1 Hz, 2H; C(O)CH2), 3.91 (t, J=8.3 Hz, 2H;
NCH2), 4.30 (t, J=8.3 Hz, 2H; OCH2), 6.70 (t, J=7.1 Hz, 1H; p-Ar-H),
6.78 (d, J=8.0 Hz, 2H; o-Ar-H), 7.22 ppm (dd, J=7.1, 8.0 Hz, 2H; m-Ar-
H); 13C NMR (126 MHz, CDCl3) d=32.6 (CH2), 38.2 (CH3), 42.5, 48.3,
62.1 (CH2), 112.6, 116.7, 129.2 (Ar-C), 149.0 (ipso-Ar-C), 153.8,
172.3 ppm (C=O); IR (neat, film): n˜ =2918, 1771, 1690, 1598, 1505, 1385,
1220, 1034 cmÀ1; HRMS [+ESI] calcd for [C13H16N2O3Na]+ : 271.1059;
found: 271.1062 [M+Na]+.
1
87%); H NMR (500 MHz, CDCl3): d=1.20 (d, J=6.3 Hz, 3H; CHCH3),
2.14 (s, 3H; C(O)CH3), 2.59 (dd, J=5.7, 16.8 Hz, 1H; C(O)CHH), 2.66
(dd, J=5.0, 16.8 Hz, 1H; C(O)CHH), 4.30 (m, 1H; NCH), 6.20 ppm (br,
1H; NH); 13C NMR (126 MHz, CDCl3): d=20.1, 30.6 (CH3), 41.9 (CH),
48.7 (CH2), 169.5, 208.2 ppm (C=O);[57] IR (neat, film): n˜ =3287, 3075,
2927, 1710, 1637, 1546, 1370, 729 cmÀ1; HRMS [EI, 70 eV] calcd for:
[C7H10D3NO2]+ : 146.1135; found: 146.1132 [M]+.
N-(1-Methyl-3-oxo-butyl)acetamide (H3-11): Compound H3-11 was pre-
pared in analogy to compound 11 with CH3CN as a solvent. The product
was obtained as a colourless oil (27 mg, 88%). 1H NMR (500 MHz,
CDCl3): d=1.21 (d, J=6.3 Hz, 3H; CHCH3), 1.92 (s, 3H; C(O)CH3),
2.14 (s, 3H; C(O)CH3), 2.60 (dd, J=5.7, 16.8 Hz, 1H; C(O)CHH), 2.67
(dd, J=5.0, 16.8 Hz, 1H; C(O)CHH), 4.30 (m, 1H; NCH), 6.14 ppm (br,
1H; NH); 13C NMR (126 MHz, CDCl3): d=20.2, 23.4, 30.6 (CH3), 41.9
(CH), 48.6 (CH2), 169.4, 208.2 ppm (C=O); IR (neat, film): n˜ =3287,
3084, 2924, 1710, 1648, 1547, 1370, 731 cmÀ1; HRMS [EI, 70 eV] calcd
for: [C7H13NO2]+ : 143.0946; found: 143.0947 [M]+.
3-(Phenylsulfanyl)cyclohexan-1-one (27): Compound 27 was prepared ac-
cording to GP1 (Tf2NH, in CHCl3, 2 h) and was obtained as a colourless
oil (99 mg, 96%). Rf 0.31 (PE/ether 1:1); 1H NMR (500 MHz, CDCl3):
d=1.70 1.75 (m, 2H; CH2), 2.12 2.17 (m, 2H; CH2), 2.29 2.40 (m, 3H;
C(O)CH2, C(O)CHH), 2.68 (dd, J=4.4, 14.3 Hz, 1H; C(O)CHH), 3.44
(m, 1H; SCH), 7.27 7.33 (m, 3H; m-Ar-H, p-Ar-H), 7.42 ppm (d, J=
5.7 Hz, 2H; o-Ar-H); 13C NMR (126 MHz, CDCl3): d=24.0, 31.2, 40.9
(CH2), 46.1 (CH), 47.8 (CH2), 127.8, 129.0 (Ar-C), 133.1 (ipso-Ar-C),
133.2 (Ar-C), 208.6 ppm (C=O); IR (neat, film): n˜ =3056, 2943, 1708,
1582, 1438, 1220, 1024, 739 cmÀ1; HRMS [+ESI] calcd for [C12H15OS]+
207.0844; found: 207.0848 [M+H]+.
:
5-Chlorohexan-3-one (12): Compound 12 was prepared by bubbling gas-
eous HCl through a solution of 2e (57 mL, 49 mg, 0.5 mmol) in CH3CN
(1 mL) for 10 min. The solvent was evaporated and preparative thin-
layer chromatography furnished a colourless oil (22 mg, 34%). The NMR
spectra of this compound and of stoichiometric mixtures of 2e and AlCl3,
TiCl4, ReCl5 in CD3CN were identical. Rf 0.50 (PE/diethyl ether 1:1);
1H NMR (400 MHz, CD3CN): d=0.96 (t, J=7.0 Hz, 3H; CH2CH3), 1.47
(d, J=6.6 Hz, 3H; CHCH3), 2.41 (q, J=7.0 Hz, 2H; CH2CH3), 2.75 (dd,
J=5.1, 17.1 Hz, 1H; C(O)CHH), 2.89 (dd, J=7.8, 17.1 Hz, 1H;
C(O)CHH), 4.45 ppm (ddq, J=5.1, 6.6, 7.8 Hz, 1H; CHCl); 13C NMR
3-[3-(Benzylsulfanyl)propionyl]oxazolidin-2-one (28): Compound 28 was
prepared according to GP1 (Tf2NH, in CH2Cl2, 2 h) and was obtained as
a colourless oil (124 mg, 93%). Rf 0.27 (ether); 1H NMR (500 MHz,
CDCl3): d=2.74 (t, J=7.0 Hz, 2H; SCH2), 3.20 (t, J=7.0 Hz, 2H;
C(O)CH2), 3.76 (s, 2H; Ar-CH2), 3.99 (t, J=8.3 Hz, 2H; NCH2), 4.38 (t,
J=8.3 Hz, 2H; OCH2), 7.24 7.36 ppm (m, 5H; Ar-H); 13C NMR
(126 MHz, CDCl3): d=25.6, 35.2, 36.3, 42.5, 62.2 (CH2), 127.0, 128.5,
128.9 (Ar-C), 138.2 (ipso-Ar-C), 153.5, 171.6 ppm (C=O); IR (neat, film):
491
Chem. Eur. J. 2004, 10, 484 493
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

J. B. Spencer et al.
FULL PAPER
n˜ =2920, 1770, 1692, 1384, 1200, 1036, 757, 703 cmÀ1; HRMS [+ESI]
calcd for [C13H15NO3SNa]+ : 288.0670; found: 288.0669 [M+Na]+.
[5] a) L. Falborg, K. A. J˘rgensen, J. Chem. Soc. Perkin Trans. 1 1996,
2823 2826; b) M. P. Sibi, J. J. Shay, M. Liu, C. P. Jasperse, J. Am.
Chem. Soc. 1998, 120, 6615 6616; c) J. K. Myers, E. N. Jacobsen, J.
Am. Chem. Soc. 1999, 121, 8959 8960; d) M. P. Sibi, M. Liu, Org.
Lett. 2001, 3, 4181 4184; e) G. Cardillo, L. Gentilucci, M. Gianotti,
H. Kim, R. Perciaccante, A. Tolomelli, Tetrahedron: Asymmetry
2001, 12, 2395 2398; f) W. Zhuang, R. G. Hazell, K. A. J˘rgensen,
Chem. Commun. 2001, 1240 1241; g) K. Nakama, S. Seki, S. Kane-
masa, Tetrahedron Lett. 2002, 43, 829 832; h) H. Sugihara, K.
Daikai, X. L. Jin, H. Furuno, J. Inanaga, Tetrahedron Lett. 2002, 43,
2735 2739.
[6] M. J. Gaunt, J. B. Spencer, Org. Lett. 2001, 3, 25 28.
[7] T. C. Wabnitz, J. B. Spencer, Tetrahedron Lett. 2002, 43, 3891 3894.
[8] N. Srivastava, B. K. Banik, J. Org. Chem. 2003, 68, 2109 2114.
[9] S. Kobayashi, K. Kakumoto, M. Sugiura, Org. Lett. 2002, 4, 1319
1322.
3-(1H-Indol-3-yl)-1-phenylpentan-1-one (29a): Compound 29a was pre-
pared according to GP1 (Tf2NH, 508C, 30 min) and was obtained as a
colourless oil (131 mg, 95%). Rf 0.38 (CH2Cl2); 1H NMR (400 MHz,
CDCl3): d=0.96 (t, J=7.4 Hz, 3H; CH2CH3), 1.93 1.99 (m, 2H;
CH2CH3), 3.41 (dd, J=7.4, 16.2 Hz, 1H; C(O)CHH), 3.51 (dd, J=6.3,
16.2 Hz, 1H; C(O)CHH), 3.70 (m, 1H; Ar-CH), 6.98 (s, 1H; hetero-Ar-
H), 7.18 (dd, J=7.7, 8.0 Hz, 1H; Ar-H), 7.22 (dd, J=7.7, 8.0 Hz, 1H; Ar-
H), 7.33 (d, J=8.0 Hz, 1H; Ar-H), 7.45 (dd, J=6.9, 7.4 Hz, 2H; m-Ar-
H), 7.56 (t, J=7.4 Hz, 1H; p-Ar-H), 7.76 (d, J=8.0 Hz, 1H; Ar-H), 7.97
(d, J=6.9 Hz, 2H; o-Ar-H), 8.13 ppm (br, 1H; NH); 13C NMR
(126 MHz, CDCl3): d=12.3 (CH3), 28.5 (CH2), 34.7 (CH), 45.0 (CH2),
111.5 (Ar-C), 119.0 (ipso-Ar-C), 119.2, 119.4, 121.5, 121.9 (Ar-C), 126.8
(ipso-Ar-C), 128.2, 128.6, 133.0 (Ar-C), 136.7, 137.4 (ipso-Ar-C),
200.4 ppm (C=O); IR (neat, film): n˜ =3419, 2963, 1677, 1597, 1456, 1336,
1209, 904 cmÀ1; HRMS [EI, 70 eV] calcd for [C19H19NO]+ : 277.1467;
found: 277.1472 [M]+.
[10] a) T. Hosokawa, T. Shinohara, Y. Ooka, S.-I. Murahashi, Chem. Lett.
1989, 2001 2004; b) S. Ganguly, D. M. Roundhill, Organometallics
1993, 12, 4825 4832; c) K. J. Miller, T. T. Kitagawa, M. M. Abu-
Omar, Organometallics 2001, 20, 4403 4412.
[11] L. Dheilly, C. Lievre, C. Frechou, G. Demailly, Tetrahedron Lett.
1993, 34, 5895 5898.
5-(1H-Indol-3-yl)hexan-3-one (29b): Compound 29b was prepared ac-
cording to GP1 (Tf2NH, 508C, 2 h) and was obtained as a colourless oil
1
(95 mg, 88%). Rf 0.25 (CH2Cl2); H NMR (400 MHz, CDCl3): d=1.02 (t,
J=7.4 Hz, 3H; CH2CH3), 1.42 (d, J=7.0 Hz, 3H; CHCH3), 2.40 (q, J=
7.4 Hz, 2H; CH2CH3), 2.72 (dd, J=8.2, 15.8 Hz, 1H; C(O)CHH), 2.94
(dd, J=6.0, 15.8 Hz, 1H; C(O)CHH), 3.68 (m, 1H; Ar-CH), 6.92 (s, 1H;
hetero-Ar-H), 7.12 (dd, J=7.1, 7.8 Hz, 1H; Ar-H), 7.23 (dd, J=7.1,
7.8 Hz, 1H; Ar-H), 7.32 (d, J=7.8 Hz, 1H; Ar-H), 7.68 (d, J=7.8 Hz,
1H; Ar-H), 8.10 ppm (br, 1H; NH); 13C NMR (126 MHz, CDCl3): d=
7.8, 21.3 (CH3), 27.2 (CH), 39.5, 50.3 (CH2), 111.4, 119.2, 119.3, 120.3
(Ar-C), 121.1 (ipso-Ar-C), 122.9 (Ar-C), 126.3, 136.6 (ipso-Ar-C),
211.5 ppm (C=O); IR (neat, film): n˜ =3409, 2973, 1702, 1457, 1339, 1098,
906 cmÀ1; HRMS [EI, 70 eV] calcd for [C14H17NO]+ : 215.1310; found:
215.1319 [M]+.
[12] M. Bandini, P. G. Cozzi, M. Giacomini, P. Mechiorre, S. Selva, A.
Umani-Ronchi, J. Org. Chem. 2002, 67, 3700 3704.
[13] a) S. Kanemasa, Y. Oderaotoshi, E. Wada, J. Am. Chem. Soc. 1999,
121, 8675 8676; b) M. Saito, M. Nakajima, S. Hashimoto, Chem.
Commun. 2000, 1851 1852; c) M. Saito, M. Nakajima, S. Hashimo-
to, Tetrahedron 2000, 56, 9589 9594; d) S. Kobayashi, C. Ogawa, M.
Kawamura, M. Sugiura, Synlett 2001, 983 985.
[14] T. C. Wabnitz, J. B. Spencer, Org. Lett. 2003, 5, 2141 2144.
[15] T. C. Wabnitz, J.-Q. Yu, J. B. Spencer, Synlett 2003, 1070 1072.
[16] Reviews: a) M. P. Sibi, N. A. Porter, Acc. Chem. Res. 1999, 32, 163
171; b) F. W. Stacey, J. F. Harris, Jr., in OrganicReatcions, Vol. 13
(Ed.: A. C. Cope), Wiley, New York, 1963, pp. 150 376.
[17] a) M. Maumy, P. Capdeville, Bull. Soc. Chim. Fr. 1995, 132, 734
738; b) B. A. Jazdzewski, P. L. Holland, M. Pink, V. G. Young, Jr.,
D. J. E. Spencer, W. B. Tolman, Inorg. Chem. 2001, 40, 6097 6107.
[18] See also Supporting Information for kinetic data.
1-(1H-Indol-3-yl)-pentan-3-one (29c): Compound 29c was prepared ac-
cording to GP1 (Tf2NH, 12 h) and was obtained as a colourless solid
(98 mg, 98%). Rf 0.25 (CH2Cl2); m.p. 92 948C; 1H NMR (400 MHz,
CDCl3): d=1.06 (t, J=7.3 Hz, 3H; CH2CH3), 2.42 (q, J=7.3 Hz, 2H;
CH2CH3), 2.84 (t, J=7.3 Hz, 2H; C(O)CH2), 3.08 (t, J=7.3 Hz, 2H; Ar-
CH2), 6.95 (s, 1H; hetero-Ar-H), 7.12 (dd, J=7.7, 8.0 Hz, 1H; Ar-H),
7.22 (dd, J=7.7, 8.0 Hz, 1H; Ar-H), 7.36 (d, J=8.0 Hz, 1H; Ar-H), 7.63
(d, J=8.0 Hz, 1H; Ar-H), 8.06 ppm (br, 1H; NH); 13C NMR (100 MHz,
CDCl3): d=7.9 (CH3), 19.6, 36.2, 42.9 (CH2), 111.3 (Ar-C), 115.4 (ipso-
Ar-C), 118.8, 119.4, 121.7, 122.1 (Ar-C), 127.3, 136.5 (ipso-Ar-C),
211.7 ppm (C=O); IR (neat, film): n˜ =3315, 3059, 2936, 1702, 1331, 1219,
1112, 1046 cmÀ1; HRMS [EI, 70 eV] calcd for [C13H15NO]+ : 201.1154;
found: 201.1148 [M]+.
[19] a) A. L. Seligson, W. C. Trogler, Organometallics 1993, 12, 744 751;
b) Z. Wang, Z. Zhang, X. Lu, Organometallics 2000, 19, 775 780;
c) M. Kawatsura, J. F. Hartwig, J. Am. Chem. Soc. 2000, 122, 9546
9547.
[20] CH2Cl2 was used for the original PdII-catalysed aza-Michael addition
reactions (see ref. [6]), but CH3CN and CH3NO2 were preferred sol-
vents due to better solubility of Lewis acid catalysts and faster reac-
tions in most cases.
[21] T. Ishikawa, E. Uedo, R. Tani, S. Saito, J. Org. Chem. 2001, 66, 186
191.
[22] CCDC-214434 contains the supplementary crystallographic data for
this paper. These data can be obtained free of charge via
www.ccdc.cam.ac.uk/conts/retrieving.hmtl (or from the Cambridge
Crystallographic Data Centre, 12 Union Road, Cambridge CB2
1EZ, UK; fax: (+44) 1223 336 033; or email: deposit@ccdc.cam.ac.
uk).
Acknowledgement
We thank Prof. Dr. C. B. W. Stark for the helpful discussions and valua-
ble suggestions. Financial support (T.C.W.) by the Fonds der Chemischen
Industrie, the Studienstiftung des deutschen Volkes and St. John×s Col-
lege, Cambridge, is gratefully acknowledged. J.-Q.Y. also thanks St.
John×s College, Cambridge, for a fellowship.
[23] To facilitate identification, H3 11 was also prepared by analogous
reaction of 2d with CH3CN and H2O (see Experimental Section).
[24] For Ritter reactions of acetonitrile and enones, see: P. J. Scheuer,
H. C. Botelho, C. Pauling, J. Org. Chem. 1957, 22, 674 676.
[25] H. C. Brown, B. Kanner, J. Am. Chem. Soc. 1966, 88, 986 992.
[26] C. F. Baes, R. E. Mesmer, The Hydrolysis of Cations, Wiley-Inter-
science, New York, 1976.
[27] L. G. Sillÿn, A. E. Martell, Stability Constants of Metal-Ion Com-
plexes, Chemical Society Special Publications 17 and 25 (Supplement
1), The Chemical Society, London, 1964, 1971.
[1] a) N. Sokoloff, P. Latschinoff, Ber. Dtsch. Chem. Ges. 1874, 7, 1384
1387; b) F. Loydl, Justus Liebigs Ann. Chem. 1878, 192, 80 88.
[2] a) T. Komnenos, Justus Liebigs Ann. Chem. 1883, 218, 145 169;
b) A. Michael, Am. Chem. J. 1887, 9, 112 129; c) A. Michael, J.
Prakt. Chem. 1887, 35, 349 356.
[3] Reviews: a) M. Liu, M. P. Sibi, Tetrahedron 2002, 58, 7991 8035;
b) E. Juaristi, H. Lopez-Ruiz, Curr. Med. Chem. 1999, 6, 983 1004.
[4] a) S. Matsubara, M. Yoshioka, K. Utimoto, Chem. Lett. 1994, 827
830; b) M. Pÿrez, R. Pleixats, Tetrahedron 1995, 51, 8355 8362;
c) T.-P. Loh, L.-L. Wei, Synlett 1998, 975 976; d) G. Bartoli, M.
Bosco, E. Marcantoni, M. Petrini, L. Sambri, E. Torregiani, J. Org.
Chem. 2001, 66, 9052 9055; e) M. Kawatsura, J. F. Hartwig, Organo-
metallics 2001, 20, 1960 1964.
[28] A correlation between catalytic activity and hydrolysis constants
and water exchange rate constants has been established for reactions
in water, see: S. Kobayashi, S. Nagayama, T. Busujima, J. Am.
Chem. Soc. 1998, 120, 8287 8288.
[29] J. Burgess, Metal Ions in Solution, Horwood, Chichester, 1978.
[30] R. N. Sylva, P. L. Brown, G. E. Batley, J. Chem. Soc. Dalton Trans.
1985, 1967 1970.
[31] P. L. Brown, J. Ellis, J. Chem. Soc. Dalton Trans. 1983, 35 36.
492
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 484 493

Lewis Acid Mediated Hetero-Michael Addition Reactions
484 493
[32] G. H. Khoe, P. L. Brown, R. N. Sylva, J. Chem. Soc. Dalton Trans.
1986, 1901 1906.
[46] Chiral HfIV-complexes have also been used. However, the mecha-
nism of this reaction is not clear (see ref [13d]).
[33] P. L. Brown, R. J. Sylva, J. Chem. Soc. Dalton Trans. 1980, 1577
1581.
[34] Due to the wide signal separation, the extent of protonation can
even be assessed in the presence of paramagnetic metal ions which
cause considerable line broadening.
[35] Yields of the [Pd(CH3CN)2Cl2]-catalysed aza-Michael addition of 3
and 2a: 93% (CH2Cl2, 12 h), 0% (CH3CN, 12 h).
[36] a) S. Kobayashi, Synlett 1994, 689 701; b) K. Mikami, M. Terada, H.
Matsuzawa, Angew. Chem. 2002, 114, 3704 3722; Angew. Chem.
Int. Ed. 2002, 41, 3554 3571.
[37] B. Shelimov, J.-F. Lambert, M. Che, B. Didillon, J. Am. Chem. Soc.
1999, 121, 545 556.
[38] F. A. Cotton, S. J. Lippard, Inorg. Chem. 1966, 5, 9 16.
[39] Approximately 10% of the b-chloroketone 12 were detected when
the aza-Michael addition reactions mediated by ReCl5 (10%) was
monitored by NMR spectroscopy (in CD3CN).
[47] It is worth noting that the p-interaction of an electron-rich double
bond, as for example in b-methylstyrene, with PdII species is still re-
sponsible for related reactivities [see: a) R. G. F. Giles, V. R. Lee
Son, M. V. Sargent, Aust. J. Chem. 1990, 43, 777 781; b) J.-Q. Yu,
M. J. Gaunt, J. B. Spencer, J. Org. Chem. 2002, 67, 4627 4629]. This
is confirmed by the observation that the isomerisation of cis-b-meth-
ylstyrene to the trans-isomer is not inhibited by the addition of the
non-coordinating ligand 2,6-di-tert-butylpyridine.
[48] K. Ishihara, S. Funahashi, M. Tanaka, Inorg. Chem. 1986, 25, 2898
2901.
[49] G. J. Kubas, Inorg. Synth. 1990, 28, 67 70.
[50] J. N‰slund, I. Persson, M. Sandstrˆm, Inorg. Chem. 2000, 39, 4012
4021.
[51] D. Enders, J. Zhu, L. A. Kramps, Liebigs Ann./Recl. 1997, 1101
1113.
[52] K. Narasaka, H. Kusama, Y. Hayashi, Bull. Chem. Soc. Jpn. 1991,
64, 1471 1478.
[53] A. C. Cope, C. M. Hofmann, C. Wyckoff, E. Hardenbergh, J. Am.
Chem. Soc. 1941, 63, 3452 3459.
[54] C. F. H. Allen, F. W. Spangler, OrganicSyntheses, Collective Volume
III 1955, pp. 377 379.
[55] R. N. Salvatore, F. Chu, A. Nagle, E. A. Kapxhiu, R. M. Cross,
K. W. Jung, Tetrahedron 2002, 58, 3329 3348.
[56] Most aza- and oxa-Michael addition reactions described in this
report were reversible.
[57] Low signal intensities did not permit detection of the CD3 moieties
in 10 and 11 due to 13C 2H coupling.
[40] F. Lengfeld, Am. Chem. J. 1901, 26, 324 332.
[41] Identical lmax (227 nm, 321 nm) have been obtained by UV/Vis spec-
troscopy for solutions of AuCl and AuCl3 in CH3CN (see Supporting
Information).
[42] a) L. Tottie, P. Baeckstrˆm, C. Moberg, J. Tegenfeldt, A. Heumann,
J. Org. Chem. 1992, 57, 6579 6587; b) G. H. Posner, H. Dai, D. S.
Bull, J.-K. Lee, F. Eydoux, Y. Ishihara, W. Welsh, N. Pryor, S.
Petr, Jr., J. Org. Chem. 1996, 61, 671 676; c) S. E. Sen, S. M. Smith,
K. A. Sullivan, Tetrahedron 1999, 55, 12657 12698; d) A. Fujii, E.
Hagiwara, M. Sodeoka, J. Am. Chem. Soc. 1999, 121, 5450 5458.
[43] B. Bonnet, J. Roziere, R. Fourcade, G. Mascherpa, Can. J. Chem.
1974, 52, 2077 2084.
[58] It is conceivable that one of these species is an oxazine derivative,
which can be formed in acid-catalysed Ritter reactions, see ref. [24]
[44] B. Coupez, C. Boehme, G. Wipff, Phys. Chem. Chem. Phys. 2002, 4,
5716 5729.
[45] Moderate ees have been obtained in aza-Michael addition reactions
with chiral TiIV and CuII catalysts, see refs [5a, e].
Received: July 29, 2003 [F5407]
493
Chem. Eur. J. 2004, 10, 484 493
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim