Enantioselective addition of cyclic enol silyl ethers to 2-alkenoyl-pyridine-N-oxides catalysed by CuII-Bis(oxazoline) complexes

Article abstract of DOI:10.1002/chem.201201214

Asymmetric reactions involving (E)-3-aryl-1-(pyridin-2-yl-N-oxide)prop-2- en-1-ones and cyclic enol silyl ethers show good yields and excellent enantioselectivities (up to 99.9 % ee) when catalysed by bis(oxazoline)-Cu ^II complexes. Different reaction pathways can be followed by different enol silyl ethers: with 2-(trimethylsilyloxy)furan, a Mukaiyama-Michael adduct is obtained, whereas a hetero Diels-Alder cycloadduct was formed by using (1,2-dihydronaphthalen-4-yloxy)trimethylsilane. In the latter reaction, the absolute configuration of the product is consistent with a reagent approach to the less hindered Re face of the coordinated substrate in the reactive complex. Copyright

Full text of DOI:10.1002/chem.201201214

FULL PAPER
DOI: 10.1002/chem.201201214
Enantioselective Addition of Cyclic Enol Silyl Ethers to 2-Alkenoyl-Pyridine-
N-Oxides Catalysed by CuII–Bis(oxazoline) Complexes
Alessandro Livieri,[a] Massimo Boiocchi,[b] Giovanni Desimoni,[a] and Giuseppe Faita*[a]
Abstract: Asymmetric reactions involving (E)-3-aryl-1-(pyridin-2-yl-N-oxide)prop-
2-en-1-ones and cyclic enol silyl ethers show good yields and excellent enantiose-
lectivities (up to 99.9% ee) when catalysed by bis(oxazoline)–CuII complexes. Dif-
ferent reaction pathways can be followed by different enol silyl ethers: with 2-(tri-
methylsilyloxy)furan, a Mukaiyama–Michael adduct is obtained, whereas a hetero
Diels–Alder cycloadduct was formed by using (1,2-dihydronaphthalen-4-yloxy)tri-
methylsilane. In the latter reaction, the absolute configuration of the product is
consistent with a reagent approach to the less hindered Re face of the coordinated
substrate in the reactive complex.
Keywords: asymmetric catalysis
·
copper · cycloaddition · enol silyl
ethers · Mukaiyama reactions · ni-
trogen heterocycles
Introduction
catalysts”.[4] Within C2-symmetric ligands, bis(oxazoline)
(BOX) and, later, pyridine bis(oxazoline) (PyBOX) deriva-
tives gained a preeminent position because of their quali-
ties.[5] Several substrates have been investigated in bis(oxa-
zoline) asymmetric catalysis, and 2-alkenoyl-pyridine-N-
oxides 1 have emerged as promising substrates in the last
five years. This structure provides access to several products
due to its wide reactivity (Scheme 1), and it also allows the
corresponding reduced pyridine derivatives to be obtained.[6]
The first paper about this topic dealt with the Diels–
Alder reaction[7] of 1 acting as a dienophile (Scheme 1a).
Later, the enantioselective Friedel–Crafts alkylation with in-
doles[8a] and pyrroles[8b] catalysed by [CuII–PyBOX] com-
plexes was investigated (Scheme 1d). Afterwards, the in-
verse-electron-demand hetero Diels–Alder reaction[9] with
electron-rich alkenes was explored, as well as the asymmet-
ric 1,3-dipolar cycloaddition,[10] which was further investigat-
ed by Barroso et al. (Scheme 1b,c).[11] More recently, the Mi-
chael addition of malonates to 2-alkenoyl-pyridine-N-oxides
was usefully catalysed by ZnII–BOX complexes (Sche-
me 1e).[12]
Pyridine-N-oxides can act as donor-forming complexes with
Lewis acids. If the pyridine ring is functionalised at posi-
tion 2 with another coordinating group, such as a carbonyl
function, this structure can behave as a bidentate ligand.
This feature allows the use of 2-carbonyl-pyridine-N-oxide
derivatives as suitable substrates in metal catalysis. Jørgen-
sen and co-workers[1a] noted that the oxidation of the pyri-
dine nitrogen atom into the corresponding N-oxide is indis-
pensable, in order to obtain an appropriate coordination
around the Lewis acid. Furthermore, the better coordination
ability of the pyridine-N-oxides instead the corresponding
pyridines in terms of reversible–irreversible coordination to
the metal cation has been demonstrated (for example, use
of the pyridine-N-oxide avoids poisoning of the Osmium-
based catalyst in the Sharpless aminohydroxylation).[1b,c] Ac-
tually, the first paper about an asymmetric reaction with
such derivatives dealt with the asymmetric carbonyl reduc-
tion by bakerꢀs yeast.[2] If the Lewis acid, usually a metal
cation, is coordinated to a chiral ligand (L*), an asymmetric
reactive complex is formed. In recent decades, a huge inves-
tigation has been made in the field of asymmetric ligands[3]
and those with C2 symmetry emerged as “privileged chiral
[a] Dr. A. Livieri, Prof. G. Desimoni, Prof. G. Faita
Department of Organic Chemistry
University of Pavia
V. le Taramelli, 10; 27100 Pavia (Italy)
Fax : (+39)0382-987323
E-mail: faita@unipv.it
[b] Dr. M. Boiocchi
Centro Grandi Strumenti
University of Pavia
V. Bassi, 8; 27100 Pavia (Italy)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201201214.
Scheme 1. Structure and reactivity in the asymmetric catalytic processes
of 2-alkenoyl-pyridine-N-oxides 1.
Chem. Eur. J. 2012, 00, 0 – 0
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
1
&
ÞÞ
These are not the final page numbers!

Scheme 3. Structures of the bis(oxazoline) ligands 2a–d used in this
work.
yl-pyridine-N-oxide 1a. The reactions were carried out in
anhydrous dichloromethane with 3 ꢂ molecular sieves, and
1,1,1,3,3,3-hexafluoro-2-propanol (HFIP) was used as an ad-
ditive (Scheme 4). All of the reactions produced discrete (at
Scheme 2. Structures of the CuII-based complex with BOX 2d and the
corresponding reactive intermediate with 1a, which reacts in a Diels–
Alder reaction to give the cycloadduct 5. TIPS: triisopropylsilyl; Tf: tri-
fluoromethanesulfonyl.
In a previous paper, the CuII complex of a new BOX was
isolated and characterised by X-ray diffraction (3,
Scheme 2).[13] By starting from this complex, the structure of
the reactive intermediate involving substrate 1a was then
determined by X-ray crystal analysis of complex 4.[11] The
reacting substrate, 1a, is coordinated to the cation by its two
oxygen atoms to form a square-pyramidal complex, as ob-
served in other structures in which oxygen coordination
leads to square-planar geometries; this reflects the larger d-
orbital splitting of the cation.[14]
Scheme 4. Mukaiyama–Michael reaction between the (E)-cinnamoyl-pyr-
idine-N-oxide 1a and 2-(trimethylsilyloxy)furan (6). TMS: trimethylsilyl.
À208C) or quantitative yields (at ambient temperature) of
the Mukaiyama–Michael products (Table 1). The reaction is
highly stereoselective, because only adduct anti-7a, the rela-
The Diels–Alder cycloaddition of cyclopentadiene to 4
gave cycloadduct 5 with quantitative yields and with almost
total control of the stereochemistry, with a stereochemical
outcome that was fully consistent with a diene approach to
the less hindered Re face of the coordinated dienophile.
In this paper, the scope of catalyst 3 to new reactions will
be explored. In particular, the addition of different cyclic
enol silyl ethers to 1, to give both cyclic and open chain ad-
ducts, will be described.
Table 1. Mukaiyama–Michael reaction between 1a and 6 in CH2Cl2 cata-
lysed by CuACHTNUTRGNE(NUG OTf)2/BOX catalysts (5 mol%).
Entry
Ligand
T
t [h]
Yield [%]
(+)-7a ee[a] [%]
1
2
3
4
2a
2d
2a
2b
À208C
À208C
RT
24
24
18
18
75
72
quant
quant
86
80
79
RT
À80[b]
[a] Determined by chiral HPLC (see the Experimental Section). [b] The
opposite enantiomer was obtained.
Results and Discussion
tive configuration of which was confirmed by X-ray crystal
analysis (Figure 1), was obtained. The absolute configuration
of (+)-7a cannot be determined from the crystal structure
because heavy-element atoms that would cause significant
anomalous dispersion effects are not present in the structure
and the diffraction data collected with MoKa radiation are
not suitable for absolute structure assignment performed
with the probabilistic approach based on Bijovet pair inten-
sity differences.[16]
(E)-Cinnamoyl-pyridine-N-oxides 1 were prepared accord-
ing to the procedure reported in the literature.[9] This prepa-
ration includes the oxidation of 2-acetylpyridine to the cor-
responding N-oxide by meta-chloroperoxybenzoic acid
(mCPBA), followed by condensation under basic conditions
with the appropriate aromatic aldehyde to give (E)-1.
All of the catalysts were CuACHTNUGRTENUNG(OTf)2 based complexes of
the BOX ligands 2a–d illustrated in Scheme 3. Ligands with
structures similar to 2d have been reported as asymmetric
catalysts giving low ee values in the epoxidation of carbonyl
compounds[15a] and from moderate to high ee values in the
cyclopropanation of furans.[15b]
The use of the catalyst based on (R)-phenyl-BOX 2a gave
anti-7a with up to 86% ee of the dextrorotatory enantiomer
(Table 1, entry 1). An increase in the temperature from
À208C to ambient temperature allowed a quantitative yield
of the adduct to be obtained, although the enantioselectivity
decreased (Table 1, entries 1 versus 3). The use of the 2d-
based catalyst, despite the opposite configuration of the C4
The first reaction investigated was the Mukaiyama–Mi-
chael addition of 2-(trimethylsilyloxy)furan (6) to cinnamo-
&
2
&
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

CuII–Bis(oxazoline)-Catalysed Enantioselective Addition to Pyridine-N-Oxides
FULL PAPER
were good and the use of the 2d-based catalyst allowed a
single enantiomer to be obtained, even if with moderate re-
action yields (Table 2, entry 4). Also in this case, the use of
either the 2a- or 2d-based catalyst, despite the opposite
configurations of the C4 stereocentres of the oxazoline
rings, produced the same dextrorotatory enantiomer of 9.
In order to better understand the reaction mechanism, the
study was extended to a benzo-condensed derivative of 8;
the reaction between 1 and 11 was then explored under the
same conditions (Scheme 6 and Table 3).
Figure 1. ORTEP view of the X-ray structure of anti-7a (ellipsoids are
drawn at the 30% probability level; red: O, blue: N and grey: C atoms).
stereocentres in the oxazoline rings, produced the same dex-
trorotatory enantiomer as that obtained by using BOX 2a
(Table 1, entries 1 and 2).
The addition of 1-(trimethylsilyloxy)cyclohexene (8) to 2-
alkenoyl-pyridine-N-oxide 1a was then investigated
(Scheme 5 and Table 2). The reaction proceeded with mod-
erate yields to give a mixture of the diastereomeric products
Scheme 6. Reaction between the (E)-cinnamoyl-pyridine-N-oxides 1a
and 1b and (1,2-dihydronaphthalen-4-yloxy)trimethylsilane (11).
Table 3. Reaction between 1a,b and 11 in CH2Cl2 catalysed by Cu-
AHCTUNRTGEG(NNNU OTf)2/BOX catalysts (5 mol%).
Entry Ligand
t
T
Yield 12/13/14
(+)-12
(+)-13
[h]
[8C] [%]
ee[a] [%] ee[a] [%]
1
2
3
4
2a
2b
2c
2d
2d
2d
2d
70 À20 99
70 À20 99
15 À20 95
65:29:6
71:25:4
49:44:7
71:22:7
20:69:11
85:13:2
20:66:14
À35[b]
36
99
99.9
99.9
99.9
99.8
65
À70[b]
75
Scheme 5. Reaction between the (E)-cinnamoyl-pyridine-N-oxide 1a and
1-(trimethylsilyloxy)cyclohexene (8).
1
À20 82
n.d.[c]
99
5
15 À20 94
140 À70 99
60 À20 67
6
n.d.[c]
98
7[d]
Table 2. Reaction between 1a and 8 in CH2Cl2 at À208C catalysed by
CuACHTUNGTRENNUNG(OTf)2/BOX catalysts (5 mol%).
[a] Determined by chiral HPLC (see the Experimental Section). [b] The
opposite enantiomer was obtained. [c] n.d.: not determined. [d] Reaction
carried out with 1b instead of 1a.
Entry
Ligand
t
Equivalents
Yield
[%]
9/10
(+)-9
[h]
of 8
ee[a] [%]
1
2
3
4
2a
2a
2b
2d
140
23
140
23
1.5
3.0
1.5
3.0
46
60
46
30
85:15
95:5
86:14
96:4
90
88
À89[b]
99
The reaction proceeded with almost quantitative yields,
and three products were isolated from the reaction mixture:
the hetero Diels–Alder adducts 12a,b and the syn/anti prod-
ucts 13a,b and 14a,b (Table 3).
[a] Determined by chiral HPLC (see the Experimental Section). [b] The
opposite enantiomer was obtained.
The use of the Ph-BOX-based catalysts did not show a
particular selectivity, because the ratio of 12 to 13+14 was
about 2:1, and all of the products were obtained with low to
moderate enantioselectivities (Table 3, entries 1 and 2). The
tBu-BOX-based catalyst strongly increased the enantioselec-
tivity of the reaction, and the major reaction product,
(+)-12, was obtained with up to 99% ee (Table 3, entry 3).
The structures of (+)-12a and (+)-13a were determined
by X-ray analysis, allowing (+)-12a to be assigned with the
(4S,4aR,10bR) absolute configuration (Figure 2). The cis
9 and 10. The two adducts were isolated, but it was not pos-
sible to assign the relative configurations of the C4, C4a and
C8a stereocentres. However, the relative ratio of the two ad-
ducts was determined, and the major one was obtained with
diastereomeric ratios ranging from 85:15 to 95:5.
Despite the unknown product structure, it was possible to
determine the enantomeric composition of the major adduct
(see the Experimental Section for details). The ee values
Chem. Eur. J. 2012, 00, 0 – 0
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
3
&
ÞÞ
These are not the final page numbers!

G. Faita et al.
with excellent enantioselectivity (Table 3, entry 4). The ap-
plication of prolonged reaction times shifted the reaction se-
lectivity towards the formation of the open chain product
(+)-13a (Table 3, entry 5), whereas if the reaction was per-
formed at a lower temperature (À708C), the favoured reac-
tion product was again the hetero Diels–Alder adduct
(+)-12a (Table 3, entry 6). A simple explanation of these re-
sults may be proposed by considering (+)-12a as a primary
reaction product that, under prolonged reaction times at
higher temperatures, opens into (+)-13a. This rationale was
easily confirmed by stirring (+)-12a at ambient temperature
for 8 h in CH2Cl2 with a catalytic amount of CuACHTUNTRGENUG(N OTf)2. The
starting product was totally converted into (+)-13a with
total retention of the enantiomeric purity (99.9% ee)
(Scheme 7). Consequently, these data allowed us to assign
the absolute configuration of structure 13 by considering the
retention of configuration at the benzylic stereocentre.
Figure 2. ORTEP view of the X-ray structure of (+)-(4S,4aR,10bR)-12a
(ellipsoids are drawn at the 30% probability level; red: O, blue: N, grey:
C and yellow: Si atoms).
configuration of the 4a and 8a substituents is in agreement
with a hetero Diels–Alder mechanism, with dienophile 11
approaching with an endo approach to the Re face of 1.
For the open-chain product (+)-13a, the absolute configu-
ration cannot be directly assigned because no heavy ele-
ments are present; however, the X-ray crystal structure
(Figure 3) clearly confirmed the syn disposition of the sub-
stituents in (+)-13a.
Scheme 7. Proposed mechanism for the conversion of (+)-12a into
(+)-13a.
Product (+)-14a,b has been also prepared by Michael ad-
dition of a-tetralone (15) to 1a,b (Scheme 8). This reaction
is slower than that reported in Scheme 6, and higher temper-
ature (from À208C to room temperature), a nucleophile
Scheme 8. Reaction between the (E)-cinnamoyl-pyridine-N-oxides 1a
and 1b and a-tetralone (15).
Figure 3. ORTEP view of the X-ray structure of (+)-(R,S)-13a (ellipsoids
are drawn at the 30% probability level; red: O, blue: N and grey: C
atoms).
excess (from 3 to 25 equiv), higher catalyst loading (from 5
to 10% mol), and prolonged reaction times are required to
obtain good reaction yields (Table 4). The 2a,b-based chiral
catalysts gave unsatisfactory results: reaction yields and ster-
eoselectivities were moderate and the enantioselectivities
More interesting results were obtained with the 2d-based
chiral catalyst. If the reaction was performed for a short re-
action time (1h), the major product obtained was (+)-12
&
4
&
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

CuII–Bis(oxazoline)-Catalysed Enantioselective Addition to Pyridine-N-Oxides
Table 4. Reaction between 1a,b and 15 in CH2Cl2 at RT catalysed by Cu-
FULL PAPER
Conclusion
ACHTUNGTRENNUNG
t
Yield 13/14 (+)-13
(+)-14
Copper triflate complexes with chiral bis(oxazolines) ligands
2 show catalytic activity in the reaction between (E)-cinna-
moyl-pyridine-N-oxides 1a,b and different enol silyl ethers.
In particular, asymmetric induction by ligand 2d gives excel-
lent results in almost all of the reactions tested, with ee
values up to 99.9%.
The reaction mechanisms have not been deeply investigat-
ed, although some spectroscopic data suggest different path-
ways for derivatives 9 and 12. The structure of (+)-12a con-
firms that the reaction catalysed by 1d is a hetero Diels–
Alder cycloaddition, in which dienophile component 11 ap-
proaches substrate 1 from the less hindered Re face.
[d] [%]
ee[a] [%] ee[a] [%]
1
2
3
4
1a
1a
1a
1b
2a
2b
2d
2d
20
20
4
56
53
99
99
73:27 À25[b]
48
À35[b]
8
71:29
92:8
26
94
92
3
82:12
n.d.[c]
[a] Determined by chiral HPLC (see the Experimental Section). [b] The
opposite enantiomer was obtained. [c] n.d.: not determined.
were negligible. By contrast, the 2d-based catalyst was more
efficient and syn selective, and (+)-13 was obtained with up
to 94% ee (Table 4, entries 3 and 4).
We focused our attention on ligand 2d, because it has
been developed by our group. Our interest further increased
from observing that it allows us to obtain from high to
almost complete asymmetric induction. Another point of in-
terest is that we had previously characterised the reactive
complex between 1, CuII and 2d. In the reaction of 1a with
enol silyl ether 11, it has been also possible to determine the
Re face as the reactive one, as previously observed in either
the Diels–Alder reaction of 1 with cyclopentadiene or the
1,3-dipolar cycloaddition with diphenylnitrone.[10] In those
reactions, opposite stereochemical outcomes were observed
with ligands 2b and 2d (or 2c). There are two explanations
for this opposite stereochemical induction, one proposed by
Evans et al.[17] and the other by Jørgensen and co-workers.[18]
Evans et al. suggested a square-planar structure for both re-
active complexes; with 2c (or 2d), the reagent approaches
the substrate from the opposite side to the bulky tert-butyl
(or TIPS) group, whereas it would approach syn to the
phenyl ring with 2b because of the electronic stabilisation of
the transition state. Jørgensen and co-workers explained the
reverse of induction with flexibility of the reactive complex
(between square-planar and tetrahedral geometries) and the
ability of the chiral ligand to rock between pseudo-axial and
pseudo-equatorial positions.
Experimental Section
General: Melting points were determined by the capillary method and
are uncorrected. 1H and 13C NMR spectra were recorded at 300 and
75 MHz, respectively, and COSY and NOE experiments were performed
in order to confirm product structures. IR spectra were registered on a
Perkin–Elmer RX I spectrophotometer. Optical rotations were measured
on a Perkin–Elmer 241 polarimeter. Separation and purification of the
products was carried out by column chromatography with Merck silica
gel 60 (230–400 mesh). The enantiomeric excesses (ee) of the products
were determined by HPLC with Daicel columns.
Materials: Dichloromethane was the hydrocarbon-stabilised Aldrich ACS
grade, distilled from calcium hydride and used immediately. Other sol-
vents were dried according to standard procedures. Copper triflate was
the Aldrich ACS reagent. Powdered molecular sieves (3 ꢂ) from Aldrich
were heated under vacuum at 3008C for 5h and kept in sealed vials in a
dryer. 2-Alkenoyl-pyridine-N-oxides 1a and 1b were prepared as report-
ed in the literature.[8,11]
General method for the enol silyl ether addition reaction: In anhydrous
glassware, 2 (0.015mmol), copper triflate (0.005g, 0.015mmol) and pyri-
dine-N-oxide derivative 1a or 1b (0.3mmol) were added to a test tube.
The reactants were dissolved in freshly dried dichloromethane (0.5mL)
and magnetically stirred for 10min at room temperature. Then, 3 ꢂ mo-
lecular sieves (40mg) and 1,1,1,3,3,3-hexafluoro-2-propanol (31 mL,
0.3mmol) were added to the reaction mixture at the appropriate reaction
temperature. After 15min, enol silyl ether 6 (66 mL, 0.4mmol), 8 (174 mL,
0.9mmol) or 11 (130 mL, 0.6mmol) was added to the tube. The reaction
was monitored by TLC and stopped after completion. The reaction mix-
ture was quenched in water and extracted with dichloromethane (3ꢃ
10mL), and the organic phase was dried over sodium sulphate. Products
were isolated by column chromatography.
In this work, we observed this inversion when 1 reacts
with 6 and 8, but not with 11 and 15. It should be noted
these latter reactions are those in which ligand 2a,b showed
the worst catalytic activity leading to low ee values. Maybe
the determination of the reactive complex between 1, CuII
and 2a,b would lead to important information about this un-
usual inversion. This goal has not been reached so far be-
cause of non-crystalline behaviour of the complex.
Mukaiyama–Michael reaction of 1 with 2-(trimethylsilyloxy)furan (6): 2-
((R)-3-((R)-5-Oxo-2,5-dihydrofuran-2-yl)-3-phenylpropanoyl)pyridine-N-
oxide (7a) was obtained by column chromatography (eluent: ethyl ace-
tate) and recrystallised from ethyl acetate/hexane to give a white solid;
m.p. 135–1378C; [a]D25 =42.2 (c=0.4, CHCl3); 1H NMR (CDCl3,
300 MHz): d=8.16 (d, 1H, J=6.3 Hz), 7.49 (dd, 1H, J=7.8, 2.2 Hz),
7.33–7.24 (m, 7H), 6.06 (dd, 1H, J=5.7, 1.9 Hz), 5.18 (dt, 1H, J=8.1,
1.7 Hz), 3.92–3.72 (m, 2H), 3.64 ppm (q, 1H, J=7.1); 13C NMR (CDCl3,
75 MHz): d=195.4, 172.2, 154.9, 146.4, 140.2, 138.4, 128.7, 128.6, 128.1,
127.9, 127.6, 127.3, 127.1, 126.9, 125.5, 122.0, 85.8, 45.1, 45.0 ppm; chiral
HPLC analysis on Chiralpak AD column: eluent: isopropanol/hexane
20:80; flow rate: 1.0 mLminÀ1; tR: 30.8 ((+)-7a), 34.0 min ((À)-7a).
At this stage, it is not possible to determine the reactive
1
pathway followed in the first two reactions. H NMR spectra
together with NOE experiments suggested different struc-
tures for the pyran derivatives 9 and 12. In particular, it is
possible that 9 arose from tandem reactions, with an initial
Mukaiyama–Michael addition followed by an intramolecular
cyclisation. This curious behaviour has been already report-
ed for analogous reactions of 8 with (E)-2-oxo-4-aryl-3-bute-
noates.[19] Hopefully, crystallisation of 9 will allow us to con-
firm these theoretical considerations.
Reaction of 1 with 1-(trimethylsilyloxy)cyclohexene (8): 2-(4-Phenyl-8a-
(trimethylsilyloxy)-4a,5,6,7,8,8a-hexahydro-4H-chromen-2-yl)pyridine-N-
oxide (9) was obtained by column chromatography (eluent: ethyl acetate/
cyclohexane 4:6) as
a
colourless oil; [a]2D5 =31.4 (c=0.8, CHCl3);
1H NMR (CDCl3, 300 MHz): d=8.30 (dt, 1H, J=6.5, 0.8 Hz), 7.90 (dd,
Chem. Eur. J. 2012, 00, 0 – 0
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
5
&
ÞÞ
These are not the final page numbers!

G. Faita et al.
1H, J=8.2, 2.0 Hz), 7.82 (d, 1H, J=5.2 Hz), 7.25 (m, 7H), 3.67 (dd, 1H,
J=8.0, 5.1 Hz), 2.16 (m, 2H), 1.74 (dt, 2H, J=9.9, 3.1 Hz), 1.63 (m, 2H),
1.41 (m, 2H), 1.04 (dq, 1H, J=12.7, 3.4 Hz), 0.06 ppm (s, 9H); 13C NMR
(CDCl3, 75 MHz): d=143.1, 141.0, 139.9, 139.6, 130.5, 126.5, 125.4, 124.5,
123.7, 122.5, 113.1, 99.1, 43.8, 39.9, 37.8, 26.9, 25.6, 22.4, 1.5 ppm; 10
(minor product): 1H NMR (CDCl3, 300 MHz): d=8.28 (d, 1H, J=
6.6 Hz), 7.87 (dd, 1H, J=8.0, 2.1 Hz), 7.69 (d, 1H, J=4.3 Hz), 7.1–7.3
(m, 7H), 3.53 (t, 1H, J=4.5 Hz), 2.14 (m, 2H), 1.8–1.2 (m, 7H), 0.95 (m,
1H), 0.06 ppm (s, 9H); chiral HPLC analysis on Chiralpak AD column:
eluent: isopropanol/hexane 10:90; flow rate: 1.0 mLminÀ1; tR: 8.9 (10),
9.4 ((+)-9), 15.3. (10), 33.5 min ((À)-9).
Reaction of 1a with (1,2-dihydronaphthalen-4-yloxy)trimethylsilane (11):
2-((4S,4aR,10bR)-4-Phenyl-10b-(trimethylsilyloxy)-4a,5,6,10b-tetrahydro-
4H-benzo[h]chromen-2-yl)pyridine-N-oxide ((À)-12a) was obtained by
column chromatography (eluent: ethyl acetate/cyclohexane 1:1) and re-
crystallised from ethyl acetate/ligroine as a white solid; m.p. 161–1638C;
[a]2D5 =À69.7 (c=0.4, CHCl3); 1H NMR (CDCl3, 300 MHz): d=8.40 (dd,
1H, J=6.5, 0.8 Hz), 8.15(dd, 1H, J=8.2, 2.0 Hz), 7.54 (dd, 1H, J=
7.8,0.8 Hz), 7.38 (d, 1H, J=2.6 Hz), 7.05–7.15 (m, 9H), 3.36 (dd, 1H, J=
10.7, 2.6 Hz), 3.08 (ddd, 1H, J=18.8, 10.5, 7.5 Hz), 2.93 (dd, 1H, J=18.8,
7.3 Hz), 2.36 (m, 2H), 1.77 (m, 1H), 0.12 ppm (s, 9H); 13C NMR (CDCl3,
75 MHz): d=142.5, 142.3, 141.1, 139.7, 136.8, 136.2, 128.8, 128.1, 127.8,
126.3, 125.7, 124.7, 124.4, 123.5, 122.8, 114.0, 99.8, 43.8, 40.5, 23.2, 19.2,
1.2 ppm; chiral HPLC analysis on Chiralpak AD column: eluent: isopro-
panol/hexane 10:90; flow rate: 1.0 mLminÀ1; tR: 7.0 ((+)-12a), 8.0 min
((À)-12a).
2-((S)-3-((R)-1-Oxo-1,2,3,4-tetrahydronaphthalen-2-yl)-3-phenylpropa-
noyl)pyridine-N-oxide ((+)-13a) was obtained by column chromatogra-
phy (eluent: ethyl acetate/cyclohexane 1:1) and recrystallised from ethyl
acetate/ligroine as a white solid; m.p. 147–1498C; [a]2D5 =25.3 (c=0.1,
CHCl3); 1H NMR (CDCl3, 300 MHz): d=8.16 (d, 1H, J=6.2 Hz), 8.00
(d, 1H, J=7.8 Hz), 7.47 (t, 1H, J=7.4 Hz), 7.34–7.16 (m, 10H), 4.13 (dt,
1H, J=11.9, 4.7 Hz), 3.82 (d, 2H, J=7.1 Hz), 3.03 (m, 1H), 2.86 (m,
2H), 2.14 (m, 1H), 1.67 ppm (m, 1H); 13C NMR (CDCl3, 75 MHz): d=
198.7, 195.9, 146.6, 143.1, 141.3, 139.7, 132.8, 132.1, 128.1, 127.84, 127.0,
127.0, 126.1, 126.1, 124.9, 51.8, 46.8, 38.7, 27.1, 25.4 ppm; chiral HPLC
analysis on Chiralpak AD column: eluent: isopropanol/hexane 10:90;
flow rate: 1.0 mLminÀ1; tR: 61.8 ((À)-13a), 65.8 min ((+)-13a).
2-((S)-3-(4-Bromophenyl)-3-((R)-1-oxo-1,2,3,4-tetrahydronaphthalen-2-
yl)-propanoyl)pyridine-N-oxide ((+)-13b) was obtained by column chro-
matography (eluent: ethyl acetate/cyclohexane 1:1); [a]2D5 =35.5 (c=0.8,
CHCl3); 1H NMR (CDCl3, 300 MHz): d=8.16 (d, 1H, J=6.0 Hz), 7.98
(d, 1H, J=7.8 Hz), 7.45–7.19 (m, 10H), 4.16 (dt, 1H, J=14.3, 7.0 Hz),
3.82 (dd, 2H, J=7.5, 2.7 Hz), 3.06–2.79 (m, 3H), 2.12 (m, 1H), 1.66 ppm
(m, 1H); 13C NMR (CDCl3, 75 MHz): d=198.1, 195.4, 146.3, 143.0, 140.3,
139.9, 132.8, 132.0, 130.8, 128.1, 127.2, 127.0, 126.3, 126.1, 125.0, 120.0,
51.6, 46.5, 38.0, 27.4, 25.2 ppm; chiral HPLC analysis on Chiralpak AD
column: isopropanol/hexane 20:80; flow rate: 1.0 mLminÀ1; tR: 30.9 ((À)-
13b), 34.6 min ((+)-13b).
2-((S)-3-(4-Bromophenyl)-3-((S)-1-oxo-1,2,3,4-tetrahydronaphthalen-2-
yl)-propanoyl)pyridine-N-oxide ((+)-14b) was obtained by column chro-
matography (eluent: ethyl acetate/cyclohexane 1:1); 1H NMR (CDCl3,
300 MHz): d=8.17 (d, 1H, J=6.5 Hz), 8.02 (d, 1H, J=7.0 Hz), 7.47–7.19
(m, 10H), 4.27 (dt, 1H, J=10.6, 4.0 Hz), 4.13 (m, 1H), 3.48 (dd, 1H, J=
17.6, 3.9 Hz), 2.98 (m, 2H), 2.81 (dt, 1H, J=12.1, 4.6 Hz), 2.15 ppm (m,
2H); 13C NMR (CDCl3, 75 MHz): d=197.6, 195.5, 146.1, 143.2, 141.1,
139.9, 132.9, 132.2, 130.9, 129.5, 128.1, 127.2, 127.0, 126.4, 126.2, 125.0,
119.8, 53.0, 42.7, 38.7, 28.5, 24.3 ppm; chiral HPLC analysis on Chiralpak
AD column: isopropanol/hexane 20:80; flow rate: 1.0 mLminÀ1; tR: 38.7
((+)-14b), 40.6 min ((À)-14b).
General method for a-tetralone (15) addition to 1a,b: In anhydrous
glassware, 2 (0.03mmol), copper triflate (0.010g, 0.03mmol) and pyri-
dine-N-oxide derivatives 1a,b (0.3mmol) were added to a test tube. The
reactants were dissolved in freshly dried dichloromethane (0.5mL) and
magnetically stirred at room temperature. After 15min, a-tetralone (15;
650 mL, 5mmol) was added to the tube. The reaction was monitored by
TLC and stopped after completion. The reaction mixture was quenched
in water and extracted with dichloromethane (3ꢃ10mL), and the organic
phase was dried over sodium sulphate. The products were isolated by
column chromatography with ethyl acetate/cyclohexane (3:7) as the
eluent. The 1H and 13C NMR spectra were identical to those found for
the products from the corresponding reaction with the enol silyl ether.
X-ray diffraction studies of 7a, 12a and 13a: All crystal structures were
solved by direct methods (SIR97)[20] and refined with full-matrix least-
Table 5. Crystallographic data for 7a, 12a and 13a.
2-((S)-3-((S)-1-Oxo-1,2,3,4-tetrahydronaphthalen-2-yl)-3-phe-
nylpropanoyl)pyridine-N-oxide ((+)-14a) was obtained by
column chromatography (eluent: ethyl acetate/cyclohexane
1:1); [a]2D5 =14.4 (c=0.1, CH3OH); 1H NMR (CDCl3,
300 MHz): d=8.17 (dd, 1H, J=6.5, 0.5 Hz), 8.03 (dd, 1H, J=
7.0, 0.5 Hz), 7.45 (dt, 1H, J=7.0, 0.5 Hz), 7.32–7.19 (m, 10H),
4.32 (dt, 1H, J=8.5, 4.2 Hz), 4.10 (dd, 1H, J=17.4, 10.6 Hz),
3.48 (dd, 1H, J=17.4, 4.3 Hz), 2.99 (m, 2H), 2.81 (m, 1H),
2.13 ppm (m, 2H); 13C NMR (CDCl3, 75 MHz): d=197.8,
195.9, 143.3, 142.0, 139.8, 132.8, 128.1, 127.8, 127.7, 127.0,
126.3, 126.1, 125.9, 124.9, 53.2, 42.6, 39.2, 28.5, 24.2 ppm; chiral
HPLC analysis on Chiralpak AD column: eluent: isopropanol/
7a
12a
13a
formula
Mr
crystal system
space group
a [ꢂ]
b [ꢂ]
c [ꢂ]
a [8]
b [8]
C18H15NO4
309.31
C27H29NO3Si
443.60
C24H21NO3
371.42
orthorhombic
P212121 (no. 19)
8.368(2)
17.598(3)
10.262(3)
90
monoclinic
P21 (no. 4)
9.752(2)
11.937(2)
10.409(1)
90
orthorhombic
P212121 (no. 19)
6.094(1)
15.840(2)
19.215(4)
90
90
90
100.91(1)
90
90
90
g [8]
hexane 10:90; flow rate: 1.0 mLminÀ1
; tR: 71.4 ((+)-14a),
V [ꢂ3]
Z
1511.2(6)
4
1189.8(3)
2
1854.7(6)
4
74.3 min ((À)-14a).
1calcd [gcmÀ3
]
1.360
1.238
1.330
Reaction of 1b with (1,2-dihydronaphthalen-4-yloxy)trimethyl-
silane (11): 2-((4S,4aR,10bR)-4-(4-Bromophenyl)-10b-(trime-
thylsilyloxy)-4a,5,6,10b-tetrahydro-4H-benzo[h]chromen-2-yl)-
pyridine-N-oxide ((À)-12b) was obtained by column chroma-
tography (eluent: ethyl acetate/cyclohexane 1:1); 1H NMR
(CDCl3, 300 MHz): d=8.24 (dd, 1H, J=6.5, 0.8 Hz), 8.14 (dd,
1H, J=8.2, 2.0 Hz), 7.52 (d, 1H, J=7.8 Hz), 7.47–7.36 (m,
4H), 7.28 (dt, 1H, J=7.7, 1.4 Hz), 7.21–7.14 (m, 5H), 3.31 (dd,
1H, J=10.9, 2.6 Hz), 3.1–2.8 (m, 2H), 2.33 (m, 2H), 1.70 (m,
1H), 0.10 ppm (s, 9H); 13C NMR (CDCl3, 75 MHz): d=144.4,
143.9, 143.3, 142.2, 138.8, 138.3, 133.4, 131.7, 131.0, 130.41,
128.0, 126.8, 126.6, 125.7, 125.1, 122.2, 115.3, 101.9, 46.0, 42.2,
25.3, 21.3, 3.4 ppm; chiral HPLC analysis on Chiralpak AD
column: isopropanol/hexane 10:90; flow rate: 1.0 mLminÀ1; tR:
8.2 ((+)-12b), 9.5 min ((À)-12b).
m (MoKa) [mmÀ1
2qmax [8]
]
0.097
0.127
0.088
25
30
30
min/max transmission
collected reflns
independent reflns
Rint
0.959/0.990
3601
0.948/0.990
8113
0.986/1.000
7144
2671
6927
3077
0.023
0.011
0.026
strong refln [Fo >2s(Fo)]
refined parameters
R1/wR2 (strong data)
R1/wR2 (all data)
goodness of fit
1925
5010
2233
208
292
253
0.0422/0.0928
0.0741/0.1082
0.962
0.0446/0.0926
0.0736/0.1045
1.019
0.0444/0.0942
0.0693/0.1040
1.017
max/min residuals [eꢂÀ3
Flackꢀs parameter
]
0.16/À0.16
0.21/À0.14
0.16/À0.20
n.d.[a]
À0.05(11)
n.d.[a]
[a] n.d.: not determined.
&
6
&
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

CuII–Bis(oxazoline)-Catalysed Enantioselective Addition to Pyridine-N-Oxides
FULL PAPER
square procedures on F2 (SHELXL-97)[21] by using all reflections collect-
ed on an Enraf–Nonius CAD4 diffractometer (l=0.71073 ꢂ, T=293 K).
Lorentz, polarisation and absorption effect (psi-scan method)[22] correc-
tions were applied. H atoms were placed at the calculated positions. Only
for 12a, the absolute configuration was established by anomalous disper-
sion effects in the diffraction data. Crystal data are reported in Table 5.
CCDC 857146 (7a), 857147 (12a) and 857148 (13a) contain the supple-
mentary crystallographic data for this paper. These data can be obtained
free of charge from the Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
[8] a) P. K. Singh, V. K. Singh, Org. Lett. 2008, 10, 4121–4124; b) P. K.
Singh, V. K. Singh, Org. Lett. 2010, 12, 80–83.
[9] S. Barroso, G. Blay, M. C. Munoz, J. R. Pedro, Adv. Synth. Catal.
2009, 351, 107–111.
[10] A. Livieri, M. Boiocchi, G. Desimoni, G. Faita, Chem. Eur. J. 2009,
15, 516–520.
[11] S. Barroso, G. Blay, M. C. Munoz, J. R. Pedro, Org. Lett. 2011, 13,
402–405.
[12] S. K. Ray, P. K. Singh, V. K. Singh, Org. Lett. 2011, 13, 5812–5815.
[13] G. Desimoni, G. Faita, M. Toscanini, M. Boiocchi, Chem. Eur. J.
2009, 15, 9674–9677.
[14] R. Rasappan, D. Laventine, O. Reiser, Coord. Chem. Rev. 2008, 252,
702–714.
[15] a) V. K. Aggarwal, L. Bell, M. P. Coogan, P. Jubault, J. Chem. Soc.
Perkin Trans. 1 1998, 2037–2042; b) M. Schinnerl, C. Bçhm, M.
Seitz, O. Reiser, Tetrahedron: Asymmetry 2003, 14, 765–771.
[16] a) R. W. W. Hooft, L. H. Straver, A. L. Spek, J. Appl. Crystallogr.
2008, 41, 96–103; b) R. W. W. Hooft, L. H. Straver, A. L. Spek, J.
Appl. Crystallogr. 2010, 43, 665–668.
Acknowledgements
This work was supported by the Ministero dellꢀUniversitꢄ e della Ricerca
(MiUR) (PRIN 2008) and by the University of Pavia.
[17] D. A. Evans, J. S. Johnson, E. J. Oliava, J. Am. Chem. Soc. 2000, 122,
1635–1649.
[18] J. Thorhauge, M. Roberson, R. G. Hazell, K. A. Jørgensen, Chem.
Eur. J. 2002, 8, 1888–1898.
[19] A. Martel, S. Leconte, G. Dujardin, E. Brown, V. Maisonneuve, R.
Rectoux, Eur. J. Org. Chem. 2002, 514–525.
[20] A. Altomare, M. C. Burla, M. Camalli, G. L. Cascarano, C. Giaco-
vazzo, A. Guagliardi, A. G. G. Moliterni, G. Polidori, R. Spagna, J.
Appl. Crystallogr. 1999, 32, 115–119.
[1] a) A. Landa, A. Minkkilꢅ, G. Blay, K. A. Jørgensen, Chem. Eur. J.
2006, 12, 3472–3483; b) G. Heimgꢅrtner, D. Raatz, O. Reiser, Tetra-
hedron 2005, 61, 643–655; c) D. Raatz, C. Innertsberger, O. Reiser,
Synlett 1999, 1907–1910.
[2] M. Takeshita, S. Yoshida, Heterocycles 1990, 30, 871–874.
[3] R. Noyori, Asymmetric Catalysis in Organic Synthesis, Wiley, New
York, 1994.
[4] T. P. Yoon, E. N. Jacobsen, Science 2003, 299, 1691–1693.
[5] a) G. Desimoni, G. Faita, K. A. Jørgensen, Chem. Rev. 2006, 106,
3561–3651; b) G. Desimoni, G. Faita, K. A. Jørgensen, Chem. Rev.
2011, 111, PR284–PR437; c) G. Desimoni, G. Faita, P. Quadrelli,
Chem. Rev. 2003, 103, 3119–3154.
[21] G. M. Sheldrick, Acta Crystallogr. Sect. A 2008, 64, 112–122.
[22] A. C. T. North, D. C. Phillips, F. S. Mathews, Acta Crystallogr. Sect.
A 1968, 24, 351–359.
[6] S. Youssif, ARKIVOC 2001, 242–268.
Received: April 10, 2012
[7] S. Barroso, G. Blay, J. R. Pedro, Org. Lett. 2007, 9, 1983–1986.
Published online: && &&, 0000
Chem. Eur. J. 2012, 00, 0 – 0
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
7
&
ÞÞ
These are not the final page numbers!

G. Faita et al.
Asymmetric Catalysis
A. Livieri, M. Boiocchi, G. Desimoni,
G. Faita* ....................... &&&& — &&&&
Enantioselective Addition of Cyclic
Enol Silyl Ethers to 2-Alkenoyl-Pyri-
dine-N-Oxides Catalysed by CuII–
Bis(oxazoline) Complexes
Out of the BOX: Asymmetric reac-
tions involving (E)-3-aryl-1-(pyridin-2-
yl-N-oxide)prop-2-en-1-ones and cyclic
enol silyl ethers show good yields and
excellent enantioselectivities (up to
99.9% ee) when catalysed by CuII–
bis(oxazoline) (BOX) complexes (see
scheme; TMS: trimethylsilyl).
Mukaiyama–Michael and hetero
Diels–Alder pathways can be followed
with the different enol silyl ethers.
&
8
&
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!