Chiral monooxazolines as modular copper(I)-heterocomplex building blocks: Investigations on the catalytic asymmetric cyclopropanation of alkenes

Article abstract of DOI:10.1016/j.tet.2011.04.066

Novel chiral monodentate oxazoline ligands have been synthesized in good yields. The catalytic activity of these monodentate oxazoline/Cu catalysts was evaluated in the catalytic asymmetric cyclopropanation of styrene and α-methylstyrene, giving moderate to good enantioselectivities (up to 74% ee for the trans-cyclopropane product) and full conversions (up to 100%). In an attempt to enhance the enantioselectivities of the cyclopropanations, heterocombinations of these ligands were used. Unfortunately, with the data set that was used in this study, no improvements were observed. However, to gain an insight into the nature of the active catalyst present under these circumstances, NMR, mass spectrometric and computational studies were carried out and indicated the presence of bidentate heterocomplexes in the equilibrium mixture. Analysis of the stereoselectivities (ees and des) did not prove very useful in pin-pointing the identity of the active chiral catalyst and only afforded a very weak conclusion. In order to ascertain the importance of the π-π interactions, the monodentate oxazoline ligands 3a and 3b were synthesized and screened in these reactions, and the resulting stereoselectivities were compared to the results obtained using ligands 1a and 1b. There seemed to be very weak π-π interactions at work.

Full text of DOI:10.1016/j.tet.2011.04.066

Tetrahedron 67 (2011) 4640e4648
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Chiral monooxazolines as modular copper(I)-heterocomplex building blocks:
investigations on the catalytic asymmetric cyclopropanation of alkenes
,
,
,b,
Elisabete P. Carreiro a b, J.P. Prates Ramalho a b, Anthony J. Burke a
*
a
ꢀ
ꢀ
~
Centro de Química de Evora, Universidade de Evora, Rua Romao Ramalho, 59, 7000 Evora, Portugal
Departamento de Química, Universidade de Evora, Rua Romao Ramalho, 59, 7000 Evora, Portugal
b
ꢀ
~
a r t i c l e i n f o
a b s t r a c t
Article history:
Novel chiral monodentate oxazoline ligands have been synthesized in good yields. The catalytic activity
of these monodentate oxazoline/Cu catalysts was evaluated in the catalytic asymmetric cyclopropanation
of styrene and a-methylstyrene, giving moderate to good enantioselectivities (up to 74% ee for the trans-
cyclopropane product) and full conversions (up to 100%). In an attempt to enhance the enantiose-
lectivities of the cyclopropanations, heterocombinations of these ligands were used. Unfortunately, with
the data set that was used in this study, no improvements were observed. However, to gain an insight
into the nature of the active catalyst present under these circumstances, NMR, mass spectrometric and
computational studies were carried out and indicated the presence of bidentate heterocomplexes in the
equilibrium mixture. Analysis of the stereoselectivities (ees and des) did not prove very useful in pin-
pointing the identity of the active chiral catalyst and only afforded a very weak conclusion. In order to
Received 22 February 2011
Received in revised form 10 April 2011
Accepted 15 April 2011
Available online 21 April 2011
Keywords:
Mono(oxazoline)
Arylid-OX
Asymmetric catalysis
Cyclopropanation
Homocomplex
Heterocomplex
ascertain the importance of the
synthesized and screened in these reactions, and the resulting stereoselectivities were compared to the
results obtained using ligands 1a and 1b. There seemed to be very weak interactions at work.
pep interactions, the monodentate oxazoline ligands 3a and 3b were
pep interactions
pep
Ó 2011 Elsevier Ltd. All rights reserved.
1. Introdution
cyclopropanations,5,6 in catalytic enantioselective [2þ2þ2] cycload-
ditions with Ni7 and in enantioselective DielseAlder reactions.8,9 All
of these reactions showed moderate stereoselectivity.
Over the last number of years chiral monodentate ligands have
proven their potential in numerous transition-metal-catalyzed
asymmetric transformations. Although there is enormous applica-
tion of bidentate chiral ligands in catalytic asymmetric synthesis,
monodentate versions have rarely had the same impact, for the main
reason that they fail to form compact catalytic manifolds, which are
pivotal for high asymmetric induction during the reaction. Mono-
dentate ligands offer the advantage that they are in general less
structurally complex than di- and multi-dentate ligands, thus the
synthesis in principle is less demanding. For example, Binol-based
modular mono-phosphonites,1 mono-phosphites2 and mono-
phosphoramidites3 so far have been the principle monodentate
ligand to be investigated, often leading to medium or high enantio-
selectivities in diverse asymmetric catalytic reactions when metals
like Rh or Cu were used. Oxazoline ligands are superior to phosphine
ligands in that they are stable to both hydrolysis and oxidative con-
ditions, a considerable advantagewhen compared to the latter family,
which are easily converted into phosphine oxides.4 Monodentade
oxazolines were applied successfully in catalytic asymmetric
Chiral olefinic ligands have been shown to be useful in catalytic
asymmetric synthesis and have been the subject of various
reviews.10aec In fact, recently, Glorius et al. introduced a new family of
modular oxazolines, which are olefin/oxazolines (abreviated Ole-
fOX).10d They were tested in Rh catalyzed conjugate additions of
phenylboronic acid to cyclohexenone, and gave very good enantio-
selectivities. However, theses ligands are true bidentate ligands as
the coordination to the Rh is via the oxazoline nitrogen and the
olefinic bond. Monodentate ligands are very empowering as they
allow for the creation of a large diversity of chiral catalysts, and are
very amenable for application in combinatorial enantioselective ca-
talysis using mixtures of such ligands. The first combinatorial ho-
mogeneous asymmetric catalysis using the concept of mixing chiral
monodentate ligands was reported simultaneously by Reetz11 and
Feringa.12 Using this concept in some catalytic asymmetric hydro-
genations11,12a and in the conjugated addition of arylboronic acids12b
it was possible to enhance the enantioselectivities, as well as the
reaction rate, by simply mixing two known monodentate ligands.
The heterocomplex formed was assumed to be the active catalyst.
In this paper, we introduce two families of olefinated mono-
dentate oxazolines, designated Arylid-OX 2 and Propen-OX 3,
(Fig. 1). The design and synthesis of the Arylid-OX 1 family (Arylid-
* Corresponding author. Tel.: þ351266745310; fax: þ351266745303; e-mail
address: ajb@dquim.uevora.pt (A.J. Burke).
0040-4020/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2011.04.066

E.P. Carreiro et al. / Tetrahedron 67 (2011) 4640e4648
4641
by Neustadt et al.16 using a simple Knoevenagel condensation with
malonic acid and the respective aldehyde. Our standard synthetic
procedure was subsequently used to transform the acids to the
corresponding ligands.14,15
X
R
O
N
O
O
R1
O
O
N
N
N
N
1 R= 2,4-(MeO)2C6H3
2 R= Ph
3 R= Me
2.1. Asymmetric catalytic cyclopropanation: homo- and
heterocombinations of mono(oxazolines)
R1
R1
R1
R1
4
5
R1=iPr, tBu, Ph, Bn
X = H, Cl, OMe
The novel monodentate mono(oxazolines) were evaluated in
the Cu(I)-catalyzed asymmetric cyclopropanation of styrene and
O
O
a
-methylstyrene using ethyl diazoacetate (EDA) (Scheme 1). The
O
catalyst was generated in situ by the addition of the pre-catalyst,
[Cu(MeCN)4]PF6, and the relevant chiral mono(oxazoline) 1e3 in
catalytic quantities, followed by the addition of excess alkene and
then by the slow addition of EDA. This method had previously been
established and optimized by our research group for the bisox-
azolines 4e5.13e15 The reactions were conducted in both CH2Cl2
and toluene to determine the influence of both polar and apolar
solvents. The reactions with toluene required heating for activation,
but, unfortunately, heating probably destroyed or deactivated the
catalyst, as the yield was poor. Given the distinct possibility of
having
Arylid-OX or Alkylid-OX ligands, namely, 1, 2 and 3 (Scheme 2). The
Arylid-OX ligands 1 contained a very electron-rich -system. In
fact, computational studies (vide infra) show that the aryl group is
co-planar with the C]C. It was hoped that this extensive -elec-
tron delocalization would promote significant interactions,
N
1a R1=iPr
1b R1=Bn
1c R1=Ph
1d R1=tBu
R1
Fig. 1. Structures of monodentate oxazolines (1e3) and bis(oxazolines) (4,5).
OXs 1c and 1d have previously been reported for the catalytic
asymmetric cyclopropanation reaction, CACP)6 was inspired by the
known Arylid-BOX 413,14 and Isbut-BOX 515 ligands developed by
our research group. These ligands were tested in the catalytic
asymmetric cyclopropanation of alkenes with ethyl diazoacetate
(EDA) using [Cu(MeCN)4]PF6 and Cu(I)OTf as the pre-catalysts
(Scheme 1).
pep interactions at play, we focused on three types of
p
p
pep
leading to more compact Cu-catalysts that would promote better
asymmetric induction.17 To test this hypothesis, the two ligand
families, type 2, with only a phenyl group in the back-bone, and
R
thus expected to show less intense
pep interactions with lower
EtOCOCHN2,
Cu(I)
(R)
(R)
(R)
(S)
reaction enantioselectivities expected, and type 3, bereft of an aryl-
backbone group, and thus expected to manifest even lower reaction
enantioselectivities, were prepared and screened.
CO2Et
CO2Et
R
R
R = H, Me
cis
trans
The homocombinations of Arylid-OXs 1a/1a and 1b/1b were
tested in some catalytic asymmetric cyclopropanations with Cu(I)
and styrene (Table 1). For comparative purposes, the Cu(I) catalyst
was applied at two loading levels: 0.36 and 2 mol %, respectively.
Unfortunately, the yields were low for all the reactions studied
(6e37%). In the case of 1a/1a the best enantioselectivity obtained
was 48% ee for the cis-cyclopropane product using 6.3 mol % of
ligand, on the other hand, with ligand 1b the best enantioselectivity
obtained was 45% ee for the cis-cyclopropane at a loading of only
Scheme 1. Asymmetric cyclopropanation of styrenesdthe benchmark reaction for this
study.
2. Results and discussion
The Arylid-OX families 1 and 2 and Propen-OX 3 were prepared
in good yields using the synthetic pathway shown in Scheme 2. The
carboxylic acid precursor was obtained via the procedure reported
OH
H
R
OH
R
Cl
R
N
(a)
(b)
O
R1
O
O
12a R= 2,4-dimethoxyphenyl; R1=iPr (81%)
12b R= 2,4-dimethoxyphenyl; R1=Bn (61%)
13 R= Phenyl; R1=Ph (52%)
6 R= 2,4-dimethoxyphenyl
7 R= Phenyl
9
10
11
8 R= Methyl
14a R= Methyl; R1=iPr (33%)
14b R= Methyl; R1=Ph (78%)
R
O
(c)
N
R1
1a R= 2,4-dimethoxyphenyl; R1=iPr (98%)
1b R= 2,4-dimethoxyphenyl; R1=Bn (55%)
2
R= Phenyl; R1=Ph (76%)
3a R= Methyl; R1=iPr (62%)
3b R= Methyl; R1=Ph (72%)
Scheme 2. Reagents and Conditions: (a) (COCl)2, DMF, CH2Cl2, 0 ꢀC; (b) (S)-Phenylglycinol or (S)-Valinol or (S)-Phenylalaninol, NEt3, CH2Cl2; (c) CH3SO2Cl, NEt3, CH2Cl2.

4642
E.P. Carreiro et al. / Tetrahedron 67 (2011) 4640e4648
Table 1
Catalytic asymmetric cyclopropanation of styrene.a
Entry
Oxazoline (mol %)
Cu(MeCN)4PF6 (mol %)
Yield (%)c
ee cis (%)d
ee trans (%)d
cis/transd
Maleate/fumerate:
cyclopropanesd
Homocombinations
1
2
3
4
5
6
7
1a/1a (1.13)
0.36b
6
12
4
10
34
10
17
27
48
45
32
51
46
42
24
37
35
25
58
32
33
37:63
39:61
38:62
37:63
31:69
34:66
44:56
29:71
3:97
1a/1a (6.3)
1b/1b (1.13)
1b/1b (6.3)
1d/1d (6.3)
3b/3b (6.3)
3a/3a (6.3)
2
0.36b
26:74
10:90
14:86
22:78
16:84
2
2
2
2
Heterocombinations
8
9
1a/1b (1.13)
1a/1d (1.13)
0.36b
0.36b
11
37
40
37
32
30
30:60
34:66
10:90
8:92
a
b
c
Styrene (4 equiv), chiral ligand, Cu(I) pre-catalyst, ethyl diazoacetate (1 equiv), CH2Cl2, rt, 24 h.
Styrene (0.7 equiv), chiral ligand, Cu(I) pre-catalyst, ethyl diazoacetate (1 equiv), CH2Cl2, rt, 24 h.
Yield determined from the mass of isolated crude product isomers.
Determined by chiral GC.
d
1.13 mol % of ligand. The formation of maleates at a loading of only
0.36 mol % of Cu(I) was considerably high with both ligands, this
was evidently due to the presence of excess EDA under these
conditions. The cis/trans ratio of the diastereomers did not change
much. On comparing the homocombinations of 3a/3a with 3b/3b, it
was 3a/3a that gave the best enantioselectivity (46% ee for cis-cy-
clopropane) and the best diastereoselectivity (32% de), but the yield
was lower and was due to the presence of a greater proportion of
maleate/fumerate (Table 1, entry 6). Surprisingly, in all cases, the
cis-cyclopropane gave a higher enantioselectivity than the trans-
cyclopropane (the predominant isomer). On increasing the loading
of pre-catalyst from 0.36 to 2 mol % there was a significant increase
in the enantioselectivity for the homocombination 1a/1a (Table 1,
entries 1 and 2). The yield doubled as expected. In the case of 1b/1b
the enantioselectivity actually decreased. The yield also more than
doubled. The lower yields obtained at the lower catalyst loading
level, were most probably a consequence of the dimerization side-
reaction, which was obviously enhanced by the presence of less
styrene in the reaction mixture.
The formation of heterocomplexes was also investigated, by
mixing two types of ligands. To this end, ligands 1a and 1b together
with Cu(I), were combined in equimolar amounts. It was expected
that the three catalyst typesdi.e., (1a)2Cu, (1b)2Cu and (1a1b)
Cudwould prevail in solution as a 1:1:2 mixture, having the het-
erocomplex (1a1b)Cu as the most active species based on literature
precedent.12b Unfortunately, in the case of both hetero-
combinations using 1a/1b and 1a/1d (Table 1, entries 8e9), the
enantioselectivities obtained were not as high as the two homo-
combinations of the respective ligands.
The catalytic asymmetric cyclopropanation of a-methylstyrene
with all homocombinations of the mono(oxazoline) ligands were
tested in the same proportion ((6.3:2) ligand: Cu(I)(CH3CN)4PF6)
(Table 2). Generally they gave very good conversions with this
substrate, the trans-cyclopropane (the predominant isomer) gave
the best enantioselectivity. The best enantioselectivity obtained
was 74% ee with only 0.36 mol % of the homocombination 1d/1d
(Table 2, entry 6), this was undoubtedly due to the presence of
the bulky tert-butyl group. In the case of 3a/3a and 3b/3b, the
Table 2
Catalytic asymmetric cyclopropanation of
a
-methylstyrenea
Entry Oxazoline (mol %)
Cu(MeCN)4PF6 (mol %)
Conversion (%)c
ee cis (%)c
ee trans (%)c
cis/transc
Maleates:
cyclopropanesc
Homocombinations
1
1a/1a (1.13)
0.36b
86
94
86
14
22
41
24
34
42
22
27
35
12
32
45
42
63
47
44
74
29
37
39
24
44
46:54
45:55
42:58
46:54
46:54
43:57
46:54
47:53
47:53
46:54
50:50
10:90
2:98
21:79
20:80
4:96
9:91
10:90
9:91
2
3
4
5
6
7
8
9
1a/1a (6.3)
1b/1b (6.3)
1c/1c (1.13)
1c/1c (6.3)
1d/1d (1.13)
2/2 (1.13)
2
2
0.36b
2
99
100
100
98
98
92
0.36b
0.36b
2
2/2 (6.3)
3b/3b (6.3)
3b/3b (6.3)
3a/3a (6.3)
2
12:88
12:88
10:90
10
11
2d
98
98
2
Heterocombinations
12
13
14
15
16
17
18
1a/1b (6.3)
2
2
2
2
2
2
2
90
80
89
87
94
31
24
38
29
38
37
37
52
46
57
44
44
44
43
44:56
44:56
44:56
44:56
45:55
45:55
47:53
6:94
57:43
26:74
35:65
18:82
7:93
1a/1c (6.3)
1b/1c (6.3)
1c/2 (6.3)
1c/3b(6.3)
2/3b (6.3)
3a/3b (6.3)
100
95
16:84
a
a
a
-Methylstyrene (4 equiv), chiral ligand, Cu(I) pre-catalyst, ethyl diazoacetate (1 equiv), CH2Cl2, rt, 24 h.
-Methylstyrene (0.7 equiv), chiral ligand, Cu(I) pre-catalyst, ethyl diazoacetate (1 equiv), CH2Cl2, rt, 24 h.
b
c
Determined by chiral GC.
Toluene, T¼40 ꢀC, t¼48 h.
d

E.P. Carreiro et al. / Tetrahedron 67 (2011) 4640e4648
4643
difference in the enantioselectivities obtained was small, ligand 3a
gave the best enantioselectivity for the trans-cyclopropane product
(Table 2, entry 11, 44% ee), but there was no diastereoselectivity.
Toluene was used with homocombination 3b/3b, but the enantio-
selectivities for both isomers decreased approximately half, this
was probably due to the higher reaction temperature leading to
lower asymmetric induction with most likely, some catalyst de-
composition. The highest diastereoselectivity achieved was 37% de
using the homocomplex derived from 1d and styrene (Table 1,
entry 5).
enantioselectivity over time (a 24 h period) were conducted with
the homocomplexof 1c and the heterocombination of 1b and 1c.18 In
the case of the homocomplex the enantioselectivity remained con-
stant throughout. In the case of the heterocombination, the ee
ascended to its maximum after about 1.5 h and remained there till
the end.
2.2. NMR study
An insightful NMR study of the homocomplexes (1a)2-Cu(I),
(1b)2-Cu(I) (as standards), and of the heterocombination of 1a, 1b
and Cu(I) was also carried out. The complexes were prepared inside
an NMR tube in dry CDCl3 and the ligands were combined in
equimolar amounts, together with 1 equiv of the Cu(I) pre-catalyst.
The results are shown in Fig. 2 (the proposed signal attributions are
shown in Table 3). The goal was to confirm exactly the types of
complexes present in solution in order to gain an insight into the
nature of the catalytic active species.
A point needs to be made regarding the potential
pep
interactions that were expected to enhance the reaction enantio-
selectivity and thus integrated into the ligand design. An analysis of
the enantioselectivity and diastereoselectivity results (only for the
homocombinations) for the cyclopropanation of both styrene and
a
-methylstyrene seemed to indicate very weak pep interactions,
as the differences in ees and des on changing the ligand were quite
small. In the case of the cyclopropanation of styrene with 1a, ees of
8.0
7.0
6.0
5.0
4.0
3.0
2.0
1.0
ppm (f1)
Fig. 2. 1H NMR spectra of the homo- and heterocomplexes.
Table 3
48% (cis) and 37% (trans) with a de of 22% were obtained (Table 1,
entry 2), but for 3a lower ees of 42% (cis) and 33% (trans), with a de
of 12% (Table 1, entry 7) were obtained, clearly supporting the
Proposed attributions for the 1H NMR spectra of the homo- and the putative
heterocomplex ((1a1b)Cu(I))
Entry
Ligand/complex
d (ppm)
possibility of stronger
the reaction with the former ligand system. However, in the case of
-methylstyrene, this was not so clear cut. For example, with 1a ees
of 22% (cis) and 42% (trans) with a de of 10% (Table 2, entry 2) were
obtained, but for 3a, ees of 32% (cis) and 44% (trans) were obtained
with no diastereoselectivity (Table 2, entry 11). This was puzzling, if
not contradicatory, as on the one hand the drop in diaster-
pep interactions being present in the case of
H-1
H-2
H-20
a
H-3
H-4
1
2
3
4
5
6
1a
1b
4.32
4.51
4.61
4.66
4.74
4.60
4.00
4.04
4.36
4.25
4.51
4.45
6.66
6.66
6.80
6.42
6.69
6.70
7.54
7.58
7.78
7.72
7.75
7.70
a
4.28
Cu(I)-(1a)2
Cu(I)-(1b)2
Cu(I)-(1a1b)b
4.07
4.05
4.31
4.22
6.57
6.60
b
Cu(I)-(1a1b)2
a
Not observed, presumably hidden by the phenyl signal.
This complex is believed to be the predominat complex in solution.
eoselectivity on going from 1a to 3a might indicate weaker
pep
b
interactions, but on the other hand, the increase in the ee of the cis-
isomer with 3a would seem to suggest otherwise. On comparing 1c
with 3b (Table 2, entries 5 and 9) there was virtually no change in
the stereoselectivities.
It was the homocomplexes that gave the highest enantiose-
lectivities and diastereoselectivities.
Analysis of the 1H NMR spectra obtained for the (2:1) mono
(oxazoline)/Cu(I) mixture, suggested the presence of the (1a1b)-
Cu(I) complex, as two well defined doublets for the H3 protons
with coupling constants of 15 Hz
(d
¼6.69 ppm) and 16.8 Hz
In order to ascertain the catalyst stability during the catalytic
(
d
¼6.57 ppm) were observed. The signal corresponding to these
reactions, two key reactions with
a
-methylstyrene as substratewere
protons in the spectrum for the (1b)2-Cu(I) complex appears to be
realized. These studies, which involved monitoring the variation of
overlapped with the aromatic peaks. The signals for H-1 in the

4644
E.P. Carreiro et al. / Tetrahedron 67 (2011) 4640e4648
spectra of the homocomplexes (1a)2-Cu(I) and (1b)2-Cu(I)
appeared at 4.61 and 4.66 ppm, respectively, and these hydrogens
in the spectrum of the heterocomplex 1a/1b-Cu(I) appeared at 4.74
and 4.07 ppm, respectively. The purported equilibrium is shown in
Scheme 3. Coordination of the ligands with Cu is expected to
increase the chemical shifts of the H-1 signals due to electron do-
nation from the oxazoline group. The H-1 signals for both ligands in
the putative heterocomplex show a greater separation. When the
ratio of the ligands to Cu was doubled, the chemical shifts de-
creased in the case of the putative H-1, H-2 and H-20 protons (see
Table 3, entry 6), indicating an equilibrium with the free ligand. Low
temperature experiments were also conducted (down to ꢁ25 ꢀC) in
an effort to try and calculate the equilibrium constant for this
process, but unfortunately, they were inconclusive.
presence of the two ligands, as well as both homocomplexes Cu(I)-
(1a)2 and Cu(I)-(1b)2 and the heterocomplex Cu(I)-(1a1b) (in all cases
without any coordinated MeCN ligands and having one methoxyl
group cleaved off). These were the principle peaks in the spectrum.
2.3. Computational study
A quantum chemical theoretical study was carried out to verify if
the heterocomplex is more likely to be the predominant complex in
the equilibrium mixture in solution. Considering the number and
size of these systems a semi-empirical method was considered the
best choice for geometry optimization. The geometries were fully
optimized in Cartesian coordinates by employing the novel PM6
Hamiltonian19 included in the recent version of MOPAC 2007
H2
H2 H2'
H1
O
H2
H2'
H2'
O
H4
O
O
H4
O
H4
O
H1
NCCH3
N
H1
N
N
PF6
NCCH3
H3
H3
H3
NCCH3
O
Cu
PF6
O
Cu
PF6
H1
O
Cu
H3CCN
H3
H3CCN
H3
H3CCN
H3
H1
H1
N
N
N
O
H2
H2
H2'
H2'
H2
O
H2'
O
O
O
O
H4
H4
H4
O
O
O
(1b)2-Cu(I)
(1a1b)-Cu(I)
(1a)2-Cu(I)
Scheme 3. Postulated equilibrium between the homocomplexes ((1a)2Cu(I) and (1b)2Cu(I)) and the putative heterocomplex ((1a1b)Cu(I)).
In the case of the heterocombination (Table 3, entry 5) some
comments need to be made regarding the increase in the chemical
shift for one of the H-1 signals (to 4.74 ppm), and the decrease in
the chemical shift in the case of the other H-1 signal (to 4.07 ppm).
This is most probably due to some deshielding/shielding effect.
Analysis of the calculated structure for Cu(I)-(1a1b) showed that
the H-1 of 1a appears to be close to the C]C bond of the other li-
gand, and perhaps sits in the deshielding zone of the latter. This is
quite likely as the distance calculated between H-1 and H-3 (other
(Fig. 3).20 The same method has already been successfully employed
in a previous study of Cu(I) complexes by us14 and others21 whose
results (in our case) showed good agreement with the X-ray crys-
tallographic data.14 Frequency calculations further performed in-
dicated that the stationary points obtained were minima. Following
the initial geometry optimization with the PM6 method, Density
Functional Theory (DFT) calculations were carried out on all com-
plexes. These calculations were performed using the B3LYP22,23
functional as included in the Gamess package.24 The Lanl2DZ
effective core basis set was employed for the metal atom while the
6-31G** basis set was used for all the other atoms. The binding
energy for each heterocomplex was obtained by calculating the
ꢁ
ligand) was only 2.662 A. In the case of the second H-1, analysis of
the same model, seems to indicate that it lies in the shielding zone
of one of the nitrile ligands.
Fig. 3. Calculated structure for the Cu(I) complexes: (a) Cu(I)-1a2(MeCNCu(I)-1a2), (b) Cu(I)-1b2(MeCN2), and Cu(I)-1a1b2(MeCN2).
An insightful mass spectrometric study was also conducted that
corroborates the observations obtained with the NMR study. Ligands
1a and 1b were mixed with [Cu(MeCN)4]PF6 in a ratio of 3:3:2 in
dichloromethane at room temperature and after 2 h the solvent was
removed and the remaining mixture was analyzed by ESI-TOF mass
spectrometry.18 Analysis of the resulting spectrum showed the
energy difference between the optimized structure for each com-
plex and the sum of the energies of the optimized isolated moieties.
Selected energy values calculated with the B3LYP/Lanl2DZ, 6-31G
*
//
PM6 method are depicted in Table 4. The calculations show that it is
the heterocomplexdCu(I)-1a1b(MeCN)2dwhich has the lowest DE
value indicating a higher stability for this complex.

E.P. Carreiro et al. / Tetrahedron 67 (2011) 4640e4648
4645
Table 4
3. Conclusions
Calculated energies for complexes: Cu(I)-1a2(MeCN)2, Cu(I)-1b2(MeCN)2 and Cu(I)-
1a1b(MeCN)2 at the B3LYP/Lanl2DZ, 6-31G*//PM6 level
In an attempt to understand if
p
ep
interactions are important
Chemical species
Total energy (au)
D
E (kcal/mol)
DErel (kcal/mol)
for maintaining catalyst integrity in such systemsdan interaction
previously believed to be operational and important in such sys-
tems5,6dthe analogous ligand system, Propen-OX 3, was prepared
and studied. A careful analysis of the results obtained appeared to
indicate that these interactions were very weak indeed, and thus
appear to be less important for achieving significant stereo-
selectivities. It seems that other factors (like the type of substituent
at the stereogenic centre, etc.) are more influential.
Cu(I)-(1a)2
Cu(I)-(1a1b)
Cu(I)-(1b)2
1a
1b
MeCN
ꢁ2265.7216795
ꢁ2418.0626597
ꢁ2570.3924010
ꢁ902.1820275
ꢁ1054.5160923
ꢁ132.6734221
ꢁ195.7599954
ꢁ157.370
0.000
ꢁ4.339
ꢁ1.626
d
ꢁ161.709
ꢁ158.996
d
d
d
d
d
d
Cu(I)
d
Whilst computational studies suggested the existence of the
heterocomplexes in the heterocombination cases, NMR and mass
spectrometric studies on the heterocombination 1a/1b, in fact,
proved the existence of the (1a1b)Cu(I) heterocomplex, which is
expected to exist in all cases.
In the case of the heterocombination of ligands, attempts were
made via a variety of chemical and analytical techniques to identify
the nature of the active chiral catalyst. These studies showed that
only three types of complex were present, two homocomplexes and
one heterocomplex. A careful analysis of the stereoselectivities
failed to allow firm conclusions to be made about the nature of the
active chiral catalyst at work in these reactions, but there is an
indicationdalbeit very slightdthat in fact, one of the homo-
complexes is the active catalyst in a number of instances.
The potential of the Arylid-OX 1e2 family for catalytic asymmetric
synthesis was probed using the standard benchmark catalytic
asymmetric cyclopropanation ofalkenes. A highestenantioselectivity
of 74% ee was obtained, demonstrating their potential. The use of
mixtures of these modular mono-oxazoline ligands is a powerful
approach for creating enormous chemical diversity.
These results support the conclusions obtained from both 1H
NMR spectroscopy and mass spectrometric experiments. Although
these calculations indicate a slightly greater stability for the het-
erocomplex (Cu(I)-1a1b(MeCN)2), this does not mean that this
complex is the most active chiral catalyst present.25
2.4. The active chiral catalyst in the heterocombination case
From the previous studies (Sections 2.2 and 2.3) there is strong
evidence that only three types of complex are present in the het-
erocombination case; i.e., two homocomplexes and one hetero-
complex. Given the impossibility of isolating the heterocomplexes,
attempts were made at trying to identify patterns in this data,
which would indicate the active chiral catalyst under these con-
ditions. However, this proved to be a very non-trivial exercise, as
there was very little variation in the enantioselectivities and dia-
stereoselectivities for the various systems studied (13 and 16%
maximum for the enantioselectivity and diastereoselectivity, re-
spectively). Only the results obtained with a-methylstyrene were
considered, as the yields obtained with styrene were very poor and
no meaningful conclusions could be drawn under these circum-
stances. However, the following key observations were made with
4. Experimental
4.1. General remarks
a
-methylstyrene leading to some weak conclusions. It was ob-
served that in some cases the enantioselectivities for the cis and
trans isomers were an average of those values obtained for the
homocomplex cases. For example, in the case of the hetero-
combination of 1a/1b ees of 31% (cis) and 52% (trans) with a de of
12% (Table 2 entry 12) were obtained, whereas for the homo-
combination cases, ees of 22% (cis) and 42% (trans) with a de of 10%
were obtained for 1a (Table 2, entry 2) and 41% (cis) and 63% (trans)
with a de of 16% for 1b (Table 2, entry 3). This situation could arise
from two scenarios: (1) the two homocomplexes are equally active
and the observed values, are in fact, the average for these two
predominate catalytic cycles or (2) the heterocomplex is the most
active catalyst, but it gives stereoselectivities that are the average of
that achieved with the two homocomplexes. However, in most
other cases the enantioselectivities approached only the values
obtained for one of the homocomplexes. For example, in the same
cyclopropanation reaction with the heterocombination 1b/1c the
results were 38% ee (cis) and 57%ee (trans) with a de of 12% (Table 2,
entry 14), and this result approximates the results obtained with
the homocomplex derived from 1b (Table 2, entry 3). In the case of
the heterocombination 1c/2 (Table 2, entry 15), there was again an
alignment towards the homocomplex formed from 1c (Table 2,
entry 5), which was again observed with the heterocombination
3a/3b (Table 2, entry 18) where there was an alignment towards the
homocomplex 3b (Table 2, entry 9, when both the ees and des are
taken into consideration).
trans-2,4-Dimethoxycinnamic acid 6 was obtained using the
literature method.16 All reagents were obtained from Aldrich, Fluka,
Alfa Aesar or Acros. Solvents were dried using common laboratory
methods. Column chromatography was carried out on silica gel
(SDS, 70e200
mm) and flash column chromatography (Merck,
40e63 m and SDS, 40e63
m
mm). TLC was carried out on aluminium
backed Kisel-gel 60 F254 plates (Merck). Plates were visualised ei-
ther by UV light or phosphomolybdic acid in ethanol. Gas chro-
matographic (GC) analyses of the products were performed on
a HewlettePackard (HP) 6890 series instrument equipped with
a flame ionization detector (FID). The chromatograph was fitted
with a cyclosil-B capillary column (30 m, 250 mm, 0.25 mm) (Agilent
112-2532). The melting point was recorded on a Barnstead Elec-
trothermal 9100 apparatus and was uncorrected. The 1H NMR
spectra were recorded on either a Bruker AMX300 (1H: 300.13 MHz
and 13C: 75 MHz) or a Bruker Avance III instrument (1H: 400 MHz
and 13C: 100 MHz) using CDCl3 as solvent and TMS as internal
standard. Mass spectra were recorded on a VG Autospec M(Wa-
ters-Micromass) spectrometer using the FAB technique, Waters-
Micromass GC-TOF and MicroTOF Focus (Bruker Daltonics)
using the TOF technique and electron spray ionization (ESI) mass
spectra were performed on
a Brucker Daltonics Apex-Qe
instrument at 300.0. Infra-red spectra were measured with
a PerkineElmer Paragon 1000 model. Specific rotations were
measured on a PerkineElmer 241 polarimeter.
This pattern was again repeated for 1a/1c (Table 2, entry 13)
where there was an alignment towards the homocomplex formed
from 1c (Table 2, entry 5). These latter results seem to indicate that
one of the homocomplexes is the most active catalyst with certain
combinations of ligands!
4.2. Synthesis of Arylid-OX 1
4.2.1. General procedure for the synthesis of cinnamamides (12a,
12b). A dry two-necked round bottom flask (50 mL) equipped with

4646
E.P. Carreiro et al. / Tetrahedron 67 (2011) 4640e4648
a magnetic stir bar was charged with trans-2,4-dimethoxycinnamic
acid 6 (2.0 g, 9.61 mmol), dimethylformamide (0.08 mL, 1.03 mmol)
and CH2Cl2 (20 mL). The solution was cooled to 0 ꢀC, and oxalyl
chloride (1.73 mL, 1.98 mmol) was added dropwise over a 30 min
period and the solution was stirred at room temperature until the
evolution of gas ended. The solvent was evaporated in vacuo to give
trans-2,4-dimethoxycinnamoyl chloride 9 as a dark green solid (due
to the unstable nature of this compound it was stored in the freezer
at ꢁ10 ꢀC). Yield: 2.17 g (100%). A two-necked round bottom flask
(50 mL) fitted with a magnetic stirring bar was charged with a so-
lution of (S)-valinol (0.796 g, 7.72 mmol) and dry CH2Cl2 (15 mL)
and the solution was cooled to 0 ꢀC using an ice bath. Dry trie-
thylamine (1.08 mL, 7.72 mmol) was added via syringe. A solution of
trans-2,4-dimethoxycinnamoyl chloride 9 (1.00 g, 4.41 mmol) in
CH2Cl2 (5 mL) was slowly added via syringe to the vigorously stirred
reaction mixture over 30 min. The ice bath was removed, and the
reaction mixture was stirred at room temperature for a further 4 h.
The reaction mixture was washed with 2 M HCl (12 mL), saturated
aqueous NaHCO3 (15 mL) and the aqueous layer was back-extracted
with CH2Cl2 (15 mL). The combined organic extracts were washed
with brine (15 mL), and the aqueous layer was back-extracted
with CH2Cl2 (15 mL). The combined organic extracts were dried
over anhydrous MgSO4, filtered and concentrated in vacuo
to give (S)-trans-N-(1-hydroxy-3-methylbutan-2-yl)-2,4-dimethox-
ycinnamamide 12a as a yellow solid. The crude product was puri-
fied by recrystallization (EtOAc/hexane) to afford amide 12a as
a white solid. Yield: 0.492 g (81%); mp 140.4e141.8 ꢀC. 1H NMR
the crude product. The crude product was purified by column chro-
matography (silica gel, EtOAc/Hex (1:1)) giving the (ꢁ)-(S)-trans-2-
(2,4-dimethoxyphenyl)-(4-isopropyloxazoline-2-yl)ethene 1a as
ayellowoil.Yield:0.374 g(70%).1HNMR(300 MHz, CDCl3):
d¼7.55 (d,
1H, J¼16.4 Hz, RCH]CHR), 7.41 (d, 1H, J¼8.7 Hz, CH(Ar)), 6.66 (d, 1H
J¼16.4 Hz, RCH]CHR), 6.51e6.43 (m, 2H, CH(Ar)), 4.33 (m, 1H, CH),
4.02e4.00 (m, 2H, CH2), 3.84 (s, 3H, eOCH3), 2.92 (s, 3H, eOCH3),
1.7e1.8 (m, 1H, CH), 1.01 (d, 1H J¼6.7 Hz, e(CH3)2), 0.913 (d, 1H
J¼6.7 Hz, e(CH3)2) ppm. 13C NMR (75 MHz, CDCl3):
¼163.9, 161.8,
d
159.0, 134.7, 129.2, 117.3, 114.0, 105.0, 98.3, 72.4, 69.7, 55.0, 32.8, 18.9,
18.2 ppm. IR (NaCl, CH2Cl2): nmax 3055, 2969, 1743, 1604, 1505, 1465,
22
1301, 1214, 1162, 1033, 777, 669 cmꢁ1. [
FAB-MS m/z: 276.16 [MþH].
a]
ꢁ51.56 (c 1.44, CHCl3).
D
Compound 1b: Using the same procedure as described pre-
viously, cinnamamide 12b (0.40 g, 1.17 mmol) was reacted with
methanesulfonyl chloride (0.20 g, 1.76 mmol) and dry triethyl-
amine (0.49 mL, 3.51 mmol) to give the (S)-trans-2-(2,4-dime-
thoxyphenyl)-(4-benzyloxazoline-2-yl)ethene 1b as a yellow oil
after purification by column chromatography (silica gel, EtOAc/Hex
(1:1)). Yield: 0.339 g (90%). 1H NMR (300 MHz, CDCl3):
d¼7.56 (d,
1H, J¼16.4 Hz, RCH]CHR), 7.42 (d, 1H, J¼8.5 Hz, CH(Ar)), 7.34e7.22
(m, 5H, CH(Ar)), 6.65 (d, 1H, J¼16.4 Hz, CRH]CHR), 6.52e6.44 (M,
2H, CH(Ar)), 4.55e4.44 (m, 1H, CH), 4.03 (t, 1H, J¼8 Hz, CHH), 4.26
(t, 1H, J¼8 Hz, CHH), 3.85 (s, 3H, eOCH3), 3.83 (s, 3H, eOCH3),
3.21e3.13 (m, 1H, CHH), 2.69 (dd, 1H J¼8.7 and 13.7 Hz, CHH) ppm.
13C NMR (75 MHz, CDCl3):
d
¼164.6, 161.9, 159.0, 138.2, 135.2, 129.4,
129.2, 128.6, 126.4, 117.3, 113.3, 105.1, 98.4, 71.4, 67.7, 55.4, 41.9 ppm.
(300 MHz, CDCl3):
d
¼7.81 (d, 1H J¼15.6 Hz, RCH]CRH), 7.39 (d, 1H
IR (NaCl, CH2Cl2): nmax 3055, 2955, 1714, 1640, 1603, 1506, 1460,
22
J¼8.4 Hz, CH(Ar)), 6.52e6.45 (m, 3H, RCH]CHR, CH(Ar)), 5.81 (d,
1H J¼8.1 Hz, NH), 3.86e3.83 (m, 1H, CH), 3.86 (s, 3H, eOCH3), 3.83
(s, 3H, eOCH3), 3.77e3.69 (m, 2H, CH2), 1.99e1.93(m, 1H, CH),
1.02(s, 3H, e(CH3)2), 1.00(s, 3H, e(CH3)2) ppm. 13C NMR (75 MHz,
1248, 1214, 1031, 939, 842, 775, 693 cmꢁ1. [
a
]
þ159.0 (c 0.52,
D
CHCl3). ESI-MS m/z: 324.16 [MþH].
4.3. Synthesis of Arylid-OX 2
CDCl3):
d
¼168.0, 162.1, 159.8, 136.6, 130.4, 119.7, 116.8, 105.0, 98.3,
63.9, 57.0, 55.3, 29.2, 19.5, 19.0 ppm. IR (KBr): nmax 3371, 3311, 2960,
A dry two-necked round bottom flask (25 mL) equipped with
a magnetic stirring bar was charged with trans-cinnamic acid 7
(1.5 g,10 mmol), dimethylformamide (0.1 mL,1.3 mmol) and CH2Cl2
(15 mL). The solution was cooled to 0 ꢀC, and oxalyl chloride (1.1 mL,
12.7 mmol) was added dropwise over a 30 min period and the so-
lution was stirred at room temperature until the evolution of gas
ended. The solvent was evaporated in vacuo to give trans-cinnamoyl
chloride 10 as a dark green solid (due to the unstable nature of this
compound it was stored in the freezer at ꢁ10 ꢀC). A two-necked
round bottom flask (25 mL) fitted with a magnetic stirring bar was
charged with a solution of (S)-phenylglycinol (2.06 g, 15 mmol) and
dry CH2Cl2 (15 mL) and the solution was cooled to 0 ꢀC using an ice
bath. Dry triethylamine (2.01 mL,15 mmol) was added via syringe. A
solution of trans-cinnamoyl chloride 10 (all quantity, 10 mmol) in
CH2Cl2 (5 mL) was slowly added via syringe to the vigorously stirred
reaction mixture over 30 min. The ice bath was removed, and the
reaction mixture was stirred at room temperature for a further 4 h.
The reaction mixture was washed with 2 M HCl (15 mL), saturated
aqueous NaHCO3 (15 mL) and the aqueous layer was back-extracted
with CH2Cl2 (15 mL). The combined organic extracts were washed
with brine (15 mL), and the aqueous layer was back-extracted with
CH2Cl2 (15 mL). The combined organic extracts were dried over
anhydrous MgSO4, filtered and concentrated in vacuo to give
(S)-trans-N-(2-hydroxy-1-phenylethyl)-cinnamamide 13 as a yel-
low solid. The crude product was purified by recrystallization
(EtOAc/hexane) toaffordamide 13 as awhite solid (1.389 g, 52%);mp
2864, 1643, 1603, 1568, 1531, 1460, 1417, 1355, 1304, 1260, 1216,
22
1163, 1138, 1075, 1042, 977, 934, 847, 800, 680, 639, 600 cmꢁ1. [
a]
D
ꢁ85.22 (c 0.67, CHCl3). FAB-MS m/z: 294.17 [MþH].
Compound 12b: The same procedure as described previously
was used in the reaction of trans-2,4-dimethoxycinnamoyl chloride
9 (1.0 g, 4.41 mmol) with (S)-phenylalalinol (1.167 g, 7.72 mmol)
and dry triethylamine (1.08 mL, 7.72 mmol) to give (S)-trans-N-(1-
hydroxy-3-propan-2-yl)-2,4-dimethoxycinnamamide
12b
as
a white solid after purification by recrystallization (EtOAc/hexane);
Yield: 0.702 g (47%); mp 145.0e146.7 ꢀC. 1H NMR (300 MHz,
CDCl3):
d
¼7.78 (d, 1H, J¼15.6 Hz, RCH]CHR), 7.36 (d, 1H, J¼8.4 Hz,
CH(Ar)), 7.31e7.21 (m, 5H,CH(Ar)), 6.48e6.36 (m, 3H, RCH]CHR,
CH(Ar)), 5.91 (d, 1H, J¼7.5 Hz, NH), 4.28 (m(br s), 1H, CH), 3.83 (s,
3H, eOCH3), 3.82 (s, 3H, eOCH3), 3.78e3.63 (m, 2H, CH2), 2.95 (d,
2H J¼7.2 Hz, CH2) ppm. 13C NMR (75 MHz, CDCl3/MeOD):
¼167.8,
d
162.0, 159.4, 137.8, 136.2, 129.9, 129.0, 128.2, 126.1, 118.3, 116.5,
104.9, 98.1, 62.8, 55.1, 52.5, 36.6 ppm. IR (KBr): nmax 3399, 3017,
2955, 2859, 1637, 1591, 1534, 1504, 1439, 1300, 1280, 1048, 980, 827,
22
759, 646, 596 cmꢁ1. [
342.1698 [MþH].
a
]
ꢁ63.37 (c 0.98, CHCl3). ESI-MS m/z:
D
4.2.2. General procedure for the synthesis of Arylid-OX (1a, 1b). A
solution of methanesulfonyl chloride (0.234 g, 2.04 mmol) in dry
CH2Cl2 (1 mL) was added dropwise over 20 min to a solution of cin-
namamide 12a (0.4 g, 1.36 mmol) and dry triethylamine (0.57 mL,
4.08 mmol) in dry CH2Cl2 (15 mL) and the solution was stirred be-
tween ꢁ5 and ꢁ10 ꢀC. The reaction mixture was allowed to warm to
room temperature and stirring was continued for 3 days. The reaction
mixture was then poured into a saturated aqueous NH4Cl solution
(15 mL). The organic layer was separated and the aqueous layer was
extracted with CH2Cl2 (2ꢂ10 mL). The combined organic layers were
washed with brine, dried (MgSO4), filtered and concentrated to afford
190.1e191.1 ꢀC. 1H NMR (400 MHz, CD3OD):
d
¼7.54e7.24 (m, 12H,
CH(Ar), NH, RCH]CRH), 6.73 (d,1H J¼15.7 Hz, RCH]CHR), 5.11 (br s,
1H, CH), 3.81e3.77 (m, 2H, CH2) ppm. 13C NMR (100 MHz, CD3OD):
d
¼168.5, 142.1, 141.3, 136.4, 130.8, 129.9, 129.6, 128.9, 128.5, 128.1,
122.0, 66.3, 57.3 ppm. IR (KBr): nmax 3304, 3062, 3028, 3062, 2953,
2859, 1654, 1623, 1547, 1494, 1450, 1355, 13,345, 1233, 1215, 1160,
21
1074, 1058, 972, 863, 700, 528 cmꢁ1. [
a
]
D ꢁ20.12 (c 0.815, MeOH).

E.P. Carreiro et al. / Tetrahedron 67 (2011) 4640e4648
4647
TOF-MS m/z: 268.14 [MþH]þ. A solution of methanesulfonyl chlo-
ride (0.411 g, 3.59 mmol) in dry CH2Cl2 (1 mL) was added dropwise
over 20 min toa solutionofcinnamamide 13 (0.64 g, 2.39 mmol) and
dry triethylamine (1 mL, 7.17 mmol) in dry CH2Cl2 (15 mL) and the
solution was stirred between ꢁ5 and ꢁ10 ꢀC. The reaction mixture
was allowed to warm to room temperature and stirring was con-
tinued for 3 days. The reaction mixture was then poured into a sat-
urated aqueous NH4Cl solution (15 mL). The organic layer was
separated and the aqueous layer was extracted with CH2Cl2
(2ꢂ10 mL). The combined organic layers were washed with brine,
dried (MgSO4), filteredand concentratedto afford the crudeproduct.
The crude product was purified by flash column chromatography
(silica, EtOAc/Hex (1:9)) giving the (þ)-(S)-trans-2-phenyl-(4-phe-
nyloxazoline-2-yl)ethene 2 as a white solid (0.455 g, 76%); mp:
(1.54 mL, 11 mmol) to give (S)-trans-N-(1-hydroxy-3,3-methyl-
butan-2-yl)-but-2-enamide 14a as a white solid after purification
by flash column chromatography (silica, EtOAc); Yield: 0.49 g
(33%); mp 92.3e93.8 ꢀC. 1H NMR (400 MHz, CDCl3):
d¼6.78 (dq, 1H
J¼6.8, 15.2 Hz, MeCH]CHR), 6.23 (d, 1H J¼9.6 Hz, NH), 5.84 (dd, 1H
J¼1.6, 15.2 Hz, MeCH]CHR), 3.76e3.70 (m, 1H, CH), 3.67e3.59 (m,
2H, CH2), 1.82e1.90 (m, 1H, CH), 1.8 (dd, 3H J¼1.5, 6.9 Hz, Me), 0.92
(d, 3H J¼6.8 Hz, CH(CH3)2), 0.89 (d, 3H J¼6.8 Hz, CH(CH3)2) ppm.
13C NMR (100 MHz, CDCl3):
18.8, 17.6 ppm. IR (KBr): nmax 3433, 3296, 3064, 3028, 2953, 2872,
d
¼167.0, 140.0, 125.1, 63.4, 56.9, 28.8,
1670, 1620, 1597, 1542, 1464, 1445, 1391, 1375, 1357, 1226, 1066, 975,
22
921, 839, 728, 685, 653, 523 cmꢁ1. [
TOF-MS m/z: 172.13 [MþH].
a]
ꢁ44.56 (c 0.86, CHCl3).
D
59e60 ꢀC. 1H NMR (400 MHz, CDCl3):
d
¼7.53e7.49 (m, 3H, CH(Ar)
4.4.2. General procedure for the synthesis of Propen-OX (3a, 3b). A
solution of methanesulfonyl chloride (0.42 g, 3.65 mmol) in dry
CH2Cl2 (1 mL) was added dropwise over 20 min to a solution of
amide 14b (0.5 g, 2.44 mmol) and dry triethylamine (1 mL,
7.32 mmol) in dry CH2Cl2 (20 mL) and the solution was stirred
between ꢁ5 and ꢁ10 ꢀC. The reaction mixture was allowed to
warm to room temperature and stirring was continued for 3 days.
The reaction mixture was then poured into a saturated aqueous
NH4Cl solution (15 mL). The organic layer was separated and the
aqueous layer was extracted with CH2Cl2 (2ꢂ10 mL). The combined
organic layers were washed with brine, dried (MgSO4), filtered and
concentrated to afford the crude product. The crude product was
purified by flash column chromatography (silica, EtOAc/Hex (1:9))
giving the (S)-trans-(4-phenyloxazoline-2-yl)prop-2-ene 3b as
a colourless oil. Yield: 0.33 g (72%). 1H NMR (400 MHz, CDCl3):
and RCH]CHR), 7.45e7.33 (m, 5H, CH(Ar)), 7.31e7.27 (m, 3H,
CH(Ar)), 6.75 (d,1H J¼16.3 Hz, RCH]CHR), 5.33 (dd,1H J¼8.4,10 Hz,
CHH), 4.72 (dd, 1H J¼8.4, 10 Hz, CHH), 4.2 (t, 1H J¼8.2 Hz, CH) ppm.
13C NMR (100 MHz, CDCl3):
128.7, 127.5, 127.5, 126.6, 115.0, 74.3, 70.0 ppm. IR (KBr): nmax 3057,
d
¼164.4,142.2,140.4,135.1,129.5,128.8,
3030, 2955, 2889,1650,1605,1492,1474,1450,1361,1244,1201,1074,
20
997, 981, 851, 760, 698, 536 cmꢁ1. [
MS m/z: 250.13 [MþH].
a]
D ꢁ12.12 (c 0.99, CHCl3). TOF-
4.4. Synthesis of Propen-OX 3
4.4.1. General procedure for the synthesis of amides (14a, 14b). A dry
two-necked round bottom flask (25 mL) equipped with a magnetic
stir bar was charged with trans-but-2-enoic acid 8 (1.0 g, 12 mmol),
dimethylformamide (0.15 mL, 1.95 mmol) and CH2Cl2 (15 mL). The
solution was cooled to 0 ꢀC, and oxalyl chloride (1.27 mL, 15 mmol)
was added dropwise over a 30 min period and the solution was
stirred at room temperature until the evolution of gas ended. The
solvent was evaporated in vacuo to give trans-but-2-enoyl chloride
11 as a green oil (due to the unstable nature of this compound it was
stored in the freezer at ꢁ10 ꢀC). A two-necked round bottom flask
(25 mL) fitted with a magnetic stirring bar was charged with a so-
lution of (S)-phenylglycinol (2.06 g, 15 mmol) and dry CH2Cl2
(15 mL) and the solution was cooled to 0 ꢀC using an ice bath. Dry
triethylamine (2.1 mL, 15 mmol) was added via syringe. A solution
of trans-but-2-enoyl chloride 11 (12 mmol) in CH2Cl2 (5 mL) was
slowly added via syringe to the vigorously stirred reaction mixture
over 30 min. The ice bath was removed, and the reaction mixture
was stirred at room temperature for a further 4 h. The reaction
mixture was washed with 2 M HCl (12 mL), saturated aqueous
NaHCO3 (15 mL) and the aqueous layer was back-extracted with
CH2Cl2 (15 mL). The combined organic extracts were washed
with brine (15 mL), and the aqueous layer was back-extracted with
CH2Cl2 (15 mL). The combined organic extracts were dried over
anhydrous MgSO4, filtered and concentrated in vacuo to give
(S)-trans-N-(2-hydroxy-1-phenylethyl)- but-2-enamide 14b as
a yellow solid. The crude product was purified by column chro-
matography (silica flash, EtOAc/Hex (7:2) and AcOEt) to afford
amide 14b as a white solid. Yield: 1.924 g (78%); mp 95.5e96.9 ꢀC.
d
¼7.35e7.26 (m, 5H, H(Ar)), 7.25e7.23 (m, 5H, H(Ar)), 6.69 (dq, 1H
J¼6.9, 13.7 Hz, MeCH]CHR), 6.1 (dd, 1H J¼6.9, 13.7 Hz, MeCH]
CHR), 5.22 (t, 1H J¼9 Hz, CHH), 4.62 (dd, 1H J¼8.4, 10 Hz, CHH), 4.1
(t, 1H J¼8.2 Hz, CH), 1.9 (dd, 3H J¼1.7, 6.8 Hz, CH3) ppm. 13C NMR
(100 MHz, CDCl3):
d
¼163.9, 142.4, 139.8, 128.4, 127.5, 126.6, 126.3,
126.2, 118.9, 74.2, 69.7, 18.4 ppm. IR (NaCl): nmax 3272, 3061, 3030,
2968, 2912, 1732, 1676, 1644, 1612, 1585, 1494, 1474, 1454, 1360,
22
1178, 997, 968, 903, 761, 734, 701, 667, 542 cmꢁ1; [
1.28, CHCl3). TOF-MS m/z: 188.11 [MþH].
a
]
D ꢁ113.80 (c
Compound 3a: Using the same procedure as described pre-
viously, amide 14a (0.35 g, 2.04 mmol) was reacted with meth-
anesulfonyl chloride (0.35 g, 3.07 mmol) and dry triethylamine
(0.86 mL, 6.13 mmol) to give the (S)-trans-(4-isopropyloxazoline-2-
yl)prop-2-ene 3a as a colourless oil after purification by flash col-
umn chromatography (EtOAc/Hex (1:5)). Yield: 0.193 g (62%). 1H
NMR (400 MHz, CDCl3):
d
¼6.54 (dq, 1H J¼6.9, 13.7 Hz, MeCH]
CHR), 5.96 (d, 1H J¼15.8 Hz, MeCH]CHR), 4.28e4.18 (m, 1H, CH),
3.94e3.89 (m, 2H, CH2), 1.82 (d, 3H J¼6.9 Hz, Me), 1.75e1.70 (m, 1H,
CH(CH2)2), 0.94 (d, 3H J¼6.8 Hz, CH(CH2)2), 0.89 (d, 3H J¼6.8 Hz,
CH(CH2)2) ppm. 13C NMR (100 MHz, CDCl3):
d
¼162.7, 138.8, 119.1,
72.2, 69.6, 32.7, 18.8, 18.2, 18.2 ppm. IR (NaCl): nmax 3272, 2961,
2912, 2874, 1722, 1679, 1648, 1614, 1468, 1446, 1357, 1309, 1250,
22
1192, 1178, 1103, 995, 970, 911, 815, 795, 733, 668, 529 cmꢁ1. [
a]
D
ꢁ69.06 (c 0.96, CHCl3). TOF-MS m/z: 154.13 [MþH].
1H NMR (400 MHz, CDCl3):
d
¼7.35e7.22 (m, 5H, H(Ar)), 6.81 (dq,1H
4.5. Representative cyclopropanation using Cu(MeCN)4]PF6
pre-catalyst
J¼6.8, 15 Hz, MeCH]CHR), 6.61 (d, 1H J¼6.8 Hz, NH), 5.83 (dd, 1H
J¼1.5, 15 Hz, MeCH]CHR), 5.06 (q, 1H J¼5.4, 6 Hz, CH), 4.07e3.77
(m, 2H, CH2), 1.83 (dd, 3H J¼6.8, 20.2 Hz, Me) ppm. 13C NMR
[Cu(MeCN)4]PF6 (0.36 mol % or 2 mol %) was added to a two-
neck round-bottomed flask containing the chiral oxazoline
(1.13e6.3 mol %) in CH2Cl2 (1 mL) and the solution was stirred at
room temperature for 15 min under a nitrogen atmosphere. Alkene
(2.66 mmol) and a solution of ethyl diazoacetate (7.45 mmol) or
(0.665 mmol) in CH2Cl2 (1 mL) was then added to the reaction
mixture over a period of 8 h using a syringe pump. After the ad-
dition of ethyl diazoacetate, the mixture was stirred for 16 h. The
reaction mixture was firstly passed through a short pad of silica gel
(100 MHz, CDCl3):
124.7, 66.2, 55.9, 17.7 ppm. IR (KBr): nmax 3292, 3080, 3030, 2918,
d
¼166.5, 140.7, 139.1, 128.7, 127.9, 127.7, 126.7,
1671, 1631, 1545, 1494, 1448, 1359, 1333, 1283, 1235, 1120, 1074,
1051, 968, 917, 850, 7550, 7000, 666, 524 cmꢁ1. [
a
]
þ68.01 (c
D22
0.965, CHCl3). TOF-MS m/z: 206.12 [MþH].
Compound 14a: The same procedure as described previously
was used in the reaction of trans-but-2-enoyl chloride 11
(8.6 mmol) with (S)-valinol (1.14 g, 11 mmol) and dry triethylamine

4648
E.P. Carreiro et al. / Tetrahedron 67 (2011) 4640e4648
6. Carreiro, E. P.; Ramalho, J. P. P.; Rodrigues, A. I.; Burke, A. J. Tetrahedron:
Asymmetry 2009, 20, 1272.
(washed with CH2Cl2) to remove the catalyst complex, the products
were then isolated by column chromatography (hexane/EtOAc 9:1).
All cyclopropane products were obtained as a mixture of cis and
trans diastereoisomers, the ratio was determined using GC analysis.
Isolated yields, diastereoselectivities and enantioselectivities are
given in Tables 1 and 2.
7. Ikeda, S.; Kondo, H.; Arii, T.; Odashima, K. Chem. Commun. 2002, 2422.
ꢀ
8. Bonini, B. F.; Capito, E.; Comes-Franchini, M.; Ricci, A.; Bottoni, A.; Bernardi, F.;
Miscione, G. P.; Giordano, L.; Cowley, A. R. Eur. J. Org. Chem. 2004, 4442.
9. Chuzel, O.; Magnier-Bouvier, C.; Schulz, E. Tetrahedron: Asymmetry 2008, 19,
1010.
10. (a) Glorius, F. Angew. Chem., Int. Ed. 2004, 43, 3364; (b) Johnson, J. B.; Rovis, T.
€
Angew. Chem., Int. Ed. 2008, 47, 840; (c) Defeiber, C.; Grutzmacher, H.; Carreira,
€
E. M. Angew. Chem., Int. Ed. 2008, 47, 4482; (d) Hahn, B. T.; Tewes, F.; Frohlich,
Acknowledgements
R.; Glorius, F. Angew. Chem., Int. Ed. 2009, 48, 1.
11. (a) Reetz, M. T.; Sell, T.; Meiswinkel, A.; Mehler, G. Angew. Chem., Int. Ed. 2003,
42, 790; (b) Reetz, M. T.; Li, X. G. Angew. Chem., Int. Ed. 2005, 44, 2959; (c) Reetz,
M. T.; Fu, Y.; Meiswinkel, A. Angew. Chem., Int. Ed. 2006, 45, 1412.
~
We are grateful for financial support from the Fundac¸ ao para
ˇ
a Ciencia e a Tecnologia (FCT) (project; PPCDT/QUI/55779/2004)
through POCI 2010, supported by the European community fund,
FEDER. FCT is also acknowledged for the award of a Ph.D. grant to EPC
(SFRH/BD/277688/2006). We acknowledge LabRMN at FCT-UNL for
the acquisition ofNMR spectra, the NMR spectrometers are partof the
National NMR Network and were purchased in the framework of the
National Programme for Scientific Re-equipment, contract REDE/
1517/RMN/2005, with funds from POCI 2010 (FEDER) and FCT.
~
12. (a) Pena, D.; Minnaard, A. J.; Boogers, J. A. F.; de Vries, A. H. M.; de Vries, J. G.;
Feringa, B. L. Org. Biomol. Chem. 2003, 1, 1087; (b) Duursma, A.; Hoen, R.;
Schuppan, J.; Hulst, R.; Minnaard, A.; Feringa, B. L. Org. Lett. 2003, 5, 3111.
13. Carreiro, E. P.; Chercheja, S.; Moura, N. M. M.; Gertrudes, S. C.; Burke, A. J. Inorg.
Chem. Commun. 2006, 9, 823.
14. Burke, A. J.; Carreiro, E. P.; Chercheja, S.; Moura, N. M. M.; Ramalho, J. P. P.;
Rodrigues, A. I.; Santos, C. I. M. J. Organomet. Chem. 2007, 692, 4863.
15. Carreiro, E. P.; Chercheja, S.; Burke, A. J.; Ramalho, J. P. P.; Rodrigues, A. I. J. Mol.
Catal. A: Chem. 2005, 236, 38.
16. Neustadt, B. R.; Smith, E. M.; Nechuta, T. L.; Bronnenkant, A. A.; Haslanger, M.;
Watkins, R. W.; Foster, C. J.; Sybertz, E. J. J. Med. Chem. 1994, 37, 2461.
17. Headen, T. F.; Howard, C. A.; Skipper, N. T.; Wilkinson, M. A.; Bowron, D. T.;
Soper, A. K. J. Am. Chem. Soc. 2010, 132, 5735.
References and notes
18. Burke, A. J.; Carreiro, E. P. unpublished results.
19. Stewart, J. J. P. J. Mol. Model. 2007, 13, 1173.
20. Stewart, J.J.P. MOPAC2007, Computational Chemistry, Version 7.319L web:
http://openmopac.net/.
1. (a) Reetz, M. T.; Sell, T. Tetrahedron Lett. 2000, 41, 6333; (b) Claver, C.; Fernandez,
E.; Gillon, A.; Heslop, K.; Hylett, D. J.; Martorell, A.; Orpen, A. G.; Pringle, P. G.
Chem. Commun. 2000, 961.
21. Meng, Q.; Li, M.; Tang, D.; Shen, W.; Zhang, J. J. Mol. Struct.: THEOCHEM 2004,
711, 193.
22. Beck, A. D. J. Chem. Phys. 1993, 98, 5648.
23. Lee, C.; Yang, W.; Parr, R. G. Physiol. Rev. 1988, 37, 785.
24. Schmidt, M. W.; Baldridge, K. K.; Boatz, J. A.; Elbert, S. T.; Gordon, M. S.; Jensen,
J. H.; Koseki, S.; Matsunga, N.; Nguyen, K. A.; Su, S. J.; Windus, T. L.; Dupuis, M.;
Montgomery, J. A. J. Comput. Chem. 1993, 14, 1347.
2. Reetz, M. T.; Mehler, G. Angew. Chem., Int. Ed. 2000, 39, 3889.
3. (a) Feringa, B. L. Acc. Chem. Res. 2000, 33, 346; (b) Yoon, T. P.; Jacobsen, E. N. Science
2003, 299,1691; (c) Hulst, R.; de Vries, N. K.; Feringa, B. L. Tetrahedron: Asymmetry
1994, 5, 699; (d) van den Berg, M.; Minnaard, A. J.; Schudde, E. P.; van Esch, J.; de
Vries, A. H. M.; de Vries, J. G.; Feringa, B. L. J. Am. Chem. Soc. 2000, 122, 11539; (e)
Teichert, J. F.; Feringa, B. L. Angew. Chem., Int. Ed. 2010, 49, 2486.
4. Braunstein, P.; Naud, F. Angew. Chem., Int. Ed. 2001, 40, 680.
5. Dakovic, S.; Liscic-Tumir, L.; Kirin, S. I.; Vinkovic, V.; Raza, Z.; Suste, A.; Sunjic, V.
J. Mol. Catal. A: Chem. 1997, 118, 27.
25. Wassenaar, J.; Jansen, E.; van Zeist, W.-J.; Bickelhaupt, F. M.; Siegler, M. A.;
Speck, A. L.; Reek, J. N. H. Nat. Chem. 2010, 2, 417.