Copper(I) and silver(I) complexes of 1,5-methylene- and diethylmethylene-bridged bis(oxazoline) ligands. In situ Cu(II)-catalyzed oxidation of methylene bridge

Article abstract of DOI:10.1016/j.tet.2004.06.117

Silver(I) and copper(I) complexes of C₂-symmetric bis(oxazoline) ligands were studied by UV, NMR, IR, EPR and ES-MS spectroscopies. The stability constants of the Ag-1a and Ag-1b complexes with 1:1 and 1:2 stoichiometries in acetonitrile were d

Full text of DOI:10.1016/j.tet.2004.06.117

Tetrahedron 60 (2004) 8079–8087
Copper(I) and silver(I) complexes of 1,5-methylene- and
diethylmethylene-bridged bis(oxazoline) ligands. In situ
Cu(II)-catalyzed oxidation of methylene bridge
a,
ˇ ´
Vesna Caplar, Zlata Raza, Marin Roje, Vladislav Tomisic, Gordan Horvat, Josip Pozar,
b
b
c
c
c
ˇ
*
ˇ
a
a
ˇ
´
Ivo Piantanida and Mladen Zinic
a
ˇ
´
Laboratory for Supramolecular and Nucleoside Chemistry, RuCer Boskovic Institute, PO Box 180, 10002 Zagreb, Croatia
b
ˇ
´
Laboratory for Stereoselective Catalysis and Biocatalysis, RuCer Boskovic Institute, PO Box 180, 10002 Zagreb, Croatia
cLaboratory of Physical Chemistry, Faculty of Science, University of Zagreb, PO Box 163, 10001 Zagreb, Croatia
Received 23 April 2004; revised 28 May 2004; accepted 24 June 2004
Available online 29 July 2004
Abstract—Silver(I) and copper(I) complexes of C2-symmetric bis(oxazoline) ligands were studied by UV, NMR, IR, EPR and ES-MS
spectroscopies. The stability constants of the Ag–1a and Ag–1b complexes with 1:1 and 1:2 stoichiometries in acetonitrile were determined
by NMR spectrometric titrations. The evidence of tetrahedral coordination for complex (Ag(1a)2(C was obtained from the complexation
induced shifts (CIS) and NOEs. Mass spectra revealed the Cu(II) mediated oxidation of methylene bridge in copper complexes of 1a and
1b, which was in accordance with the UV, NMR, IR and EPR findings. The efficiency of Cu(I) complexes of methylene-bridged
1,5-bis(oxazoline)s 1 as chiral catalysts in stereoselective cyclopropanation of styrene with ethyl diazoacetate, was compared to that of the
dialkylmethylene-bridged 1,5-bis(oxazoline)s 2.
q 2004 Elsevier Ltd. All rights reserved.
1. Introduction
Over the last years a wide variety of chiral, C2-symmetric
bis(oxazoline) ligands has been synthesized and their
metallic complexes used as stereoselective catalysts in
different reactions,1–4 including cyclopropanations5–8 and
aziridinations of olefins and imines,9–11 Diels–Alder and
hetero Diels–Alder reactions,12–14 1,3-dipolar cycloaddi-
tions,15 allylic displacement,16,17 addition of dialkylzinc to
aldehydes,18 organolithium addition to imines,19 hydrosilyl-
ative reduction,20,21 allylic oxidation of olefins,22 hydro-
silylation of ketones,23 Friedel–Crafts reaction,24 diene
Scheme 1.
cyclization/hydrosilylation,25 glyoxylate–ene reaction,26
Canizzaro reaction27 etc. The Lewis basicity of the nitrogen
Both types of the ligands are capable to form six-membered
donor atoms and the conformational rigidity of chelates
formed represent important structural features of C2-sym-
metric bis(oxazoline) ligands. Among 1,5-bis(oxazoline)
ligands, the most frequently used ones in catalytic
transformations can be divided into two groups: the
bis(oxazolines) with methylene spacer (I) and the bis(oxa-
zolines) with the dialkylmethylene spacer (II) (Scheme 1).
metal chelates, which fixes their conformation in nearly
planar geometry. Recently, we have prepared a series of
macrocyclic28 and acyclic29 1,5-bis(oxazoline) derivatives,
the latter possessing elongated aromatic arms of variable
length and flexibility attached on chiral centres. We have
also performed a detailed study of the structure, stoichi-
ometry and conformation of Cu(I) and Ag(I) complexes
with type II bis(oxazoline) ligands in solution.30 In the type
II of the 1,5-bis(oxazoline) ligands, the presence of
dialkylmethylene spacer fixes two oxazoline double bonds
in an isolated position. However, if the type I ligands are
Keywords: Bis(oxazoline); Metal complex; Chiral catalysts.
* Corresponding author. Tel.: C385-1-45-61-111; fax: C385-1-46-80-
195; e-mail: vcaplar@irb.hr
0040–4020/$ - see front matter q 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2004.06.117

ˇ
V. Caplar et al. / Tetrahedron 60 (2004) 8079–8087
8080
considered, the presence of the methylene bridge allows
formation of two tautomeric forms, Ia and Ib, the latter
possessing 4p conjugated system (Scheme 1). It should be
noted that removal of the NH proton in Ib may give an
anionic ligand which should exhibit different affinity toward
metal cations and as the consequence the properties of such
catalytic complex may also be altered. PE and NMR spectra
of bis(oxazolines) 1f and 1 h (Chart 1) revealed that the Ia
tautomeric form presents the dominant species in the gas
phase as well as in the solution. On the other hand, a UV
spectroscopic investigation of ligand 1h, indicated the
presence of a small amount of the enamine Ib in solution.31
To investigate possible differences between methylene- (1a,
1b) and the previously studied diethylmethylene (2a, 2b)30
bridged 1,5-bis(oxazoline) ligands (Chart 1) we undertook a
detailed spectroscopic study of 1a, 1b–Ag(I) and Cu(I)
complexes and the comparative study of the catalytic
properties of type I and type II (Chart 2) Cu(I) complexes in
the cyclopropanation of styrene with ethyl diazoacetate.
Chart 1.
For comparison, the cyclopropanations were also performed
by using the Cu(I) catalytic complexes of alkoxyphenyl
substituted ligands 1c, 1d, 1e29 and commercially available
ligands 1f, 1g, 1h and 2c, 2d, 2e lacking long substituents on
chiral centres. All of the 1,5-bis(oxazoline) ligands studied
are based on S-tyrosine and RK4-hydroxyphenylglycine.
The aromatic arms of variable length and flexibility attached
onto stereogenic centers of the ligands may influence both
their complexation and catalytic properties and may result
by different stereochemical outcome.
2. Results and discussion
2.1. Spectroscopic studies of 1a and 1b silver(I) and
copper(I) complexes in solution
Chart 2.
2.1.1. Silver(I) complexes. The UV spectral changes in the
titrations of 1a and 1b with AgC in MeCN were insufficient
Figure 1. Complexation-induced shifts (CIS) for 1a–AgC complexes.

ˇ
V. Caplar et al. / Tetrahedron 60 (2004) 8079–8087
8081
to enable accurate analysis, and changes remained only very
slight over longer period of time. Therefore, the formation
of AgC complexes of 1a and 1b in MeCN-d3 was studied by
1H NMR. The spectra of the free ligands were found
concentration-independent in the concentration range of
2!10K3 to 2!10K2 mol dmK3 showing lack of ligand
self-association.
are strongly shielded in (Ag(1a)2(C complex. In the 1:1
complex the protons close to the metal centre [C(4)–H,
C(5)–H and C(20)–H] should be strongly deshielded upon
AgC coordination by the through bond effects; in the 1:2
AgC–ligand complex, such effects are less pronounced
being distributed between two bis(oxazoline) units. Exactly
the same trend of induced shifts was observed for
(Ag(2a)2)C and (Ag(2a))C complexes and explained by
formation of the pseudotetrahedral AgC complex of 1:2
stoichiometry, with the cation bound to four nitrogen atoms
of two orthogonally oriented bis(oxazoline) units.30 In such
a complex, each bis(oxazoline) unit is located between two
C(4)–phenyl groups of the second ligand molecule, which
results through shielding of C(4)–H and C(20)–H. The
bridging methylene protons are shielded by two distant
phenyls of the C(40)–O-benzyl arms in the 1:2 complex.
Additional support for the structure of the 1:2 complex
comes from the analysis of the NOESY spectrum taken at
the c(AgC)/c(1a) ratio of 0.5.
Addition of AgC produced significant shifts in all of the
ligand resonances, without significant line broadening,
indicating fast equilibrium kinetics on the NMR time
scale. The observed complexation induced shifts (CIS;
DdZdHiligKdHicompl, ppm), were plotted against the
c(AgC)/c(1a,1b) ratio (Fig. 1). In the case of ligand 1a,
additions of AgC induced upfield shifts of C(2)–CH2
(methylene bridge), C(4)–H and C(20)–H (aromatic) signals,
until the ratio of 0.5 was reached followed by downfield
shifts of all resonances at higher ratios. The upfield shifts
observed up to the molar ratio of 0.5 suggests formation of
the (Ag(1a)2(C complex, whereas downfield shifts of all of
the ligand resonances in the c(AgC)/c(1a) range of 0.5–2.0
shows transformation of the first complex into (Ag(1a))C
one. Strong deshielding effects were observed for C(4)–H,
C(5)–Hcis and C(20)–H protons (see Figure 2 for numera-
tion) on going from 1:2 to 1:1 complex (Fig. 1). Also the
methylene bridge protons experienced the strongest
deshielding effect on going from the (Ag(1a)2)C towards
the (Ag(1a))C complex. This suggests that the same protons
Examination of the CPK model of 1a shows that the
majority of the observed NOE crosspeaks could be
explained by the favourable distances between respective
hydrogen atoms in a single 1a molecule. However, a clear
crosspeak corresponding to NOE interaction between
C(40)–OCH2 and C(4)–H is observed. The closest possible
distance between these protons in the model of 1a is over
32
˚
˚
6 A, well exceeding the limiting value of 2.2–5 A,
necessary for observation of the NOE effect. Thus, this
crosspeak could be explained only by mutual interaction of
two 1a ligands in the pseudotetrahedral [Ag(1a)2]C
complex (Fig. 2).
The additions of AgC to ligand 1b also induced upfield
shifts of C(4)–H, C(4)–CHa, C(20)–H (aromatic), and
C(40)–OCH2 signals until the ratio of 0.5 was reached;
further additions of the cation produced downfield shifts of
all resonances (Fig. 3). A clear maximum at the c(AgC)/
c(1b) ratio of 0.5 suggests initial formation of the AgC–1b
complex of a 1:2 stoichiometry (Fig. 4) which at higher
ratios transforms to the 1:1 complex. The strongest
Figure 2. Schematic presentation of (Ag(1a)2(C pseudotetrahedral
complex and intermolecular NOE interaction between C(4)H and OCH2–
Ph observed in the NOESY spectrum of [Ag(1a)2]C
.
Figure 3. Complexation induced shifts for 1b–AgC complexes.

ˇ
V. Caplar et al. / Tetrahedron 60 (2004) 8079–8087
8082
Figure 5. 1H NMR spectra of 1a in CD3CN at different c(CuC)/c(1a)
ratios.
Figure 4. Experimental (dots) and calculated (line) chemical shifts of
C(20)–H of 1b induced by addition of AgBF4 (solvent: MeCN-d3), rZ
c(AgC)/c(1b).
complexes with AgC in MeCN; generally, their stability
constants are for more than one order of magnitude higher
than those of the corresponding diethylmethylene bridged
derivatives 2a and 2b (except that of Ag(1b)2 being
somewhat lower than that of Ag(2b)2).30
shielding effects upon formation of (Ag(1b)2(C showed
C(4)–H, C(4)–CHa, the aromatic C(20)–H proton, as well as
the benzylic methylene group protons. The observed
downfield trend of the complexation induced shifts at
the c(AgC)/c(1b) range between 0.5 and 2.0, could be
explained by the disappearance of the [Ag(1b)2]C complex
and predominant formation of [Ag(1b)]C complex.
2.1.2. Copper(I) complexes. 1H NMR spectroscopic
titrations of 1a and the corresponding diethylmethylene
bridged 2a with (Cu(MeCN)4(PF6 showed significant
differences. The methylene bridged bis(oxazoline) 1a gave
at least two sets of signals upon addition of Cu(I) salt
indicating formation of mixture of complexes (Fig. 5). In
contrast, ligand 2a showed a single set of signals with
significant downfield shifts of the oxazoline protons in
accord with fast complex formation and simultaneous
coordination of the cation with both nitrogens (Fig. 6). In
addition, Cu(I) salt induced severe broadening of 1a lines
(as shown in Figure 5) and besides the signals of the free
ligand, new resonances shifted downfield for 0.3 ppm could
be observed. In the spectrum at c(CuC)/c(1a) molar ratio
of only 0.1 new resonances with integrals approximately
20% of those of the free ligand indicated the initial
formation of (Cu(1a)2(C complex. However, further
additions of CuC, (Fig. 5; c(CuC)/c(1a)Z0.3) resulted in
2.1.1.1. Stability constants.. Among several chemically
reasonable speciations used in the fitting of the spectro-
metric data, two (set by assuming the presence of [AgL]C
and [AgL2]C or [AgL2]C and [Ag2L2]C; L standing for
either 1a or 1b) gave satisfactory results. As the electrospray
mass spectrometry (ES-MS) of 1a and 1b AgC complexes
unambiguously indicated the formation of [AgL]C and
[AgL2]C, and no evidence of the presence of [Ag2L2]2C
species, we could exclude the formation of [Ag2L2]2C (see
Figures 1 and 2 in the Supplementary Material). The
electrospray mass spectra of AgC(1a) and AgC(1b)
solutions in MeCN showed peaks at m/zZ625 for
[Ag(1a)]nnC and 1145 for [Ag(1a)2]C and 653 for
[Ag(1b)]nnC and 1201 for [Ag(1b)2]C (Figs. 1 and 2,
Supplementary Material). The isotopic abundances of the
peaks at m/zZ625 and 653 showed peak separation of 1 Da,
corresponding to the singly charged species [Ag(1a)]C and
[Ag(1b)]C, respectively. The stability constants of the AgC
complexes calculated from the 1H NMR titration data
are collected in Table 1. Both, 1a and 1b form quite stable
1:1 (KAgLz105 MK1) and 1:2 (KAgL2z103K104 MK1
)
Table 1. Stability constants of AgC complexes with 1a, 1b, 2a and 2b in
MeCN determined by H NMR titrations
1
log Ka
Ag(1a)
Ag(1a)2
Ag(1b)
Ag(1b)2
Ag(2a)
Ag(2a)2
Ag(2b)
Ag(2b)2
4.8G0.4
4.1G0.2
5.0G0.2
2.8G0.2
3.5b
3.3b
3.9b
3.1b
Figure 6. 1H NMR spectra (600 MHz) obtained by titration of 2a in CD3CN
with [Cu(MeCN)4]PF6 at different ratios rZc(CuC)/c(2a).
a
KZ[AgLn]/[AgLnK1][L], L standing for 1a, 1b, 2a or 2b; nZ1,2.
See Ref. 30.
b

ˇ
V. Caplar et al. / Tetrahedron 60 (2004) 8079–8087
8083
severe line broadening and diminished intensity of oxazo-
line proton resonances, being characteristic for the presence
of a paramagnetic species; presumably air oxidation of CuC
to paramagnetic Cu2C occurred. The process can be also
observed visually through appearance of violet-blue colour.
The corresponding diethylmethylene bridged ligand 2a
could be titrated with CuC up to the c(CuC)/c(2a) molar
ratio of 1.0 lacking any line broadening (Fig. 6) or colour
change.
(broadening of the band) in the region of carbon–heteroatom
double bonds (about 1670 cmK1) indicating an oxidative
process on methylene bridge of 1a to carbonyl group by air-
oxygen, thus generating the new chromophore with three
conjugated double bonds. All other bands remained
unaffected. Opposite to this, IR spectra of AgC complexes
did not show any change during 4 weeks of standing.
The FTIR spectra of the 1a–AgC complex did not show any
change during 4 weeks, showing that the presence of copper
ions was necessary for oxidation. The mass spectra of the
freshly prepared acetonitrile solutions of (Cu(MeCN)4(PF6
and 1a and 1b, respectively, show the simultaneous
presence of the free ligand and 1:1 and 1:2 complexes
(Fig. 7a here and Fig. 3a in Supplementary Material).
However, the same solutions after standing for 24 h at room
temperature gave different mass spectra; the peaks of free
ligands and 1:1 and 1:2 complexes disappeared and new
peaks corresponding to 1:1- and 1:2-complexes increased
for 14 and 28 mass units, respectively, were observed
(Fig. 7b, here and Fig. 3b, Supplementary Material). In
contrast, the mass spectra of AgBF4 and 1a and 1b
acetonitrile solutions did not show any changes after 24 h
at room temperature (see Figures 1 and 2 in Supplementary
Material).
In order to shed more light on the transformations of copper
complexes of 1a and 1b the IR- and MS-investigations were
performed. The IR spectra of freshly prepared and aged
(four weeks) Cu–1a complexes showed changes
The increase of masses of the complexes for 14 and 28 mass
units clearly show that oxidation of methylene bridge into
carbonyl group took place accompanied presumably by the
reduction of Cu(II) back to Cu(I), as outlined in Scheme 2.
The presence of Cu(II) was additionally confirmed by EPR
spectrum which showed the formation of a para-magnetic
Cu(II) immediately after mixing the solutions of Cu(I) salt
and ligand 1a. These results are in agreement with those of
NMR and FTIR experiments. Similar oxidation of methyl-
ene bridge in Cu(II)–benzimidazole complex was recently
reported.33
All our attempts to avoid oxidation by working under argon
atmosphere failed, since even the very small amount of
residual oxygen was sufficient to initiate the deleterious
oxidative processes.
The UV spectral changes observed upon mixing of
equimolar amounts of 1a and [Cu(MeCN)4]PF6 (cZ1!
10K4 mol dmK3) in acetonitrile at 25 8C can be grouped
into two types.
As can be seen in Figure 8, the addition of CuC to 1a
solution caused a rather significant hyperchromic effect on
the ligand UV spectrum, accompanied by the occurrence of
a shoulder at z300 nm. According to ES-MS results
(Fig. 7) and the previously reported spectrophotometric
Figure 7. ES-MS spectra of Cu–1a complex in acetonitrile, freshly
prepared (a) and after 24 h standing at room temperature (b).
Scheme 2.

ˇ
V. Caplar et al. / Tetrahedron 60 (2004) 8079–8087
8084
Figure 8. Time dependence of the UV spectra of acetonitrile solution containing 1a (cZ1!10K4 mol dmK3) and [Cu(MeCN)4]PF6 (cZ1!10K4 mol dmK3).
Spectra are labelled with the approximate times after mixing of 1a and CuC. lZ1 cm, tZ(25.0G0.1) 8C. (a) Spectrum of the free ligand solution and its
changes during the first 10 min after the addition of CuC. (b) Spectra corresponding to the ‘second process’, recorded in the period of several hours.
titrations of 2a and 2b with CuC,30 these spectral changes
can be attributed to the formation of [Cu(1a)]C and
[Cu(1a)2]C complexes. The initial increase in the spectrum
intensity (up to approx. 5 min from mixing) was followed by
a drop in absorbance at spectral maximum (at about 280 nm)
of several hours, by a hypsochromic shift of the spectrum
and its broadening, and by disappearance of a shoulder at
z300 nm (Fig. 8).
[Cu(MeCN)4]PF6 at higher concentrations, the resulting
acetonitrile solution appeared pronouncedly violet-blue.
The colour changed with time to yellow-green, and finally
became dark brown. The starting blue colour was assumed
to be due to the presence of Cu2C in the solution. For this
reason, an experiment was performed in which a solution
containing 5.5!10K3 mol dmK3 of both CuC and 1a was
prepared, and its spectrum dependence on time was
followed simultaneously in the UV and visible regions.
Since the absorption coefficients in the UV region were
about 100 times higher than in the visible one, the
measuring cells of 0.01 and 1 cm path lengths were used,
respectively. The spectra collected during three hours are
displayed in Figure 9. The changes of the UV spectrum were
faster, but basically similar to those described above for the
case of the lower reactant concentrations. The visible
absorption band recorded z1 min after mixing was centred
at app. 620 nm. The spectrum then started to broaden, the
This indicated that another process took place in the
solution. We believe that it was the Cu2C mediated
oxidation of the methylene bridge in the [Cu(1a)]C and
[Cu(1a)2]C complexes, which is in accordance with the
EPR results and the observed time-dependent changes of the
corresponding ES-MS spectra. The CuCC1b system
showed a similar behaviour. The UV spectra recorded for
this system can be found in the Supplementary Material
(Fig. 4). As already mentioned, upon mixing of 1a and
Figure 9. Time dependence of the absorption spectra of an equimolar acetonitrile solution of 1a and [Cu(MeCN)4]PF6 (cZ5.5!10K3 mol dmK3). Spectra are
labelled with the approximate times after mixing of 1a and CuC. tZ(25.0G0.1) 8C. (a) Spectra recorded in the UV region; lZ0.01 cm. (b) Spectra recorded in
the visible region; lZ0.01 cm.

ˇ
V. Caplar et al. / Tetrahedron 60 (2004) 8079–8087
8085
absorbances at higher wavelengths decreased, whereas at
lower wavelengths the spectrum intensity increased for a
quite long period of time (Fig. 9). According to the NMR,
EPR and ES-MS findings, these observations can be
tentatively explained by the initial oxidation of coordinated
CuC to Cu2C, followed by the reversible reaction, i.e.
reduction of copper and the oxidation of the ligand bridge
from the methylene to the carbonyl one. Obviously, oxygen
plays an important role in these reactions. The main final
reaction products are CuC complexes with oxidized 1a
(stoichiometries 1:1 and 1:2), in agreement with the species
observed by ES-MS (Fig. 7).
with O-benzyl substituents gave lower ees than the
unsubstituted ligands (compare 2a/2b and 2c/2d). Taken
together, the results of cyclopropanation studies reveal only
a small influence of elongated substituents on stereogenic
centres on the stereochemical outcome of the reaction. The
most important result, arising from Table 2, are superior ees
achieved by substituted ligands 2 in comparison to
methylene-unsubstituted ligands 1. This is in agreement
with clearly evidenced mixtures of Cu complexes formed
with methylene bridged ligands in solution, while the
dialkylmethylene bridged ligands under the same conditions
produce better defined Cu(I) catalytic complexes.
2.2. Catalytic cyclopropanation
3. Conclusion
The results of cyclopropanation of styrene with diazoacetate
catalyzed by Cu(I) complexes of the ligands 1a–h and 2a–e
are collected in Table 2. The results show that the catalytic
complexes with the methylene bridged ligands 1 give
somewhat better chemical yields of cyclopropanes than
those with the dialkylmethylene bridged ligands 2, while the
cis/trans ratios are similar or slightly lower. However, ees
achieved with ligands 2 are in all cases superior to those
obtained with methylene bridged ligands 1. The comparison
of the ees of the cis- and trans-cyclopropanes by using
O-benzyl substituted (1a, 1b) and unsubstituted (1f, 1g)
ligand complexes revealed the significant increase of ee
(18% for cis- and 10% for trans-products) for 1b compared
to 1g. For the 1a/1f pair of catalytic complexes lower ee
increase for 1a compared to 1f was observed (15% for cis-
and 4% for trans-products). The analysis of enantioselec-
tivities obtained by ligands 2a–e revealed that the ligands
Combined spectroscopic studies of silver(I) and copper(I)
complexes of 1,5-dinitrogen ligands with C2-symmetric 1a
and 1b have shown that they form stable 1:2 metal/ligand
complexes with a pseudotetrahedral arrangement of coordi-
nating N atoms around the central metal ion, which
transform on ulterior addition of metal ions to 1:1
complexes. The stability constants (Table 1) were deter-
mined for [Ag–1a]C and [Ag–1b]C complexes of 1:2 and
1:1 stoichiometries. All spectroscopic investigations of
copper complexes of 1a and 1b revealed the oxidative
transformation of methylene bridge concomitant with
oxidation of Cu(I) to Cu(II). C2-Symmetric bis(oxazolines)
with methylene bridge (1) and dialkylmethylene bridge (2)
in the form of Cu(I) catalytic complexes were tested in
enantioselective cyclopropanations of styrene with ethyl
diazoacetate. Most of the ligands exhibited modest
Table 2. Enantioselectivity in cyclopropanation of styrene catalyzed by Cu(I) complexes of 1a–h and 2a–2e
Ligand
Molar ratioa
Yield (%)
cis/trans
cis (ee%)b
trans (ee%)b
1a
1a
1b
1b
1c
1c
1d
1d
1e
1e
1f
1.2
2.0
1.2
2.0
1.2
2.0
1.2
2.0
1.2
2.0
1.2
2.0
2.0
1.2
2.0
2.0
1.2
2.0
2.0
75
81
57
67
87
81
58
62
72
69
78
73
81
71
67
76
85
78
80
38/62
36/64
40/60
42/58
36/64
35/65
40/60
39/61
35/65
36/64
36/64
39/61
30/70
39/61
38/62
29/71
37/63
32/68
25/75
45.7(1S,2R)
43.2(1S,2R)
53.2(1R,2S)
52.7(1R,2S)
45.3(1R,2S)
46.3(1R,2S)
47.6(1R,2S)
46.1(1R,2S)
30.6(1S,2R)
34.2(1S,2R)
30.3(1R,2S)
38.7(1R,2S)
52.0(1R,2S)
35.3(1R,2S)
31.3(1R,2S)
15.0(1R,2S)
53.2(1R,2S)
65.9(1R,2S)
77.0(1R,2S)
45.6(1S,2S)
45.1(1S,2S)
55.8(1R,2R)
53.4(1R,2R)
43.7(1R,2R)
50.3(1R,2R)
47.4(1R,2R)
47.7(1R,2R)
37.8(1S,2S)
40.2(1S,2S)
41.9(1R,2R)
47.1(1R,2R)
60.0(1R,2R)
46.1(1R,2R)
49.4(1R,2R)
36.0(1R,2R)
67.1(1R,2R)
78.5(1R,2R)
90.0(1R,2R)
1f
1f5
1g
1g
1g5
1h
1h
1h5
2a
2a
2b
2b
2c
2c
2d
2d
2e
2e
2e6
1.2
2.0
1.2
2.0
1.2
2.0
1.2
2.0
1.2
2.0
2.0
62
72
59
53
77
79
76
73
75
80
77
37/63
34/66
40/60
40/60
30/70
31/69
33/67
35/65
28/72
29/71
27/73
54.6(1S,2R)
54.8(1S,2R)
60.4(1R,2S)
62.0(1R,2S)
54.0(1S,2R)
54.2(1S,2R)
67.3(1S,2R)
67.8(1S,2R)
95.0(1R,2S)
94.9(1R,2S)
97.0(1R,2S)
59.7(1S,2S)
59.7(1S,2S)
54.5(1R,2R)
55.8(1R,2R)
64.8(1S,2S)
65.3(1S,2S)
70.8(1S,2S)
71.3(1S,2S)
95.7(1R,2R)
96.2(1R,2R)
99.0(1R,2R)
a
rZc(ligand)/c(CuC).
Enantiomeric excesses were determined by gas chromatography using GLC chiral CP-Chirasil-Dex CB capillary column.
b

ˇ
V. Caplar et al. / Tetrahedron 60 (2004) 8079–8087
8086
enantioselectivity with the highest enantioselectivity of
62% ee for cis- and 56% ee for trans-isomer obtained by the
Cu(I) complex of 2b (Table 2). The observed oxidative
transformation of the Cu(I) complexes of methylene bridged
ligands may account for the lower enantioselectivities
obtained in the cyclopropanations with this type of catalytic
complexes.
Acknowledgements
The financial support from the Croatian Ministry of Science
and Technology (Programs 009807 and 0119620) is grate-
fully acknowledged. The authors thank Dr.sc. Vesna
¨
Nothig-Laslo for EPR measurements, and Ms. Tamara
Gopurenko for the help with mass spectra.
4. Experimental
Supplementary data
Preparation of compounds 1a–e and 2a,b was reported
previously,29 and compounds 1f–h and 2c–e were obtained
from Aldrich. Reagents were purchased from Aldrich and
Fluka and were used without further purification. All
solvents were purified and dried according to standard
procedures. IR spectra were taken in KBr pellets on a Perkin
Elmer 297 spectrometer. NMR spectra were recorded on the
Bruker spectrometer, at 300 or 600 MHz. Mass spectra were
recorded by means of Finnigan LCQ Deca instrument. EPR
spectra were recorded at Varian E-9 spectrometer.
Supplementary data associated with this article can
be found, in the online version, at doi:10.1016/j.tet.2004.
06.117
References and notes
1. Jacobsen, E. N., Pfaltz, A., Yamamoto, H., Eds.; Compre-
hensive Asymmetric Catalysis. Springer: Berlin, 1999.
2. Noyori, R. Asymmetric Catalysis in Organic Synthesis; Wiley:
New York, 1993.
4.1. Spectrometry
The UV/Vis absorption spectra were recorded at (25.0G
0.1) by means of a Varian Cary 5 spectrophotometer
equipped with a thermostating device. Quartz cells of 0.01
and 1 cm path lengths were used. Absorbances were
sampled at 1 nm intervals.
3. Fache, F.; Schulz, E.; Tommasino, M. L.; Lemaire, M. Chem.
Rev. 2000, 100, 2159–2231.
4. Ghosh, A. K.; Mathivanan, P.; Cappiello, J. Tetrahedron:
Asymmetry 1998, 9, 1–45.
5. Lowenthal, R. E.; Abiko, A.; Masamune, A. Tetrahedron Lett.
1990, 31, 6005–6008.
1H NMR titrations of 1a, 1b and 2a were carried out at
ambient temperature in CD3CN (data taken in Dd/ppm
according to the signal of solvent used as internal standard)
with Bruker 300 and 600 MHz. c(1a,1b,2a)Z2!10K3
6. Evans, D. A.; Woerpel, K. A.; Himman, M. M.; Faul, M. M.
J. Am. Chem. Soc. 1991, 113, 726–728.
7. Bedekar, A. V.; Koroleva, E. B.; Andersson, P. G. J. Org.
Chem. 1997, 62, 2518–2526.
mol dmK3),V0Z0.5 mL,c(AgBF4)Z0–4!10K2 mol dmK3
,
c[Cu(MeCN)4]PF6Z0–2!10K2 mol dmK3. Aliquots of
the metal ion solution were added into the solution of the
ligand in a NMR probe with Hamilton syringe. The obtained
spectrometric data were processed using the SPECFIT
program.34
8. Imai, Y.; Zhang, W. B.; Kida, T.; Nakatsuji, Y.; Ikeda, I.
J. Org. Chem. 2000, 65, 3326–3333.
9. Evans, D. A.; Faul, M. M.; Bilodeau, M. T.; Andersson,
B. A.; Burnes, D. M. J. Am. Chem. Soc. 1993, 115, 5328–5329.
10. Takada, H.; Nishibayashi, Y.; Ohe, K.; Uemura, S. Chem.
Commun. 1996, 931–932.
4.2. Catalytic cyclopropanation
11. Hansen, K. B.; Finney, E. S.; Jacobsen, E. N. Angew. Chem.
Int. Ed. Engl. 1995, 34, 676–678.
To an excess of styrene (0.52 g, 0.57 ml, 5.0 mmol) the
precatalytic Cu(I) trifluoromethanesulphonate benzene
complex (15 mmol, available from Fluka) and the corre-
sponding quantity of chiral oxazoline ligand were added.
Then ethyl diazoacetate (1.0 mmol, 1.0 ml of 1 mol dmK3
solution in 1,2-dichlorethane) was added dropwise by a
syringe pump over a period of 4.5 h. The reaction mixture
was stirred under inert argon atmosphere overnight at room
temperature. Diastereomeric mixture of cis/trans ethyl
2-phenylcyclopropan-1-carboxylates was isolated by
chromatography on a silica gel column (1!15 cm) with
ethylacetate–light petroleum (gradient 0–10%) as eluent.
Diastereomeric composition and chemical yield were
determined by gas chromatography on the HP-1 capillary
column with biphenyl as an internal standard. Enantiomeric
excesses were determined by gas chromatography using
GLC chiral CP-Chirasil-Dex CB capillary column. Cyclo-
propanation products were characterized by independent
synthesis as described.6,35
12. Corey, E. J.; Ishihara, K. Tetrahedron Lett. 1992, 33, 6807–
6810.
13. Yao, S. L.; Johannsen, M.; Audrain, H.; Hazell, R. G.;
Jorgensen, K. A. J. Am. Chem. Soc. 1998, 120, 8599–8605.
14. Evans, D. A.; Johnson, J. S.; Olhava, E. J. J. Am. Chem. Soc.
2000, 122, 1635–1649.
15. Gothelf, K. V.; Hazell, R. G.; Jorgensen, K. A. J. Org. Chem.
1998, 63, 5483–5488.
16. von Matt, P.; Pfaltz, A. Angew. Chem. Int. Ed. Engl. 1993, 32,
566–568.
17. Chelucci, G. Tetrahedron: Asymmetry 1997, 8, 2667–2670.
18. Imai, Y.; Matsuo, S.; Zhang, W. B.; Nakatsuji, Y.; Ikeda, I.
Synlett 2000, 239–241.
19. Denmark, S. E.; Stiff, C. M. J. Org. Chem. 2000, 65, 5875–
5878.
20. Nishiyama, H.; Yamaguchi, S.; Park, S. B.; Itoh, K.
Tetrahedron: Asymmetry 1993, 4, 143–150.
21. Imai, Y.; Zhang, W. B.; Kida, T.; Nakatsuji, Y.; Ikeda, I.
Tetrahedron: Asymmetry 1996, 7, 2453–2462.

ˇ
V. Caplar et al. / Tetrahedron 60 (2004) 8079–8087
8087
ˇ
Portada, T.; Piantanida, I.; Zinic, M.; Eur J. Inorg. Chem.
´
22. Andrus, M. B.; Asgari, D. Tetrahedron 2000, 56, 5775–5780.
23. Helmchen, G.; Krotz, A.; Ganz, K. T.; Hansen, D. Synlett
1991, 257–259.
2002, 1738–1748.
ˇ
31. Kovac, B.; Klasinc, L.; Raza, Z.; Sunjic, V. J. Chem. Soc.,
ˇ
´
24. Gathergood, N.; Zhang, W.; Jorgensen, K. A. J. Am. Chem.
Soc. 2000, 122, 12517–12522.
Perkin. Trans. 2 1999, 2455–2459.
32. Neuhaus, D.; Williamson, M. P. The Nuclear Overhauser
Effect in Structural and Conformational Analysis, 2nd ed.;
Wiley-VCH: New York, 2000; p 494.
25. Perch, N. S.; Pei, T.; Widenhoefer, R. A. J. Org. Chem. 2000,
65, 3836–3845.
26. Evans, D. A.; Burgey, C. S.; Paras, N. A.; Vojkovsky, T.;
Tregay, S. W. J. Am. Chem. Soc. 1998, 120, 5824–5825.
27. Russell, A. E.; Miller, S. P.; Morken, J. P. J. Org. Chem. 2000,
65, 8381–8383.
33. El Azzaoui, B.; Fifani, J.; Tjiou, E. M.; Essassi, E. M.; Jaud, J.;
Lopez, L.; Bellan, J. Tetrahedron Lett. 1999, 40, 4677–4680.
34. Binstead, R. A.; Zuberbu¨hler, A. D. SPECFIT Programe,
Global Least Squares Fitting by Factor Analysis and Marquart
Minimization; Spectrum Software Associates: Chapel Hill,
NC, 1994.
ˇ
ˇ
ˇ
28. Portada, T.; Roje, M.; Raza, Z.; Caplar, V.; Zinic, M.; Sunjic,
´
´
V. Chem. Commun. 2000, 1993–1994.
ˇ
ˇ
´
´
29. Caplar, V.; Raza, Z.; Katalenic, D.; Zinic, M. Croat. Chem.
Acta 2003, 76, 23–36.
ˇ
35. Kirin, S. I.; Vinkovic, V.; Sunjic, V. Chirality 1995, 7,
´
´
115–120.
ˇ
30. Hollosi, M.; Majer, Z.; Vass, E.; Raza, Z.; Tomitic, V.;
´