Investigation of the structure and catalytic activity in olefin cyclopropanation of neutral and cationic dicopper complexes of 3,5-bis(pyridinylimino)benzoic acid

Article abstract of DOI:10.1016/j.ica.2011.10.055

Three neutral and one cationic copper(I) complexes with 3,5-bis(pyridinylimino)benzoic acid are synthesized and characterized in solution and in the solid state by a variety of spectroscopic techniques and X-ray crystallography. The compounds are monomeric in nature possessing two Cu(I) centers each coordinated to one of the two diimine sites of the ligand. The chromophores are either N₂PX, where X is a halide (Cl, Br or I) for the neutral or N₂P₂ for the ionic compounds, respectively. The phosphorus donor is triphenylphosphane. The crystal structures of the isostructural bromido and iodido compounds are solved and discussed. The compounds are tested for their catalytic activity in olefin cyclopropanation reactions by means of ethyl diazoacetate decomposition and prove to be moderately active with the ionic one being the most active and the most promising since for cyclohexene it reveals a considerable diastereoselectivity and a 90:10 exo:endo ratio of the final product.

Full text of DOI:10.1016/j.ica.2011.10.055

Inorganica Chimica Acta 383 (2012) 105–111
Contents lists available at SciVerse ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
Investigation of the structure and catalytic activity in olefin cyclopropanation
of neutral and cationic dicopper complexes of 3,5-bis(pyridinylimino)benzoic acid
b,
Dimitrios Tzimopoulos a, Stefano Brenna b, Agnieszka Czapik c, Maria Gdaniec c, , Attilio Ardizzoia
,
⇑
⇑
Pericles D. Akrivos a,
⇑
a Aristotle University of Thessaloniki, Department of Chemistry, Laboratory of Inorganic Chemistry, P.O. Box 135, 541 24 Thessaloniki, Greece
b Dipartimento di Scienze Chimiche ed Ambientali, Universita’ degli Studi dell’Insubria, Via Valleggio 11, 22100 Como, Italy
c Faculty of Chemistry, Adam Mickiewicz University, ul. Grunwaldzka 6, 60 780 Poznan´, Poland
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 6 August 2011
Received in revised form 15 October 2011
Accepted 21 October 2011
Available online 29 October 2011
Three neutral and one cationic copper(I) complexes with 3,5-bis(pyridinylimino)benzoic acid are
synthesized and characterized in solution and in the solid state by a variety of spectroscopic techniques
and X-ray crystallography. The compounds are monomeric in nature possessing two Cu(I) centers each
coordinated to one of the two diimine sites of the ligand. The chromophores are either N2PX, where X
is a halide (Cl, Br or I) for the neutral or N2P2 for the ionic compounds, respectively. The phosphorus donor
is triphenylphosphane. The crystal structures of the isostructural bromido and iodido compounds are
solved and discussed. The compounds are tested for their catalytic activity in olefin cyclopropanation
reactions by means of ethyl diazoacetate decomposition and prove to be moderately active with the ionic
one being the most active and the most promising since for cyclohexene it reveals a considerable
diastereoselectivity and a 90:10 exo:endo ratio of the final product.
Keywords:
Copper complexes
Crystal structure
Catalytic cyclopropanation
Iminopyridine complexes
Ó 2011 Elsevier B.V. All rights reserved.
1. Introduction
A large variety of sp2-nitrogen based ligands [23–28] have been
used with the aim of enhancing the selectivity. Recently, high
Transition metal catalyzed C–C bond-forming reactions always
represent a primary goal for organometallic chemists [1–7], and
among the others cyclopropanation of alkenes has attracted enor-
mous research interest due to the importance of cyclopropanes as
starting materials or intermediates in a wide range of organic syn-
thetic reactions [8]. Thus, the search for new synthetic routes to
functionalized cyclopropanes is a continuous stimulus, especially
as regards the synthesis of new transition metal-based catalytic
systems. Actually, one of the most effective method to convert ole-
fins into the corresponding cyclopropanes is the metal-assisted
decomposition of diazocompounds [9–12] and very often the
catalytic system incorporates copper because it offers, besides high
efficiency [13,14], relatively low cost with respect to other second-
row transition metals. Both experimental [15,16] and theoretical
[17–21] studies have been accomplished in order to elucidate the
mechanism of olefin cyclopropanation promoted by Cu-based
catalysts and it is well known and established that the active
catalyst is a Cu(I) system either introduced directly or formed
in situ from an appropriate Cu(II) pre-catalyst [22].
enantioselectivities were reached with binaphthyldiimines [29]
and chiral bispidines [30]. Copper(I) species containing azolate
bridging ligands [31,32] or imine-type ligands [33], which proved
to be highly selective towards the formation of the trans isomer
in both terminal and internal olefin cyclopropanation, were also
recently reported by some of us.
Herein we report on the synthesis and full characterization of
new bisimine copper(I) complexes containing 3,5-bis(pyridinyli-
mino)benzoic acid as building block (Fig. 1). All these compounds
proved to be active catalysts in olefin cyclopropanation reactions,
in mild conditions. Specially, the cationic species 4 demonstrated
to be the most active among the series considered. Finally, the
X-ray crystal structure of the iso-structural compounds 2 and 3
is reported.
2. Experimental
2.1. Reagents and methods
The starting reagents were obtained from ACROS and were used
without any further purification. The solvents used were of reagent
grade and were not subjected to any further drying process prior to
their use. Elemental analyses for C, H and N were performed on a
Perkin–Elmer 240B elemental analyzer. Infrared spectra were
⇑
Corresponding authors. Tel.: +30 2310997706; fax: +30 2310997738 (P.D.
Akrivos).
E-mail addresses: magdan@amu.edu.pl (M. Gdaniec), attilio.ardizzoia@uninsu-
bria.it (A. Ardizzoia), akrivos@chem.auth.gr (P.D. Akrivos).
0020-1693/$ - see front matter Ó 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2011.10.055

106
D. Tzimopoulos et al. / Inorganica Chimica Acta 383 (2012) 105–111
PPh3
Cu
PPh3
Ph3P
Cu
Ph3P
Cu
Ph3P
PPh3
X
X
Cu
N
N
N
N
N
N
N
N
(ClO4)2
O
OH
O
OH
X= Cl (1), Br (2) and I (3)
4
Fig. 1. Schematic representation of the studied compounds.
recorded in KBr pellets on a Perkin–Elmer Spectrum One FTIR spec-
trometer with a resolution of 2 cmꢀ1 following the collection of 16
2.4. Synthesis of complex 4
scans over the range 4000–360 cmꢀ1. Solution NMR for 1H and 13
C
The synthetic procedure was identical to the above except that
2.0 mmol of [Cu(MeCN)2(PPh3)2][ClO4] were used. Yield: 78%. Anal.
Calc. for C91H74Cl2Cu2N4O10P4: C, 64.09; H, 4.37; N, 3.29. Found: C,
63.84; H, 4.30; N, 3.20%. IR (KBr, cmꢀ1): 3442 (O–H), 1618 (C@N),
1724, 1435 (COO). 1H NMR (DMSO-d6, d): 11.87 (COOH), 9.10
(CH@N), 7.69 (o-C6H3), 7.05 (p-C6H3). 13C NMR (DMSO-d6, d):
165.2 (COOH), 162.2 (CH@N), 132.7 (C–COOH). 31P NMR (DMSO-
d6, d): 0.43 ppm (s).
were measured at 300 and 75 MHz, respectively in CDCl3 solutions
on a Bruker 300 spectrometer using TMS as internal standard.
Absorption spectra were recorded with a Perkin–Elmer Spectrum
One spectrophotometer.
2.2. Synthesis of the compounds
The pyridinylimino substituted benzoic acids were synthesized
according to literature method [34] with slight variations, which
include the use of ethanol as the single solvent and without use
of formic acid as catalyst. The compounds were identified by ele-
mental analyses and IR spectra.
The copper precursor complexes [Cu(MeCN)4][ClO4] and [Cu(-
MeCN)2(PPh3)2][ClO4] were prepared according to literature meth-
ods [35,36].
2.5. Olefin cyclopropanation
In a standard experiment, to a solution of copper catalyst
(0.006 mmol) and olefin (3 mmol) in dichloromethane (10 mL) at
room temperature and under inert atmosphere, EDA (1.2 mmol)
was added dropwise. The consumption of EDA was monitored by
infrared spectroscopy. The mixture was then worked up by remov-
ing the solvent and the crude product was purified by column
chromatography (dichloromethane:hexane = 6:4). All the cyclo-
propanes obtained were characterized by 1H NMR and GC–MS.
Quantitative analyses of products were performed on a Finningan
Trace GC with a DB-5MS UI capillary column (30 m, 0.25 mm)
equipped with a Finningan Trace MS.
2.3. Synthesis of complexes 1–3
In a typical procedure, to a suspension of the appropriate cop-
per(I) halide (1.0 mmol) in 10 mL of acetonitrile, an equimolar
amount of triphenylphosphane was added, and the oligomeric
[CuX(PPh3)]n formed as a white precipitate. Then a suspension of
the acid (1.0 mol) in dichloromethane (10 mL) was added and the
resulting dark red solution was left stirring for 1 h at room temper-
ature. The deposited precipitate was filtered and recrystallized
from dichloromethane–diethylether. Samples for subsequent
studies were obtained after prolonged stay under vacuum while
crystals were deposited after slow evaporation of the solvent in
the dark.
2.6. X-ray crystal structure analysis
The crystal structures were determined using an Oxford Diffrac-
tion Xcalibur E diffractometer. Data collection and cell refinement
were carried out with CRYSALIS PRO [37]; SHELX-97 programs [38] were
used to solve and refine the structures. In the crystal structure of 2
it was not possible to model solvent molecules due to their exten-
sive disorder. Since the solvent molecules are beyond the scope of
Complex 1: Yield: 82%. Anal. Calc. for C55H44Cl2Cu2N4O2P2: C,
56.71; H, 5.95; N, 6.01. Found: C, 56.60; H, 5.82; N, 6.06%. IR
(KBr, cmꢀ1): 3422 (O–H), 1636 (C@N), 1715, 1434 (COO). 1H
NMR (DMSO-d6, d): 12.01 (COOH), 8.87 (CH@N), 7.66 (o-C6H3),
7.16 (p-C6H3). 13C-NMR (DMSO-d6, d): 166.4 (COOH), 160.8
(CH@N), 132.8 (C–COOH). 31P NMR (DMSO-d6, d): ꢀ3.80 ppm (s).
Complex 2: Yield: 86%. Anal. Calc. for C55H44Cl2Cu2N4O2P2: C,
51.77; H, 5.43; N, 5.49. Found: C, 51.60; H, 5.30; N, 5.56%. IR
(KBr, cmꢀ1): 3432 (O–H), 1636 (C@N), 1718, 1435 (COO). 1H
NMR (DMSO-d6, d): 11.14 (COOH), 8.93 (CH@N), 7.55 (o-C6H3),
7.03 (p-C6H3). 13C NMR (DMSO-d6, d): 166.1 (COOH), 159.8
(CH@N), 132.9 (C–COOH).. 31P NMR (DMSO-d6, RT): ꢀ3.78 ppm (s).
Complex 3: Yield: 88%. Anal. Calc. for C55H44Cl2Cu2N4O2P2: C,
47.41; H, 4.97; N, 5.03. Found: C, 47.52; H, 4.82; N, 4.98%. IR
(KBr, cmꢀ1): 3442 (O–H), 1636 (C@N), 1700, 1435 (COO). 1H
NMR (DMSO-d6, d): 11.86 (COOH), 9.03 (CH@N), 7.56 (o-C6H3),
7.08 (p-C6H3). 13C NMR (DMSO-d6, d): 165.49 (COOH), 158.9
(CH@N), 132.9 (C–COOH). 31P NMR (DMSO-d6, RT): ꢀ3.87 ppm (s).
Table 1
Catalytic cyclopropanation of olefins mediated by copper complexes.a
Entry
Olefin
Catalyst
Yield (%)b
cis:trans ratio
1
2
3
4
5
6
7
8
9
styrene
styrene
styrene
styrene
styrene
4
1
2
3
78
62
63
46
<5
75
72
69
81
<5
<5
32:68
31:69
31:69
30:70
–
44:56
21:79
26:74
10:90c
-
1/Clꢀ
a
-methylstyrene
4
4
4
4
4
4
1-octene
ethyl vinyl ether
cyclohexene
2-cyclohexen-1-one
trans-b-nitrostyrene
10
11
-
a
Reaction conditions: solvent CH2Cl2, 25 °C, Cu:EDA:olefin molar ratio
1:100:250.
b
Based on EDA conversion into products.
In this specific case, the diastereoselectivity is referred to endo:exo ratio.
c

D. Tzimopoulos et al. / Inorganica Chimica Acta 383 (2012) 105–111
107
Table 3
this study, the SQUEEZE procedure implemented in PLATON [39] was
Selected bond lengths (Å) and angles (°).
used to remove solvent contribution to the structure factors. In
the unit cell of 2 the solvent molecules are enclosed in centrosym-
metric voids having the volume of 502 Å3. In the crystal structure
of 3 the solvent molecules, acetonitrile and five water molecules,
are disordered and were refined with partial occupancies. The N
atom of acetonitrile and water molecule O1W occupy the same po-
sition in which they accept hydrogen bond from the carboxylic
group. In structures 2 and 3 the carbon bound hydrogen atoms
were added in calculated positions and refined as riding on their
carrier atoms. The carboxylic group H atoms were located in differ-
ence Fourier maps and in case of 2 refined in riding model approx-
imation. In 3 the carboxylic H atom was freely refined. Positions of
the H atoms from disordered solvent molecules were not deter-
mined. Some further details concerning crystal data and structure
refinement are given in Table 2. Selected bond lengths and angles
(a) Compound 2
Cu3–N3 2.0622(17)
Cu3–N33 2.0880(18)
Cu3–P3 2.1825(7)
Cu3–Br3 2.4123(4)
Cu5–N5 2.0974(19)
Cu5–N53 2.0781(18)
Cu5–P5 2.1947(7)
Cu5–Br5 2.4451(4)
N33–Cu3–N3 79.72(7)
N33–Cu3–P3 108.14(5)
N3–Cu3–P3 129.28(5)
N33–Cu3–Br3 108.99(5)
N3–Cu3–Br3 111.67(5)
P3–Cu3–Br3 112.27(2)
N53–Cu5–N5 78.86(7)
N53–Cu5–P5 125.46(5)
N5–Cu5–P5 109.79(5)
N53–Cu5–Br5 105.54(5)
N5–Cu5–Br5 108.68(5)
P5–Cu5–Br5 120.19(2)
(b) Compound 3
Cu3–N3 2.062 (2)
Cu3–N33 2.083(2)
Cu3–P3 2.1895(9)
Cu3–I3 2.5765(4)
Cu5–N5 2.093(2)
Cu5- N53 2.075(2)
Cu5–P5 2.2074(9)
Cu5–I5 2.5955(4)
are listed in Table 3. Molecular graphics were generated with ORTEP
3 for Windows [40].
-
N33–Cu3–N3 80.05(9)
N33–Cu3–P3 109.34(7)
N3–Cu3–P3 130.36(7)
N33–Cu3–I3 109.46(7)
N3–Cu3–I3 111.36(6)
P3–Cu3–I3 110.58(3)
N53–Cu5–N5 78.97(9)
N53–Cu5–P5 126.48(8)
N5–Cu5–P5 110.35(7)
N53–Cu5–I5 109.52(8)
N5–Cu5–I5 108.39(6)
P5–Cu5–I5 115.91(3)
3. Results and discussion
3.1. Syntheses of complexes
3.2. Spectroscopic investigation
When a suspension of the appropriate halide CuX (X = Cl, Br, I)
in acetonitrile is treated with an equimolar amount of PPh3, the
precipitation of the white oligomeric [CuX(PPh3)]n immediately
occurs. This intermediate species can be easily converted into the
desired dinuclear copper(I) complexes 1–3, formulated as [Cu2X2-
(PPh3)2(3,5-pybac)] (3,5-pybac = 3,5-bis(pyridinylimino) benzoic
acid ligand; 1, X = Cl; 2, X = Br; 3, X = I), by addition of a dichloro-
methane suspension of the ligand. The work up of the resulting
dark red solution allowed the isolation of complexes 1–3 as orange
crystalline powders.
The infrared spectra of the compounds are indicative of their
stoichiometry. They all present the characteristic bands due to
the hydroxyl group 3420–3440 cmꢀ1 and the imine C@N bond in
the region 1620–1635 cmꢀ1, respectively [41]. These values are
very close to the ones observed for the free ligand (observed at
3410 and 1635 cmꢀ1, respectively) and indicate that the coordina-
tion is rather weak and occurs through the diimine site and not the
carboxylic group. The asymmetric carboxylic band is observed in
the region 1700–1724 cmꢀ1 in a small margin around the value
of 1704 realized for the free acid. For compound 4 there is observed
a broad and featureless intense band with an approximate peak at
1094 cmꢀ1 revealing that the perchlorate anion is not coordinated
Compound 4 is prepared from a solution of [Cu(MeCN)2-
(PPh3)2][ClO4] in dichloromethane, by replacement of the acetoni-
trile molecules with one equivalent of 3,5-bis(pyridinylimino)ben-
zoic acid ligand.
Table 2
Crystallographic data and refinement details for 2 and 3.
2
3
Empirical formula
Formula weight
T (K)
[Cu2Br2[P(C6H5)3]2(C19H14N4O2)]ꢁsolvent
[Cu2I2[P(C6H5)3]2(C19H14N4O2)]ꢁ0.62 (CH3CN)ꢁ1.89(H2O)
998.69
130
1291.93
130
Crystal system
Space group
Unit cell dimensions
a (Å)
b (Å)
c (Å)
triclinic
P1
triclinic
P1
ꢀ
ꢀ
13.6330(2)
15.5608(3)
15.8732(4)
60.999(2)
70.612(2)
86.583(2)
2757.1(1)
2, 1
13.7035(5)
15.6164(6)
15.9500(5)
61.391(4)
72.566(3)
86.708(3)
2843.8(2)
2, 1
a
(°)
b (°)
c
(°)
Volume (Å3)
Z, Z0
Absorption coefficient (mmꢀ1
Reflections collected
Independent reflections
)
2.332
37236
11195
1.936
47251
11582
Completeness up to theta (max) = 26.37° (%)
Data/restraints/parameters
99.4
11195/0/608
0.995
99.7
11582/0/642
0.883
Goodness-of-fit (GOF) on F2
Final R indices [I > 2
r
(I)]
R1 = 0.0271,
wR1 = 0.0674
R2 = 0.0409
wR2 = 0.0699
0.460, ꢀ0.410
R1 = 0.0315,
wR1 = 0.0623
R2 = 0.0561
wR2 = 0.0653
1.044, ꢀ0.785
R indices (all data)
Largest difference in peak and hole

108
D. Tzimopoulos et al. / Inorganica Chimica Acta 383 (2012) 105–111
to the metal and does not participate in any kind of electrostatic
interaction with any other part of the molecule [42]. The com-
pounds are probably monomeric since the hydroxyl group bond
stretch is observed as a well defined band of medium intensity
for all the studied compounds indicating that the carboxylate re-
gion is not involved in the typical hydrogen bonding of carboxylic
acids. Furthermore, the asymmetric COO band appears at higher
wavenumbers than when the group is involved in hydrogen bond-
ing [43]. This has been observed in the spectra of the correspond-
ing compounds with the 3-pyridinylimino derivatives where the
analogous frequencies are observed around 1680 cmꢀ1 [44].
The resonances in the 1H solution NMR spectra are observed in
the expected regions. The imine proton resonance in particular,
which in the free acid is observed at 8.70 ppm as a singlet is shifted
downfield and appears at 8.87, 8.94 and 9.03 for compounds 1–3 in
order with the expected halide atom effect [45]. The resonance ap-
pears even more downfield shifted for compound 4 at 9.10 ppm. In
all cases the signal is a single sharp line indicating symmetric envi-
ronment for both regions due to the fluxional character of the
methylimino group. The 13C NMR signals are again indicative of
the coordination to the metal center [46]. The carboxylic carbon
resonance remains very close to the free ligand value of
166.60 ppm indicating no significant change in the carboxylate re-
gion upon coordination to the metal center. Very small shifts are
observed also for the imino carbon atoms in the complexes relative
to the free ligand value of 162.40 ppm. The resonances realized for
compounds 1–3 are 158.9, 159.8 and 160.8 ppm, respectively,
revealing again the halide influence to the shielding while for 4
the corresponding signal is at 162.2 ppm.
centered LUMOs [48]. This was further verified by semi-empirical
calculations carried out by means of the PM6 Hamiltonian [49]
as implemented in the MOPAC2009 package [50].
Calculations were carried out at the ground electronic state
assuming a single molecule in the gas phase and the structural
optimization was full except for the assumption of perfectly planar
aromatic rings. The octet of frontier molecular orbitals obtained for
compound 4 are presented in Fig. 3. As it is evident the occupied
orbitals are mainly metal-centered with some delocalization to-
wards the phosphorus and the halide atoms, while the virtual ones
are located over the entire or in part of the bis(imino) benzoate.
Subsequent configuration interaction taking into consideration
the above frontier MOs revealed, in agreement with the experi-
mental findings, that the CT bands in the halide compounds are
considerably extended towards the lower energy side of the visible
spectrum reaching even 610 nm whereas the first excitation for the
ionic complex is calculated to be at 415 nm, in close analogy to the
experimental findings.
3.3. Crystal structures of 2 and 3
ꢀ
The solvated forms of compounds 2 and 3, crystallising in the P1
space group are, to a large extent, isostructural. The asymmetric unit
of 2 consists of the dinuclear [Cu2Br2[P(C6H5)3]2(C19H14N4O2)] com-
plex molecules (Fig. 4a) and unidentified solvent molecules. In turn
the asymmetric unit of 3 contains dinuclear [Cu2I2[P(C6H5)3]2-
(C19H14N4O2)] molecules (Fig. 4b) and disordered solvent molecules
that were identified as acetonitrile and water molecules. The two
Cu(I) centers with the N2PX (X = Br, I) donor set show a strongly dis-
torted tetrahedral coordination geometry (Table 3) and intramolec-
ular distances between the metal centers are 7.867 and 7.861 Å in 2
and 3, respectively. Moreover, they constitute two independent chi-
rality centers as the Cu atoms have four different substituents. Ow-
ing to the identical configuration of the two chiral centers in the
studied molecules, the compounds 2 and 3 are racemic. The complex
molecules adopt an asymmetric shape with one of the planar pyr-
idylimino groups strongly twisted relative to the plane of the central
C1–C6 benzene ring (dihedral angles 65.9° in 2 and 65.0° in 3) and
The visible spectra of the compounds were recorded in metha-
nol and dichloromethane solutions. The spectra recorded in dichlo-
romethane are presented in Fig. 2. In every case the spectra are
dominated by the intense
p–
p⁄ ligand centered (LC) transitions
but their common feature extends further to the existence of a
broad low energy charge transfer absorption extending from 350
to 600 nm [47] as it is inferred by the calculated excitation coeffi-
cients (e). This envelope is known to incorporate several discrete
excitations from mainly metal-based HOMOs to mainly diimine-
Fig. 2. Visible spectra of the studied compounds recorded in 1 ꢂ 10ꢀ4 M solutions in CH2Cl2. 1 dashed line; 2 dotted line; 3 mixed line and 4 solid line.

D. Tzimopoulos et al. / Inorganica Chimica Acta 383 (2012) 105–111
109
Fig. 3. Schematic representation of the four HOMOs (left column) and four LUMOs (right column) of the optimized structure of compound 4. The orbitals are placed in order
of increasing energy from bottom to top.
the other group approximately coplanar(dihedral angles 4.0 in 2 and
5.3° in 3). The bulky substituent at the benzoic acid unit incorporat-
ing the coordinated Cu3 atom is oriented syn relative to the carbox-
ylic group resulting in a short intramolecular contact Cu3ꢁꢁꢁH2
contact (2.67 Å in 2 and 2.69 Å in 3). In turn, owing to a strong twist
relative to the benzoic acid unit, the substituent at position 5 of the
benzene ring, incorpoprating the Cu5 atom, is able to place one of its
triphenylphosphane phenyl rings directly above the aromatic ring of
the benzoic acid fragment. The geometrical parameters of the intra-
molecular p–p stacking interactions between the rings C1–C6 and
C5A1–C5A6 in 2 and 3, respectively, are as follows: dihedral angles
between the aromatic rings are 2.0° and 2.4°, the centroid-to-
centroid distances 3.57 and 3.56 Å and the ring offset 0.46 and
0.42 Å.

110
D. Tzimopoulos et al. / Inorganica Chimica Acta 383 (2012) 105–111
Fig. 4. Molecular structures of compound 2 (top) and 3 (bottom). All hydrogen atoms except carboxylic H atom have been omitted for clarity. Only one C atom in each phenyl
ring of the triphenylphosphane ligand is labeled. Displacement ellipsoids are drawn at the 50% probability level.
The intramolecular
p
–
p
stacking interactions in 2 and 3 are
ported in the literature, in the conversion of styrene into the
corresponding cyclopropane esters, most of the copper(I) catalysts
provided the trans isomer as the main product [18,52], the com-
mon trans:cis ratio ranging from 50:50 to 75:25. Increasing the size
of R on N2CHCOOR enhances enantiocontrol, while the diastereose-
lectivity is less affected by steric effects, except when very bulky
substituents are used, as in the case of 2,6-di-tert-butyl-4-methyl
phenyl (BDA) diazoacetate [9,53]. The studied species confirmed
this trend, providing a 70:30 trans:cis ratio in the catalytic cyclo-
propanation of styrene.
The cationic species 4 displayed the highest activity, the
consumption of EDA being complete within 4 h, whereas the 1–3
species all required a longer reaction time (>7 h), with an initial
induction period necessary to reach the complete dissolution.
Moreover, when employing the halide-containing compounds the
yields obtained were lower, especially in the case of 3. As discussed
hereafter, this could probably be ascribed to the relative strength
of the Cu–X bonds (Cl < Br < I). In fact, among the halide-containing
ones we observed an interesting trend related to the nature of the
anion: the induction period increase moving from chlorine to
more pronounced than in the coordination cation in 4 [44] where
the interacting aromatic rings formed dihedral angle of 16.8° and
their centroids were at a distance of 4.36 Å.
3.4. Cyclopropanation reactions
The catalytic activity of complexes 1–4 in olefin cyclopropana-
tion was finally tested. In order to monitor the possibility to em-
ploy the studied complexes as catalysts, styrene was first used as
reference substrate, and all the compounds demonstrated to be ac-
tive in olefin cyclopropanation (see entries 1–4 in Table 1), giving
moderate-to-high yields under mild conditions. However, together
with the desired products, diethyl maleate and diethyl fumarate,
deriving from ethyl diazoacetate self-coupling [51], were always
detected. In order to minimize these side-reactions, a large excess
of olefin (namely, a Cu:EDA:olefin = 1:100:250 M ratio) was em-
ployed and EDA was added dropwise during a 15-min period.
The copper (I) compounds showed comparable diastereoselectivi-
ties, the cis:trans ratio being about 30:70 in all the runs. As re-

D. Tzimopoulos et al. / Inorganica Chimica Acta 383 (2012) 105–111
111
[6] J.R. Khusnutdinova, N.P. Rath, L.M. Mirica, J. Am. Chem. Soc. 132 (2010) 7303.
[7] T. Ikariya, I.D. Gridnev, Top. Catal. 53 (2010) 894.
[8] H.N.C. Wong, M.-Y. Hon, C.-W. Tse, Y.-C. Yip, J. Tanko, T. Hudlicky, Chem. Rev.
89 (1989) 165.
[9] A. Caballero, A. Prieto, M.M. Díaz-Requejo, P.J. Pérez, Eur. J. Inorg. Chem. (2009)
1137.
iodine. Probably, taking into account the activity of the ionic spe-
cies 4, and the relative strength of the Cu–X bonds (Cl < Br < I), a
prerequisite in order to have the catalytic active species is the dis-
sociation of the halide ion from the copper centers.
To prove this assumption, we performed a catalytic run in the
presence of exogenous chloride anions (entry 5). In fact, when
ethyl diazoacetate was added to a CH2Cl2 solution containing sty-
rene, complex 4 as catalyst and an equivalent (with respect to me-
tal) of benzyltriethylamonium chloride, no formation of
cyclopropanation products was detected even after a long reaction
time (48 h). This outcome corroborated the supposition that the
first step in the catalytic cycle involving the halide complexes is
most probably the anion dissociation, in this specific case pre-
vented by the presence of exogenous chloride.
We then decided to deepen the activity of the most active spe-
cies, 4, in the presence of different olefins as substrate. Thus, com-
plex 4 was tested as catalyst in the cyclopropanation of both
terminal (entries 6–8) and internal olefins (entries 9–11). The dia-
stereoselectivities reached using terminal olefins ranged from
[10] A. de Meijere, Chem. Rev. 103 (2003) 931.
[11] M.P. Doyle, A. McKervey, T. Ye, Modern Catalytic Methods for Organic
Synthesis with Diazo Compounds: From Cyclopropanes to Ylides, Wiley,
Chichester, 1998.
[12] T. Ye, A. McKervey, Chem. Rev. 94 (1994) 1091.
[13] A. Pfaltz, Cyclopropanation and C–H Insertion with Cu, in: E.N. Jacobsen, A.
Pfaltz, H. Yamamoto (Eds.), Comprehensive Asymmetric Catalysis, vol. 2,
Springer-Verlag, Berlin, 1999, pp. 513–538.
[14] W. Kirmse, Angew. Chem., Int. Ed. 42 (2003) 1088.
[15] T. Rasmussen, J.F. Jensen, N. Østergaard, D. Tanner, T. Ziegler, P. Norrby, Chem.
Eur. J. 8 (2002) 177.
[16] B.F. Straub, P. Hofmann, Angew. Chem., Int. Ed. Engl. 40 (2001) 1288.
[17] J.I. García, G. Jiménez-Osés, J.A. Mayoral, Chem. Eur. J. 17 (2011) 529.
[18] J.M. Fraile, J.I. García, A. Gissibl, J.A. Mayoral, E. Pires, O. Reiser, M. Roldán, I.
Villalba, Chem. Eur. J. 13 (2007) 8830.
[19] Q. Meng, M. Li, J. Mol. Struct. THEOCHEM 765 (2006) 13.
[20] J.A.S. Howell, Dalton Trans. (2006) 545.
[21] Q. Meng, M. Li, D. Tang, W. Shen, J. Zhang, J. Mol. Struct. THEOCHEM 711
(2004) 193.
[22] J.M. Fraile, J.I. García, V. Martínez-Merino, J.A. Mayoral, L. Salvatella, J. Am.
Chem. Soc. 123 (2001) 7616. and references therein.
44:56 cis:trans ratio in the case of
a-methylstyrene (entry 6) to
21:79 in the cyclopropanation of 1-octene. On the contrary, in
the case of an internal olefin like cyclohexene, species 4 showed
a higher diastereoselectivity and a 90:10 exo:endo ratio was
achieved. Unfortunately, with deactivated internal olefins such as
2-cyclohexen-1-one (entry 10) or trans-b-nitrostyrene (entry 11),
the cyclopropanation products were detected only in traces, thus
limiting the activity of 4 to not deactivated substrates.
[23] H. Fritschi, U. Leutenegger, A. Pfaltz, Helv. Chim. Acta 71 (1988) 1553.
[24] A.J. Burke, E. da Palma Carreiro, S. Chercheja, N.M.M. Moura, J.P.P. Ramalho, A.I.
Rodrigues, C.I.M. dos Santos, J. Organomet. Chem. 692 (2007) 4863.
[25] C. Ricardo, T. Pintauer, J. Organomet. Chem. 692 (2007) 5165.
[26] G. Chelucci, R.P. Thummel, Chem. Rev. 2002 (102) (2002) 3129.
[27] M.M. Diaz-Requejo, A. Caballero, T.R. Belderrain, M.C. Nicasio, S. Trofimenko,
P.J. Perez, J. Am. Chem. Soc. 124 (2002) 978.
[28] L. Beaufort, A. Demonceau, A.F. Noels, Tetrahedron 61 (2005) 9025.
[29] H. Suga, A. Kakehi, S. Ito, T. Ibata, T. Fudo, Y. Watanabe, Y. Kinoshita, Bull.
Chem. Soc. Jpn. 76 (2003) 189.
[30] G. Lesma, C. Cattenati, T. Pilati, A. Sacchetti, A. Silvani, Tetrahedron:
Asymmetry 18 (2007) 659.
4. Conclusion
[31] G.A. Ardizzoia, S. Brenna, F. Castelli, S. Galli, C. Marelli, N. Masciocchi, J.
Organomet. Chem. 693 (2008) 1870.
The neutral and one cationic dicopper(I) complexes with 3,5-
bis(pyridinylimino)benzoic acid are synthesized and their
structure studied. The proximity of the copper centers give rise
to asymmetric compounds as far as the orientation of the diimine
sites of the bridging ligand. The compounds are potential catalysts
in olefin cyclopropanation reactions and in fact they reveal overall
moderate-to-high yields when tested with styrene. The cationic
complex is the better and is further applicable to analogous reac-
tions where it proves to be efficient with internal olefins affording
a high diastereoselectivity of 90:10 with cyclohexene.
[32] A. Maspero, S. Brenna, S. Galli, A. Penoni, J. Organomet. Chem. 672 (2003) 123.
[33] G.A. Ardizzoia, S. Brenna, F. Castelli, S. Galli, Inorg. Chim. Acta 362 (2009) 3507.
[34] B.P. Buffin, E.B. Fonger, A. Kundu, Inorg. Chim. Acta 355 (2003) 340.
[35] P. Hemmerich, C. Sigwart, Experentia 19 (1963) 488.
[36] P.F. Barron, J.C. Dyason, L.M. Engelhardt, P.C. Healy, A.H. White, Aust. J. Chem.
38 (1985) 261.
[37] Oxford Diffraction, CRYSALIS CCD and RED, Ver. 1.171.33, Oxford Diffraction Ltd.,
Yarnton, Oxfordshire, UK, 2009.
[38] G.M. Sheldrick, Acta Crystallogr., Sect. A 64 (2008) 112.
[39] A.L. Spek, PLATON, A Multipurpose Crystallographic Tool, Utrecht University,
Utrecht, The Netherlands, 2009.
[40] L.J. Farrugia, J. Appl. Crystallogr. 30 (1997) 565.
[41] D. Tzimopoulos, A. Czapik, M. Gdaniec, T. Bakas, A.A. Isab, A.-C. Varvogli, P.D.
Akrivos, J. Mol. Struct. 965 (2010) 56. and references therein.
[42] K. Nakamoto, Infrared and Raman Spectra of Inorganic and Coordination
Complexes, fourth ed., John Wiley & Sons, USA, 1992.
Appendix A. Supplementary material
[43] I. Wolfs, H.O. Desseyn, Appl. Spectrosc. 50 (1996) 1000.
[44] S. Shakhatreh, J. Mohanraj, A. Czapik, D. Tzimopoulos, S. Kotoulas, M. Gdaniec,
P.D. Akrivos, J. Mol. Struct. 1002 (2011) 51.
[45] A.M. Kantola, P. Lantto, J. Vaara, J. Jokisaari, Phys. Chem. Chem. Phys. 12 (2010)
2679.
[46] R.J. Abraham, M. Mobli, R.J. Smith, Magn. Reson. Chem. 42 (2004) 436.
[47] W.-F. Fu, X. Gan, J. Jiao, Y. Chen, M. Yuan, S.-M. Chi, M.-M. Yu, S.-X. Xiong, Inorg.
Chim. Acta 360 (2007) 2758. and references cited therein.
[48] N. Armaroli, G. Accorsi, F. Cardinali, A. Listorti, Top. Curr. Chem. 280 (2007) 69.
[49] J.J.P. Stewart, J. Mol. Mod. 13 (2007) 1173.
[50] MOPAC2009, James J.P. Stewart, Stewart Computational Chemistry, Colorado
Springs, CO, USA, 2009, <http://www.OpenMOPAC.net>.
[51] I. Shepperson, S. Quici, G. Pozzi, M. Nicoletti, D. O’Hagan, Eur. J. Org. Chem.
(2004) 4545.
CCDC 821562 and 821563 contain the supplementary crystallo-
graphic 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. Supplementary data associ-
ated with this article can be found, in the online version, at
doi:10.1016/j.ica.2011.10.055.
References
[1] K. Huang, C.-L. Sun, Z.-J. Shi, Chem. Soc. Rev. 40 (2011) 2435.
[2] S. Dastgir, K.S. Coleman, M.L.H. Green, Dalton Trans. 40 (2011) 661.
[3] I.D. Gridnev, M. Watanabe, H. Wang, T. Ikariya, J. Am. Chem. Soc. 132 (2010)
16637.
[4] S. Dastgir, K.S. Coleman, A.R. Cowley, M.L.H. Green, Organometallics 29 (2010)
4858.
[52] M.P. Doyle, W. Hu, J. Org. Chem. 65 (2000) 8839.
[53] M.P. Doyle, V. Bagheri, T.J. Wandless, N.K. Harn, D.A. Brinker, C.T. Eagle, K.-L.
Loh, J. Am. Chem. Soc. 112 (1990) 1906.
[5] S. Reimann, J.-D. Grunwaldt, T. Mallat, A. Baiker, Chem. Eur. J. 16 (2010) 9658.