Multinuclear Cu(II) Schiff base complex as efficient catalyst for the chemical coupling of CO2 and epoxides: Synthesis, X-ray structural characterization and catalytic activity

Article abstract of DOI:10.1007/s10562-010-0529-3

The synthesis, X-ray structure, spectroscopic and catalytic properties of sterically hindered Schiff-base ligands (L₁H = N-[allylamine]-3,5-di- tert-butyl salicylaldimine, L₂H=N-[2-amino-5-methyl pyridine]-3,5-di-tert-butyl salicylal

Full text of DOI:10.1007/s10562-010-0529-3

Catal Lett (2011) 141:717–725
DOI 10.1007/s10562-010-0529-3
Multinuclear Cu(II) Schiff Base Complex as Efficient Catalyst
for the Chemical Coupling of CO2 and Epoxides: Synthesis, X-ray
Structural Characterization and Catalytic Activity
•
Mahmut Ulusoy Onur S¸ahin Ahmet Kilic
•
•
¨
Orhan Bu¨yu¨kgu¨ngor
Received: 25 August 2010 / Accepted: 24 November 2010 / Published online: 14 December 2010
ꢀ Springer Science+Business Media, LLC 2010
Abstract The synthesis, X-ray structure, spectroscopic
and catalytic properties of sterically hindered Schiff-base
ligands (L1H = N-[allylamine]-3,5-di-tert-butyl salicylaldi-
mine, L2H=N-[2-amino-5-methyl pyridine]-3,5-di-tert-butyl
salicylaldimine and L3H=N-[2-amino-6-methyl pyridine]-
3,5-di-tert-butyl salicylaldimine), and their mononuclear
Cu(II) complex for L1H with multinuclear Cu(II) complexes
for L2H and L3H, were described. The copper(II) complexes
of these ligands were synthesized by treating an methanolic
solution of the appropriate ligand with an appropriate amount
of CuCl2ꢀ2H2O. The ligands and their copper(II) complexes
were characterized by FT-IR, UV–Vis, 1H-NMR, elemental
analysis, measurement of room temperature magnetic
moment, and X-ray structural determination. The reaction of
the L2H and L3H ligands in a 1:1 mol ratio with CuCl2ꢀ2H2O
afforded ionic copper metal(II) complexes in the presence of
NEt3. The Cu(II) metal complexes tested as catalysts for the
formation of cyclic organic carbonates from carbon dioxide
and liquid epoxides which served as both reactant and solvent.
[Cu3(L2)4]Cl2ꢀCuCl2 complex which has 5-methyl substituent
on the pyridine ring showed high catalytic activity for
chemical coupling carbon dioxide with epoxides (propylene
oxide (PO), epichlorohydrine (EC) and 1,2-epoxy butane
(EB)) selectively.
Keywords Schiff base ꢀ Cu(II) complexes ꢀ X-ray
structure ꢀ CO2 ꢀ Cyclic carbonate
1 Introduction
In the current global climate, with high oil prices and
increasing concern over global warming, and consumption
petroleum resources, the development of renewable carbon
sources is of the utmost importance. CO2 is a particularly
attractive alternative feedstock as it is inexpensive, highly
naturally abundant, and the byproduct of many industrial
processes, including combustion [1]. The reactions of CO2
with metal complexes has been extensively studied,
revealing potential pathways for catalytic reactions [2–5].
However the thermodynamic stability of CO2 has ham-
pered its utility as a reagent for chemical synthesis; in fact
its high stability makes it an ideal medium for many
chemical processes [2, 6]. In particular, the catalytic cou-
pling of CO2 with heterocycles has received considerable
attention over the past 35 years [1, 7–10]. A majority of
these publications involve the reaction of CO2 with epox-
ides to generate polycarbonates and-or cyclic carbonates.
Cyclic carbonates are used excellent aprotic polar solvents,
electrolytes in secondary batteries, precursors for polycar-
bonates and biomedical applications [11–14]. These prod-
ucts have a high boiling point and have therefore found
many applications as solvents. However, the carbamates
can obtained via synthesis of ammonia or amines. The
carbamates can then be converted into polyurethane, a
versatile material with a high commercial value [15–18].
There is a continuing interest in copper metal complexes
of Schiff bases because of the presence of both nitrogen
and oxygen donor atoms in the backbones of these ligands
[19]. Lots of studies about metal based complexes as the
M. Ulusoy (&) ꢀ A. Kilic
Department of Chemistry, University of Harran, 63190
Sanliurfa, Turkey
e-mail: ulusoymahmut@harran.edu.tr
¨
O. S¸ahin ꢀ O. Bu¨yu¨kgu¨ngor
Department of Physics, Ondokuz Mayis University, 55139
Kurupelit, Samsun, Turkey
123

718
M. Ulusoy et al.
catalysts for the catalytic conversion of CO2 into valuable
organic products were published in recent years [20–22].
This study aims at obtaining new coordinative com-
pounds with similar properties. Therefore, the present
manuscript describes the synthesis, structural character-
ization and the catalytic activity of three new steric
hindered Schiff base ligands (L1H=N-[allylamine]-3,5-di-t-
butyl salicylaldimine, L2H=N-[2-amino-5-methyl pyridine]
-3,5-di-t-butyl salicylaldimine and L3H=N-[2-amino-6-
methyl pyridine]-3,5-di-t-butyl salicylaldimine) involving
nitrogen and oxygen donor sites and their Cu(II) metal
complexes derivatives with general formula presented
Scheme 2.
time a yellow solid precipitated from solution which was
isolated by filtration, washed with cold methanol and dried
vacuum. Then, the products were recrystallized from eth-
anol. The products are soluble in common solvents such as
CHCl3, CH3CH2OH, DMF and DMSO.
For (L1H) ligand: Color: Yellow; m.p: 102 ꢁC; Yield
(%): 88; Anal. Calc. for C18H27NO (F.W: 273.4 g/mol): C,
79.07; H, 9.95; N, 5.12. Found: C, 79.12; H, 9.89; N,
5.15%. 1H NMR (400 MHz, CDCl3, Me4Si, ppm):
d = 13.62 (s, 1H, –OH, D-exchangeable), d = 8.21 (s, 1H,
HC=N), d = 7.41 (s, 1H, Ar–CH), d = 7.19 (s, 1H,
Ar–CH), d = 6.04–5.98 (m, 1H, C=CH), d = 5.28–5.17
(m, 2H, C=CH2), d = 4.22 (s, 2H, N–CH2), d = 1.43 (s,
9H, C–CH3) and d = 1.26 (s, 9H, C–CH3). IR (KBr pellets,
t
max/cm-1): 3520–2480 t(OH…N), 3044 t(Ar–CH), 2955-
2 Experimental
2785 t(Alip–CH), 1630 t(C=N), 1458–1432 t(C=C), 1174
t(C–O). UV–Vis (kmax, nm, * = shoulder peak): 240, 264,
331 and 417* (in CHCl3).
2.1 Materials and Measurements
For (L2H) ligand: Color: Yellow; m.p: 144 ꢁC; Yield
(%): 91; Anal. Calc. for C21H28N2O (F.W: 324.5 g/mol):
C, 77.74; H, 8.70; N, 8.63. Found: C, 77.72; H, 8.64; N,
8.57%. 1H NMR (400 MHz, CDCl3, Me4Si, ppm):
d = 13.94 (s, 1H, –OH, D-exchangeable), d = 9.48 (s, 1H,
HC=N), d = 8.32 (s, 1H, Ar–CH), d = 7.59–7.56 (d, 1H,
J=12, Ar–CH), d = 7.47 (s, 1H, Ar–CH), d = 7.34 (s, 1H,
Ar–CH), d = 7.21 (s, 1H, Ar–CH), d = 2.41 (s, 3H,
Ar–CH3), d = 1.46 (s, 9H, C–CH3) and d = 1.28 (s, 9H,
C–CH3). IR (KBr pellets, tmax/cm-1): 3629 t(OH), 3052
t(Ar–CH), 2958–2742 t(Alip–CH), 1613 t(C=N),
1471–1458 t(C=C), 1170 t(C–O). UV–Vis (kmax, nm,
* = shoulder peak): 241, 281, 317, 368, 484* (in CHCl3).
For (L3H) ligand: Color: Yellow; m.p: 106 ꢁC; Yield
(%): 93; Anal. Calc. for C21H28N2O (F.W: 324.5 g/mol):
C, 77.74; H, 8.70; N, 8.63. Found: C, 77.68; H, 8.67; N,
8.66%. 1H NMR (400 MHz, CDCl3, Me4Si, ppm):
d = 13.96 (s, 1H, –OH, D-exchangeable), d = 9.81 (s, 1H,
HC=N), d = 7.81–7.79 (d, 1H, J = 12, Ar–CH), d = 7.60
(s, 1H, Ar–CH), d = 7.47 (s, 1H, Ar–CH), d = 7.35 (s, 1H,
Ar–CH), d = 7.22 (s, 1H, Ar–CH), d = 2.38 (s, 3H,
Ar–CH3), d = 1.47 (s, 9H, C–CH3) and d = 1.29 (s, 9H,
C–CH3). IR (KBr pellets, tmax/cm-1): 3580–2528 t(OHꢀꢀꢀN),
3048 t(Ar–CH), 2954–2749 t(Alip–CH), 1612 t(C=N),
1550–1457 t(C=C), 1171 t(C–O). UV–Vis (kmax, nm,
* = shoulder peak): 241, 279, 317, 368, 480* (in CHCl3).
All reagents and solvents were of reagent-grade quality and
obtained from commercial suppliers (Aldrich or Merck).
The elemental analyses were carried out in the Laboratory
of the Scientific and Technical Research Council of Turkey
(TUBITAK). IR spectra were recorded on a Perkin Elmer
Spectrum 100 FT-IR Spectrometer as KBr pellets. 1H-NMR
spectra were recorded on a Varian AS-400 MHz instrument
CDCl3 at room temperature. Chemical shifts were given in
parts per million from tetramethylsilane. Magnetic Sus-
ceptibilities were determined on a Sherwood Scientific
Magnetic Susceptibility Balance (Model MK1) at room
temperature (20 ꢁC) using Hg[Co(SCN)4] as a calibrant;
diamagnetic corrections were calculated from Pascal’s
constants [23]. Electronic spectral studies were conducted
on a Varian Carry 100 model UV–visible spectrophotom-
eter at the wavelength range of 200–1100 nm. Melting
points were measured in open capillary tubes with an
Electrothermal 9100 melting point apparatus and uncor-
rected. Catalytic tests were performed in a PARR 4843
50 mL stainless pressure reactor.
2.2 Synthesis of Ligands (L1H, L2H and L3H):
N-[allylamine]-3,5-di-t-butyl salicylaldimine (L1H), N-[2-
amino-5-methyl pyridine]-3,5-di-tert-butyl salicylaldimine
(L2H) and N-[2-amino-6-methyl pyridine]-3,5-di-tert-butyl
salicylaldimine (L3H) ligands were synthesized by the
reaction of 4.3 mmol (1.0 g) 3,5-di-tert-butyl-2-hydro-
xybenzaldehyde in 30 mL absolute methanol with
4.3 mmol (0.25 g) allylamine for L1H, 4.3 mmol (0.46 g)
2-amino-5-methyl-pyridine for L2H and 4.3 mmol (0.46 g)
2-amino-6-methyl-pyridine for L3H in 15 mL absolute
methanol. Also, 3–4 drops of formic acid were added as
catalyst. The mixtures were refluxed for 3 h. During this
2.3 Synthesis of the Cu(II) Metal Complexes
0.50 g, 1.84 mmol ligand (L1H), 0.50 g, 1.54 mmol ligand
(L2H) or 0.50 g, 1.54 mmol ligand (L3H) was dissolved in
methanol (30 cm3). A solution of 0.16 g, 0.92 mmol of the
metal salt [CuCl2ꢀ2H2O] for mononuclear [Cu(L1)2] com-
plex and 0.26 g, 1.54 mmol of the metal salt [CuCl2ꢀ2H2O]
for multinuclear [Cu3(L2,3)4]Cl2ꢀCuCl2 complexes in
123

Multinuclear Cu(II) Schiff Base Complex as Efficient Catalyst
719
methanol (20 cm3), was added dropwise under a N2
atmosphere with continuous stirring. A few drops of NEt3
was added while stirring. The stirred mixtures were then
heated to the reflux temperature for 3 h and were main-
tained at this temperature. Then, the mixture was evapo-
rated to a volume of 5 mL in vacuum and left to cool to
room temperature. The precipitated products were filtered
in vacuum and washed with a small amount of ethanol and
water. The products were recrystallized from CH2Cl2–
hexane. The resulting green solutions were cooled to room
temperature and allowed to stand for two week during
which green crystals formed. The crystals were filtered off,
washed with cold methanol and dried in air.
reaction. The vessel was then cooled to 5–10 ꢁC in ice bath
after the expiration of the desired time of reaction. The
pressure was released, then, the excess gases were vented.
The yields of epoxides to corresponding cyclic carbonates
were determined by comparing the ratio of product to
1
substrate in the H NMR spectrum of an aliquot of the
reaction mixture.
2.5 Crystallography
A dark-green colour crystal of the copper complex, suitable
for data collection was mounted on a glass fiber and data
collection was performed on a STOE IPDS II diffractom-
eter with graphite monochromated Mo Ka radiation at
296 K. Details of crystal data, data collection and refine-
ment were given in Table 2. The structure was solved by
direct-methods using SHELXS-97 [24] and refined by full-
matrix least-squares methods on F2 using SHELXL-97
from within the WINGX [25] suite of software. All non-
hydrogen atoms were refined with anisotropic parameters.
Hydrogen atoms bonded to carbon were placed in calcu-
For [Cu(L1)2]: Color: Green; m.p: 225 ꢁC; Yield (%):
75; Anal. Calc. for [C42H54N4O2Cu] (F.W: 608.4 g/mol):
C, 71.07; H, 8.62; N, 4.60. Found: C, 70.94; H, 8.71; N,
4.48%. IR (KBr pellets, tmax/cm-1): 3042 t(Ar–CH),
2955–2833 t(Alif–CH), 1609 t(C=N), 1481–1422 t(C=C),
1168 t(C–O), 519 t(Cu–N) and 486 t(Cu–O). leff = 1.73
[B.M]. UV–Vis (kmax, nm, * = shoulder peak): 242, 269,
389, 485*, 704* (in CHCl3).
´
˚
lated positions (C–H = 0.93–0.96 A) and treated using a
For [Cu3(L2)4]Cl2ꢀCuCl2: Color: Dark Green;
m.p: [ 300 ꢁC; Yield (%): 72; Anal. Calc. for
[C84H110N8O4Cl4Cu3] (F.W: 1628.3 g/mol): C, 61.96; H,
6.81; N, 6.88. Found: C, 61.83; H, 6.72; N, 6.72%. IR (KBr
pellets, tmax/cm-1): 3048 t(Ar–CH), 2954–2826 t(Alif–
CH), 1605 t(C=N), 1460–1435 t(C=C), 1165 t(C–O), 537
t(Cu–N) and 490 t(Cu–O). leff = 1.78 [B.M]. UV–Vis
(kmax, nm, * = shoulder peak): 241, 306, 355, 463, 511*,
734* (in CHCl3).
riding model with U = 1.2 times the U value of the parent
atom for CH and CH3. The molecular drawing was
obtained using ORTEP-III [26]. Geometric calculations
were performed on PLATON [27].
3 Results and Discussion
The ligands (L1H, L2H and L3H) were prepared in mod-
erate yields by refluxing the 3,5-di-tert-butyl-2-hy-
droxybenzaldehyde with allylamine, 2-amino-5-methyl-
pyridine or 2-amino-6-methyl-pyridine (one equivalent
For [Cu3(L3)4]Cl2ꢀCuCl2: Color: Dark green; m.p:
264 ꢁC; Yield (%): 70; Anal. Calc. for [C84H110
N8O4Cl4Cu3] (F.W: 1628.3 g/mol): C, 61.96; H, 6.81; N,
6.88. Found: C, 61.88; H, 6.79; N, 6.74%. IR (KBr pellets,
t
max/cm-1): 3046 t(Ar–CH), 2955–2868 t(Alif–CH), 1603
t(C=N), 1461–1423 t(C=C), 1166 t(C–O), 529 t(Cu–N)
and 472 t(Cu–O). leff = 1.78 [B.M]. UV–Vis (kmax, nm,
* = shoulder peak): 215, 241, 269, 368, 415, 518*, 728*
(in CHCl3).
2.4 General Procedure for the Cycloaddition
of Epoxides to CO2
A 50 mL stainless pressure reactor was charged with
Cu(II) complexes (1.125 9 10-5 mol), epoxide (4.5 9
10-2 mol), 4-dimethylamino-pyridine (DMAP) (11 mg,
9.0 9 10-5 mol) or NEt3 (0.013 mL, 9.0 9 10-5 mol), if
used CH2Cl2 (5.0 mL). The reaction vessel was placed
under a constant pressure of carbon dioxide for 2 min to
allow the system to equilibrate (except propylene oxide
(PO) as the epoxide) and CO2 was charged into the auto-
clave with desired pressure, then heated to the desired
temperature. The pressure was kept constant during the
Scheme 1 Synthetic route for preparation of proposed ligands (LnH)
123

720
M. Ulusoy et al.
chemical shifts of the different types of protons were listed in
experimental section. The proton NMR spectra of the free
ligands show a peak as singlet at 13.62 ppm for L1H, at
13.94 ppm for L2H and at 13.96 ppm for L3H, respectively.
The peak corresponding to this proton disappears while
addition a few drops of D2O. The chemical shifts observed at
d = 8.21 ppm for L1H, at d = 9.48 ppm for L2H and at
d = 9.81 ppm for (L3H) is assigned to the proton of azo-
methine (CH=N) as singlet [28, 29]. It is remarkable to see
that the downfield chemical shift (d = 9.48 or 9.81 ppm)
corresponds to the azomethine proton of the pyridine CH3
group, which has the high electron donating power, while the
upperfield value (d = 8.21 ppm) is of the allyl group with
the high electron withdrawing power. Thus, it can be con-
cluded that the position of the azomethine proton is strongly
affected by the electron withdrawing/donating character of
the groups on the aryl ring and double bound. Also, the
protons of the tert-butyl groups of ligands exhibit singlet
peak at d = 1.43 and 1.26 ppm for L1H, d = 1.46 and
1.28 ppm for L2H, and d = 1.47 and 1.29 ppm for L3H,
indicating that the tert-butyl protons of these compounds are
magnetically nonequivalent [30].
3.2 IR Spectra
The IR spectra of the Cu(II) complexes are compared with
those of the free ligand in order to determine the coordi-
nation sites that may be involved in chelation. There are
some guide peaks in the spectra of the ligands, which are of
good help for achieving this goal. The position and/or the
intensities of these peaks are expected to be changed upon
chelation. Coordination of the steric hindered ligands to the
metal through the nitrogen atom is expected to reduce
the electron density in the azomethine link and lower
the t(C=N) absorption frequency. The very strong and
sharp bands located at 1630–1612 cm-1 are assigned to the
t(C=N) stretching vibrations of the azomethine of the
ligands. These bands are shifted 21–8 cm-1 to a lower
wavenumber which supports the participation of the azo-
methine group of these ligands in binding to the copper ion
[31–33]. A strong peak observed at range 1174–1170 cm-1
in the free ligands has been assigned to phenolic C–O
stretching. On the complexation this band is shifted to a
lower frequency range 1168–1165 cm-1, indicating coor-
dination through the phenolic oxygen. The coordination of
the azomethine nitrogen, phenolic oxygen for [Cu(L1)2]
complex and azomethine nitrogen, phenolic oxygen and
pyridine nitrogen for [Cu3(L2)4]Cl2ꢀCuCl2 and [Cu3
(L3)4]Cl2ꢀCuCl2 complexes are further supported by the
appearance of peaks at range 537–519 cm-1 and at range
490–472 cm-1 due to t(Cu / O) and t(Cu / N) stre-
tching vibrations that are not observed in the infrared
spectra of the ligands [34]. Thus, it is clear that the ligands
Scheme 2 The structure of the proposed Cu(II) metal complexes
ratio) in absolute methanol (Scheme 1). The isolated ligands
were characterized by their elemental analysis, melting
point, FT-IR, UV–Vis and 1H-NMR spectroscopy. The
copper metal complexes (Scheme 2) were synthesized by
treating CuCl2ꢀ2H2O with one or two equivalents of the
corresponding ligand in methanol at reflux temperature. The
Cu(II) metal complexes are single crystalline or powder-like
and they are stable at atmospheric conditions. To confirm the
identity of the pre-catalysts prepared in the present work, a
variety of techniques including elemental analysis, infrared
spectroscopy, UV–Vis spectroscopy, X-ray structural
determination for [Cu3(L3)4]Cl2ꢀCuCl2 complex and mag-
netic susceptibility (leff, BM) have been utilized. The metal
to ligand ratios in the mono and multinuclear Cu(II) metal
complexes were found to be 1:2 or 4:4 (Scheme 2). Results
are presented in the experimental section.
3.1 NMR Spectra
The 1H-NMR spectral data of the salen ligands were recor-
ded in CHCl3 solution using tetramethylsilane (TMS) as
internal standard. The 1H NMR spectra of the ligands and the
123

Multinuclear Cu(II) Schiff Base Complex as Efficient Catalyst
721
L1H, L2H and L3H are bonded to the per copper metal ion
in a N2O2 (for L1H) or N2O2 with N4 fashion (for L2H and
L3H) through the deprotonated phenolate oxygen, pyridine
nitrogen and salicylaldimine nitrogen.
Table 2 Crystal data and structure refinement parameters for
[Cu3(L2)4]Cl2ꢀCuCl2
Empirical formula
Formula weight
Temperature (K)
C84H108Cu4N8O4Cl4
1689.74
296
3.3 UV–Vis Spectra
´
˚
Wavelength (A)
0.71073
Crystal system
Space group
Monoclinic
C 2/c
Electronic spectra of ligands (LnH) and their mono- or
multinuclear Cu(II) metal complexes have been recorded
in the 200–1100 nm range in CHCl3 solution and their
corresponding data are given in experimental part. In the
electronic spectra of the ligands and their Cu(II) com-
plexes, the wide range bands seem to be due to both the
p ? p*and n ? p* of C=N chromophore or change-
transfer transition arising from p electron interactions
between the metal and ligand which involves either a
metal-to-ligand or ligand-to-metal electron transfer and
d–d transitions [35–37]. The electronic spectra of the
free ligands in CHCl3 showed strong absorption bands in
the ultraviolet region (264–368 nm) that could be
attributed respectively to the p ? p*and n ? p* transi-
tions in the benzene ring or azomethine (–C=N) groups.
The free ligands also exhibited different absorption bands
between 240–241 nm and 417–484 nm that attributed to
the n ? p*and n ? r* transitions [38]. In the spectra of
the mono- and multinuclear Cu(II) complexes, the posi-
tion and intensity of the absorption band characteristic of
the ligand appeared to be modified with respect to those
of the free ligand. Also, these spectra presented new
bands between 518–511 and 734–704 nm that were
characteristic of formed Cu(II) complexes. The absorp-
tion bands located in the 485–415 nm region were
attributed to the d ? p* charge-transfer transitions,
which overlap with the p ? p*and n ? p* transitions of
the ligands [38]. The new absorption bands observed at
lower energy and in lower intensities between 518–511
and 734–704 nm may be attributed to the d–d transitions
(dxy ? dx2–y2 and dz2 ? dx2–y2). These modifica-
tions in shifts and intensity for the absorption bands
supported the coordination of the ligand to the central
Cu(II) ion [38].
Unit cell dimensions
˚
a (A)
12.9385(5)
˚
b (A)
24.6741(7)
˚
c (A)
28.7391(11)
a (8)
b (8)
c (8)
90
93.834(3)
90
´
3
˚
Volume (A )
9154.3(6)
Z
4
Dcalc (Mg/m-3
Absorption coefficient (mm-1
)
1.226
)
1.08
F(000)
3536
Crystal size (mm)
h range for data collection (o)
0.41 9 0.31 9 0.21
1.4–26.1
Independent reflection
9005
Collected reflection
63,499
Absorption correction
Integration
Tmin
0.660
Tmax
0.841
Rint
0.097
h
-15 ? 15
k
-30 ? 30
l
-35 ? 35
Full-matrix least-squares on F2
Refinement method
wR(F2)
Goodness-of-fit on F2
Final R indices [I [ 2r(I)]
R indices (all data)
(D/r)max
0.118
0.92
0.045
0.084
0.003
0.45
´
-3
˚
Dqmax (e A
)
´
-3
)
˚
Dqmin (e A
-0.62
´
˚
Table 1 Selected bond lengths (A) and angles (ꢁ) for the copper complex with e.s.d.s. in parameters
O1–Cu1
1.899 (2)
1.911 (2)
1.936 (2)
1.936 (2)
153.03 (10)
92.30 (9)
94.50 (10)
N3–Cu2
2.148 (2)
2.131 (2)
1.304 (4)
91.45 (9)
157.94 (11)
97.28 (9)
94.60 (13)
C16–N1
C28–N2
C37–N2
Cu1i–Cu2–Cu1
Cl1–Cu3–Cl1i
N4–Cu2–N3i
O1–Cu1–N2
1.411 (4)
1.297 (4)
1.411 (4)
O2–Cu1
N4–Cu2
N1–Cu1
C7–N1
N2–Cu1
O2–Cu1–N2
N1–Cu1–N2
N4–Cu2–N3
N4–Cu2–N4i
175.94 (3)
178.83 (8)
144.55 (10)
91.96 (9)
O1–Cu1–O2
O1–Cu1–N1
O2–Cu1–N1
i
Symmetry code: 1-x, y, 1/2-z
123

722
M. Ulusoy et al.
magnetic moment of all the Cu(II) complexes showed a
normal magnetic moment (in experimental part). The mag-
netic moment data of the Cu(II) complexes range 1.78–1.73
B.M. for per Cu(II) molecule, corresponding to one unpaired
electron [39]. It is obvious that the Cu(II) complexes do not
possess antiferromagnetic properties at room temperature.
So, the Cu(II) complexes may be considered to have
tetragonal geometry.
Table 3 Hydrogen bonds and pꢀꢀꢀring interactions for [Cu3(L2)4]
Cl2ꢀCuCl2
´
´
´
D-HꢀꢀꢀA
D–H (A) HꢀꢀꢀA (A)
DꢀꢀꢀA (A) D–HꢀꢀꢀA (o)
˚
˚
˚
C9–H9BꢀꢀꢀO1
0.96
2.38
2.34
2.32
2.41
2.88
2.90
3.010(4)
2.977(5)
2.966(5)
3.039(5)
3.748(3)
3.791(4)
123
123
124
122
156
160
C10–H10AꢀꢀꢀO1 0.96
C30–H30AꢀꢀꢀO2 0.96
C31–H31BꢀꢀꢀO2 0.96
C7–H7ꢀꢀꢀCl1
0.93
C38–H38ꢀꢀꢀCl1ii 0.93
3.5 X-ray Structural Determination
´
˚
XꢀꢀꢀH(I)
Cg(J)
Cg(4)
Cg(5)i
Cg(3)
Cg(6)i
HꢀꢀꢀCg (A) X–HꢀꢀꢀCg (o)
C(9)–H(9B)
C(21)–H(21A)
C(31)–H(31B)
C(42)–H(42A)
2.7986
2.7876
2.9119
2.9068
142.68
134.75
143.26
164.79
The compound of [Cu3(L2)4]Cl2ꢀCuCl2 (Fig. 2) crystallizes
with Z0= 1/2 in the space group C2/c, and important
interatomic data are listed in Table 1. The crystallographic
data are summarized in Table 2. Details of hydrogen-bonds
and C–Hꢀꢀꢀp interactions are given in Table 3. Compound
[Cu3(L2)4]Cl2ꢀCuCl2, has got three Cu(II) ions. Firstly, the
Cu2 atom is located on a symmetry center and is coordi-
nated by four N atoms (N3, N4, N3i, N4i) [Symmetry code:
i = 1-x, y, 1/2-z] from pyridine groups of two L2H
ligand. Secondly, Cu1 atom has got four coordination
number with two phenolate O atom (O1 and O2) and two
azomethine-N atoms (N1 and N2) from two L2H ligands.
Finally, the Cu3 atom is located on a symmetry center and
is coordinated by two Cl atoms (Cl1 and Cl1i) [Symmetry
code: i = 1-x, y, 1/2-z]. Furthermore, there are two Cl
atoms (Cl2 and Cl3) which are not coordinated (Fig. 1).
Molecules of (I) are linked into sheets by a combination
of C–HꢀꢀꢀCl hydrogen bonds (Table 3). Atom C7 in the
Cg(3): N3–C16–C17–C18–C19–C20, Cg(4): N4–C37–C38–C39–
C40–C41, Cg(5): C1–C2–C3–C4–C5–C6, Cg(6): C22–C23–C24–
C25–C26–C27
i
Symmetry code: -x, y, 1/2-z; 1/2-x, y-1/2, 1/2-z
ii
3.4 Magnetic Moments
Magnetic susceptibility measurements provide sufficient
data to characterize the structure of the metal complexes.
Magnetic moments measurements of compounds were car-
ried out at 25 ꢁC. The room temperature effective magnetic
moments (leff) of all the complexes were measured on
polycrystalline samples after necessary diamagnetic cor-
rections using Pascal’s table. The room temperature
Fig. 1 Molecular structure of
[Cu3(L2)4]Cl2ꢀCuCl2,
asymmetric unit. [Symmetry
code: i = 1-x, y, 1/2-z]
123

Multinuclear Cu(II) Schiff Base Complex as Efficient Catalyst
723
Fig. 2 Part of the crystal
structure of
[Cu3(L2)4]Cl2ꢀCuCl2, showing
the formation of R22(12) and
R46(30) rings. H atoms not
involved in these interactions
have been omitted for clarity
hydrogen-bond donor to the benzene ring C22–C27 in the
molecule at (1-x, y, 1/2-z), so forming a R11(12) ring
(Fig. 3).
3.6 Catalytic Properties
The coupling of carbon dioxide and epoxides catalyzed by
salicylaldiminato-Cu(II) (Scheme 2) was investigated
under different reaction conditions. The results were shown
in Fig. 4 and Table 4. The yields of cyclic carbonates with
respect to the corresponding epoxides were determined by
comparing the ratio of product to substrate in the 1H NMR
spectrum of an aliquot of the reaction mixture. It can be
seen that the [Cu3(L2)4]Cl2ꢀCuCl2 has a better catalytic
activity than from the other Cu(II) complexes for the for-
mation of cyclic organic carbonates from carbon dioxide
(Fig. 4). Selectively very high conversion (TON = 30,335)
was obtained while using this catalyst ([Cu3(L2)4]
Cl2ꢀCuCl2) and epichlorohydrine as the substrate. The
results shown that the multinuclear Cu(II) salicyaldiminato
complexes [Cu3(L2)4]Cl2ꢀCuCl2 and [Cu3(L3)4]Cl2ꢀCuCl2
are more efficient than the other [Cu(L1)2] metal complex
which differs from the containing the ally substituent
instead of pyridine ring. Additional solvent such as
dichloromethane are found to be useless for catalytic runs
(compare Table 4, entry 5 and 6) [41].
Fig. 3 Part of the crystal structure of [Cu3(L2)4]Cl2ꢀCuCl2, showing
the C–Hꢀꢀꢀp interactions. [Symmetry code: i = 1-x, y, 1/2-z]
reference molecule at (x, y, z) acts as hydrogen-bond
donor, to atom Cl1 in the molecule at (x, y, z), so forming a
centrosymmetric R22(12) ring [40]. Similarly, atom C38 in
the reference molecule at (x, y, z) acts as hydrogen-bond
donor, to atom Cl1 in the molecule at (1/2-x, y-1/2, 1/2-
z), so forming a C22(9) chain running parallel to the [0-10]
direction. The combination of C–HꢀꢀꢀCl hydrogen bonds
generates R46(30) rings parallel to the ab plane (Fig. 2).
Compound (I) also contains two intramolecular C–Hꢀꢀꢀp
interactions and two intermolecular C–Hꢀꢀꢀp interactions
(Table 3). The intramolecular C–Hꢀꢀꢀp interactions produce
S(10) rings. The first is the intermolecular C–Hꢀꢀꢀp inter-
action, C21 atom in the molecule at (x, y, z) acts as a
hydrogen-bond donor to the benzene ring C1–C6 in the
molecule at (1-x, y, 1/2-z), so forming a R11(12) ring. The
second is the C42 atom in the molecule at (x, y, z) acts as a
The catalyst [Cu3(L2)4]Cl2ꢀCuCl2 shows low activity
without additives of DMAP that can enhance the Lewis
acidity of Cu metal centre. To ascertain the activity of
DMAP alone in this reaction under the given conditions the
catalytic blank-run without Cu(II) catalyst was done and
found to be 5% conversion (Table 4, entry 1). According to
this result it can be understood metal catalyst must be used
with a Lewis base such as DMAP. A comparative study
with DMAP and NEt3 was investigated. DMAP is more
active base with 47% yield (Table 4, entry 7), the yield is
123

724
M. Ulusoy et al.
Fig. 4 The comparison of
metal catalysts for the formation
of 1,2-epoxy butane to
corresponding cyclic carbonate
at the same catalytic conditions;
reaction conditions: Cu(II)
complexes (1.125 9
10-5 mol), 1,2-epoxy butane
(4.5 9 10-2 mol), DMAP
(9 9 10-5
mol), CO2 (1.6 Mpa),
100 ꢁC, 2 h
Table 4
O
O
O
O
0. 025 % [Cu3(L2)4]Cl2 .CuCl2
CO2
R
0.1% DMAP,
1.6 Mpa
R
Yield (%)a
Selectivity (%)
TONb
˚
Temp. (C)
Entries
Epoxide
Time (hour)
Solvent
1.
PO
PO
EC
EC
EB
EB
PO
PO
PO
PO
PO
PO
SO
SO
2
1
100
80
–
5.0c
7.3d
98.3
90.7
60.0
76.3
46.9
53.0
64.7
87.0
88.6
34e
95
96
98
98
98
96
99
99
98
98
97
94
92
89
50
7
2.
–
3.
2
100
100
100
100
100
100
100
100
100
100
100
100
–
3932
30,335*
2400
3052
1876
2120
2588
3480
3544
340
4.
6
–
5.
2
CH2Cl2
6.
2
–
–
–
–
–
–
–
–
–
7.
2
8.
4
9.
6
10.
11.
12.
13.
14.
10
12
2
2
83.0
95.0
3320
3800
4
R Propylene oxide (PO), epichlorohydrine (EC), styrene oxide (SO) and 1,2-epoxy butane (EB)
*
2.99 9 10-5 mol [Cu3(L2)4]Cl2ꢀCuCl2 was used as the catalyst; 100 mmol (7.8 mL) epichlorohydrin was used as the substrate;
5.98 9 10-5 mol DMAP was used as the base
a
Yield of epoxides to corresponding cyclic carbonates was determined by comparing the ratio of product to substrate in the 1H NMR spectrum
of an aliquot of the reaction mixture
b
Moles of cyclic carbonate produced per mole of catalyst
c
Blank run, without any Cu(II) complex as catalyst
d
Et3N was used as base
CuCl2ꢀH2O (4.5 9 10-5 mol) was used as the catalyst
e
decreased to 7% yield (Table 4, entry 2) when NEt3 was
replaced as base while using the PO as the substrate.
Meanwhile the electron-withdrawing groups at the
2-position of the epoxide activated the substrate, while
electron-donating substituents deactivate the epoxide
(Table 4). The epoxides (1,2-epoxybutane, propylene
oxide, styrene oxide, epichlorohydrin) were tested as
substrates and epichlorohydrin was found to be the most
reactive epoxide, while propylene epoxide exhibited the
lowest activity of the epoxides surveyed in accordance with
previous study [21]. The active catalyst contains trimetallic
species of the copper in it. In order to see the effect of
[CuCl2] on the activation of CO2, the blank run was per-
formed (Table 4, entry 12). As can be seen from the table
123

Multinuclear Cu(II) Schiff Base Complex as Efficient Catalyst
725
2. Coates GW, Moore DR (2004) Angew Chem Int Ed 43:6618
only 34% cyclic carbonate yield was obtained. This result
shows that the independent CuCl2.H2O salt is not effective
alone. The whole complex [Cu3(L2)4]Cl2ꢀCuCl2 showed
activity. The influence of time on the yield of propylene
carbonate (PC) was investigated at CO2 pressures of
1.6 MPa. The yield increased with increasing the catalytic
reaction time (Table 4).
3. Darensbourg DJ, Ulusoy M, Karronnirun O, Poland RR, Rei-
benspies JH, C¸ etinkaya B (2009) Macromolecules 42:6992
4. Gibson DH (1996) Chem Rev 96:2063
5. Miao CX, Wang JQ, Wu Y, Du Y, He LN (2008) ChemSusChem
1:236
6. Kendall JL, Canelas DA, Young JL, DeSimone JM (1999) Chem
Rev 99:54
7. Darensbourg DJ, Holtcamp MW (1996) Coord Chem Rev
153:155
8. Inoue S (1976) Chemtech 6:588
9. Beckman EJ (1999) Science 283:946
10. Lu XB, Shi L, Wang Y, Zhang R, Zhang YJ, Peng XJ, Zhang ZC,
Li B (2006) J Am Chem Soc 128:1664
11. Jung JH, Ree M, Chang T (1999) J Polym Sci Part A Polym
Chem 37:3329
12. Aida T, Ishikawa M, Inoue S (1986) Macromolecules 19:8
13. Shaikh AAG, Sivaram S (1996) Chem Rev 96:951
14. Sakakura T, Kohno K (2009) Chem Commun 11:1312
15. Mette M, Mikkel J, Frederik CK (2010) Energy Environ Sci 3:43
16. Xiaoding X, Moulijn JA (1996) Energy Fuels 10:05
17. Zeverhoven R, Eloneva S, Teir S (2006) Catal Today 115:73
18. Pachero MA, Marshall CL (1997) Energy Fuels 11:2
19. Kanmani Raja K, Easwaramoorthyb D, Kutti Ranib S, Rajeshc J,
Jorapurd Y, Thambidurai S, Athappanc PR, Rajagopala G (2009)
J Mol Catal A Chem 303:52
20. Choi JC, Kohno K, Ohshima Y, Yasuda H, Sakakura T (2008)
Cataly Commun 9(7):1630
21. Miller JA, Gross BA, Zhuravel MA, Jin W, Nguyen ST (2005)
Angew Chem Int Ed 44:3885
22. Kilic A, Ulusoy M, Durgun M, Tasci Z, Yilmaz I, Cetinkaya B
(2010) Appl Organometal Chem 24:446
23. Earnshaw A (1968) Introduction to magnetochemistry. Academic
Press, London, p 4
24. Sheldrick GM(1997) SHELXS-97—a program for automatic
solution of crystal structures. University of Gottingen, Germany
25. Farrugia LJ (1998) WINGX—a windows program for crystal
structure analysis. University of Glasgow, Glasgow
26. Farrugia LJ (1997) J Appl Crystallogr 30:565
27. Spek AL (2005) PLATON—a multipurpose crystallographic tool.
Utrecht University, Utrecht,
28. Kilic A, Tas E, Deveci B, Yilmaz I (2007) Polyhedron 26:4009
29. Tas E, Kilic A, Durgun M, Yilmaz I, Ozdemir I, Gurbuz N (2009)
J Organomet Chem 694:446
Although the metal centres are different (copper and
ruthenium), the reaction mechanism we envisioned the
occurrence of a similar pathway to the one proposed pre-
viously [42], i.e. activation of the epoxide by a Lewis acid
and attacks of Lewis base (DMAP) to the sterically less
hindered carbon atom to open the epoxide ring. The gen-
erated oxy anion species then reacts with CO2 to give the
cyclic carbonate.
4 Conclusion
New salicylaldimine ligands (L1H, L2H and L3H) with
their mono or multinuclear Cu(II) complexes were syn-
thesized. Besides the classical methods such as FT-IR,
1
UV–Vis, H-NMR, elemental analysis, and X-ray struc-
tural determination in addition to magnetic susceptibility
techniques for structural characterization. The synthesized
steric hindered salen type complexes tested as catalysts for
the formation of cyclic carbonates from CO2 and liquid
epoxides. The results showed that the multinuclear Cu(II)
salicyaldiminato complexes [Cu3(L2)4]Cl2ꢀCuCl2 and
[Cu3(L3)4]Cl2ꢀCuCl2 are more efficient than the other
[Cu(L1)2] metal complex which differs from the containing
the ally substituent instead of pyridine ring. Very high
conversion (TON = 30,335) was obtained while using this
catalyst ([Cu3(L2)4]Cl2ꢀCuCl2) and epichlorohydrine as the
substrate.
30. Ulusoy M, Sahin O, Buyukgungor O, Cetinkaya B (2008) J Or-
ganomet Chem 693:1895
31. Kılınc¸arslan R, Karabıyık H, Ulusoy M, Aygu¨n M, C¸ etinkaya B,
5 Supplementary Material
¨
Bu¨yu¨kgu¨ngor O (2006) J Coord Chem 59:1649
32. Kilic A, Tas E, Gumgum B, Yilmaz I (2007) Heteroatom Chem
18:657
33. Li Y, Yang ZY (2009) Inorg Chim Acta 13:4823
34. Shit S, Chakraborty J, Samanta B, Slawin AMZ, Gramlich V,
Mitra S (2009) Struct Chem 20:633
CCDC 759981 contains the supplementary crystallographic
data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre
via www.ccdc.cam.ac.uk/data_request/cif.
35. Sacconi L, Ciampolini M, Maffio F, Cavasino FP (1962) J Am
Chem Soc 84:3246
Acknowledgments The authors wish to acknowledge the Faculty of
Arts and Sciences, Ondokuz Mayis University, Turkey, for the use of
the Stoe IPDS-II diffractometer (purchased from Grant No. F279
of the University Research Fund).
36. Carlin RL (1965) Trans Met Chem. Marcel Dekker, New York,
vol 1
37. Tas E, Ulusoy M, Guler M, Yilmaz I (2004) Trans Met Chem
29:180
38. Pui A, Mahy JP (2007) Polyhedron 26:3143
39. Ilhan S, Temel H, Kilic A (2008) J Coord Chem 61(2):277
40. Bernstein J, Davis RE, Shimoni L, Chang NL (1995) Angew
Chem Int Ed Engl 34:1555
References
41. Ulusoy M, Cetinkaya E, Cetinkaya B (2009) Appl Organometal
Chem 23:68
42. Bu Z, Qin G, Cao S (2007) J Mol Cat A 277:35
1. Kember MR, White AJP, Williams CK (2009) Inorg Chem
48:9535
123