Synthesis and characterization of novel positively charged organocobaloximes as catalysts for the fixation of CO2 to cyclic carbonates

Article abstract of DOI:10.1016/j.jorganchem.2018.01.029

A series of cobaloxime complexes (1-6) and corresponding to positively charged phosphonium linked cobaloximes (called cobaloximes-PCPL) (1a-6a) have been prepared in high yields and characterized by means of NMR (¹H, ¹³C, and ³¹

Full text of DOI:10.1016/j.jorganchem.2018.01.029

Journal of Organometallic Chemistry 858 (2018) 78e88
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
Synthesis and characterization of novel positively charged
organocobaloximes as catalysts for the fixation of CO2 to cyclic
carbonates
Ahmet Kilic*, Mustafa Durgun, Emine Aytar, Rahime Yavuz
Harran University, Chemistry Department, 63290, Osmanbey Campus, Sanliurfa, Turkey
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of cobaloxime complexes (1-6) and corresponding to positively charged phosphonium linked
cobaloximes (called cobaloximes-PCPL) (1a-6a) have been prepared in high yields and characterized by
means of NMR (1H, 13C, and 31P) spectra, FT-IR spectra, UVeVis spectra, mass spectra, melting point as
well as elemental analysis. Following a full characterization, the study of spectroscopic properties of
these novels cobaloximes/cobaloximes-PCPL compounds was done in detail. Under optimized reaction
conditions, these novel cobaloximes (1-6) and cobaloximes-PCPL (1a-6a) were applied to the as a mo-
lecular catalyst for the cycloaddition of epoxides with CO2 without organic solvent. In a combination with
(4-bromo butyl)triphenylphosphonium bromide (BTP) and cobaloxime compounds, namely
cobaloximes-PCPL were identified as highly active catalysts for the formation of cyclic carbonates
compared to previously reported cobaloximes catalysts by our groups. As an axial group bounding to the
Co(III) center of cobaloximes-PCPL, (BTP) was barely revealed that it has an effect on catalyst activities.
Also, a systematic investigation of reaction parameter effects such as temperature, time and pressure was
done.
Received 6 December 2017
Received in revised form
15 January 2018
Accepted 20 January 2018
Available online 31 January 2018
Keywords:
Spectroscopy
Cobaloximes-PCPL
Carbon dioxide
Epoxide
Cyclic carbonate
© 2018 Elsevier B.V. All rights reserved.
1. Introduction
aprotic polar solvents that are used widely in chemical and phar-
maceutical industries [5], electrolytes for lithium-ion batteries,
It was a long time goal for numerous scientists in chemical
sector to develop superior performance, high stability, reactivity,
inexpensive and environmental friendly catalysts for the coupling
reactions of CO2 and epoxides which produce five-membered cyclic
carbonates in a 100% atom economic manner, due to the fact that
CO2 is an abundant, cheap, renewable, nonflammability, easy
accessibility and non-toxic C1 feedstock [1,2]. But, the thermody-
namic and kinetic stability of CO2 and its negative adiabatic elec-
tron affinity and large ionization potential [3] make effective
fixation and conversion of CO2 into high-value chemicals under
normal conditions arduous [4]. Currently, in light of those re-
searches, many ways to solve this problem have been suggested. In
this context, one of the promising and mostly preferred route is the
chemical utilization of CO2 to synthesize cyclic carbonates as
valuable products, owing to widespread applications of cyclic car-
bonates as organic intermediates, monomers for polymerization,
precursors of polymeric materials, crude materials for plastics and
fine chemicals [6]. For the formation of cyclic carbonates via
cycloaddition of CO2 to epoxides that are currently used as useful
chemicals in numerous applications, diverse homogeneous and
heterogeneous catalytic systems have been reported.
To date, numerous metal-based homogeneous catalysts,
including metal-organic frameworks (MOFs) [7,8], metal-salen
complexes [9e19], metalloporphyrin [20e22], metal chloride
[23e26], polyoxometalate [27], metal oxide [28,29] and metal-free
catalysts containing hydroxyl [30e34], carboxyl [35e37] have been
used to encourage CO2 conversion under ambient conditions. But,
the production of the cyclic carbonates through the cycloaddition
of CO2 to epoxides was the factor for which the research of the
effective, robust, cheap, and stable catalyst has been continued.
With this in mind, one of the novel classes of catalysts are the
cobaloximes and corresponding to cobaloximes-PCPL form have
undoubtedly attracted much attention probably due to the intrinsic
properties of various dioxime ligands such as easy-to-handle, air-
stable, different structures, strong complexation ability with almost
* Corresponding author.
E-mail address: kilica63@harran.edu.tr (A. Kilic).
https://doi.org/10.1016/j.jorganchem.2018.01.029
0022-328X/© 2018 Elsevier B.V. All rights reserved.

A. Kilic et al. / Journal of Organometallic Chemistry 858 (2018) 78e88
79
all metal ions, structural diversity, and tenability. To the best of our
knowledge, the use of PCPL-based and, in general, the use of truly
organometallic complexes (containing a C-metal bond) for the
cycloaddition of CO2 to epoxides is very rare. The nature of the
metal center in metal complexes plays an important role in their
catalytic performance and cobaloxime with corresponding to
cobaloximes-PCPL are maybe one of the most efficient catalyst
systems for desirable CO2 transformation. Especially, the covalent
linked (BTP) containing cobaloximes-PCPL (1a-6a) were identified
as highly active catalysts for the formation of cyclic carbonates
compared to previously reported cobaloxime catalysts by our
groups [38]. To that end, we hope that the cobaloximes-PCPL
continue to generate a lot of interest compared to the cobalox-
imes due to synergistic effects of multiple sites on epoxide activa-
tion which Lewis acidic cobalt, protonated triphenyl phosphonium
able to activate epoxide and bromide anion acting as a nucleophile
source. For our study, six cobaloxime complexes (1-6) and six cor-
responding to cobaloximes-PCPL (1a-6a) were prepared and char-
acterized by means of NMR (1H, 13C, and 31P) spectra, FT-IR spectra,
UVeVis spectra, mass spectra, melting point as well as elemental
analysis and evaluated the catalytic activity for the cycloaddition
between epoxides and CO2.
obtained through using an Agilent LC-MS/MS spectrometer by ESI
technique. Melting points of all complexes have been determined
in open capillary tubes on an Electrothermal 9100 melting point
apparatus and are uncorrected. Catalytic experiments were carried
out in a PARR 4560 50 mL stainless steel high-pressure reactor. CO2
with a purity of 99.999% was commercially available and used
without further purification. GC analysis was performed on an
Agilent 7820A model for conversion and selectivity of the mixture.
The mixture was separated by centrifugation, and the liquid phase
was subjected to GC analysis with ethylene glycol dibutyl ether as
an internal standard and hydrogen as the carrier gas.
(Ph3PþBr-C4H8Br) (BTP) was synthesized according to the following
procedure [39] (Scheme 2). The ligand precursors L1H2 and L2H2
were synthesized according to optimization of literature methods
[38]. The dimethylglyoxime is preferred as ligand L3H2 and pur-
chased from Sigma-Aldrich Company. The preferred ligands (L1H2),
(L2H2), and (L3H2) for this study are given in the Scheme 1.
2.1. General procedure for the cycloaddition of epoxides to CO2
A 25 mL stainless pressure reactor was charged with cobalox-
imes
or
cobaloximes-PCPL
(4.5 ꢁ 10ꢀ8 mmol),
epoxides
(4.5 ꢁ 10ꢀ5 mmol), Lewis base (9 ꢁ 10ꢀ8 mmol) and an appropriate
amount of ethylene glycol dibutyl ether (as an internal standard for
GC analysis). The reaction vessel was placed under a constant
pressure of CO2 for 2 min to allow the system to equilibrate and
then the autoclave was charged with CO2 to the desired pressure.
The reaction mixture was heated to the desired temperature. The
pressure was kept constant during the reaction. After the desired
reaction time the vessel was cooled to 5-10 ꢂC in the ice bath. The
pressure was released by venting the excess gases. The yields of the
corresponding cyclic carbonates were determined by using an
Agilent 7820A Gas Chromatograph.
2. Experimental
Organic solvents and chemicals with analytical and starting
reagents obtained from commercial sources were used as received
for this study any additional chemical process. All FT-IR measure-
ments were acquired in the range of 4000e400 cmꢀ1 using the ATR
accessory on a Perkin-Elmer Two UATR-FT spectrophotometer.
UVeVis spectra were recorded on a Perkin-Elmer model Lambda 25
spectrophotometer in the range of 200e1100 nm using quartz cu-
vettes at room temperature. The CH2Cl2 and CH3OH have been
selected as solvents for electronic spectra. Elemental analyses (C, H,
and N) were performed on a LECO CHNS model 932 of the
elemental analyzer. The 1H and 13C NMR spectra were acquired on a
Bruker 300 and 75 MHz spectrometer with DMSO-d6 as solvent
referenced to 2.50 ppm, consecutively. 31P NMR spectra were
recorded at 242.9 MHz in DMSO-d6. The example was allowed to
equilibrate for at least 20 min at room temperature before the 31P
NMR spectrum was obtained. The mass spectra (LC-MS) were
2.2. Synthesis of the cobaloximes [ClCo(L(1-3)H)2(B1-2)] (1-6)
The ligand precursors L1H2 (1.15 g, 6.0 mmol), L2H2 (1.15 g,
6.0 mmol), and L3H2 (0.70 g, 6.0 mmol) with CoCl2.6H2O (0.72 g,
3.0 mmol) were mixed in a 100 mL reaction flask for 30 min at room
temperature under an N2 atmosphere. The solutions turned green
immediately. The green reaction mixtures were stirred for another
Scheme 1. The selected ligands (L1H2), (L2H2), and (L3H2) for this study.
Scheme 2. The synthesis of (Ph3PþBr-C4H8Br) (BTP) compound.

80
A. Kilic et al. / Journal of Organometallic Chemistry 858 (2018) 78e88
4 h at reflux temperature. Then, the mixtures were cooled to room
temperature and 1-Amino-4-methyl piperazine (B1) (0.35 g,
3.0 mmol) or 3-Morpholinopropylamine (B2) (0.43 g, 3.0 mmol)
was added to mixtures as a neutral base in the presence of air. The
initially green solution turned brown upon a neutral base and
5.0 mL of water addition. The air was passed through the mixture
for about 4 h. Recrystallization from CH2Cl2/C2H5OH by slow
evaporation afforded pure compounds obtained. The obtained dark
brown crystals were washed three times with diethyl ether, fol-
lowed by drying at 60 ꢂC for 24 h in vacuum to give desired com-
pounds. The spectroscopic results are given as supplementary
information for the cobaloxime complexes [ClCo(L(1-3)H)2(B1-2)]
(1-6).
structural diversity, which makes them potential materials for
further chemical, spectroscopic and catalytic applications. The ac-
tivity of these cobaloxime (1-6) and corresponding cobaloximes-
PCPL (1a-6a) in the coupling of CO2 and epoxides for formed of
cyclic carbonate is also presented. In light of the obtained catalytic
results, our previous publication show lower conversion for the
coupling of CO2 and epoxides produce five-membered cyclic car-
bonates catalyzed by cobaloxime complexes compared with
cobaloximes-PCPL. In the design of the cobaloximes-PCPL (1a-6a)
complexes, we are just removing the axial chlorine group and (BTP)
as new axial group connect to the Co(III) center in order to further
improve the catalytic activity due to synergistic effects of multiple
sites on epoxide activation which Lewis acidic cobalt, protonated
triphenyl phosphonium able to activate epoxide and bromide anion
acting as a nucleophile source.
2.3. Synthesis of the cobaloximes-PCPL [(BTP)Co(L(1-3)H)2(B1-2)]
(1a-6a)
3.2. Spectroscopic studies
The each cobaloxime [ClCo(L1-3H)2(B1-2)] (1-6) complexes
(3.0 mmol) was carefully added to an aqueous solution of NaOH
(0.4 g, 10.0 mmol, 4 mL) were stirred in 50 mL of absolute ethanol.
The mixtures were stirred to for 40 min under dry nitrogen and
then was cooled to 0 ꢂC. The mixtures immediately turned blue,
followed by the subsequent addition of an aqueous solution of
sodium borohydride (NaBH4) (0.76 g, 20.0 mmol, 5 mL) in the
presence of the N2 atmosphere. Later, the (BTP) (3.83 g, 8.0 mmol)
immediately was added to the reaction flask and dark blue color
turned brown. The resulting mixtures were stirred at 0 ꢂC for 4 h in
the N2 atmosphere and in the absence of light. After that, the
mixture was exposed to air and 5 mL of acetone and 5 mL of water
were added. The obtained crude precipitate was filtered, washed
with diethyl ether and hexane. The crude product was then
recrystallized from CH2Cl2/CH3OH (1:4) solutions and to give the
expected product as white brown. The spectroscopic results are
given as supplementary information for the cobaloximes-PCPL
[(BTP)Co(L(1-3)H)2(B1-2)] (1a-6a).
To have information about all complexes, FT-IR spectra of the
cobaloximes (1-6) and cobaloximes-PCPL (1a-6a) were compared
in detail. The FT-IR stretching vibration data of all the cobaloxime
complexes are listed in the Supporting Information section and
Figs. 1Se3S. Our attention was focused on the disappearance of the
free
y(O-H) peaks of all ligands, and also appearance of the inter-
molecular hydrogen bond
y(O-H/O) peaks at around 3547-
3117 cmꢀ1, which on an encapsulation of the NH2 groups, confirm
formation of the (1-6) and (1a-6a) complexes, respectively [40].
The FT-IR spectra of the (1a-6a) complexes compare with the (1-6)
complexes, a new stretching frequency was observed at range
3057e3053 cmꢀ1, indicate the presence of aromatic triphenyl
phosphonium groups and also support to the (BTP) as axial group
connect to the Co(III) center of the cobaloximes-PCPL (1a-6a).
When the characteristic
y
(C¼N) peaks at range 1652e1608 cmꢀ1 of
the cobaloximes (1-6) were compared with that of the corre-
sponding cobaloximes-PCPL (1a-6a), a small frequency shift (3-
5 cmꢀ1) is exhibits which may be attributed to the formation of the
cobaloxime complexes (Supporting Information section and
Figs. 1Se3S). The N-O stretching frequencies appear as a strong
peak in the 1271-1230 cmꢀ1 range as reported for similar coba-
loxime complexes. Another important observation for the all
cobaloxime complexes, a new peak appears at low frequencies
which are absent in the free ligand spectrum. These stretching vi-
brations are observed in the 482-510 cmꢀ1 range, are due to the
CoIII-N vibrations, an indication of coordination of the ligand to the
Co(III) center presumably through the oxime nitrogen lone pair [3].
Furthermore, the other stretching vibrations in the FT-IR spectra of
all cobaloxime complexes also support the formation of the pro-
posed compounds.
3. Results and discussion
3.1. General properties
A summary of our selected dioxime ligands (L1H2, L2H2, and
L3H2) and their cobaloximes (1-6) and cobaloximes-PCPL (1a-6a)
compounds are given in Schemes 1 and 3. As shown in Scheme 1,
the asymmetric dioxime ligands (L1H2 and L2H2) are prepared ac-
cording to previously reported literature procedure by our group
[38] to high yield and making some modifications, whereas the
symmetric ligand L3H2 (dimethyl glyoxime) are purchased from
commercial firms. The differently substituted target cobaloxime
complexes (1-6) have been synthesized via CoCl2.6H2O and two
equivalents of dioxime ligands with one equivalent of 1-Amino-4-
methyl piperazine (B1) or 3-Morpholinopropylamine (B2) in the
same reaction flask and were open to the air in EtOH, fairly good
yields after recrystallization from CH2Cl2/C2H5OH solvents (Scheme
3). The cobaloximes-PCPL (1a-6a) has been synthesized using the
cobaloximes (1-6) and (BTP) in absolute EtOH and the presence of
NaOH and NaBH4 under nitrogen atmosphere, to afford
cobaloximes-PCPL (1a-6a) in high yields (Scheme 3). The structure
of the cobaloxime and cobaloximes-PCPL was confirmed by NMR
(1H, 13C, and 31P) spectra, FT-IR spectra, UVeVis spectra, mass
spectra, melting point as well as elemental analysis. The found and
the calculated percentages of all cobaloxime complexes were
confirmed by C, H, N analysis and results agree with each other and
these prove the proposed molecular formulas. The obtained coba-
loxime and cobaloximes-PCPL are easy-to-handle, air-stable,
different structures, strong complexation ability with Co(III) ions,
Further evidence for the formation of the cobaloxime (1-6) and
cobaloximes-PCPL (1a-6a) were obtained from the 1H and 13C NMR
spectra and 31P NMR for only the cobaloximes-PCPL (1a-6a). The 1H
and 13C NMR data of all the complexes are listed in the Supporting
Information section and Figs. 4Se15S. The NMR spectra results of
all complexes were recorded in DMSO-d6 taking TMS as an internal
standard. The most characteristic 1H NMR peaks that confirm the
formation the cobaloxime (1-6) and cobaloximes-PCPL (1a-6a)
were observed and identified one singlet (the containing of sym-
metric ligands) or two singlet (the containing of asymmetric li-
gands) resonances at range 19.06e18.37 ppm attributed to
intramolecular D2O exchangeable H-bridge (OꢀH/O) resonance
[40e42]. The aromatic ring protons of ligands and the aromatic
protons of triphenyl phosphonium group were observed with the
expected chemical shift and integral values in the different chem-
ical shift region. These resonances were detected at range 7.41e7.08
and 7.90e7.58 ppm, respectively. It can be clearly seen that absence

A. Kilic et al. / Journal of Organometallic Chemistry 858 (2018) 78e88
81
Scheme 3. The structure of the proposed the cobaloximes (1-6) and cobaloximes-PCPL (1a-6a).
of these aromatic proton signals of triphenyl phosphonium group
which replacement of the chlorine atom by (BTP) in the
cobaloximes-PCPL (1a-6a), as expected. Among the three pairs of
new displayed peak, the chemical shifts in 4.68e3.35, 2.51e2.10
and 1.24e0.95 ppm correspond respectively to (-CH2-CH2-Co), (P-
CH2), and (P-CH2-CH2) of the cobaloximes-PCPL (1a-6a), indicating
the formation of cobaloximes-PCPL (1a-6a) from cobaloxime (1-6)
complexes by losing the axial Cl ligand. Moreover, the NMR data
show that the chemical shift of 13C peaks supports the formation of
the cobaloxime (1-6) and cobaloximes-PCPL (1a-6a) in DMSO gave
signals in good agreement with the proposed structures. The
cobaloxime (1-6) and cobaloximes-PCPL (1a-6a) showed signals in
range 154.96e144.46 ppm assigned to dioxime (C¼NOH) carbons
[3,43]. These chemical signals from dioxime (C¼NOH) carbons
were in the expected range. The aromatic ring carbon signals and
the aromatic carbon of triphenyl phosphonium group were
centered in 142.66e117.99 ppm with the expected chemical shift
d
in the different shift region. Also, other important results that
confirm the formation of cobaloximes-PCPL (1a-6a) is the presence
of a new displayed carbon peak in range 41.94e39.56, 32.63e25.86,
30.01e20.64 and 21.54e15.07 ppm belonging to the (P-CH2), (-CH2-
CH2-Co), (-CH2-CH2-Co), and (P-CH2-CH2) carbon signals of (BTP),
respectively, in their 13C NMR spectra. Furthermore, the other
proton and carbon signals in the NMR spectra of all complexes also
support the formation of the proposed compounds (see experi-
mental part). Another decisive event, the cobaloximes-PCPL (1a-6a)
showed a characteristic peak in 31P NMR (Figs. 1e2 and S16-S19)
compared to cobaloxime (1-6) complexes, suggesting the alteration
of axial Cl ligand because of the presence of the (4-bromobutyl)
triphenylphosphonium in cobaloximes-PCPL compounds. These

82
A. Kilic et al. / Journal of Organometallic Chemistry 858 (2018) 78e88
new phosphorus peaks of the triphenyl phosphonium group of the
axial ligand are detected in 24.07e23.73 ppm in 31P NMR spectra
(Figs. 1e2), respectively, confirmed the covalent attachment of
(BTP) to the Co(III) center of cobaloxime complexes.
To investigate the electronic properties of the cobaloxime (1-6)
and cobaloximes-PCPL (1a-6a), the UVeVis absorption spectra of
all complexes in CH2Cl2 and CH3OH were measured at room tem-
perature and the experimental data results are given in the
experimental part. Regarding the absorption spectroscopy, there
are some obvious bands clearly separated from the other transition
in the electronic spectrum of the cobaloxime (1-6) and
cobaloximes-PCPL (1a-6a) and these absorption bands centered at
under 382 nm in CH2Cl2 and under 385 nm in CH3OH could be
m/z corresponding to cobaloxime (1-6) appeared at 592.1 for
complex (1), at 621.2 for complex at (2), 592.2 for complex (3), at
621.2 for complex (4), at 439.9 for complex (5), at 468.9 for complex
(6), consecutively. Whereas, the molecular ion peaks of
cobaloximes-PCPL (1a-6a) were displayed at m/z ¼ 954.9 for com-
plex (1a), at 984.0 for complex (2a), at 954.7 for complex (3a), at
983.9 for complex (4a), at 802.7 for complex (5a), and at 831.8 for
complex (6a), consecutively. This LC-MS speciation indicates that
the axial Cl ligand is replacement by (BTP) in the cobaloximes-PCPL
(1a-6a), as expected. These mass results are in a good agreement
with the proposed molecular formula for these compounds.
3.3. Catalytic performance
ascribed to the
p / p* or n / p* charge transfer transitions of Co-
dioximate (C¼N/Co) and H2N-Co(III) ligand-to-metal charge
transfer (LMCT) transition [44]. Another decisive absorption bands
for the occurrence of these cobaloxime complexes, which displayed
new absorption peaks at range 391e469 nm because of the inter-
molecular transition from the dioxime ligands to the vacant orbitals
localized on the coordinated Co(III) center [45]. Additionally, one of
the most important absorption bands for these cobaloxime com-
plexes is the d-d transition in the visible region. But, the charac-
teristic forbidden d-d transition of Co(III) center not observed in the
visible spectra of the complexes, most probably due to their very
low molar absorption coefficient. However, the cobaloxime (2) and
(3) in CH2Cl2 have displayed new broad d-d transitions at 706 and
702 nm respectively, at high concentrations with the lowest en-
ergies, due to Co(III) center of cobaloximes, characteristic of their
geometries.
To evaluate the effects of cobaloxime and (4-bromo butyl) tri-
phenyl phosphonium bromide (BTP) as axial group connect to
cobaloximes-PCPL as molecular catalysts with Lewis bases as a co-
catalysts on the CO2 fixation reaction, we studied the activity of the
prepared various catalysts and the obtained results are given in
Table 1. Our catalytic system does not work as a single component
system likely because of the presence of the cyclic amino ligands.
The density of Lewis acid/base, steric and electronic effects of co-
catalyst, axial or equatorial ligands, absorbed CO2 quantities and
reaction conditions may influence the catalytic activity in the
cycloaddition of CO2 to epoxides for obtained the desired cyclic
carbonate. Owing to the high density of Lewis acidic cobalt center,
various axial neutral base, quaternary phosphonium bromide able
to activate epoxide and bromide anion acting as a nucleophile,
cobaloximes-PCPL (1a-6a) exhibits a better catalytic efficiency
compared to cobaloximes (1-6) in the cycloaddition of different
epoxides with CO2 into cyclic carbonate under suitable conditions
(100 ꢂC, 1.6 MPa, and ambient CO2 pressure) using only 0.1 mol%
catalyst amount under solvent-free conditions (Table 1). The first
catalytic experiments were realized at 100 ꢂC, with an initial
The LC-MS spectrum has also provided evidence for the for-
mation of the cobaloxime (1-6) and cobaloximes-PCPL (1a-6a);
whose peak locations and the isotopic distributions confirming
their molecular weights (Figs. 3e6). The LC-MS analyses were
measured at 25 ꢂC and the data are given in experimental part. The
Fig. 1. The 31P NMR spectra of [(BTP)Co(L1H)2(B2)] (2a) complex.

A. Kilic et al. / Journal of Organometallic Chemistry 858 (2018) 78e88
83
Fig. 2. The 31P NMR spectra of [(BTP)Co(L2H)2(B2)] (4a) complex.
Fig. 4. The LC-MS spectra of [ClCo(L2H)2(B1)] (3) complex.
Fig. 3. The LC-MS spectra of [ClCo(L1H)2(B2)] (2) complex.
other epoxides [46]. However, the use of the cobaloxime (1-6) or
cobaloximes-PCPL (1a-6a) catalysts alone the absence of co-
catalysts was not effective for the cycloaddition reaction, indi-
cating that the cocatalysts play a key role in catalyzing this reaction.
The possible side reactions give diols, occurring via the decompo-
sition of the epoxides. According to recently published papers [47],
propane-1,2-diol was obtained as a by-product from the reaction of
propylene oxide and water and was analyzed by GC-MS and NMR
techniques.
After showing high catalytic performance for cobaloxime-PCPL
(1a), the effect of reaction parameters (epoxide, base, tempera-
ture, CO2 pressure and time) was investigated for the cobaloxime-
PCPL (1a) catalyst. First, in order to study the efficiency and general
applicability of the catalyst (1a), the cycloaddition of CO2 with
different epoxides (epichlorohydrin (ECH), 1, 2-epoxybutane (EB),
propylene oxide (PO), styrene oxide (SO), and cyclohexene oxide
pressure of CO2 of 1.6 MPa, for 2 h with 0.1 mol% of the cobaloxime
(1-6) and cobaloximes-PCPL (1a-6a) catalysts (Table 1). Using 1, 2-
epoxybutane (EB) as a model epoxide, the coupling reaction of CO2
and 1, 2-epoxybutane (EB) to yield 4-(ethyl)-1,3-dioxolane-2-one
(EBHC) was examined. For example, EB was used as both sub-
strate and solvent, catalyst loading was 0.1 mole %, in combination
with 4-dimethylamino pyridine (DMAP) as a co-catalyst was
employed and the reaction was carried out with 1.6 MPa CO2 at
100 ꢂC. Among the cobaloxime/cobaloximes-PCPL catalysts,
cobaloxime-PCPL (1a) and DMAP as a co-catalyst showed high
catalytic activity (61.8%) and selectivity (99.7%) for the coupling of
CO2 and EB (Table 1, entries 7). Excluding cobaloxime-PCPL (1a)
complex, when the other cobaloximes/cobaloximes-PCPL was
employed as a catalyst at 100 ꢂC, the lower catalytic conversions
was observed (Table 1, entry 1-12). We assume that this could be
explained by the low solubility of them in the 1, 2-epoxybutane and

84
A. Kilic et al. / Journal of Organometallic Chemistry 858 (2018) 78e88
Fig. 5. The LC-MS spectra of [(BTP)Co(L1H)2(B2)] (2a) complex.
Fig. 6. The LC-MS spectra of [(BTP)Co(L2H)2(B1)] (3a) complex.
(CHO)) under reaction conditions which was carried out at 100 ꢂC,
2 h, DMAP, 1.6 MPa and without using any solvent, and the results
were shown in Fig. 7. D'Elia et al. explored various catalytic com-
plexes for the cycloaddition of pure, diluted, and waste CO2 to ep-
oxides under ambient conditions [48]. In a related work, the
catalysts with bromide as a counterion were more effective than
those bearing a chloride. Thus, the quaternary phosphonium bro-
mide linked cobaloximes-PCPL have been preferred in this study. In
this study, 74% SO could be converted to SC in 6 h using a catalytic
loading of 2.5 mol %, whereas 54% SO could be converted to SC in
2 h, 100 ꢂC, DMAP, 1.6 MPa and without using any solvent using a
catalytic loading of 0.1 mol % in this study. Also, similar the phos-
phonium linker directly to the metal center as single component
homogeneous catalysts have been prepared and used for the syn-
thesis of cyclic carbonates by Melendez et al. and Martinez et al.
[49,50].
In this study, it was found that epichlorohydrin (ECH) was the
most reactive epoxide due to the fact that the electron-
withdrawing chloromethyl group of epichlorohydrin could assist
weakening of the C-O bond of the epoxide, thus favoring the
achieving the ring opening, while cyclohexene oxide (CHO)
exhibited the lowest activity. This result may be due to the more
electron donating substituents linked to C2-atom of these epoxides
(SO, PO, EB, and CHO) which could be coordinately bonded to the
Co(III) center and was responsible for the catalyst poisoning. In
addition, cyclohexene oxide (CHO), a challenging 1,2-disubstituted

A. Kilic et al. / Journal of Organometallic Chemistry 858 (2018) 78e88
85
Table 1
The results of the cycloaddition reaction of EB with CO2 catalyzed by cobaloxime (1-6) and cobaloximes-PCPL (1a-6a) catalysts.
Entry
Cat.
Yielda (%)
Selectivitya (%)
TONb
TOFc (hꢀ1
)
1
2
3
4
5
6
7
8
1
2
3
4
5
6
1a
2a
3a
4a
5a
6a
53.6
17.9
18.4
10.4
12.5
13.1
61.8
56.7
17.4
22.5
19.4
47.3
99.7
98.6
98.4
96.6
98.9
99.2
99.7
99.1
96.7
97.9
95.2
99.8
536
179
184
104
125
131
618
567
174
225
194
473
268
90
92
52
63
66
309
284
87
113
97
9
10
11
12
237
Reaction conditions: Cat. (4.5 ꢁ 10ꢀ5 mol), 1,2-Epoxybutane (4.5 ꢁ 10ꢀ2 mol), DMAP (9 ꢁ 10ꢀ5 mol), CO2 (1.6 MPa), 100 ꢂC, 2 h.
a
Yield and selectivity of epoxide to corresponding cyclic carbonate were determined by GC.
b
Moles of cyclic carbonate produced per mole of catalyst.
c
The rates expressed in terms of the turnover frequency {TOF [mol of product (mol of catalyst h)ꢀ1] ¼ turnovers/h}.
Fig. 7. Cycloaddition of CO2 to different epoxides catalyzed by cobaloxime-PCPL (1a).
epoxide, was converted to the corresponding cyclic carbonate in
only 3.8% yield because of the steric and electronic effect of the
cyclohexene rings [38,51].
important role in this coupling reaction [52]. For this purpose,
organic bases (DMAP, CH3CN, C5H5N, NEt3, and PPh3) and inorganic
bases (KOH, Na2CO3, Na2SO4 and Cs2CO3) were used as co-catalyst.
Interestingly, the use of CH3CN resulted in the yield of 57.7%, and
the yield was remarkably increased up to 97.0% when DMAP was
used as the base (Fig. 8) in high selectivity. The catalytic efficiencies
of these co-catalysts were found to be in the following order:
The catalytic results show that in the absence of catalyst or co-
catalyst no reaction progress is observed which indicates that the
presence of the catalyst as Lewis acid (electrophile) and cocatalyst
as Lewis base (nucleophile) is necessary for obtaining the highest
yield. However, we have carried out the blank reaction without the
catalyst. To ascertain the activity of DMAP alone in this reaction
under the given conditions the catalytic blank-run without
cobaloximes/cobaloximes-PCPL catalysts was done and found to be
5% conversion. Since cobaloximes-PCPL (1a) catalyst is a good Lewis
acid, various Lewis bases as cocatalyst were checked in the reaction
of 1, 2-epoxybutane (EB) with CO2 in the presence of this catalyst
(Table 1). The activities of Lewis base as co-catalyst played an
CH3CN < Cs2CO3 < Na2CO3
e
Na2SO4 < KOH < PPh3 < C5H5N < NEt3 < DMAP. In general, organic
bases were found to be more effective than inorganic bases, which
is probably due to the solubilities of bases in the epoxides. A similar
effect was observed in our previous papers [53].
Various factors including the temperature, CO2 pressure and
time were studied to get the optimal reaction conditions and
evaluate the catalytic performance. The cycloaddition reaction

86
A. Kilic et al. / Journal of Organometallic Chemistry 858 (2018) 78e88
Fig. 8. Influence of bases on ECHC yield and selectivity catalyzed by cobaloxime-PCPL (1a).
Fig. 9. Influence of temperature on ECHC yield and selectivity catalyzed by cobaloxime-PCPL (1a).
between CO2 and ECH catalyzed by cobaloxime-PCPL (1a) as a
catalyst and DMAP as a co-catalyst was investigated as a model
reaction system (Figs. 9e11). First, we investigated the effect of
temperature for the cobaloximes-PCPL (1a) catalyst. The lower
yield was detected for temperatures below 75 ꢂC, indicating that
the temperature effect must play a role in the activation of the CO2.
Increasing the temperature from 75 to 100 ꢂC resulted in a quan-
titative conversion of the epoxide and the optimum outcome was
obtained at 100 ꢂC, 1.6 MPa and 2 h. There was only a 65.8% yield of
ECHC obtained at 75 ꢂC under a CO2 pressure of 1.6 MPa within 2 h,
but the ECHC yield jumped to 97.0% at 100 ꢂC (Fig. 9). However, a
further increase in the reaction temperature caused a slight
decrease of ECHC yield as well as a slight decrease in selectivity. The
origin of the decreasing yield and selectivity may be due to
decomposition of the product to yield byproducts such as 1-
chloroethane-1, 2-diol [54].
and reaches a maximum (97%) at 1.6 MPa. However, a further in-
crease in CO2 pressure causes a slight decrease of ECHC yield as well
as a slight decrease in selectivity because the concentration of ECH
decreases along with an increased amount of dissolved CO2 at
higher pressure. This indicates that 1.6 MPa is the optimal CO2
pressure for the coupling reaction of ECH and CO2 to form ECHC
catalyzed by cobaloxime-PCPL (1a) catalyst and DMAP as a co-
catalysts. A similar trend of CO2 pressure on catalytic activity was
reported in our previous studies and by others [38,55,56].
The dependence of ECHC yield and selectivity on reaction time
was also investigated under identical reaction conditions. The
cycloaddition reaction proceeds and almost done rapidly within the
first 2 h, reaching an ECHC yield of 97.0% (Fig. 11). A further increase
in the reaction time led to a slight increase in ECHC yield. As seen in
Fig. 11, the selectivity of ECHC increased in the initial stage and
stayed steady after 2 h. In a conclusion, a reaction time of 2 h was
obtained as the optimal choice for the synthesis of ECHC.
The reaction pressure of CO2 was an important parameter to
obtain higher product conversions and the best selectivity. The
effect of CO2 pressure on the synthesis of ECHC is shown in Fig. 10.
The ECHC yield increases as pressure increases from 0.5 to 1.6 MPa
For possible mechanism of the catalytic cycle, it has been pro-
posed that the utilization of epoxides and CO2 as starting materials
to prepare the corresponding cyclic carbonate probably requires

A. Kilic et al. / Journal of Organometallic Chemistry 858 (2018) 78e88
87
Fig. 10. Influence of pressure on ECHC yield and selectivity catalyzed by cobaloxime-PCPL (1a).
Fig. 11. Influence of time on ECHC yield and selectivity catalyzed by cobaloxime-PCPL (1a).
the activation through both Co(III) as a Lewis acid (electrophile) and
a Lewis base (nucleophile) the former activates the epoxide, while
the latter attacks the less sterically hindered carbon atom to open
the epoxide ring. A ring-opening reaction is a solution to the
detailed reaction. The generated oxy anion species then reacts with
CO2 to give the corresponding cyclic carbonate and the cycle goes
on [57e59].
6) and corresponding to positively charged phosphonium linked
cobaloximes (called cobaloximes-PCPL) (1a-6a) complexes have
been prepared in high yields and characterized by means of NMR
(1H, 13C, and 31P) spectra, FT-IR spectra, UVeVis spectra, mass
spectra, melting point as well as elemental analysis. After a full
characterization, the detailed spectroscopic properties of these
novel cobaloximes/cobaloximes-PCPL compounds have also been
studied. As mentioned above, the synthesis of new cobaloximes
and cobaloximes-PCPL of a various dioxime ligands bearing chlo-
rine or (4-bromo butyl) triphenylphosphonium as axial groups is
presented to perform as efficient catalysts for chemical fixation of
4. Conclusion
We have reported a new class of the cobaloxime complexes (1-

88
A. Kilic et al. / Journal of Organometallic Chemistry 858 (2018) 78e88
M. Cokoja, F.E. Kühn, Catal. Sci. Technol. 4 (2014) 1638e1643.
[24] A. Barthel, Y. Saih, M. Gimenez, J.D.A. Pelletier, F.E. Kühn, V. D'Elia, J.-M. Basset,
Green Chem. 18 (2016) 3116e3123.
[25] M.J. Kelly, A. Barthel, C. Maheu, O. Sodpiban, F.-B. Dega, S.V.C. Vummaleti,
E. Abou-Hamad, J.D.A. Pelletier, L. Cavallo, V. D'Elia, J.-M. Basset, J. CO2 Utili-
zation 20 (2017) 243e252.
[26] S. Arayachukiat, C. Kongtes, A. Barthel, S.V.C. Vummaleti, A. Poater,
S. Wannakao, L. Cavallo, Valerio D'Elia, ACS Sustain. Chem. Eng. 5 (2017)
6392e6397.
CO2 to synthesize value-added chemicals in the absence of a sol-
vent. Among the cobaloximes/cobaloximes-PCPL, the cobaloxime-
PCPL (1a) and DMAP as a co-catalyst showed high yield and
selectivity for the coupling of CO2 and 1, 2-epoxybutane (EB) under
suitable conditions. Then, various reaction parameters such as the
epoxide, base, reaction time, temperature, and CO2 pressure were
investigated. Ongoing investigations are focused on the synthesis of
the novel various axial and equatorial active groups containing
different cobaloximes-PCPL catalysts for the synthesis of cyclic
carbonates from CO2 under ambient conditions and will be re-
ported in due course.
[27] H. Yasuda, L.N. He, T. Sakakura, C. Hu, J. Catal. 233 (2005) 119e122.
[28] M.B. Gawande, R.K. Pandey, R.V. Jayaram, Catal. Sci. Technol
2 (2012)
1113e1125.
[29] K. Yamaguchi, K. Ebitani, T. Yoshida, H. Yoshida, K. Kaneda, J. Am. Chem. Soc.
121 (1999) 4526e4527.
[30] C.J. Whiteoak, A. Nova, F. Maseras, A.W. Kleij, ChemSusChem
5 (2012)
2032e2038.
Acknowledgement
ꢀ
[31] C.J. Whiteoak, A.H. Henseler, C. Ayats, A.W. Kleij, M.A. Pericas, Green Chem. 16
(2014) 1552e1559.
[32] M.E. Wilhelm, M.H. Anthofer, M. Cokoja, I.I.E. Markovits, W.A. Herrmann,
F.E. Kuhn, ChemSusChem 7 (2014) 1357e1360.
[33] J.Q. Wang, W.G. Cheng, J. Sun, T.Y. Shi, X.P. Zhang, S.J. Zhang, RSC Adv. 4 (2014)
2360e2367.
This work was supported by the Research Fund of Harran Uni-
versity (HUBAK Projects No: 16068) Sanliurfa, Turkey.
[34] H. Büttner, J. Steinbauer, T. Werner, ChemSusChem 8 (2015) 2655e2669.
[35] J. Sun, L.J. Han, W.G. Cheng, J.Q. Wang, X.P. Zhang, S.J. Zhang, ChemSusChem 4
(2011) 502e507.
Appendix A. Supplementary data
[36] T. Chang, X.R. Gao, L. Bian, X.Y. Fu, M.X. Yuan, H.W. Jing, Chin. J. Catal. 3 (2015)
408e413.
[37] X.F. Liu, Q.W. Song, S. Zhang, L.N. He, Catal. Today 263 (2016) 69e74.
[38] (a) A. Kilic, M.V. Kilic, M. Ulusoy, M. Durgun, E. Aytar, M. Dagdevren, I. Yilmaz,
J. Organomet. Chem. 767 (2014) 150e159;
Supplementary data related to this article can be found at
https://doi.org/10.1016/j.jorganchem.2018.01.029.
References
(b) A. Kilic, A.A. Palali, M. Durgun, Z. Tasci, M. Ulusoy, M. Dagdevren, I. Yilmaz,
Inorg. Chim. Acta. 394 (2013) 635e644;
(c) A. Kilic, M. Ulusoy, E. Aytar, M. Durgun, J. Ind. Eng. Chem. 24 (2015)
98e106;
(d) A. Kilic, M. Ulusoy, M. Durgun, E. Aytar, Inorg. Chim. Acta. 411 (2014)
17e25.
[1] P. Gao, Z. Zhao, L. Chen, D. Yuan, Y. Yao, Organometallics 35 (11) (2016)
1707e1712.
[2] Y.-Hu Han, Z.-Yuan Zhou, C.-Bin Tian, and S.-Wu Du, Green Chem. https://doi.
org/10.1039/C6GC00413J.
[3] M.Y. Wani, S. Kumar, C.T. Arranja, Carlos M.F. Dias, Abilio J.F.N. Sobral, New J.
Chem. 40 (2016) 4974e4980.
[4] L. Liu, S.-Ming Wang, Z.-Bo Han, M. Ding, D.-Qiang Yuan, H.-Long Jiang, Inorg.
Chem. 55 (2016) 3558e3565.
[39] W.L. Dai, B. Jin, S.L. Luo, X.B. Luo, X.M. Tu, C.T. Au, Catal. Today 233 (2014)
92e99.
[40] R. Yamuna, S. Sarath, K. Kubandiran, M. Umadevi, G. Chakkaravarthi,
J. Organomet. Chem. 811 (2016) 40e47.
[41] A. Kilic, M. Ulusoy, M. Durgun, E. Aytar, A. Keles, M. Dagdevren, I. Yilmaz,
J. Coord. Chem. 67 (16) (2014) 2661e2679.
[42] S. Mirra, M. Strianese, C. Pellecchia, V. Bertolasi, G. Monaco, S. Milione, Inorg.
Chim. Acta. 444 (2016) 202e208.
[43] A. Panagiotopoulos, K. Ladomenou, D. Sun, V. Artero, A.G. Coutsolelos, Dalton
Trans. 45 (2016) 6732e6738.
[44] M. Strianese, S. Mirra, V. Bertolasi, S. Milione, C. Pellecchia, New J. Chem. 39
(2015) 4093e4099.
[45] (a) Z.L. Chu, W. Huang, J. Mol. Struct. 837 (2007) 15e22;
(b) W.K. Dong, L.C. Zhu, Y.J. Dong, J.-C. Ma, Y. Zhang, Polyhedron 117 (2016)
148e154.
[46] A. Kilic, S. Akdag, E. Aytar, M. Durgun, M. Ulusoy, New J. Chem. 40 (2016)
7901e7910.
[5] (a) B. Mousavi, S. Chaemchuen, B. Moosavi, Z. Luo, N. Gholampour and F.
Verpoort, New J. Chem. https://doi.org/10.1039/c6nj00128a. (b) Y. Zhou, S. Hu,
X. Ma, S. Liang, T. Jiang, B. Han, J. Mol. Catal. A Chem. 284 (2008) 52;
(c) K. Biggadike, R.M. Angell, C.M. Burgess, R.M. Farrell, A.P. Hancock,
A.J. Harker, A.J. Irving, W.R. Irving, C. Ioannou, P.A. Procopiou, R.E. Shaw,
Y.E. Solanke, O.M.P. Singh, M.A. Snowden, R. Stubbs, S. Walton, H.E. Weston,
J. Med. Chem. 43 (2000) 19;
(d) A.A.G. Shaikh, S. Sivaram, Chem. Rev. 96 (1996) 951.
[6] T. Sakakura, K. Kohno, Chem. Commun. (2009) 1312e1330.
[7] J.L. Song, Z.F. Zhang, S.Q. Hu, T.B. Wu, T. Jiang, B.X. Han, Green Chem. 11 (2009)
1031e1036.
[8] W.Y. Gao, Y. Chen, Y.H. Niu, K. Williams, L. Cash, P.J. Perez, L. Wojtas, J.F. Cai,
Y.S. Chen, S.Q. Ma, Angew. Chem. Int. Ed. 53 (2014) 2615e2619.
[9] M. North, R. Pasquale, Angew. Chem. Int. Ed. 48 (2009) 2946e2948.
[10] C.X. Miao, J.Q. Wang, Y. Wu, Y. Du, L.N. He, ChemSusChem 1 (2008) 236e241.
[11] R.C. Luo, X.T. Zhou, S.Y. Chen, Y. Li, L. Zhou, H.B. Ji, Green Chem. 16 (2014)
1496e1506.
[12] Y.W. Ren, J.G. Chen, C.R. Qi, H.F. Jiang, ChemCatChem 7 (2015) 1535e1538.
[13] W.M. Ren, Y. Liu, X.B. Lu, J. Org. Chem. 79 (2014) 9771e9777.
[14] S.H. Duan, X.Y. Jing, D.D. Li, H.W. Jing, J. Mol. Catal. A Chem. 411 (2016) 34e39.
[15] M. North, P. Villuendas, ChemCatChem 4 (2012) 789e794.
[16] X.B. Lu, D.J. Darensbourg, Chem. Soc. Rev. 41 (2012) 1462e1484.
[17] X.B. Lu, Y.J. Zhang, B. Liang, X. Li, H. Wang, J. Mol. Catal. A Chem. 210 (2004)
31e34.
[47] K.M.K. Yu, I. Curcic, J. Gabriel, H. Morganstewart, S.C. Tsang, J. Phys. Chem. A
114 (2010) 3863e3872.
[48] R.R. Shaikh, S. Pornpraprom, V. D'Elia, ACS Catal. 8 (2018) 419e450.
[49] J. Melendez, M. North, P. Villuendas, Chem. Commun. (2009) 2577e2579.
[50] J. Martinez, J.A. Castro-Osma, C. Alonso-Moreno, A. Rodriguez-Dieguez,
M. North, A. Otero, A. Lara-Sanchez, ChemSusChem 10 (2017) 1175e1185.
[51] D.J. Darensbourg, J.C. Yarbrough, J. Am. Chem. Soc. 124 (2002) 6335e6342.
[52] D.S. Bai, H.W. Jing, Q. Liu, Q.Q. Zhu, X.F. Zhao, Catal. Commun. 11 (2009)
155e157.
[53] A. Kilic, H. Fırat, E. Aytar, M. Durgun, A.K. Baytak, M. Aslanoglu, M. Ulusoy,
Chem. Pap. 71 (2017) 1705e1720.
[54] L. Xiao, Q. Yue, C. Xia, L. Xu, J. Mol. Catal. A Chemical 279 (2008) 230e234.
[55] Y. Xie, Z. Zhang, T. Jiang, J. He, B. Han, T. Wu, K. Ding, Angew. Chem. Int. Ed. 46
(2007) 7255.
[56] D.J. Darensbourg, R.M. Mackiewicz, D.R. Billodeaux, Organometallics 24
(2005) 144.
[57] Z. Bu, G. Qin, S. Cao, J. Mol. Catal. 277 (2007) 35e39.
[58] Y.M. Shen, W.L. Duan, M. Shi, J. Org. Chem. 68 (2003) 1559e1562.
[59] Y. Chen, R. Luo, Q. Xu, J. Jiang, X. Zhou, H. Ji, ChemSusChem 10 (2017)
2534e2541.
[18] X. Zhang, Y.B. Jia, X.B. Lu, B. Li, H. Wang, L.C. Sun, Tetrahedron Lett. 49 (2008)
6589e6592.
[19] E.K. Noh, S.J. Na, S.W. Kim, B.Y. Lee, J. Am. Chem. Soc. 129 (2007) 8082e8083.
[20] T. Ema, Y. Miyazaki, J. Shimonishi, C. Maeda, J. Hasegawa, J. Am. Chem. Soc.
136 (2014) 15270e15279.
[21] C. Maeda, T. Taniguchi, K. Ogawa, T. Ema, Angew. Chem. Int. Ed. 54 (2015)
134e138.
[22] T. Ema, Y. Miyazaki, S. Koyama, Y. Yano, T. Sakai, Chem. Commun. 48 (2012)
4489e4491.
[23] M.E. Wilhelm, M.H. Anthofer, R.M. Reich, V.D. Elia, J.M. Basset, W.A. Herrmann,