Synthesis and characterization of a new poly α-amino acid Co(II)-complex supported on magnetite graphene oxide as an efficient heterogeneous magnetically recyclable catalyst for efficient free-coreductant gram-scale epoxidation of olefins with molecular oxygen

Article abstract of DOI:10.1016/j.jorganchem.2019.05.030

A novel magnetic nanocomposite was prepared by immobilization of a cobalt complex of a synthetic poly α-amino acid on Fe₃O₄-doped graphene oxide (GO/Fe₃O₄@PAA Co(II)) and was demonstrated to be a highly efficient catalyst for the epoxidation of olefins in mild conditions. PAA was synthesized through a multi-step synthesis, first by a poly condensation reaction of salicylaldehyde followed by the Strecker synthesis. The synthesized nanocomposite was characterized by various analytical and spectroscopic methods including FTIR, ICP, XRD, EDX, XPS, FE-SEM, TEM, TGA, VSM and DLS analyses. A wide variety of olefins could be tolerated toward epoxidation in the presence of molecular oxygen without the need for any co-reductant. The magnetic nanocomposite could be readily separated by a magnet from the mixture and reused for several times without any significant reactivity loss, which represents its potential for practical and industrial application. Also, the scalability of the process was investigated in this work.

Full text of DOI:10.1016/j.jorganchem.2019.05.030

Accepted Manuscript
Synthesis and characterization of a new poly α-amino acid Co(II)-complex supported
on magnetite graphene oxide as an efficient heterogeneous magnetically recyclable
catalyst for efficient free-coreductant gram-scale epoxidation of olefins with molecular
oxygen
Milad Kazemnejadi, Boshra Mahmoudi, Zeinab Sharafi, Mohammad Ali Nasseri, Ali
Allahresani, Mohsen Esmaeilpour
PII:
S0022-328X(19)30229-3
DOI:
https://doi.org/10.1016/j.jorganchem.2019.05.030
JOM 20812
Reference:
To appear in: Journal of Organometallic Chemistry
Received Date: 18 April 2019
Revised Date: 28 May 2019
Accepted Date: 31 May 2019
Please cite this article as: M. Kazemnejadi, B. Mahmoudi, Z. Sharafi, M.A. Nasseri, A. Allahresani, M.
Esmaeilpour, Synthesis and characterization of a new poly α-amino acid Co(II)-complex supported on
magnetite graphene oxide as an efficient heterogeneous magnetically recyclable catalyst for efficient
free-coreductant gram-scale epoxidation of olefins with molecular oxygen, Journal of Organometallic
Chemistry (2019), doi: https://doi.org/10.1016/j.jorganchem.2019.05.030.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Synthesis and characterization of a new poly α-amino acid Co(II)-complex
supported on magnetite graphene oxide as an efficient heterogeneous
magnetically recyclable catalyst for efficient free-coreductant gram-scale
epoxidation of olefins with molecular oxygen
Milad Kazemnejadi*a, Boshra Mahmoudi,b Zeinab Sharafi,c Mohammad Ali Nasseri,a Ali
Allahresani,a Mohsen Esmaeilpourd
a Department of Chemistry, Faculty of Sciences, University of Birjand, Birjand 97175-615,
Iran
b Research Center, Sulaimani Polytechnic University, Sulaimani 46001, Kurdistan Region,
Iraq
c Razi Herbal Medicines Research Center, Lorestan University of Medical Sciences,
Khorramabad, Iran
d Department of Chemistry, College of Science, Shiraz University, Shiraz 7194684795, Iran
*E-mail: miladkazemnejadi@birjand.ac.ir (M. Kazemnejadi)
Abstract
A novel magnetic nanocomposite was prepared by immobilization of a cobalt complex of a
synthetic poly α-amino acid on Fe3O4-doped graphene oxide (GO/Fe3O4@PAA Co(II)) and
was demonstrated to be a highly efficient catalyst for the epoxidation of olefins in mild
conditions. PAA was synthesized through a multi-step synthesis, first by a poly condensation
reaction of salicylaldehyde followed by the Strecker synthesis. The synthesized
nanocomposite was characterized by various analytical and spectroscopic methods including
FTIR, ICP, XRD, EDX, XPS, FE-SEM, TEM, TGA, VSM and DLS analyses. A wide variety
of olefins could be tolerated toward epoxidation in the presence of molecular oxygen without
the need for any co-reductant. The magnetic nanocomposite could be readily separated by a
magnet from the mixture and reused for several times without any significant reactivity loss,
which represents its potential for practical and industrial application. Also, the scalability of
the process was investigated in this work.
Keywords: Graphene oxide, Epoxidation, Fe3O4, Nanocatalyst, Magnetically recyclable,
Gram-scale
1. Introduction
Epoxides are one of the key and versatile intermediates for manufacture of fine chemicals
including plasticizer, perfumes, cosmetics, epoxy resins, pharmaceuticals, etc. [1].
Traditionally, epoxides are synthesized by stoichiometric amount of oxidizing reagent such as
potassium monoperoxysulfate [2] and organic peracids [3]. These peroxides are not only
toxic, corrosive and harmful, but also generate toxic waste products, which limited their
application in industrial aspects as well as in laboratory due to the safety issues. So, the
researches were conducted to use more convenient, easily handling and environmentally
friendly oxidants such as molecular oxygen, TBHP, air and hydrogen peroxide [3,4]. In order
to exploit the advantage of these oxidants for epoxidation, as the easily accessible and green
oxidants, there is a need for a transition metal in order to oxygen transfer [5]. Various
catalytic systems based on transition metals have been developed recently such as: (1)
Fe2/mpg-C3N4 (mesoporous carbon nitride) [6], (2) mesoporous cobalt oxide [7], (3) zeolites
[8], (4) nickel complexes [9], (5) Mn(II) and (III) complexes [10], (6) Mn-porphyrin [11], (7)

ACCEPTED MANUSCRIPT
Cu2(oxalate)(1,10-phen)2Cl2 [12], (8) [Mo(η3-C3H5)Br(CO)2{iPrN=C(H)C5H4N}] [13], and
(9) Rh2(tpa)4 (tpa = triphenylacetate) [14].
Despite various progress in this topic, however, most of the reported methods are involve
expensive reagents, unsatisfactory yield, low selectivity, non-environmentally benign, using
co-reductant, harsh reaction conditions, etc. It worth noted that two excess possible products
could be obtained in result of the epoxidation of an olefins bearing Hα (Scheme 1) [15],
which is not suitable for practical aspects. So, finding a highly efficient catalytic system with
high selectivity and toxicity, which involves green chemistry considerations seems essential.
Application of nanomaterials as a solid support due to their high ratio of surface to volume
are well known for decades [16,17]. Among them, graphene oxide (GO) with the interesting
properties such as sheet structure, nano-size dimension with high surface area, high thermal/
chemical stability and possibility of functionalization with various organic compounds
become as one of the most suitable solid supports for organic synthesis [3,4,18]. Further
improvement in heterogeneous catalysts led to the preparation of magnetic nanoparticles that
facilitates the separation of catalyst with an external magnetic field [19]. Fe3O4 nanoparticles
are one of the most applicable and popular magnetic nanoparticles due to their widespread
applications in various fields of area such as catalyst and medicine [5,20]. Some advantages
are associated with them including low cost, high thermal stability, non-toxic, chemical
stability, strong magnetization, and high crystalline structure, which conducted them for
magnetization of other solid supports; this is a smart strategy for magnetization of solid
supports (in this case nano graphene oxide), which has been used extensively in the last
decades [3,19].
Scheme 1. Epoxidation of cyclohexene catalyzed by the transition metal catalytic systems
Cobalt complexes and generally Co-based catalytic systems have received attention due to
their well-known potential for oxidation programs, being cheap and accessible, and non-toxic
nature [21]. Herein, in accordance with green chemistry considerations and also importance
and application of useful epoxides in organic synthesis, we plan a multiphase nanocomposite
with decoration of a Co(II) complex of a synthetic poly α-amino acid onto magnetic
graphene oxide as an highly active heterogeneous magnetically recoverable catalyst for the
efficient free-coreductant gram-scale O2-epoxidation of olefins in mild conditions. This
system not only benefits from advantages of a solid nano-support with high aspect ratio
(graphene oxide), but also utilizes from the presence of PAA as a strong ligand on its surface
that is responsible for high durability and activity of the catalyst.
2. Experimental
2.1. Materials and apparatus

ACCEPTED MANUSCRIPT
All chemicals were obtained from Sigma and Fluca supplier and used as received without
further purification. All the solvents were distilled and dried before use. Progress of the
reactions and purity of products were monitored by thin layer chromatography (TLC) on
silica gel and/or gas chromatography (GC) on a Shimadzu-14B gas chromatography equipped
with HP-1 capillary column and N2 as carrier gas. For GC analyses, anisole was used as an
internal standard. Purification of imines and oximes was achieved by recrystallization from
ethanol. FTIR spectra were obtained using a JASCO FT/IR 4600 spectrophotometer using
1
13
KBr pellet. The H NMR (250 MHz) and CNMR (62.9 MHz) spectra were recorded on a
Bruker Avance DPX-250 spectrometer in CDCl3 and DMSO-d6 as solvent. Transmission
electron microscopy (TEM) images were taken on a Philips EM208 microscope and was
operated at 100 kV. The presence of the elements was detected by EDX spectroscopy using
field emission scanning electron microscope (FE-SEM, JEOL 7600F), equipped with a
spectrometer of energy dispersion of X-ray from Oxford instruments. Size distribution of the
nanoparticles were measured by dynamic light scattering (DLS) analysis on a HORIBA-
LB550 instrument. TGA of the samples have been performed on a NETZSCH STA 409
PC/PG in nitrogen atmosphere with a heating rate of 10 °C /min in the temperature ranges of
25-1000 °C. The magnetic behavior of the samples was conducted on Lake Shore vibrating
sample magnetometer (VSM) at room temperature. ICP experiments were accomplished
using VARIAN VISTA-PRO CCD simultaneous ICP-OES instrument. Conversion and
epoxide selectivity was measured according to the following equations 1,2 [1,5]:
(initial mol%)-(final mol%)
conversion (mol%)=
×100 (1)
initial mol%
GC peak area of styrene epoxide
Styrene epoxide selectivity=
×100 (2)
GC peak area of all products
2.2. Preparation of PAA
In order to preparation of poly(α-amino acid) (PAA), firstly, poly salicylaldehyde (PSA) was
prepared according to a previously reported procedure [1,5,20]. PSA (0.1 g, M n =2226) was
added to 20 mL of water, then NaCN (0.5 g, 10 mmol) and NH4Cl (0.5 g, 10 mmol) were
added to the reaction mixture. The reaction was refluxed at 50 °C for 12 h. The resultant
orange product (mp > 400 °C, M n = 2400), poly(α-amino nitrile) 3, was isolated by simple
filtration followed by drying into oven (50 °C) for 5 h. Poly(α-amino nitrile) was transformed
to poly(α-amino acid) by hydrolysis of nitrile groups; 0.2 g of 3 was added to 15 mL of water
along with several drops of H2SO4, and the mixture was refluxed at 80 °C for 12 h. The
mixture reaction was cooled to room temperature and neutralized with 0.5 N KOH solution.
Poly(α-amino acid) 4, was filter off, drying and isolated as a stable powder (mp > 400 °C).
Scheme 2 shows the preparation of PAA 4 via a multistep synthesis.
2.3. Preparation of GO
Graphene oxide (GO) was prepared according to a modified Hummer’s method from purified
natural graphite [22,23].
2.4. Preparation of GO/Fe3O4 hybrids
GO–Fe3O4 hybrid was prepared via a chemical co-precipitation method according to the
slightly modified literature procedure [24,25]. GO/Fe3O4 NPs were prepared with a mass ratio
of 1: 5 (GO: FeSO4). This premium ratio provides a suitable responsive to a magnetic field
for the readily separation of the NPs [4]. In a typical process, GO (25 mg) was first sonicated

ACCEPTED MANUSCRIPT
in 150 mL deionized water for 10 min. The mixture was stirred and heated to 80 °C in an oil
bath. Then, NH4OH 25% solution was added dropwise to the mentioned solution until the pH
adjusted to 11. Then, FeSO4.7H2O (125 mg) along with sodium dodecyl sulfate (SDS) were
added into the aqueous dispersion of GO solution. The resulting mixture was stirred under N2
atmosphere for 24 h at 80 °C. The GO/Fe3O4 nanoparticles were collected by an external
magnetic field followed by washing with deionized water several times in order to eliminate
the excess iron salts. Finally, the product was separated by a simple external magnet, then
dried in vacuum (40 °C) for 4 h (Scheme 2).
2.5. Preparation of GO/Fe3O4@PAA-Co(II) complex
For the preparation of GO/Fe3O4@PAA complex, 20 mg of GO/Fe3O4 in 50 mL DMF was
ultra-sonicated for 1 h, and then 0.2 g of N,N'-dicyclohexylcarbodiimide (1.0 mmol, DCC)
and 4-dimethylaminopyridine (DMAP) along with 0.5 g of PAA were added into the
solution. The resultant mixture was stirred for 8 h at 80 °C under N2 atmosphere. The
product, GO/Fe3O4@PAA 7, was separated using an external magnetic field and washed with
water for several times until the pH reached to 7, and then dried into oven (60 °C).
Complexation of cobalt to GO/Fe3O4@PAA was takes place by sonication of
GO/Fe3O4@PAA (1.0 g) in 20 mL EtOH followed by addition of Co(OAc)2 (10 mg) to the
solution. The mixture was stirred at room temperature for 2 h. The product 8 was purified as
same as the previous step using an external magnetic field.
2.6. General procedure for catalytic epoxidation of olefins by GO/Fe3O4@PAA-Co(II)
In a typical run, in a three-necked round-bottom flask equipped with a O2 (g) inlet (bubbling
15 mL/min, 1.0 atm.), a condenser and a magnet stirrer bar, a mixture of styrene (10.0 mmol)
and catalyst 8 (50 mg, 0.2 mol%) were stirred in CH3CN (10 mL) at 60 °C. The reaction
progress was monitored by TLC. Upon reaction completion, the catalyst was magnetically
filtered, and the residue was directly subjected to GC instrument to quantify the epoxide
product. Moreover, the desired epoxide product was purified by column chromatography
1
followed by identification by H NMR and 13C NMR spectra. Activity of the catalyst was
expressed by turnover frequency (TOF) as well as turnover number (TON) [1,5].
Scheme 2. Synthesis of poly(AA)
3.
Results and discussion
3.1. Catalyst characterization
According to GPC analysis of the prepared poly(α-amino nitrile), Mw = 2400, it could be
calculated that the present nitrile functions per mole of poly(α-amino nitrile) is 17/2400 = 7

ACCEPTED MANUSCRIPT
mmol. So, assuming that all of the nitrile groups are converted to acid; the poly(α-amino
acid) contain 7.0 mmol acid functions per gram of PAA.
FTIR spectra of 5-chloromethyl salicylaldehyde, 2, 3, 4, GO (5), 6, 7, and 8 are shown in Fig.
1. The characteristic peaks at 725 cm-1, 1481 cm-1, and some series vibrations at 2900-2950
cm-1 are related to C-Cl (Str.), C-H (Aliphatic, Bend.), and C-H (Aliphatic, Str.) vibrations,
respectively, which completely confirmed the chloromethylation of salicylaldehyde (Fig. 1a)
[1,5,20]. The polymerization of 5-chloromethyl salicylaldehyde causes to a large reduction of
O-H stretching vibration intensity at 3400 cm-1, confirming that the polymerization takes
place through the ether bond formation (Fig. 1b) [20]. The preparation of α-amino nitrile via
the Strecker synthesis was characterized by a strong peak at 2110 cm-1 related to CN (nitrile)
groups (Fig. 1c). Moreover, two characteristic peaks at 3500 and 3550 cm-1 demonstrating the
presence of primary amine groups (Stretching vibrations). Elimination of the sharp peak refer
to nitrile at the spectrum of 4, confirmed clearly their transformation to carboxylic acids (Fig.
1d). This reduction along with a red shift of the amine groups in FTIR spectrum of poly(α-
+
amino acid), could be related to the formation of COO- NH3 , that is characteristic for a
typical amino acid [26]. Fig. 1e, shows the FTIR spectrum of graphene oxide. A peak at
1735 cm-1 was assigned to the stretching vibration of C=O for carboxylic groups (Fig. 1e).
The intense bond at 3422 cm-1 is attributed to the stretching vibration of hydroxyl groups of
GO surface. The peak at 1626 cm-1 represents C=C stretching vibrations for aromatic rings
related to the non-oxidized graphene oxide framework [27]. Advent of a strong absorption at
572 cm-1 at GO/Fe3O4 FTIR spectrum indicates the Fe-O (Str.) nanoparticles, confirming the
incorporation of Fe3O4 NPs on the GO. Another characteristic peak was the reduction of O-H
bond intensity due to the decoration of Fe3O4 on the graphene oxide (Fig. 1f). Fig. 1g shows
the amide bond linkage in 7 at 1622 cm-1. A shoulder at 1635 cm-1 could be attributed to free
C=O carboxylic bonds. Also, vibrations about 1400 cm-1 represents the bending vibration of
methylene groups in the polymer (Fig. 1g). Reduction in the intensity of peaks at carboxylic
region, confirmed the successful coordination of cobalt cations through these functional
groups as indicated in Scheme 3 (Fig. 1h). Furthermore, advent of a series of packs at ~479
cm-1 shows the Co-O bond stretching vibration [28], confirming the successful coordination
of Co (Scheme 3).
Scheme 3. Synthesis of GO/Fe3O4@PAA Co(II) complex
The magnetic property of the samples was studied by VSM analysis. The samples give
suitable and satisfactory response to the applied magnetic field (Fig. 2A). No hysteresis
phenomenon was found for the samples, which is characteristic for the superparamagnetic
nanoparticles [29]. These results confirmed the incorporation of Fe3O4 NPs on GO. For

ACCEPTED MANUSCRIPT
comparison, Fe3O4 NPs were separately prepared and studied by VSM. The analyses show 78
emu.g-1, 65 emu.g-1, and 32 emu.g-1 saturation magnetization for Fe3O4, GO/Fe3O4, and
GO/Fe3O4@PAA-Co(II) respectively (Fig. 2A, a-c). The results indicate a large reduction in
magnetization for the catalyst 8, which represent the functionalization of GO in each step
(Fig. 2A-c).
Fig. 1 FTIR spectra of (a) 5-chloromethyl salicylaldehyde, (b) poly salicylaldehyde (2), (c) poly(α-
amino nitrile) (3), (d) poly(α-amino acid) (4), (e) GO (5), (f) Fe3O4/GO (6), (g) GO/Fe3O4@PAA (7),
(h) GO/Fe3O4@PAA-Co(II) (8)
The samples were further characterized by TGA analysis. TGA spectrum of GO/Fe3O4
indicates three main weight losses (Fig. 2B-a). In the first stage, a slow weight loss takes
place near 80 °C correspond to 9% weight loss due to desorption of water [18]. Another
weigh loss occurs around 220 ° with a mild slope, which last to 440 °C. This 18% weight loss
exhibited the vaporization and decomposition of various functional groups (Fig. 1a) on GO
framework (Fig. 2B-a). Finally, decomposition of –COO- groups provide 12% weight loss
above 600 °C in agreement with the literature [3,18]. An enhancement in thermal stability of
GO/Fe3O4@PAA was observed, demonstrating the influence of the coated PAA on
GO/Fe3O4 hybrid. Again, the small weigh loss is due to the release of water on the
GO/Fe3O4@PAA at 180 °C (10% weight loss). The main weight loss in temperature range of
550-820 °C represents the decomposition of PAA on the GO/Fe3O4 (Fig. 2B-b). This weigh
loss is equal to 40% and confirming the loading amount of PAA on the NPs. Coordination of
cobalt to GO/Fe3O4@PAA-Co(II) showed an enhance in its thermal stability, which the
overall weight loss was found to be 52% at 1000 °C (compared to GO/Fe3O4@PAA-Co(II)

ACCEPTED MANUSCRIPT
with 57% weight loss). This difference may be attributed to the coordinated cobalt in the
GO/Fe3O4@PAA-Co(II) (Fig. 2B-c). Also, the peak refer to the decomposition of polymer
takes place in lower temperatures with a milder slope, that could be attributed to the effect of
coordinated cobalt to PAA.
XRD patterns of GO, GO/Fe3O4 and GO/Fe3O4@PAA-Co are shown in Fig. 3A. GO presents
an amorphous peak at 2θ=10.8° related to (001) plane, that is characteristic for the graphene
oxide [4]. Fig. 3A-a, shows the XRD pattern of GO/Fe3O4 with six characteristic peaks at 2θ=
30.3°, 35.5°, 43.2°, 53.8°, 57.5°, 62.9°, which were assigned to their indices (220), (311),
(440), (442), (511), and (440) respectively [3,4]. These reflections were in agreement with the
crystal structure of Fe3O4 NPs (JCPDS no. 19-0629). As shown in Fig. 3A-b, the intensity of
the peak at 10.8° is largely reduced. This behavior suggests that the Fe2+ ions act as a
reducing agent for graphene oxide [3,19]. The crystal structure of the catalyst 8, was also
demonstrated the peaks for Fe3O4 crystal structure as well as a new amorphous peak at 2θ=
11.5°.This peak could be assigned to the amorphous PAA in comparison with the GO/Fe3O4
XRD pattern, that is coated on the GO framework. Furthermore, the results also proved that
the functionalization of GO/Fe3O4 with PAA doesn’t lead to the phase change of Fe3O4 NPs.
Fig. 2. (A) VSM curves of (a) Fe3O4, (b) GO/Fe3O4, and (c) GO/Fe3O4@PAA-Co(II). (B) TGA
spectra of (a) GO/Fe3O4, (b) GO/Fe3O4@PAA and (c) GO/Fe3O4@PAA-Co(II).

ACCEPTED MANUSCRIPT
Fig. 3. (A) XRD patterns of (a) GO, (b) GO/Fe3O4, (c) GO/Fe3O4@PAA-Co. (B) EDX spectrum of
GO/Fe3O4@PAA-Co (II). (C) High resolution normalized XPS Co2p spectrum of GO/Fe3O4@PAA-
Co (II)
The presence of elements in the catalyst 8 was detected by EDX analysis. As shown in Fig.
3B, the elements C, O, N, Co, and Fe were detected, which confirming the structure of 8. The
high resolution XPS Co2p spectrum of GO/Fe3O4@PAA-Co(II) revealed the oxidation state
of Co (II) ions in the catalyst. As shown in Fig. 3C, the Co2p1/2 and Co2p3/2 lines were
appeared at 797.2 eV and 781 eV respectively, which confirms that Co(II) is the only
component in GO/Fe3O4@PAA-Co(II) and all Co ions are in +2 oxidation state.
The amount of cobalt in GO/Fe3O4@PAA-Co(II) was also determined by ICP analysis which
is 0.40 mmol.g-1. This amount was completely in agreement and consistence with the results
from TGA analyses (Fig. 2B-c).
Fig. 4. (a), (b) FE-SEM images, (c) TEM image, and (d) DLS analysis of GO/Fe3O4@PAA-Co(II)
The morphology of the NPs was studied by FE-SEM technique. As shown in Fig. 4a,b, the
images clearly demonstrated the chemical deposition of Fe3O4 NPs on GO framework. The
results from TEM analysis, also, confirmed the preparation of GO/Fe3O4@PAA-Co(II)
hybrid with an average size of 28 nm for the particles (Fig. 4c). The results obtained from
DLS analysis (Fig. 4d), revealed that the main size distribution of the NPs are in the range of
25-30 nm. The mean size of the NPs was found to be 30 nm with a deviation of 8.2, in
agreement with the corresponding TEM image.
3.2. Optimization of reaction parameters

ACCEPTED MANUSCRIPT
To explore the catalytic activity of GO/Fe3O4@PAA-Co(II), the epoxidation reaction of
olefins was conducted in the presence of GO/Fe3O4@PAA-Co(II). First, for optimization of
the reaction conditions, effective parameters involve in the reaction such as temperature,
solvent, oxygen flow rate, and catalyst amount were studied over the epoxidation of styrene
as a model reaction. As shown from the spectra in Fig. 5, all mentioned parameter played a
crucial role in the epoxidation reaction. Acetonitrile and DMF afforded the most efficiencies
among the used solvents with 93% conversion (Fig. 5a). It seems that high dielectric
constants associated with them (ε=36.46, 38.25 for CH3CN and DMF respectively) are
responsible for this excellent efficiencies. This property is in agreement with the proposed
radical mechanism in the next section involves polar substrates and intermediates. Low
conversions for epoxy styrene were obtained in DEE, EtOH, and MeOH. It worth noted that
the selectivity is strongly depend on the solvent. The highest selectivity was obtained for
EtOH (Fig. 5a, 98%). Then, the model reaction was performed in various amount of the
catalyst (Fig. 5b). The most possible efficiency was obtained when the reaction was carried
out in the presence of 0.2 mol% of GO/Fe3O4@PAA-Co(II). Didn’t found any significant
influence on selectivity at other catalyst amounts. Oxygen flow rate of 15 mL/min and 60 °C
(CH3CN) give epoxy styrene with high efficiency (Fig. 5c). The selectivity linearly dropped
in higher O2 concentrations or temperatures. It may be due to promotion of side oxidation
products in these conditions.
Catalytic performance of GO/Fe3O4@PAA-Co(II) was investigated toward a variety of
olefins including aliphatic, conjugated, cyclic, and in other hand, terminal or internal types
under obtained premium conditions. The results including time, conversion, epoxide
selectivity and TOF were tabulated in Table 1. The method demonstrated high efficiency in
terms of TOF and epoxide selectivity for all substrates (Table 1, 10a-10m). Surprisingly,
epoxidation of substrates with two alkene functionality takes place selectively on one of
alkene functional groups (Table 1, 10j and 10m). Epoxidation of 3-methyl-1-(oxiran-2-
yl)but-3-en-2-one was selectively take placed on the terminal alkene functional group (Table
1, 10l), which represent chemoselectivity of the present method toward terminal alkenes than
the corresponding conjugate types. Also, the preparation of 10i and 10k with high conversion
demonstrated that the method is suitable for epoxidation of sterically hindrance olefins (Table
1, 10i and 10k). It was observed that internal alkenes afforded higher selectivity than other
terminal alkenes (Table 1, 10b, 10d, 10e).
3.3. Control experiments
Some control experiments were conducted to show the exclusive catalytic activity of
GO/Fe3O4@PAA-Co(II). For this goal, PAA-Co(II) complex, graphene oxide, Fe3O4, and
Co(OAc)2 were used as catalyst in epoxidation reaction of styrene (10.0 mmol) in CH3CN in
the presence of molecular O2 (15 mL/min) at 60 °C for 3 h. All the control experiments were
conducted under the same conditions. Table 2 shows the corresponding results. GO and
Fe3O4 afforded inconsiderable efficiency for styrene epoxidation. Also, the selectivity was
completely loss for them (Table 2).

ACCEPTED MANUSCRIPT
Fig. 5. Screening (a) solvent, (b) catalyst amount, (c) O2 bubbling flow rate, and (d) temperature on
the epoxidation of styrene catalyzed by GO/Fe3O4@PAA-Co(II) for 3 h. Green curves demonstrate
epoxide selectivity for each study. Conditions: styrene (10.0 mmol), solvent (10 mL).
Table 1 Epoxidation of alkenes by GO/Fe3O4@PAA-Co(II)a
O
R'
R'
Catalyst (0.2 mol%)
CH3CN, O2 (1 atm.)
R' = Alkyl, Aryl
R = Alkyl, H
R
R
9
10
Epoxide
selectivity
(%)b
Time
(h)
Conversion Yield
TOF
(h-1)d
Entry
1
Product 10
(%)b
(%)c
3
93
88
97
155
10a
2
3
3.5
2.2
96
94
92
90
97
95
137
213
10b
10c
4
5
2.5
1
96
99
93
96
98
99
192
495
10d
10e
6
7
8
1
96
95
90
93
95
99
60
10f
475
10g
8
9
7.5
24
98
92
94
86
79
86
65
19
10h
10i
10
11
1
9
96
94
93
82
80
85
480
52
10j
10k

ACCEPTED MANUSCRIPT
12
13
45 min 92
92
88
86
84
613
460
10l
1
90
10m
a Reaction conditions: alkene (10 mmol), catalyst 8 (50 mg, 0.2 mol%), 60 °C, CH3CN (10 mL), O2
( 1 atm. bubbling 15 mL/min), time.
b Isolated yield
c GC analysis; using an internal standard technique (Anisole) and is based on alkenes.
d Based on conversion
Co(OAc)2 provides a moderate efficiency and selectivity (Table 2). Coordination of cobalt
into PAA provides only 77 % conversion with low selectivity (88%). In this regard, given the
presence of free Fe3O4 NPs on GO surface in the catalyst, a synergetic effect arising from
Fe3O4, Co and PAA-Co complex immobilized on GO could be considered. There is not any
detectable product in the absence of any catalyst. The results illustrated the advantage and
superiority of the support and the designed structure on the catalytic behavior over all applied
reagents, which selectively conduct the epoxidation of alkene.
Table 2 The conversions and selectivities of styrene epoxidation catalyzed by PAA-Co(II) complex,
CO, Fe3O4, and Co(OAc)2 along with blank experiment
b
Catalyst
PAA-Co(II)
GO
Fe3O4
Co(OAc)2
Blank
complexa
Conversion (%) 77
Selectivity (%) 88
8
35
38
57
62
75
Trace
-
a PAA-Co(II) complex was prepared by addition of CoCl2.6H2O (10 mg) to 1.0 g of PAA in 25 mL
ethanol at reflux conditions. The reaction was stirred for 3 h, then the complex filtered, washed and
dried at room temperature.
b Fe3O4 was prepared according to a previously reported procedure [29]
3.4. Mechanism studies
According our observations and other researchers’ work [1,30-32], a plausible mechanism
was suggested for the epoxidation of olefins with molecular oxygen catalyzed by
GO/Fe3O4@PAA-Co(II). The suggested mechanism is illustrated in Scheme 4. In first, Co-
superoxo active radicals were generated by interaction of the catalyst with O2 (Scheme 4, Int.
I). This was proved by taking XPS analysis of the mixture in situ, which the presence of both
Co(II) and Co(III) was detected to the mixture (Fig. 6A). Styrene is added to this Co-
superoxo intermediate through an oxidative addition reaction, which led to the cyclic
intermediate II (Scheme 4). A cyclic peroxide radical (Intermediate III) could be formed from
intermediate II and the catalyst is regenerated for the next cycle. This peroxide interacts with
another styrene molecule to produce the desired epoxide product. However, the active
peroxide may be decomposed to form aldehyde and formaldehyde by-products (Scheme 4),
that is not completely consistent with the observed high selectivity in this work for all
substrates. Another rote is possible via the collapse of II (Scheme 4, bold arrow), which led to
a Co-oxo radical (Int. IV) and a desired epoxide molecule, based on the previously reported
mechanisms [33,34]. Intermediate IV could be also interacting with another styrene molecule
to form epoxide product. Given that high selectivity observed for the substrates, this rote
seems to be more probable. However, the formation of epoxide could be takes place through
each routes and or a mixture of them. It worth noted that the Co(II) species are only species
in the mixture in end of the reaction according to XPS analysis of the recovered catalyst after
reaction completion (Fig. 6A-b), which corroborates that the highly active plausible
intermediates III and IV are consumed and/or decomposed during reaction.

ACCEPTED MANUSCRIPT
O
O2
+
O
(III)Co
O
Co(II)
IV
2
O
O
(III)Co
O2
I
n
o
i
t
i
III
s
o
p
O
m
o
By-product
O
c
e
O
D
H
O
H
o
C
)
I
I
I
(
II
Scheme 4. A plausible reaction mechanism for catalytic epoxidation of olefins using
GO/Fe3O4@PAA-Co(II)
To support of the suggested radical mechanism, in another control experiment, hydroquinone
as a strong and efficient radical scavenger was added to the reaction mixture of styrene at
beginning (t=0) as well as 1.5 h of the reaction. Progress of the reactions were monitored
every 30 min and the results were gathered in Fig. 6B. As shown in Figure 6B, the reaction is
stopped immediately after addition of hydroquinone. Furthermore, addition of hydroquinone
at start of the reaction, almost suppress the reaction progress. These results completely
confirm that the epoxidation reaction of olefins catalyzed by GO/Fe3O4@PAA-Co(II)
accomplished by active radical species as shown in Scheme 4.
Fig. 6. (A) (a) In situ normalized high-resolution XPS spectrum of the epoxidation reaction mixture of
styrene catalyzed by the GO/Fe3O4@PAA-Co(II) and (b) XPS spectrum of the recovered catalyst
(first cycle). (B) Reaction kinetics (green curve) and effect of hydroquinone on reaction progress at
beginning (black curve) and at 62% conversion (blue curve) for the epoxidation reaction of styrene.
The magnetic nanocatalyst 8 could be easily recovered from the mixture and reused for the
subsequent reactions without any significant reactivity loss (Fig. 7a). The recyclability of the
catalyst was investigated for 6 consecutive runs and as a result, didn’t found any significant
drop in selectivity, epoxide efficiency and catalyst yield. In addition, a small metal leaching
was detected in each run that was negligible. The results clearly demonstrate the stability and
durability of the catalyst during the reaction. To prove this, the recovered catalyst after 6th run

ACCEPTED MANUSCRIPT
was characterized by some techniques. VSM analysis of the recovered catalyst demonstrated
a similar magnetic behavior as same as the corresponding fresh catalyst. There is not any
decrease in saturation magnetization. The results were also confirmed the absence of any
magnetite leaching from the catalyst surface. Furthermore, the FE-SEM and TEM images
from the recovered catalyst revealed that the catalyst preserves its morphology and
distribution during several recycling.
It is well known that some catalytic reactions would anyway take place even if an actual
catalytic species is leached off from their material [35,36]. Although, a negligible metal
leaching was found for the catalyst, to gain more insight into heterogeneous nature of the
catalyst, hot filtration test was applied on the model reaction. The catalyst was magnetically
removed in situ from the epoxidation reaction of styrene in CH3CN in the presence of
molecular oxygen (15 mL/min bubbling) after 1 h of the reaction mixture (44% epoxy styrene
conversion). Reaction progress was monitored every 15 min by GC instrument. Didn’t
observe any significant progress in epoxidation, which the conversion reached to 46 % after 3
h. These results are in agreement with leaching studies and confirm that the catalyst
completely act heterogeneously during the reaction.
Fig. 7. (A) Recycling of GO/Fe3O4@PAA-Co for the oxidation reaction of styrene in CH3CN at 60 °C
in the presence of molecular O2. (B) VSM curve, (C) FE-SEM, and (D) TEM analysis of the
recovered catalyst after 6th run.
3.5. Scalability of the method
To investigate the potential of the process for larger scales, the epoxidation reaction of
styrene was selected as a model reaction. The reaction was performed under the obtained
premium conditions with 200, 500, and 700 mmol of styrene. As presented in Table 3, the
results clearly demonstrated the direct scalability of the process, which didn’t found any
considerable loss in efficiency in term of time, conversion or selectivity even at large
quantities (Table 3). However, a reasonable insignificant loss in conversion was observed as
the amount of styrene rises; nevertheless, the selectivity was independent than the scale and
remained constant thoroughly.

ACCEPTED MANUSCRIPT
Table 3 Gram scale production of epoxy styrene using GO/Fe3O4@PAA-Coa
Entry mmol of
styrene
Time Conversion Selectivity
(h)
(%)b
90
(%)
1
2
3
200
500
700
3
97
3
88
97
3
86
95
a The reactions were performed in a 1 L round bottom flask and the reactions were mechanically
stirred. All experiments were performed in the presence of 15 mL.min-1 O2 bubbling (~ 1 atm.) at 60
°C. For each entry, the amounts of CH3CN and catalyst 8 were proportional to the amount of styrene
(see “General procedure for catalytic epoxidation of olefins by GO/Fe3O4@PAA-Co(II)”).
b GC yield.
4. Conclusion
In conclusion, a novel synthetic poly α-amino acid-Co(II) complex was immobilized on
magnetite graphene oxide (GO/Fe3O4@PAA Co(II)) and used as an efficient nanocatalyst for
selective epoxidation of olefins. A wide variety of olefins were subjected to epoxidation
using the catalyst, which high to excellent conversions and selectivities were achieved for all
of them. The catalyst has a mean size of about 30 nm (from FE-SEM, DLS and TEM
analyses) with saturation magnetization of 32 emu.g-1 (from VSM analysis) containing 0.2
mol%Co2+ (from ICP and XPS analyses). Further characterization of the nanocatalyst was
obtained from FTIR, TGA, EDX, and XRD analyses. Free coreductant, gram scale, mild
reaction conditions (CH3CN, 60 °C), fast performance, selectivity toward epoxide product,
durability (insignificant activity loss during six consecutive recycling), thermal stability,
magnetically recoverable, and high efficiency epoxide products were some of the advantages
of the present protocol, which make it suitable alternative method for the epoxidation
program. Furthermore, direct scalability for the process in large quantities was found that is
applicable for industrial goals.
Conflicts of interest
There are no conflicts to declare.
Acknowledgment
Authors gratefully acknowledge the financial support of this work by the Research Council of
University of Birjand. We also thank Roxana Nayerlou for her critical reading and comment
on the manuscript.
References
[1] M. Kazemnejadi, A. Shakeri, M. Nikookar, M. Mohammadi, M. Esmaeilpour, Res.
Chem. Intermediat. 43 (2017) 6889-6910.
[2] S.H. Lee, L. Xu, B.K. Park, Y.V. Mironov, S.H. Kim, Y.J. Song, S.J. Kim, Chem. Eur.
J. 16 (2010) 4678-4685.
[3] D. Swern, Org. React. 7 (2004) 378-434.
[4] P.S. Teo, H.N. Lim, N.M. Huang, C.H. Chia, I. Harrison, Ceram. Int. 38 (2012) 6411-
6416.
[5] M. Keshavarz, A. Zarei Ahmady, L. Vaccaro, M. Kardani, Molecules 23 (2018) 330.
[6] S. Tian, Q. Fu, W. Chen, Q. Feng, Z. Chen, J. Zhang, W.C. Cheong, R. Yu, L. Gu, J.
Dong, J. Luo, Nat. Commun. 9 (2018) 2353.
[7] C. Weerakkody, S. Biswas, W. Song, J. He, N. Wasalathanthri, S. Dissanayake, D.A.
Kriz, B. Dutta, S.L. Suib, Appl. Catal. B 221 (2018) 681-690.

ACCEPTED MANUSCRIPT
[8] D.T. Bregante, A.M. Johnson, A.Y. Patel, E.Z. Ayla, M.J. Cordon, B.C. Bukowski, J.
Greeley, R. Gounder, D.W. Flaherty, J. Am. Chem. Soc. 141 (2019) 7302-7319.
[9] D. Ghosh, B. Febriansyah, D. Gupta, L.K.S. Ng, S. Xi, Y. Du, T. Baikie, Z. Dong, H.
S. Soo, ACS Nano 12 (2018) 5903-5912.
[10] R. Egekenze, Y. Gultneh, R. Butcher, Polyhedron 144 (2018) 198-209.
[11] I. Bernar, F.P. Rutjes, J.A. Elemans, R.J. Nolte, Catalysts 9 (2019) 195.
[12] S. Goswami, S. Singha, R. Saha, A.S. Roy, M. Islam, S. Kumar, Inorg. Chim. Acta 486
(2019) 352-360.
[13] M. Vasconcellos-Dias, J. Marreiros, R. Sales, V. Félix, P. Brandão, C. D. Nunes, M.J.
Calhorda, Molecules 24 (2019) 578.
[14] A. Nasrallah, G. Grelier, M.I. Lapuh, F.J. Duran, B. Darses, P. Dauban, Eur. J. Org.
Chem. 2018 (2018) 5836-5842.
[15] J. Jiang, K. Ma, Y. Zheng, S. Cai, R. Li, J. Ma, Appl. Clay Sci. 45 (2009) 117-122.
[16] S. Ghiami, M.A. Nasseri, A. Allahresani, M. Kazemnejadi, React. Kinet. Mech. Catal.
126 (2019) 383-398.
[17] M. Kazemnejadi, S.A. Alavi, Z. Rezazadeh, M.A. Nasseri, A. Allahresani, M.
Esmaeilpour, J. Mol. Struct. 1186 (2019) 230-249.
[18] X. Yang, X. Zhang, Y. Ma, Y. Huang, Y. Wang, Y. Chen, J. Mater. Chem. 19 (2009)
2710-2714.
[19] R. Dalpozzo, Green Chem. 17 (2015) 3671-3686.
[20] M. Kazemnejadi, A. Shakeri, M. Nikookar, R. Shademani, M. Mohammadi, Royal Soc.
Open Sci. 5 (2018) 171541.
[21] X.H. Zhang, R.J. Wei, Y.Y. Zhang, B.Y. Du, Z.Q. Fan, Macromolecules 48 (2015)
536-544.
[22] W.S. Hummers Jr, R.E. Offeman, J. Am. Chem. Soc. 80 (1958) 1339-1339.
[23] X. Sun, Z. Liu, K. Welsher, J. T. Robinson, A. Goodwin, S. Zaric, H. Dai, Nano Res. 1
(2008) 203-212.
[24] J. Shen, Y. Hu, M. Shi, N. Li, H. Ma, M. Ye, J. Phys. Chem. C 114 (2010) 1498-1503.
[25] Y.L. Dong, H.G. Zhang, Z.U. Rahman, L. Su, X.J. Chen, J. Hu, X.G. Chen, Nanoscale
4 (2012) 3969-3976.
[26] P. Bilalis, L.A. Tziveleka, S. Varlas, H. Iatrou, Polym. Chem. 7 (2016) 1475-1485.
[27] H. Tourani, M.R. Naimi Jamal, M.G. Dekamin, ChemistrySelect 3 (2018) 8332-8337.
[28] M. Kazemnejadi, S.A. Alavi, Z. Rezazadeh, M.A. Nasseri, A. Allahresani, M.
Esmaeilpour, Green Chem. 21 (2019) 1718-1734.
[29] R. Sardarian, M. Kazemnejadi, M. Esmaeilpour, Dalton Trans. 48 (2019) 3132-3145.
[30] M.J. Beier, W. Kleist, M.T. Wharmby, R. Kissner, B. Kimmerle, P.A. Wright, J.D.
Grunwaldt, A. Baiker, Chem. Eur. J. 18 (2012) 887-898.
[31] D. Saha, T. Maity, R. Bera, S. Koner, Polyhedron 56 (2013) 230-236.
[32] P. Buranaprasertsuk, Y. Tangsakol, W. Chavasiri, Catal. Commun. 8 (2007) 310-314.
[33] Y.L. Hu, Y.W. Liu, D.J. Li, J. Iran. Chem. Soc. 12 (2015) 2179-2184.
[34] W. Nam, H.J. Kim, S.H. Kim, R.Y. Ho, J.S. Valentine, Inorg. Chem. 35 (1996) 1045-
1049.
[35] M. Kazemnejadi, A. Shakeri, M. Mohammadi, M. Tabefam, J. Iran. Chem. Soc. 14
(2017) 1917-1933.
[36] M. Kazemnejadi, M. Nikookar, M. Mohammadi, A. Shakeri, M. Esmaeilpour, J.
Colloid Interface Sci. 527 (2018) 298-314.

ACCEPTED MANUSCRIPT
Highlights
► A novel Co(II)-poly
α
-amino acid-modified graphene oxide was used for epoxidation of alkenes
► Epoxidation of olefins was performed in absence of any co-reductant
► The catalytic system controls epoxide selectivity and provide high efficiency
► Epoxidation occurs through a radical mechanism.