Green oxidation of alkenes in ionic liquid solvent by hydrogen peroxide over high performance Fe(III) Schiff base complexes immobilized on MCM-41

Article abstract of DOI:10.1007/s12039-014-0619-7

A series of Fe(III) Schiff base complexes immobilized on MCM-41 were prepared and characterized by various physicochemical and spectroscopic methods. The complexes were used for oxidation of cyclohexene by 30% hydrogen peroxide in the presence and absence of ethylmethyl imidazolium chloride (EMIM) ionic liquid as solvent. The immobilized complexes proved to be effective catalysts and generally exhibited much higher catalytic performance than their homogeneous analogue. Catalytic performance of the complexes was also found to be closely related to the Schiff base ligands used. Additionally, ion liquid solvent efficiently improved all the catalytic performances. Finally, the reaction was extended to different alkenes using the heterogeneous complex 2-L₄. Among all the alkenes, those containing π-electron-withdrawing groups and trans-orientations exhibited lower tendency for oxidation.

Full text of DOI:10.1007/s12039-014-0619-7

c
J. Chem. Sci. Vol. 126, No. 3, May 2014, pp. 801–805. ꢀ Indian Academy of Sciences.
Green oxidation of alkenes in ionic liquid solvent by hydrogen peroxide
over high performance Fe(III) Schiff base complexes immobilized
on MCM-41
MOHAMMAD TAGHI GOLDANIa,∗, ALI MOHAMMADIb,∗ and REZA SANDAROOSc
aDepartment of Chemical Engineering, Birjand Branch, Islamic Azad University, Birjand 9717851367, Iran
bDepartment of Chemistry, Faculty of Sciences, Ferdowsi University of Mashhad, PO Box 91775,
Mashhad, Iran
cDepartment of Chemistry, Faculty of Sciences, University of Birjand, Birjand, Iran
e-mail: mt.goldani@yahoo.com; a.mohammady62@yahoo.com
MS received 11 March 2012; revised 24 January 2014; accepted 27 January 2014
Abstract. A series of Fe(III) Schiff base complexes immobilized on MCM-41 were prepared and character-
ized by various physicochemical and spectroscopic methods. The complexes were used for oxidation of cyclo-
hexene by 30% hydrogen peroxide in the presence and absence of ethylmethyl imidazolium chloride (EMIM)
ionic liquid as solvent. The immobilized complexes proved to be effective catalysts and generally exhibited
much higher catalytic performance than their homogeneous analogue. Catalytic performance of the complexes
was also found to be closely related to the Schiff base ligands used. Additionally, ion liquid solvent effi-
ciently improved all the catalytic performances. Finally, the reaction was extended to different alkenes using
the heterogeneous complex 2-L4. Among all the alkenes, those containing π-electron-withdrawing groups and
trans-orientations exhibited lower tendency for oxidation.
Keywords. Immobilized complexes; alkene oxidation; hydrogen peroxide; ionic liquid; MCM-41.
1. Introduction
a surface-bound tether containing a functional group.
This methodology has been employed successfully in
the functionalization of various traditional types of
silica supports.26–28 More recently, MCM-41 materi-
als have been used extensively as ideal inorganic sup-
ports due to their high surface area and sharply dis-
tributed pore dimensions in the mesoporous ranges,
which allow easy access of reactants to the active sites
and also functionalization of bulky catalytic sites within
the pores.29–31
It is well-known that the widespread use of tradi-
tional organic solvents in many chemical processes is an
issue of great environmental concern. Hence, solvent-
free reactions and performing chemical reaction in ionic
liquid solvents are topics of current interest in the
context of environment-friendly chemical reactions.32
In the present investigation, a series of conventional
Fe(III) Schiff base complexes were immobilized on
MCM-41, and the resulting immobilized complexes
were used for selective oxidation of alkenes by 30%
hydrogen peroxide in the presence and absence of ethyl-
methyl imidazolium chloride (EMIM) ionic liquid as
solvent. The role of Schiff base ligands and solvent on
catalytic performances of the immobilized catalysts was
also explored.
In the area of metal-catalysed oxidation, epoxida-
tion of alkenes by heterogeneous and homogenous
catalysts have attracted enormous attention, because
expoxides are very precious intermediates in mod-
ern organic synthesis, and pharmacology and poly-
mer industry.1–8 Among the efficient oxidants used
commonly, molecular oxygen and hydrogen peroxide
are economically and ecologically preferred. However,
H2O2 is easier to handle, cheap and forms water as
a byproduct.9 In comparison to supported catalysts,
separation of catalyst from the reaction mixture is a
typical disadvantage of homogenous catalysts.10,11 The
supported metal complexes also show high catalytic
activity12,13 and selectivity14 in comparison to unsup-
ported complexes. Unsupported metal complexes of ß-
diketones15 and salen16 show low activity and selec-
tivity in epoxidation of olefins in comparison to sup-
ported metal complexes.10,14,17–25 A popular approach
for preparation of heterogeneous catalyst is to immobi-
lize the homogeneous catalyst on the support through
∗For correspondence
801

802
Mohammad Taghi Goldani et al.
immobilized analogue (abbreviated as 2-L1–L4) were
2. Experimental
prepared (figure 1).
2.1 General
3-Aminopropyltriethoxysilane (APTES), salicylalde-
hyde, ethylenediamine, diethyltriamine, 1,2-diamino-
cyclohexane (chair conformation) and o-phenylene-
diamine supplied by Merck Chemical Co. (Darmstadt,
Germany) and Fluka Co. and were used as received.
Ligands and metal-complexes were characterized by
spectroscopic measurements.
2.3 Characterization of catalysts
Chemical analysis revealed that a small amount of car-
bon and nitrogen was present in the MCM-41 due to the
remnant surfactant. Much increased carbon and nitro-
gen contents were observed in APTES-MCM-41, con-
firming its successful preparation. Elemental contents
in all five homogeneous complexes were quite com-
parable with the calculated values, indicating that the
as-prepared homogeneous complexes held the expected
elemental composition (table 3). Diagnostic bands of
FTIR spectra of the resulting immobilized complexes
are listed in table 3. Successful preparation of homoge-
neous complexes could be confirmed by the presence
of the characteristic bands at about 1600, 600 cm−1,
depending on the different ligands used, assigned to
the C=N stretching vibration of the imine groups. In
addition, presence of two or three bands in the low
frequency region (450–550 cm−1) indicated success-
ful coordination of phenolic oxygen and azomethine
nitrogen to the transition metal. In the FTIR spectra of
immobilized complexes, apart from the bands in over-
lapping region of the silica backbone, the other bands
of homogeneous complexes were all clearly observed
though there were some marginal shifts in the posi-
tions of these bands due to immobilization. This indi-
cates that quadridentate coordination structures of the
homogeneous complexes survived the immobilization
procedure.
2.2 Preparation of catalysts
Synthetic procedure for pure siliceous MCM-41
(Si-MCM-41) is followed as described in litera-
ture.33 3-Aminopropyltriethoxysilane-modified MCM-
41 (APTES-MCM-41) was prepared according to a
known procedure.34 Homogeneous Schiff base com-
plexes were prepared by condensation of salicylalde-
hyde and appropriate diamines in 2:1 molar ratio. In a
typical synthesis, diamine (10 mmol) was suspended in
absolute ethanol (60 mL). A solution of salicaldehyde
(20 mmol) in absolute ethanol (60 mL) was added drop-
wise with vigorous stirring of suspension. The result-
ing mixture was refluxed for 3 h, then neutralized with
the appropriate amount of NaOH to adjust the pH close
to 7, prior to the addition of solid FeCl3 (12 mmol).
After refluxing for another 6 h, the mixture was cooled
and filtered under reduced pressure. The collected solid
was washed with absolute ethanol and dried in air,
then recrystallized from absolute ethanol. The homoge-
neous complexes were bound to MCM-41 by one axi-
ally coordinating propyl chain spacer, which allowed
maximum conformational mobility of the active com-
plexes. In order to prepare typical supported cata-
lyst, a suspension of freshly dried APTESMCM-41
(2 g) in dry toluene (50 mL) was added to a solu-
tion of the prepared homogeneous complex (8 mmol)
in dry toluene (50 mL). The mixture was vigorously
stirred under reflux for 6 h. The resulting suspension
was cooled and filtered through a fine-porous filter
paper. The collected powder was washed overnight in
a Soxhlet extractor using absolute ethanol and ace-
tonitrile with an ethanol/acetonitrile molar ratio of 1:1
to remove free ligands and homogeneous complexes
adsorbed on the external surface of support. The solid
sample was further stirred in 0.01 mol/L NaCl solution
for 24 h to remove the remaining uncoordinated metal
ions. Finally, the mixture was filtered, and the obtained
solid was dried in air at 80◦C for 10 h. By changing
the diamine, a series of homogeneous Fe(III) Schiff
base complexes (abbreviated as 1-L1–L4) and their
The UV–vis spectra of the homogeneous complexes
displayed one characteristic peak at about 390–450 nm
typical of the metal–ligand band and one broad peak at
about 570–610 nm associated with d-d transitions. The
R
N
O
N
O
O
O
O
O
O
N
N
Fe
Cl
Si
N
Fe
R
H
1
2
Figure 1. Homogeneous Fe(III) Schiff base complexes (1-
L1-L4) and their immobilized analogues (2-L1-L4). L1: R =
-CH2CH2-; L2: R = -CH2CH2NHCH2CH2-; L3: R =
;
L4: R =
.

Alkene oxidation
803
UV–vis spectra of all the immobilized complexes were and selectivity of immobilized catalyst. In 2-L1and 2-
quite similar to those of their corresponding homoge- L2, the bridge groups were all alkyl groups; however,
neous complexes, further confirming that the immobi- 2-L2 showed higher catalytic performance due to the
lized Fe(III) Schiff base complexes were successfully additional N atom with isolated electron pair increas-
prepared.
ing donation capacity of the ligand. The bridge group in
2-L4 was phenyl which contributed to formation of the
p-extended coordination structure. This could facilitate
oxidation and so afford the best catalytic performance
of 2-L4. The worst catalytic performance, over 2-L3,
could be due to the higher stearic encumbrance around
the Fe (III) centre, originating from the chair conforma-
tion of the bridge group. Consequently, catalytic perfor-
mance observed over the immobilized complexes with
different Schiff base ligands was in the following order:
2-L4>2-L2>2-L1>2-L3. Supported catalysts exhibit bet-
ter selectivity compared to their homogeneous con-
gener. The epoxide ring-opening side reaction is cata-
lysed mainly by the acidic metal sites. Probably,
changes in the surface properties produced by elec-
tron donation of support decrease acidity of the surface,
preventing epoxides from converting to diols.
Totally, all the catalysts showed much higher cata-
lytic performance when ionic liquid was used as solvent
(table 1).
The reaction was extended to different alkenes
using catalyst 2-L4 in ionic liquid solvent from which
good to excellent catalytic performances were obtained
(tables 2 and 3).
2.4 Catalytic performance
A mixture of substrate and catalyst in the required
molar ratio was stirred for about 10 min and then the
corresponding amount of H2O2 was added. Required
amount of solvent was also added when ionic liquid
was used as solvent. After the mixture was heated to the
desired temperature for the desired time, the products
were filtered and purified by column chromatography
using hexan/ethyl acetate 9:1 as eluent.
3. Results and discussion
Epoxidation of cyclohexene using homogeneous cata-
lysts of 1-L1-L4and heterogeneous catalysts 2-L1-L4
was selected as a template reaction (table 1). In order
to compare catalytic performances (selectivity × con-
version) which originate from reaction media, all the
reactions were performed in the absence and pres-
ence of EMIM ionic liquid. When the reaction was
performed over the immobilized complexes, on the
whole, much higher catalytic performance was obtained
than over their corresponding homogeneous complexes.
Such obviously enhanced catalytic performance could,
in part, be attributed to the ability of the support to
highly disperse the active complexes and/or to con-
trol local concentrations of the reactant and product
around active sites. The ligand structure was also, as
usual, an important factor in determining the activity
Conjugated π-electron-withdrawing groups with
double bond of alkenes decrease reactivity for epoxi-
dation (compare entries 1 and 5). Although, the oxy-
gen atom is an σ-acceptor, π-resonance will increase
electron density on reaction centre through the oxy-
gen lone pairs. Accordingly, presence of a methoxy
group in methyl cinnamate (entry 6) probably causes
enhanced reactivity of this alkene with respect to
Table 1. Epoxidation of cyclohexene with H2O2 in the presence of supported and unsupported Fe (salen) complexes.a
Cyclohexene
Selectivityc
epoxide (%)
Cyclohexene
Selectivityc
epoxide (%)
Entry
Complex
conversionb(%)
Entry
Complex
conversionb(%)
1
1-L1
1-L1
1-L2
1-L2
1-L3
1-L3
1-L4
1-L4
38
55
44
60
33
51
53
69
66
75
72
82
61
73
77
90
9
2-L1
2-L1
2-L2
2-L2
2-L3
2-L3
2-L4
2-L4
73
88
74
91
43
57
78
95
96
100
100
100
65
75
100
100
2d
3
10d
11
4d
5
12d
13
6d
7
14d
15
8d
16d
aMolar ratios for Fe: cyclohexene: oxidant = 0.02:1:2, Time = 20 min, Temperature = 50oC
bConversion = (mol of cyclohexene reacted/mol of cyclohexene in the feed) × 100
cSelectivity = (mol of cyclohexene converted to the epoxide/mol of cyclohexene reacted) × 100
dIonic liquid was used as solvent (3 mL)

804
Mohammad Taghi Goldani et al.
Table 2. Epoxidation of alkenes with H2O2 in the presence of supported complex 2-La4.
Alkene
Alkene
Entry
1
Substrate
conversionb(%) Selectivityc(%) Entry
Substrate
conversionb(%) Selectivityc(%)
O
92
100
8
74
100
2d
3e
91
92
100
100
9
79
98
100
100
10
O
4f
91
69
67
64
100
100
100
100
11
12
13
14
75
62
86
66
100
100
100
100
O
O
Ph
5
6
O
OMe
O
7
aMolar ratios for Fe: substrate: oxidant = 0.02:1:2, Time = 20 min, Temperature = 50◦C, ionic liquid = 3 mL
bConversion = (mol of alkene reacted/mol of alkene in the feed) × 100
cSelectivity = (mol of alkene converted to the products/mol of alkene reacted) × 100
d−fReused complex 2-L4
Table 3. Elemental compositions and spectroscopic data for the complex samples. 1
Found (calcd.) %
C
Spectroscopic data
Samples
N
FTIR (cm−1
)
λmax(nm)
MCM – 41
APTES- MCM -41
1-L1
1-L2
1-L3
1-L4
2-L1
2-L2
2-L3
2.0
5.9
0.1
2.3
–
–
–
–
53.68 (53.74)
53.88 (53.96)
58.30 (58.35)
59.17 (59.22)
2.6
1.9
3.1
7.80 (7.83)
10.46 (10.49)
6.75 (6.80)
6.86 (6.91)
1.5
1.8
1.4
1615, 585, 465
1620, 612, 522
1620, 602, 535
1619, 539, 462
1612, 585, 465
1608, 612, 522
1611, 602, 535
1610
390, 600
390, 609
416, 606
446, 575
390, 600
390, 610
417, 606
447, 578
2-L4
3.0
1.5
1FTIR spectra of all samples were recorded in KBr disks at room temperature on a Shimadzu (model 8201 PC)
spectrophotometer. UV-vis spectra were recorded on a Shimadzu (model 2501 PC)

Alkene oxidation
805
5. Hashmi A S K, Graf K, Ackermann M and Rominger F
2013 Chem. Catal. Chem. 5 1092–1095
6. Gaspar H, Andrade M, Pereira C, Pereira A M, Rebelo S
L H, Araújo J P, Pires J, Arvalho A P and Freire C 2013
Catal. Today 203 103–110
7. Prat I, Font D, Company A, Junge K, Ribas X, Beller M
and Costas M 2013 Adv. Synth. Catal. 355 947–956
8. Backvall J E 2004 Modern oxidation methods (Wein-
heim: Wiley-VCH)
9. Grigoropoulou G, Clark J H and Elings J A 2003 Green
Chem. 5 1–7
10. Andrea B, Martino S and Beniditto C 2006 J. Mol.
Catal. A: Chem. 250 1
11. Shahram T, Hossein H M, Valiollah M, Mazid M and
Gholamhossein G 2006 Inorg. Chem. 9 575
12. Yoo D W, Han J H, Nam S H, Kim C and Lee J K 2006
Inorg. Chem. Commun. 9 654
13. Akelah A, Rehab A, Kenawy E R and Zeid A M S 2006
J. Appl. Polym. Sci. 101 1121
14. Stefan B, Agniesszker S and Beata S 2006 Appl. Catal.
A: Gen. 309 85
benzylideneacetone (entry 7). Decreased reactivity of
2,4,4-trimethyl-1-pentene (entry 9) seems to be due to
the stearic bulk of tert-butyl substituents of the alkene.
Comparison of entries 9 and 10 may indicate domi-
nance of stearic factors in epoxidation. More reactivity
of cis-alkenes in comparison to trans-ones may be due
to the cis-orientation of the substituents which facili-
tate the approach of the alkene to the active oxidant
(compare entries 7 with 8 and 13 with 14). In gene-
ral, since alkenes act as nucleophiles attacking the per-
oxide (or the active oxidant) bond, the rate of epoxi-
dation is enhanced by increasing alkyl substitutions
(compare entries 7 and 11). It should be noted that
the base-catalysed epoxidation of alkenes with H2O2
prefers electron-deficient alkenes, i.e., α,β-unsaturated
carbonyl compounds, to the electron-rich ones, because
the rate determining step seems to be the nucleophilic
addition of HO-O− to the C,C double bond.7 We next
examined reusability of catalyst 2-L4 for epoxidation
of cyclohexene. The obtained results revealed no neg-
ligible activity loss over consecutive recycling (table 2,
entries 2–4).
15. Blaz E and Pielichowski J 2006 Molecules 11 115
16. Lutsa T, Supruna W, Hofmannb D, Klepela O and Pap
H 2007 J. Mol. Catal. A: Chem. 261 16
17. Zsigmond A, Boger K and Notheisz F 2003 J. Catal.
213 103
18. Yulong Q and Daodao H 2005 Gaofenzi Tongbao 1 31
19. Shahram T, Hossein
H M, Moghadam M V
and Gholamhossein M G 2006 J. Mol. Catal. A: Chem.
255 249
20. Beloglazkina E K, Majangu A G, Romashkima R B and
Zyk N V 2006 Tetrahedron Lett. 47 2957
21. Kamata K, Yonehara K, Sumida Y, Hirata K, Nojima S
and Mizuno N 2011 Angew. Chem. Int. Ed. 50 12062
22. Ahn J H, Kim J C, Ihm S K, Oh C G and Sherrington D
C 2005 Ind. Eng. Chem. Res. 44 8560
23. Wang Q, Wang L, Chen J, Wu Y and Mi Z 2007 J. Mol.
Catal. A: Chem. 273 73
24. Ziolek M 2004 Catal. Today 90 145
25. Olason G and Sherrington D C 1999 React. Funct.
Polym. 42 163
26. Zhou X G, Yu X Q, Huang J S, Li S G, Li L S and Che
C M 1999 Chem. Commun. 1789
27. Zhang H, Xiang S and Li C 2005 Chem. Commun. 1209
28. Xiang S, Zhang Y, Xin Q and Li C 2002 Chem. Com-
mun. 2696
29. Oliveira P, Machado A, Ramos A M, Fonseca I, Braz
Fernandes F M, Botelho do Rego A M and Vital J 2009
Micropor. Mesopor. Mat. 120 432
30. Demicheli G, Maggi R, Mazzacani A, Righi P, Sartori
G and Bigi F 2001 Tetrahedron Lett. 42 2401
31. Singh U G, Williams R T, Hallam K R and Allen G C
2005 J. Solid State Chem. 178 3405
32. Formentin P, Garcia H and Leyva A 2004 J. Mol. Catal.
A: Chem. 214 137
33. Kresge C T, Leonowicz M E, Roth W J, Vartuli J C and
Beck J S 1992 Nature 359 710
4. Conclusion
A series of immobilized Fe(III) Schiff base com-
plexes were successfully prepared and they exhibited
promising catalytic performance toward selective oxi-
dation of cyclohexene by 30% H2O2. Significant dif-
ferences in catalytic performance of the immobilized
complexes could be attributed to their different ligand
structures, and the immobilized complex 2-L4 with π-
extended coordination structure was found to exhibit
the best catalytic performance. Among the alkenes,
those containing π-electron-withdrawing groups and
trans-orientations exhibited lower reactivity. Addition-
ally, ionic liquid solvent efficiently improved all the
catalytic performances.
References
1. Hauser A, Cokoja M and Kühn E 2013 Catal. Sci.
Technol. 3 552
2. Sherry B D and Fürstner A 2008 Acc. Chem. Res. 41
1500
3. Saisaha P, de Boer J W and Browne W R 2013 Chem.
Soc. Rev. 42 2059–2074
4. Visuvamithiran P, Palanichamy M, Shanthi K and
Murugesan V 2013 Appl. Catal. A: Gen. 462 31–38
34. Baleiza C, Gigante B, Sabater M J, Garcia H and Corma
A 2002 Appl. Catal A: Gen. 228 279