Mn-Schiff base modified MCM-41, SBA-15 and CMK-3 NMs as single-site heterogeneous catalysts: Alkene epoxidation with H2O2 incorporation

Article abstract of DOI:10.1016/j.molcata.2015.12.015

The development of new functional catalytic materials prepared via appropriate chemical modification of mesoporous silica SBA-15, MCM-41 or carbon nanomaterials CMK-3, are presented. Their synthesis has been carried out via two synthetic approaches: (a) a two steps procedure which includes grafting of the Schiff base ligand 1,3-bis[3-aza-3-(1-methyl-3-oxobut-1-enyl)-prop-3-en-1-yl]-2-(4-hydroxy-phenyl)-1,3-imidazolidine (L) onto the suppors and subsequent metalation of the so-formed hybrid material, and (b) an one step procedure which allows covalent grafting of the entire [Mn^II-Schiff base] catalyst onto the carbonaceous support. The resulting single-site heterogeneous catalysts were characterized and evaluated for alkene epoxidation with H₂O₂ in the presence of CH₃COONH₄ as additive. They are efficient and selective towards formation of epoxides. The highest TONs have been achieved by L@MCM-41-Mn^II and Mn^II-L@CMK-3. Moreover, Mn^II-L@CMK-3 is operative for a second use and kinetically very fast, demonstrating remarkably high TOFs 65-634 h^-1 that is correlated to its practically zero porosity. Based on the present data, the textural features of the obtained catalysts are discussed in correlation with their catalytic performance.

Full text of DOI:10.1016/j.molcata.2015.12.015

Accepted Manuscript
Title: Mn-Schiff base modified MCM-41, SBA-15 and
CMK-3 NPs as Single-Site Heterogeneous catalysts: Alkene
Epoxidation with H O incorporation
2
2
Author: A. Mavrogiorgou M. Baikousi V. Costas E.
Mouzourakis Y. Deligiannakis M.A. Karakassides M.
Louloudi
PII:
S1381-1169(15)30178-3
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.molcata.2015.12.015
MOLCAA 9717
To appear in:
Journal of Molecular Catalysis A: Chemical
Received date:
Revised date:
Accepted date:
8-10-2015
15-12-2015
21-12-2015
Please cite this article as: A.Mavrogiorgou, M.Baikousi, V.Costas, E.Mouzourakis,
Y.Deligiannakis, M.A.Karakassides, M.Louloudi, Mn-Schiff base modified MCM-
4
1, SBA-15 and CMK-3 NPs as Single-Site Heterogeneous catalysts: Alkene
Epoxidation with H2O2 incorporation, Journal of Molecular Catalysis A: Chemical
http://dx.doi.org/10.1016/j.molcata.2015.12.015
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.

Mn-Schiff base modified MCM-41, SBA-15 and CMK-3 NPs as
Single-Site Heterogeneous catalysts: Alkene Epoxidation with
H O incorporation
2
2
a
b
b
c
c
A. Mavrogiorgou , M. Baikousi , V. Costas , E. Mouzourakis , Y. Deligiannakis , M.
b
a
A. Karakassides and M. Louloudi *
a
Department of Chemistry, University of Ioannina, 45110 Ioannina, Greece
b
Department of Materials Science and Engineering, University of Ioannina, 45110
Ioannina, Greece
c
Physics Department, University of Ioannina, 45110 Ioannina, Greece
e-mail: mlouloud@uoi.gr

Graphical abstract
Highlights



Mn-catalysts were synthesized by covalent grafting onto SBA-15, MCM-41 and
CMK-3.
These catalysts are effective and selective towards epoxides with H
2
2
O and
CH
3
COONH .
4
II
Mn -L@CMK-3 provides high TONs and is kinetically very fast demonstrating
extremely high TOFs.

Abstract
The development of new functional catalytic materials prepared via appropriate chemical
modification of mesoporous silica SBA-15, MCM-41 or carbon nanomaterials CMK-3
are presented. Their synthesis has been carried out via two synthetic approaches: a) a two
steps procedure which includes grafting of the Schiff base ligand 1,3-bis[3-aza-3-(1-
methyl-3-oxobut-1-enyl)-prop-3-en-1-yl]-2-(4-hydroxy-phenyl)-1,3-imidazolidine
(L)
onto the suppors and subsequent metalation of the so-formed hybrid material, and b) an
II
one step procedure which allows covalent grafting of the entire [Mn -Schiff base]
catalyst onto the carbonaceous support. The resulting single-site heterogeneous catalysts
were characterized and evaluated for alkene epoxidation with H
2
2
O in the presence of
CH COONH as additive. They are effective and selective towards formation of
3
4
II
II
epoxides. The highest TONs have been achieved by L@MCM-41-Mn and Mn -
II
L@CMK-3. Moreover, Mn -L@CMK-3 is operative for a second use and kinetically
-1
very fast, demonstrating remarkably high TOFs 65-634 h that is correlated to its
practically zero porosity. Based on the present data, the textural features of the obtained
catalysts are discussed in correlation with their catalytic performance.
Keywords: catalytic epoxidation; supported manganese complexes; H
2
2
O activation;
mesoporous MCM-41, SBA-15, CMK-3

1
. Introduction
During previous decade, the replacement of stoichiometric and homogeneous oxidation
processes by environmentally friendly technologies which employ i) heterogeneous
catalysts and ii) “green” oxidants, e.g. such as H
2
2
O , has become a widely accepted
strategy. Rational design for effective and selective heterogeneous catalysts requires well
defined spatially-separated active sites i.e. as in conventional homogeneous systems, and
easy separation, recover and recyclability, as offered by heterogeneous systems [1]. The
development of ‘single-site heterogeneous catalysts’ can be achieved by heterogenization
of homogeneous single-site catalysts on mesoporous inorganic solids, combining by this
way the advantages of both molecular/homogeneous catalysts and heterogeneous ones [2-
6
]. In catalytic oxidation reactions, when “green” oxidants are concerned, hydrogen
peroxide, which is probably the best terminal oxidant -after dioxygen- is highly desirable
i.e. taking into account both environmental and economic considerations [7]. Within this
context, it has been demonstrated that alkene epoxidation by hydrogen peroxide catalyzed
by cheap, ‘single-site heterogeneous catalysts’ is potentially promising for applications in
chemical process and for large scale production of epoxides [3,7].
The goals of the present study are i) the development of ‘single-site heterogeneous
II
catalysts’ by covalently grafting of [Mn -Schiff base] complexes onto mesoporous
nanomaterials MCM-41, SBA-15 and CMK-3 and ii) the evaluation of these hybrid
organic-inorganic catalysts in alkenes’ oxidation with H O2. Among evaluation criteria for
2
the developed catalysts are catalyst’s efficiency, i.e. high substrate conversions, high
epoxide yields, high TONs, recovery and re-usability. For successful ‘single-site
heterogeneous catalysts’, key-parameters should comprise optimal synthetic
heterogenization process and appropriate textural/chemical features of the used inorganic
supports, i.e., surface area, porosity, polarity or hydrophobicity [3-6].
2
In this context, MCM-41 exhibits high specific surface area (~1000 m /g), high
crystallinity, high thermal stability, uniformity of hexagonal cylindrical pores, narrow
pore distribution and regulated pore diameter from 15 to 100Å [8]. SBA-15 with its
hexagonal 2D-porous structure of almost uniform cylindrical mesopores and
supplementary micropores, located into the amorphous silica walls, exhibits high surface

2
area (~700 m /g) and tunable large uniform pore sizes (~100 Å) [9]. Ordered mesoporous
carbon CMK-3 is a unique porous carbonaceous material that possess a well-ordered
hexagonal structure composed from uniformly sized carbon nanorods, with large surface
2
area (~1520 m /g) and high specific pore volume, along with tunable pore size (~70 Å)
[
10,11]. CMK-3 can be synthesized by applying the mesoporous silica sieve SBA-15 as a
template and sucrose as the carbon source [12,13]; it has gained considerable interest as
new nanoscaffold due to its potential applications as adsorbent, catalyst support and
material for advanced electronics applications [14-19].
In the front of development of alkene epoxidation catalysts, so far few studies have been
reported for covalently grafting of Schiff base metal-complexes onto mesoporous MCM-
4
1 [20-28], SBA-15 [29,30] and CMK-3 [31]. In these cases, the main adopted synthetic
strategy involved Schiff base immobilization via reaction with the surface modified
silicon-based materials [20-24, 27-30]; alternatively, the entire Schiff base metal complex
-
was tethered via axial ligation to H
2
N-modified MCM-41 [25-26] or OOC-modified
CMK-3 [31]. Among these studies, there are three examples which report H
2
2
O activation
II
[
[
20,25,31] and three catalytic systems, i.e. [Mn -Schiff base/MCM-41/t-BuOOH] [28],
III
III
Mn -Schiff base/SBA-15/t-BuOOH] [30] and [Mn -Schiff base/CMK-3/ H
2
2
O ] [31],
which use Mn-Schiff base complexes as active catalytic components.
II
Our group has developed synthetic strategies to prepare active [Mn -Schiff base]
oxidation catalysts covalently grafted onto silica-based [32-34] and carbon-based
materials [35-36]. Capitalizing our findings in catalytic oxidations, we point out that for
successful H
2
O
2
incorporation by grafted Mn-complexes, competitive H
2
2
O dismutation,
by either the Mn-center or the inorganic support, should be minimized [32-35]. We have
II
demonstrated also that the catalytic efficiency of such [Mn -Schiff base] catalysts is
switched-on by CH
epoxides [32-35].
3
COONH
4
achieving remarkable effectiveness and selectivity towards
II
Here we present the covalent grafting of Mn -Schiff base catalysts onto mesoporous
nanomaterials MCM-41, SBA-15 and CMK-3 via two synthetic approaches. For the
silicon-based MCM-41 and SBA-15 inorganic supports, the adopted approach involves
two steps: first, Schiff base grafting onto MCM-41 and SBA-15 and second, metalation

of the so-formed [Schiff base@MCM-41] and [Schiff base@SBA-15] hybrid materials.
In the case of CMK-3, our present result revealed that to obtain an active and highly rapid
II
[
[
Mn -Schiff base@CMK-3] catalyst, we had to covalently graft, in one step, the entire
II
Mn -Schiff base] catalyst onto CMK-3. For comparison reasons, the ‘two-steps’
synthetic procedure is also discussed herein for CMK-3. We present detailed catalytic
data demonstrating that the so-obtained ‘single-site heterogeneous catalysts’ are efficient
in alkene epoxidations by H
2
2
O incorporation in the presence of ammonium acetate and
selective towards alkene epoxides formation. Considerations/implications concerning the
textural features of the obtained catalysts and their catalytic performkance are also
discussed herein.
2
. Experimental
Tetraethylorthosilicate (TEOS) 98%, cetyltrimethylammonium bromide (CTAB) 95%,
triblock copolymer Pluronic P123 (EO20PO70EO20, EO = ethylene oxide, PO = propylene
oxide) and sucrose were purchased from Sigma-Aldrich. Hydrochloric acid 37%, sulfuric
acid 95−97% and sodium hydroxide were purchased from Merck, whereas ethanol
(
EtOH) 99.5% and aqueous ammonia solution (NH ) 25 wt% were purchased from
3
Panreac and Fluka, respectively. All substrates were purchased from Aldrich, in their
highest commercial purity, stored at 5 °C and purified by passage through a column of
basic alumina with methanol as the eluting solvent, prior to use. Hydrogen peroxide was
3
0% aqueous solution and was purchased from Sigma-Aldrich.
Infrared spectra were recorded on a Spectrum GX Perkin-Elmer FT-IR System.
Thermogravimetric analyses were carried out by a Shimadzu DTG-60 analyser using a
-1
3
-1
heat rate of 5 K min and a flow rate of synthetic air carrier gas of 50 cm min . The N
2
adsorption–desorption isotherms were measured at 77 K on a Sorptomatic 1990, thermo
Finnigan porosimeter. Specific surface areas were determined with the Brunauer–
Emmett–Teller (BET) method using adsorption data points in the relative pressure (P/Po)
range of 0.05–0.30 and assuming a closely packed BET monolayer. Raman spectra were
recorded with a micro-Raman system RM 1000 Renishaw using a laser excitation line at

−
1
5
32 nm (Nd:YAG) in the range 400−2000 cm . X-ray powder diffraction data were
collected on a D8 Advance Bruker diffractometer using Cu Kα (40 kV, 40 mA, λ = 1.541
8 Å) radiation and a secondary beam graphite monochromator. Elemental analyses (C,
7
H, N) were obtained using a Perkin Elmer Series II 2400 elemental analyzer. The
manganese amount was determined by flame atomic absorption spectroscopy on a
Perkin-Elmer AAS-700 spectrometer. GC analysis was performed using an 8000 Fisons
chromatograph with a flame ionization detector and a Shimadzu GC-17A gas
chromatograph coupled with a GCMS-QP5000 mass spectrometer. H
digitally controlled syringe pump type SP101IZ WPI over 1 h under stirring. Solution
potential E was measured by a Metrohm platinum redox electrode (type 6.0401.100).
Electron Paramagnetic Resonance (EPR) spectra were recorded with a continuous-wave
Brucker ER200D spectrometer at liquid N temperature (77K), equipped with an Agilent
310A X-Band frequency counter. The spectrometer was running under homemade
software based on Lab-View. Al measurements were conducted in quartz EPR tubes of
2
O
2
was added by a
h
2
5
3
mm at liquid nitrogen 77K temperature.
2
.1 Synthesis and characterization of materials
Schiff base Synthesis. The used Schiff base ligand 1,3-bis[3-aza-3-(1-methyl-3-oxobut-
-enyl)-prop-3-en-1-yl]-2-(4-hydroxy-phenyl)-1,3-imidazolidine, so-called L herein, has
1
been synthesized and characterized by our group [37].
MCM-41 Synthesis. The MCM-41 sample was synthesized using CTAB as the
surfactant and TEOS as the silica source [38]. In a typical synthesis, 50 g of TEOS added
to one liter polyethylene bottle containing 417.5 g of H
2
O, 268.5 g of NH (25 wt%) and
3
1
0.5 g of CTAB. The resulting mixture was stirred for 30 min and then was heated for 96
h at 353 K under static condition. The product was filtered, rinsed with cold EtOH, dried
in air and calcined at 823 K for 5 h with a 2 K/min heating rate to remove the template.
-1
DRIFT-IR (cm , selected peaks): 1081: ν(Si-O-Si); 803 (stress), 463 (bend), 1699
destorted), 3441 (free): ν(Si-OH). Thermal analysis shows a weight loss of 0.7 %. This
(
2
-1
material had an average surface area of ca. 1033 m g , see Table 1.

SBA-15 Synthesis. The SBA-15 sample was prepared using the triblock copolymer
pluronic P123 as the surfactant and tetraethylorthosilicate as the silica source, following
synthesis procedure reported by Zhao et al. [9,39]. Specifically, 8.50 g of TEOS added to
a solution of 4.0 g P123 copolymer in 150 g of 1.6 M HCl. The mixture was stirred until
TEOS was dissolved and then was heated for 22 h at 311 K and subsequently for 24 h at
3
68 K under static condition. The product was filtered, dried in air (without washing),
and calcined at 773 K for 6 h with a 1.5 K/min heating rate to remove the template.
-1
DRIFT-IR (cm , selected peaks): 1087: ν(Si-O-Si); 804 (stress), 468 (bend), 3446 (free),
637 (destorted): ν(Si-OH). Thermal analysis exhibits a weight loss of 1.7 %. This
1
2
-1
material showed an average surface area of ca. 712 m g , see Table 1.
CMK-3 Synthesis. The synthesis of CMK-3 carbon was performed according to a hard
template procedure reported from Jun et al. [40,41]. In particular, 1 g of SBA-15 was
added to a solution obtained by dissolving 1.25 g of sucrose and 0.14 g of H
of H O. The mixture was dried for 6 h at 373 K, and subsequently for 6 h to 433 K. The
material was treated again at 373 and 433 K using the same drying procedure after the
addition of 0.8 g of sucrose, 0.09 g of H SO and 5 g of H O. The carbonization was
2
SO in 5 g
4
2
2
4
2
completed by pyrolysis with heating to typically 1173 K under vacuum. After the
carbonization, the silica/carbon material was treated twice with 1 M NaOH in 1/1 (v/v)
solution of EtOH/H
2
O at 373 K to remove the silica template. The obtained product was
filtered, washed with ethanol, and dried at 393 K.
-1
DRIFT-IR (cm , selected peaks): 3426: ν(C-OH); 1707: ν(C=O); 1572: ν(C=C); 1356:
ν(C=O); 1163: ν(C-H) or ν(C-O); 585: ν(O-C=O). Thermal analysis exhibits a weight
loss of 9.4 %, which is attributed to the decomposition of the surface oxygen groups. The
2
-1
average surface area of this material was ca. 1260 m g , as listed in Table 1.
Synthesis of modified mesoporous materials. (a) 1 mmol of ligand L and 0.85 mmol of
(
3-chloropropyl)-trimethoxysilane were refluxed for 48 h in 50 ml of MeOH under N
2
at
0 C leading to L-OS silane-precursor. (b) To this, 1.5 g of silica nanoparticles MCM-
1 or SBA-15 and 5 ml EtOH were added and refluxed at 60 C for 24 h. After cooling at
6
4
room temperature, the resulting material L@MCM-41 or L@SBA-15 respectively was
separated by filtration, washed several times with MeOH and EtOH and dried under

vacuum at 60 C for 12 h. In the case of L@CMK-3, 1.2 mmol of ligand L, 1 mmol of
3-chloropropyl)-trimethoxysilane and 0.5 g of CMK-3 nanoparticles were used applying
the same synthetic protocol.
L@MCM-41: Ligand loading: 0.26 mmol g based on elemental and thermogravimetric
(
-1
-1
analysis. DRIFT-IR (cm , selected peaks): 1088: ν(Si-O-Si); 813(stress), 456(bend),
3
1
1
441(free): ν(Si-OH); 2965, 2861: ν(C-H); 1707: ν(C=O); 1624: ν(C=N); 1446: ν(C=C);
2
-1
371: ν(C-O). This material showed an average surface area of ca. 860 m g , see Table
.
-1
L@SBA-15: Ligand loading: 0.24 mmol g based on elemental and thermogravimetric
-1
analysis. DRIFT-IR (cm , selected peaks): 1088: ν(Si-O-Si); 813(stress), 463(bend),
3
1
1
456(free) ν(Si-OH); 2965,2853: ν(C-H); 1707: ν(C=O); 1624: ν(C=N); 1453 ν(C=C);
2
-1
371: ν(C-O). This material showed an average surface area of ca. 442 m g , see Table
.
-1
L@CMK-3: Ligand loading: 0.34 mmol g based on elemental and thermogravimetric
-
1
analysis. DRIFT-IR (cm , selected peaks): 3404: ν(C-OH); 2935: ν(C-H); 1705: ν(C=O);
1
569: ν(C=N); 1432: ν(C=C); 1354: ν(C-O); 1167: ν(C-C) or ν(C-OH); 620: ν(O-C-O).
2
-1
This material showed an average surface area of ca. 792 m g , see Table 1.
Synthesis of single-site heterogeneous catalysts. 0.3 mmol of MnCl
2
.4H O dissolved in
2
1
0 ml MeOH were added to a suspension of 0.5 g L@MCM-41 in 40 ml MeOH and
stirred overnight at room temperature. The resulting heterogeneous catalyst L@MCM-
II
4
1-Mn was separated by filtration, washed several times with MeOH, EtOH and Et
2
O
and dried at 60 C for 3 h. The amount of the bound manganese was determined by back-
titration of the remaining amount of metal-ion into the solution and by Atomic
Absorption Spectroscopy. Accordingly, the metalation of the hybrid material L@SBA-15
II
II
was achieved resulting in L@SBA-15-Mn catalyst. For L@CMK-3-Mn catalyst, 1.2
mmol of MnCl .4H O were used for 0.5 g of L@CMK-3.
2
2
L@MCM-41-Mn : Metal loading: 0.25 mmol g-1 by flame atomic absorption
II
-1
spectroscopy. DRIFT-IR (cm , selected peaks): 1090: ν(Si-O-Si); 808(stress), 457(bend),
458(free) ν(Si-OH); 2966, 2855: ν(C-H); 1709: ν(C=O); 1633: ν(C=N); 1458: ν(C=C);
3

1
370: ν(C-O). Thermal analysis exhibits a weight loss of 12.9 %, which corresponds to
-1
the combustion of organic ligand L. Ligand loading was found to be 0.28 mmol g based
on elemental and thermogravimetric analysis. This material showed an average surface
2
-1
area of ca. 766 m g , see Table 1.
II
-1
L@SBA-15-Mn : Metal loading: 0.22 mmol g by flame atomic absorption
-1
spectroscopy. DRIFTS-IR (cm , selected peaks): 1079: ν(Si-O-Si); 811(stress),
4
58(bend), 3450(free): ν(Si-OH); 2983, 2854: ν(C-H); 1709: ν(C=O); 1633: ν(C=N);
1
440: ν(C=C); 1370: ν(C-O). Thermal analysis exhibits a weight loss of 10.0 %, which is
attributed to the decomposition of ligand L. Ligand loading was found to be 0.20 mmol g-
1
based on elemental and thermogravimetric analysis. This material showed an average
2
-1
surface area of ca. 414 m g (Table 1).
II
-1
L@CMK-3-Mn : Metal loading: 0.26 mmol g by flame atomic absorption
-1
spectroscopy. DRIFTS-IR (cm , selected peaks): 3405: ν(C-OH); 2983,2927: ν(C-H);
1
5
703: ν(C=O); 1565: ν(C=N); 1432: ν(C=C); 1353: ν(C-O); 1170: ν(C-C) or ν(C-OH);
32: ν(O-C-O). Thermal analysis exhibits a weight loss of 13.4 %, corresponding to the
-1
decomposition of ligand L. Ligand loading was found to be 0.29 mmol g based on
elemental and thermogravimetric analysis. This material showed an average surface area
-1
of ca. 724 g , as listed in Table 1.
Organo-silane metalation. In the case of CMK-3 an alternative method has also been
applied with the metalation step being detoured: The L-OS silane-precursor was formed
according to the procedure described above. (a) 1.2 mmol of ligand L and 1.0 mmol of
(3-chloropropyl)-trimethoxysilane were refluxed for 48 h in 10 ml of MeOH under N
2
at
6
0 C leading to L-OS silane-precursor. (b) Then, after cooling at room temperature, 1.2
mmol of MnCl
2
.4H
2
O dissolved in 10 ml MeOH were added to this solution and stirred
II
overnight at room temperature leading to metal precursor Mn -L-OS. (c) To this 0.5 g of
CMK-3 and 5 ml EtOH were added and refluxed at 60C for 24 h. The resulting material
II
Mn -L@CMK-3 was separated by filtration, washed several times with MeOH, EtOH
II
and Et
2
O dried under vacuum at 60 C for 12 h. The catalyst Mn -L@CMK-3 was

II
formed through hydrolysis and co-condensation of the precursor Mn -L-OS with the –
OH surface functional groups of CMK-3.
II
-1
Mn -L@CMK-3: Metal loading: 0.28 mmol g by flame atomic absorption
-1
spectroscopy. DRIFTS-IR (cm , selected peaks): 3364: ν(O-H); 2922: ν(C-H); 1576:
ν(C=N); 1438: ν(C=C); 1366: ν(C-O); 1244: ν(C-C); 1172: ν(C-C); 614: ν(O-C-O).
Thermal analysis exhibits a weight loss of 14.5 %, corresponding to the decomposition of
-1
ligand L. Ligand loading was found to be 0.32 mmol g based on elemental and
2
-1
thermogravimetric analysis. The average surface area of this material was ca. 35 m g ,
see Table 1.
2
.2 Catalytic Experiments.
Catalytic Protocol. The catalytic solutions were prepared according to the experimental
conditions [catalyst : H : alkene : CH COONH ] = [1:2000:1000:1000 μmol]. The
alkene (as substrate, 1 mmol), acetophenone or bromobenzene (as internal standard, 1
mmol), catalyst (1 μmol of Mn-complex) and CH COONH (as additive, 1 mmol) were
added in a solvent mixture of acetone/MeOH (450 μl /400 μl). H (2 mmol) was added
2
O
2
3
4
3
4
2
O
2
by a digitally controlled syringe pump [SP101IZ WPI] over 30 min to this catalytic
solution. The total volume of the catalytic solution was 1 ml. According to the catalysts’
II
II
loading and their molecular weights, 1 μmol of L@MCM-41-Mn , L@SBA-15-Mn ,
II
II
L@CMK-3-Mn and Mn -L@CMK-3 corresponds to 3.8 mg, 3.7 mg, 3.3 mg and 3.5
mg respectively. The reactions were complete within 24 h, 21 h, 5 h and 1 h for
II
II
II
II
L@MCM-41-Mn , L@SBA-15-Mn , L@CMK-3-Mn
and Mn -L@CMK-3,
respectively. The progress of the reactions was analyzed by GC-MS, for 20μl samples
periodically taken from the reaction mixtures. Quantitative GC-analysis provided the
substrate conversion and product yield by comparing the integrals of the GC peaks vs. the
internal standard integral. When alcohol and ketone were detected as allylic oxidation
products, the sample was treated with PPh and measured again by GC-MS according to
3
Shul’pins’ procedure [42]. To ensure the identity of the epoxide product, the retention
time and spectral data were compared to those of commercially available products. Blank
experiments showed that without Mn-catalyst or CH
not occur.
3
COONH , oxidation reactions do
4

Recycling experiments. The heterogeneous catalysts were tested for their recyclability in
new catalytic experiments using cyclooctene as substrate, under the same experimental
conditions [catalyst: H
2
O
2
: substrate: CH
3
COONH ] = [1:2000:1000:1000 μmol] as
4
mentioned above. After the end of each catalytic reaction, 24 h for reactions catalysed by
II
II
L@MCM-41-Mn , 21 h in the case of the catalyst L@SBA-15-Mn and 5 h and 1 h for
II
II
reactions catalysed by L@CMK-3-Mn and Mn -L@CMK-3, respectively, the solid
catalysts were separated from the reaction mixture by centrifugation, washed three times
o
with MeOH and dried under vacuum (40 C, 12 h). The mass of each recovered solid
catalyst was estimated by mass-weighting. The same mass of each solid was used in each
cycle, by running 5 identical catalytic batches followed by centrifugation/washing/drying
as described above. Thus for the next use we could weigh exactly 3.8 mg, 3.7 mg, 3.3 mg
II
II
II
II
and 3.5 mg of L@MCM-41-Mn , L@SBA-15-Mn , L@CMK-3-Mn and Mn -
L@CMK-3 respectively. The recovered solid catalysts were reused for new catalytic
systems under the same catalytic conditions.
2
.3 EPR study
5
3
mg powder of each material was used for EPR measurements, that were inserted in
mm ID quartz tubes (Willmad Glass). For the estimation of the Mn-leaching after
catalyst use, the catalytic reaction mixture sample was centrifuged, and 100 μL of the
liquid supernatant was used in 3mm ID quartz tubes that were frozen at 77K in liquid N
2
.
II
Quantitative estimation of the Mn -concentration was done by comparison with a MnCl
standard in MetOH, using the double integral of the EPR signal.
2
3
3
. Results and discussion
.1 Synthesis of the catalysts
II
The covalent grafting of Mn -Schiff base catalysts onto mesoporous nanoparticles
MCM-41 and SBA-15 was achieved by applying synthetic steps depicted in Scheme 1: i)
Schiff base ligand L grafting onto MCM-41 and SBA-15 and ii) metalation of the hybrid
[
L@MCM-41] and [L@SBA-15] materials. In the first step (see Scheme 1),

derivatization of the Schiff base ligand L via its phenolic group with (3-chloropropyl)-
trimethoxysilane results to L-OS silane-precursor which is grafted -via hydrolysis and
co-condensation- onto MCM-41 and SBA-15 providing the hybrid materials L@MCM-
4
1 and L@SBA-15 respectively. Subsequently, during the second synthetic step, MnII-
II
II
complexation to grafted ligands furnished the L@MCM-41-Mn and L@SBA-15-Mn
catalysts. The same procedure was applied to CMK-3 giving the corresponding
II
L@CMK-3-Mn catalyst.
However, very recently we have found that metalation of grafted ligands onto carbon-
based supports results in totally inactive materials in catalytic epoxidations [35]. This was
II
attributed to Mn -spoliation by the free –COOH groups of the activated carbon support
II
blocking the desirable Mn -complexation with the grafted ligands in order to form the
active catalytic centre [35]. Here to exclude such possibility when dealing with CMK-3,
an alternative synthetic strategy was adopted which allowed the covalent grafting of the
II
entire [Mn -Schiff base] catalyst onto CMK-3 in one step (see Scheme 2). More
specifically derivatization of the Schiff base ligand L forms the L-OS silane-precursor;
II
II
then its metalation with Mn -ions results to Mn -L-OS. The so-formed metal complex
II
Mn -L-OS is grafted to the solid support CMK-3 by its propyl-trimethoxysilane
moieties which allow covalent attachment via hydrolysis and co-condesation with the -
OH surface functionalities of CMK-3 (Scheme 2). This catalyst, prepared by grafting of
II
II
the preformed Mn -complex to CMK-3, is herein codenamed as Mn -L@CMK-3
II
(
Scheme 2). Characterisation and catalytic behaviour of both L@CMK-3-Mn and
II
Mn -L@CMK-3 materials is detailed and discussed hereafter.
3
.2 Structural and textural characterisation
FT-IR study: The DRIFT-IR spectra of the developed catalysts show characteristic bands,
which are attributed to various structural components of the present materials, i.e. to the
support scaffold as well as to the grafted Schiff base ligand L. That is, when the DRIFT-
IR spectra of L@MCM-41, L@SBA-15 and L@CMK-3 were compared to those of the

parent materials MCM-41, SBA-15 and CMK-3 respectively, new bands can be
observed which are attributed to the anchored ligand L. More specifically, in the spectra
of grafted ligand L@MCM-41, L@SBA-15 and L@CMK-3, the bands observed at
-1
1
707, 1707 and 1705 cm respectively are attributed to (C=O) stretching vibrations of
the Schiff base ligand L. The band, which is observed in both L@MCM-41 and
-1
L@SBA-15 hybrids at 1624 cm , is assigned to the ν(C=N) stretching of the Schiff base
-1
L as well; this band, in the case of L@CMK-3 material, is shifted at 1569 cm and
highly contributed by conjugated ν(C=C) vibrations from the carbonaceous CMK-3
II
II
support. In the spectra of both catalysts L@MCM-41-Mn and L@SBA-15-Mn , the
-1
abovementioned vibrations (C=O) and ν(C=N) are observed at 1709 and 1633 cm ,
being clearly shifted when compared with the non-metalated materials L@MCM-41 and
II
L@SBA-15; this behaviour indicates coordination of the Mn with the imine-nitrogen
II
and keto-oxygen of the attached ligand L [33,35,43]. In the spectra of Mn -L@CMK-3,
the (C=O) band was slightly shifted when compared with the corresponded band of the
non-metalated L@CMK-3 material; however, the ν(C=N) stretching, which is more
-1
sensitive in metalation, is observed at 1576 cm taking in to account that, in L@CMK-3,
-1
II
is observed at 1569 cm . Based on these, we conclude that, in Mn -L@CMK-3 catalyst,
II
the Mn is selectively bound to the grafted Shiff base ligand L ensuring the presence of
the entire active catalytic centre onto the CMK-3 surface [35]. As concerned the
II
vibrations (C=O) and ν(C=N) of L@CMK-3-Mn , they are shifted as well, when
compared with the corresponding bands of the non-metalated ligand L@CMK-3,
II
indicating also coordination of the Mn with attached ligand L [33,35,43].
II
Thermal analysis: The thermograms of the silica-based catalysts L@MCM-41-Mn and
II
L@SBA-15-Mn show a 13.6 % and 13.5 % weight loss respectively, in the range of
1
30-400 °C, which are attributed to the decomposition of the anchored organic molecules
(Figure 2 A, B). The pristine CMK-3 material is dominantly composed by carbon and
oxygen, with a percentage content of 87.9% and 9.4%, respectively. The surface oxygen
groups are generated from the incomplete dehydration of the sucrose carbon source [31].
II
II
The thermograms of the L@CMK-3-Mn and Mn -L@CMK-3 (Figure 2 C, D)

materials show a 22.3 % and 23.9 % weight loss respectively, in the range of 130-400 °C.
Given that CMK-3 support bears 9.4-10.0 % surface oxygen functionalities which are
decomposed in the same temperature region (130-400 °C), the rest weight loss of 13.4
II
II
and 14.5 % for L@CMK-3-Mn and Mn -L@CMK-3 respectively can be reasonably
ascribed to the decomposition of the anchored organic ligand L (See Table 1).
Nitrogen Sorption Isotherms: N adsorption-desorption isotherms for all materials are
2
shown in Figures 3a, 4a and 5a; the Brunauer-Emmett-Teller (BET) surface area values
as well as the pore characteristics of pristine and modified materials are listed in Table 1.
In case of the silicon-based catalysts (Figure 3a and 4a), all isotherms are type IV which
is characteristic of mesoporous materials; moreover, the adsorption curve exhibit a well-
defined capillary condensation step indicating filling of uniform mesopores. The
beginning of this step for the L@MCM-41, L@MCM-41-MnII and L@SBA-15,
II
L@SBA-15-Mn materials occurs at a lower pressure range compare to the pristine
materials MCM-41 and SBA-15 respectively. This behavior suggests a decrease in pore
II
size [44,45] due to the Shiff base ligand L grafting and subsequent Mn -coordination.
Indeed, from the BJH pore size distribution (Figure 3b and 4b), a significant decrease in
pore diameter and total pore volume is observed in modified materials compare to the
pristine materials (Table 1). In case of the pristine CMK-3 and the carbon-based
II
materials L@CMK-3 and L@CMK-3-Mn , which were prepared by the same synthetic
II
II
procedure as the L@MCM-41, L@MCM-41-Mn and L@SBA-15, L@SBA-15-Mn
materials (see Scheme 1), the isotherms are also type IV. However, the isotherm of MnII-
II
L@CMK-3, this was prepared after grafting of the entire Mn -Schiff base catalyst onto
CMK-3 in one step, indicates a nonporous material. The specific surface area for this
II
2 -1
Mn -L@CMK-3 catalyst shows a dramatic collapse, e.g., from 1260 m g in pristine
2
-1
II
CMK-3 to 35 m g in Mn -L@CMK-3. We notice that analogous behaviour was
II
showed when the entire Mn -Schiff base catalyst was grafted by similar method onto the
surface of activated carbon ACox [35] where the specific surface area of ACox of 687
2
-1
2
-1
II
m g became 21 m g in Mn -L@ACox indicating formation of compacted carbon
layers with practically zero porosity [35].

2
-1
The specific surface area (SSA in Table 1) of 1033, 712 and 1260 m g for the
pristine MCM-41, SBA-15 and CMK-3 materials respectively, was decreased
significantly after Shiff base ligand L grafting onto their surface, exhibiting surface area
2
-1
8
60, 442 and 792 m g for L@MCM-41, L@SBA-15 and L@CMK-3 respectively. Α
II
further decrease was observed after Mn -coordination, calculated to 766, 414 and 724
2
-1
II
II
II
m g for L@MCM-41-Mn , L@SBA-15-Mn and L@CMK-3-Mn respectively, see
Table 1. This 26-43% decrease of the specific surface area provides evidence that the
II
pores are filled by the Mn -complex without blocking the pores [8].
II
II
Raman study: Raman spectra of CMK-3, L@CMK-3, L@CMK-3-Mn and Mn -
L@CMK-3 materials are shown in Figure 6. The spectra of all samples exhibit the two
−
1
−1
characteristic peaks of carbon-based materials at 1598 cm and 1340 cm corresponding
to the well-known G- and D-band vibrations, respectively. More specifically, the first
−
1
band at 1598 cm (G-band) derives from the in-plane-bond stretching motion of pairs of
sp carbon atoms, while the second band at 1340 cm−1 (D-band) which only becomes
2
2
active in the presence of disorder, derives from sp edge plane breathing vibrations [46].
The relative intensity ratio of D and G band (I
D
/I ) has been constantly used as a measure
G
of structural graphitization order and it is a tool to identify structural changes in carbon
materials due to their treatment. This ratio has similar value ~0.9 for all samples and it is
typical for CMK-3 structures [47] indicating a mainly disordered carbon structure.
Overall these Raman data suggest that the graphitized features of carbon remains
unimpaired after the grafting procedure.
Powder X-ray diffraction (XRD): The XRD patterns of all samples are shown in Figure 7.
In case of pristine materials MCM-41, SBA-15 and CMK-3 the patterns exhibit three
characteristic peaks which can be indexed to the (100), (110) and (200) reflections of the
hexagonal space group (P6mm). This indicates a hexagonal arrangement of the
mesopores of MCM-41 and SBA-15 and a hexagonal arrangement of interconnected
carbon nanorods in case of CMK-3, which is the negative replica of SBA-15. These
peaks are still observed in L@MCM-41, L@SBA-15, L@CMK-3, L@MCM-41-MnII,
II
II
L@SBA-15-Mn and L@CMK-3-Mn indicating that the ordered structure of pristine

II
materials was retained after the grafting of the Mn -Schiff base catalyst on their surface.
II
On the other hand, these reflections are not visible in the pattern of Mn -L@CMK-3.
However, such behavior does not necessarily suggest the damage of the structural order
of CMK-3 support; it could be correlated to some density effects originating from the
material constituting the carbon walls and the material filling/coating the pore regions.
Similar effect has already been discussed in previous studies dealing with metal oxide
incorporation into the pores of silicate and carbon mesostructures [41, 48-51]. According
to these studies the loss in intensity was due to the introduction of scattering material into
the pores which leads to an increased phase cancellation between scattering form walls
and the pore regions [41, 48-51]. That is, the absence of XRD reflections can be
II
attributed to a very high filling of the pores. In our case here, L@CMK-3-Mn presents
II
reflections indicating a hexagonal arrangement as expected, while Mn -L@CMK-3
losses these peaks. Given that the experimental conditions for the synthesis of MnII-
II
L@CMK-3 is analogous for that of L@CMK-3-Mn , putative damage of structural
order seems to be unfounded. Thus, the loss of the reflections probably originated from
II
the pore filling of CMK-3 with the active Mn -L catalyst. The direct question, why the
ligand L doesn’t fill the CMK-3 pores –see for example the L@CMK-3 material-, while
II
the entire Mn -L catalyst does, it could be addressed taking into account a possible better
II
mobility of the Mn -L complex within the pores vs. the organic ligand L alone.
3
.3 Catalytic Study
II
II
II
The developed Mn -containing materials L@MCM-41-Mn , L@SBA-15-Mn ,
L@CMK-3-MnII and Mn -L@CMK-3 were evaluated as catalysts for alkene
II
epoxidation with H
2
O
2
. Typical reaction conditions employed in these experiments were
30% delivered by the syringe pump during 30
1
equiv. of catalyst, 2000 equiv. of H
2
O
2
min into an [acetone:methanol] (0.45ml:0.40ml) solution, 1000 equiv. of substrate and
ammonium acetate (1000 equiv.) at room temperature. The heterogeneous catalysts
described here required ammonium acetate as additive to generate high catalytic activity
II
in analogous way to previous catalysts based on non-heme-Mn -complexes [32-35,52].

II
Catalytic Oxidation Yield and TONs: The heterogeneous Mn -based catalysts presented
significant catalytic activity and high selectivity towards formation of epoxides (the mass
balance is 98-100%) in most of the cases (see Table 2). Thus, cyclooctene oxidation
provides a 100% selectivity for cis-cyclooctene epoxide with 56.0%, 24.5%, 24.0% and
II
II
5
0.6% yield (100% m.b.) catalysed by L@MCM-41-Mn , L@SBA-15-Mn ,
II
II
L@CMK-3-Mn and Mn -L@CMK-3 respectively (see Figure 10 A). Cyclohexene
epoxide formation, from cyclohexene oxidation, was found to be 50.2%, 46.0%, 33.0%
II
II
and 37.8% catalysed respectively by L@MCM-41-Mn , L@SBA-15-Mn , L@CMK-
II
II
II
3
-Mn and Mn -L@CMK-3; however, in epoxidation catalysed by Mn -L@CMK-3, a
small amount of 2-cyclohexene-1-ol (3.0%) was detected derived probably by allylic
oxidation of cyclohexene. This amount of 2-cyclohexene-1-ol was identical to that for
PPh -treated sample following the Shul'pins’ procedure [42]; this treatment ensures that if
3
any hydroperoxides have been formed as primary-product, this has been completely
transformed into the alcohol- and ketone-oxidation products [42,53]. Two epoxides (cis-
and trans-) were the main products derived from epoxidation of the electron-rich double
bond in 1,2-position of limonene. The total yield of 1,2-epoxides was 70.0%, 51.7%,
II
II
II
4
1.6% and 63.4% catalysed by L@MCM-41-Mn , L@SBA-15-Mn , L@CMK-3-Mn
II
and Mn -L@CMK-3 respectively. Considerable amounts of products formed by allylic
oxidation of the limonene ring have also been observed and have been identified as the
corresponding derivative of cyclohexene-1-ol (7.9-13.7%). No formation of epoxides
derived from the more accessible but less electron-rich double bond in 8,9-position has
been observed. However, the epoxidation was clearly the major reaction path resulting
mainly in 1,2-epoxides (see Table 2, Figure 8 A). Thus, the total yield of catalytic
limonene oxidation (1,2-epoxide plus cyclohexene-1-ol derivative) was varied from
4
9.7% to 77.9%. Hexene-1, which contains a less electron-rich double bond, is a rather
hard oxidation substrate, showing low epoxide yields 6.9%, 6.0%, 4.9% and 6.5% for
L@MCM-41-MnII, L@SBA-15-MnII, L@CMK-3-MnII and Mn -L@CMK-3
II
II
respectively. Oxidation of cis-stilbene catalysed by L@MCM-41-Mn , L@SBA-15-
II
II
II
Mn , L@CMK-3-Mn and Mn -L@CMK-3 provides only cis-epoxide with 46.0%,
8.4%, 45.8% and 18.5% yield respectively. All catalysts studied herein showed 100%
selectivity towards the formation of epoxides in catalytic oxidations of styrene, trans-β-
1

methylstyrene, a-pinene, cyclopentene and 1-methyl-1-cyclohexene with yields 3.0-
1.2%, 10.0-31.5%, 27.9-41.8%, 19.5-37.3% and 13.3-36.6% respectively (Table 2,
Figure 8 A, B). In these cases a superior catalytic performance has been achieved by
1
II
II
L@MCM-41-Mn and Mn -L@CMK-3 heterogeneous catalysts. Detailed catalytic
data are provided in Table 2.
II
II
Comparing the TONs achieved by L@MCM-41-Mn , L@SBA-15-Mn , L@CMK-3-
II
II
II
Mn and Mn -L@CMK-3, we observe that L@MCM-41-Mn provides the highest
TON values for all evaluated substrates with exception trans-β-methylstyrene and
cyclopentene epoxidation (Figure 9 A, B). This remarkable catalytic performance is
II
outcompeted by the carbon-based catalyst Mn -L@CMK-3. When we compare the
II
II
TONs achieved by silicon-based Mn -catalysts L@MCM-41-Mn and L@SBA-15-
II
II
Mn , we notice in general, a superior catalytic performance of L@MCM-41-Mn
Figure 9 A, B). When, in analogous way, we compare the TONs yielded by catalysts
(
II
II
attached on CMK-3 matrix, L@CMK-3-Mn and Mn -L@CMK-3, we observe higher
II
TONs values in epoxidations catalysed by Mn -L@CMK-3 with except the cis-stilbene
II
II
epoxidation. Mn -L@CMK-3 was prepared by grafting of the preformed entire Mn -
II
Schiff base complex to CMK-3, while L@CMK-3-Mn was developed by a two-step
synthetic procedure constituted by ligand grafting (i) and followed by metalation (ii). It is
highlighted that when this two-step synthetic strategy was adopted for prepared
II
analogous Mn -catalysts supported on activated carbon surface ACox, the obtained
catalysts were totally inactive in alkene epoxidations with H
2
2
O . This was attributed to
II
Mn -spoliation by the free –COOH groups of the activated carbon support blocking the
II
II
formation of the active Mn -L [35]. However, here the corresponding L@CMK-3-Mn
hybrid developed on carbonaceous CMK-3 matrix presents considerable catalytic
II
performance which is equivalent to/or outcompetes L@SBA-15-Mn catalyst (Figure 9
II
A, B). The observed catalytic activity of L@CMK-3-Mn further strengths our spectral
II
data that the active Mn -L catalyst has been formed. This happened because the surface
II
oxygen functionalities offered by the non-oxidized CMK-3, are not able to trap Mn , i.e.
inhibiting in this way, the formation of active catalyst. Thus, probably the lack of

carboxylates on the surface of CMK-3 permitted to develop the active carbon-based
II
catalyst L@CMK-3-Mn by the two-step synthetic procedure.
Catalytic Kinetics and TOFs: Figure 10 presents the time-course profiles of the catalysts
II
II
II
II
L@MCM-41-Mn , L@SBA-15-Mn , L@CMK-3-Mn and Mn -L@CMK-3 during
oxidation reaction of cyclohexene, in conjunction with the redox potential of solution E
h
II
(
versus standard hydrogen electrode SHE). In the case of L@MCM-41-Mn , E
h
was
+388 mV at the beginning of the reaction; after 1h, it dropped at +265 mV with a 31.6%
total oxidation yield and eventually, after 24 h reaction time, it approached a E =+223
h
II
mV and a 50.2% yield. The L@SBA-15-Mn -catalysed reaction gave at t=0, +388 mV,
dropping at +263 mV after 1 h with a yield 26.7% and finally, after 21 h reaction time,
II
the values were E
h
=+225 mV with a 46.0% yield. In the case of L@CMK-3-Mn , at t=0,
h
E was +388 mV, decreasing to +248 mV after 1 h with a total yield 27.2% and finally
after 5 h, E
h
approached +221 mV with a 33.0% cyclohexene epoxide formation. And at
II
the beginning of the reaction catalysed by Mn -L@CMK-3, E
h
was +385 mV and after
1
h reaction time, it decreased at E =+225 mV with a total 37.8% epoxide yield. The
h
time-course profiles of the catalysts are analogous for all the substrates that have been
used. That is, the oxidation reactions catalysed by L@CMK-3-MnII and MnII-
L@CMK-3 were practically accomplished within 5 h and 1 h respectively and the
II
II
reactions catalysed by L@MCM-41-Mn and L@SBA-15-Mn were accomplished
within 24 h and 21 h respectively.
Taking into account these kinetic data and TONs achieved by each catalyst, we observe
II
that the carbon-based heterogeneous catalyst Mn -L@CMK-3 shows remarkably higher
-1
II
TOFs (ranged from 65-634 h ) vs. the carbon-based catalyst L@CMK-3-Mn (TOFs
-1
9
.8-99.6 h ) and even higher when compared to those achieved by the silicon-based
II
-1
II
-1
L@MCM-41-Mn (TOFs 2.9-32.5 h ) and L@SBA-15-Mn (TOFs 1.4-29.6 h )
catalysts (Figure 11). These findings are in accordance with our previous studies, where
carbon-based heterogeneous catalysts showed remarkably higher TOFs vs. the
corresponding silica-based catalysts [35]. Moreover, the present carbon-based catalyst
II
onto CMK-3, Mn -L@CMK-3, shows even higher TOFs than the corresponding

II
II
catalyst onto activated carbon ACox, Mn -L@ACox [35]. In general, Mn -L@CMK-3
II
catalyst provides slightly lower TONs that the Mn -L@ACox, however, it is more rapid
II
than Mn -L@ACox resulting finally in higher TOFs. It is possible that this feature is
correlated with the practically zero porosity of these catalysts, i.e. specific surface area of
2
-1
II
2
-1
II
2
1 m g for Mn -L@ACox and 35 m g for Mn -L@ACox. This suggestion is further
II
2 -1
supported by the fact that L@CMK-3-Mn catalyst with SSA=724 m g exhibited
II
II
activity comparable to Mn -L@ACox provides 5 to 10-times lower TOFs than Mn -
L@ACox.
Recyclability of the Catalysts: Reuse of heterogeneous catalysts remains always a
challenge in catalytic technology. Here, the heterogeneous catalysts L@MCM-41-MnII,
II
II
II
L@SBA-15-Mn , L@CMK-3-Mn and Mn -L@CMK-3 were tested for their
recyclability by repeated re-covering and re-using of the solid catalysts in new catalytic
systems
catalyst:H
acetone/MeOH (450 μl /400 μl) solvent mixture. After each catalytic run, which was
for
cyclooctene
epoxidation.
A
[
2
O
2
:substrate:additive]=[1:2000:1000:1000] molar ratio was used in an
II
II
completed in 24 h for L@MCM-41-Mn , 21 h for L@SBA-15-Mn and 5 h and 1 h for
II
II
L@CMK-3-Mn and Mn -L@CMK-3 respectively, each catalyst was recovered by
centrifugation, washed, dried and reused under the same catalytic conditions given in
Table 3.
II
According to Table-3, the carbon-based catalysts Mn -L@CMK-3 and L@CMK-3-
II
Mn presented some recyclability, in contrast to the silicon-based catalysts L@MCM-
II
II
4
1-Mn and L@SBA-15-Mn which showed a total loss of their catalytic activity
II
(
>99%). More specifically, Mn -L@CMK-3 presented <50% loss of the catalytic
activity during the 2nd run achieving overall 721 TONs for cyclooctene oxidation and
II
L@CMK-3-Mn showed some activity during the 2nd catalytic run.
Monitoring of the state of the solid bearing the Mn-catalyst and Mn-leaching in has been
done with two independent protocols as follows:

[
i] monitoring leaching of catalytically active species in solution: a) filtering out the
solid catalyst after 1 or 1/2h of reaction and then b) monitoring the eventual catalytic
prformance of the filtrate's solution by GC [33]. The data showed that no evolution of the
II
studied reactions was observed in the filtrate. This proved that no leaching of [Mn -L]
occurs.
[
ii] In accordance with the results in method [i], EPR spectroscopy, revealed that MnII-
atoms (not Mn-L) did leach in the solution. More specifically, EPR was used to study the
Mn-centers in our Mn-catalysts, unused and recycled, as well as the filtrate’s solution of
the catalytic reactions.
In Figure 12A (solid lines) we present 77K EPR spectra of the ‘as prepared’ unused
II
II
II
catalysts L@CMK-3-Mn , Mn -L@CMK-3, L@SBA-15-Mn and L@MCM-41-
II
Mn . The carbon-based materials are characterized by a broad Gaussian derivative
centered at g ~2, while the Si-based materials, SBA and MCM-41 show 6-line EPR
II
spectra with resolved Mn (S=5/2, I=5/2) hyperfine splittings. Thus EPR shows that the
II
L@CMK-3-Mn sample bears highly clustered Mn-complexes resulting in the strong
II
II
broadening that masks the Mn (S=5/2, I=5/2) hyperfine splittings. In contrast Mn -
L@CMK-3 shows a Mn-spectrum with a moderate surface clustering. The L@SBA-15-
II
II
(II)
Mn and L@MCM-41-Mn show resolved Mn (S=5/2, I=5/2) spectra. The resolved
5
5
semiforbidden Mn-hyperfien features i.e. the weak sharp lines between the main sextet
II
lines, are typical for the monomeric Mn -L complex, showing that these signals
II
originate mainly from the active catalytic component [Mn -Schiff base] grafted on a solid
matrix [36].
II
The EPR data in Figure 12A show that the Mn-concentration on L@SBA-15-Mn and
II
II
L@MCM-41-Mn is lower than in CMK-3 based materials. L@SBA-15-Mn has the
lower Mn-content. These trends are in accordance with the loadings reported in the Table
1
.
II
In Figure 12B we present the % of Mn leached in the filtrate’s solution after a first use
II
II
of our catalysts, i.e. calculated as the ratio of Mn released in solution vs. the initial Mn
in each catalytic material. From the lineshape of the EPR spectra we conclude that the

II
II
II
leached Mn -species corresponds to free Mn solvated ion. The Mn remaining on the
recycled materials is detected by EPR in Figure 12A(dotted-lines). Accordingly, the data
II
II
in Figure 12B reveal that L@CMK-3-Mn shows the highest percentage of Mn -ions
II
II
released in the catalytic solution, while L-Mn -L@CMK-3 shows the lowest Mn -ions
II
II
leaching. This indicates that in L@CMK-3-Mn a significant amount of Mn -ions were
loosely associated with the CMK-3 surface groups, other than L. That is, .addition of Mn
ions in the L@CMK-3 material, results in casual Mn-coordination on CMK-3 surface
II
sites, not only on L-ligands. In contrast Mn -L @ CMK-3 seems to be more resilient to
leaching i.e. since all the Mn-centers are anchored on the CMK-3 surface via the
II
II
preformed [Mn -Schiff base] complex. This specific Mn-coordination in Mn -
L@CMK-3 and its lower-loss of Mn-ions is in accordance with the better activity and
(
II)
(II)
recyclability respectively of the L-Mn @ CMK-3 vs, the L@ CMK-3-Mn .
II
II
As concerned the L@SBA-15-Mn and L@MCM-41-Mn , considerable amounts of
II
Mn -ions have been detected in their filtrate’s solution after a first use (Figure 12B). For
II
example, the L@MCM-41-Mn despite its good catalytic performance during the first
II
catalytic use shows 55% of Mn-loss after this use, while L@SBA-15-Mn has 22% Mn-
II
loss of Mn . At this point we notice, however, this Mn-ions leaching from L@SBA-15-
II
II
Mn and L@MCM-41-Mn does not fully explain their complete inhibition inactivity
during the second catalytic run i.e. since a non negligible fraction of Mn remained on the
5
5
solid matrix. In this context we undertook a close inspection of the Mn-EPR spectrum,
II
II
and this revealed the in L@MCM-41-Mn the Mn centers remaining on the solid, have
an EPR hyperfine pattern where the semiborbidden, weak-sharp signals superimposed on
the main hyperfine line are lost see inset in Figure 12A. This indicates that these MnII-
centrers although still associated to the MCM-41 matrix after catalytic use, they are not
any more coordinated to L, but have a no-specific coordination environment. This crucial
II
observation points to an oxidative destruction of the active [Mn -Schiff base] catalytic
II
II
center of L@MCM-41-Mn and L@SBA-15-Mn i.e. caused by their longer residence
under strongly oxidative conditions i.e. 24 and 21 h respectively. In contract in the
II
II
recycled L-Mn @CMK-3, the remaining Mn -centers seem to retain their coordination

to L, as in the unused catalyst, further substantiating the observed catalytic activity in a
second run.
Performance considerations/-implications:
II
From the synthetic point of view, the covalent grafting of Mn -Schiff base catalysts onto
mesoporous silicon-based supports MCM-41 and SBA-15 was achieved by applying two
discrete synthetic steps: i) Schiff base ligand L grafting onto MCM-41 and SBA-15 and
II
ii) metalation of the hybrid materials obtaining the L@MCM-41-Mn and L@SBA-15-
II
Mn catalysts. This two-step procedure, when applied on carbon-based supports, resulted
in totally inactive materials in catalytic epoxidations. This was because of MnII
coordination to the free carboxylate groups of the activated by oxidation carbon supports.
To affront this, we have developed an alternative synthetic strategy which allowed the
II
covalent grafting of the entire [Mn -Schiff base] catalyst onto the carbonaceous matrix in
one step. Here, both one- and two-step procedures have been applied to CMK-3 resulting
II
II
in Mn -L@CMK-3 and L@CMK-3-Mn materials respectively; based on our findings
here, both are active in alkene catalytic epoxidations with H . This is attributed to the
2
O
2
nature of the CMK-3 carbonaceous support; the non-oxidized CMK-3 offers oxygen
functionalities for surface grafting, at the same time being devoid of carboxylates that
II
would trap Mn , thus allowing the formation of the active catalyst.
II
II
Here we can make a distinction for the materials L@MCM-41-Mn , L@SBA-15-Mn
II
and L@CMK-3-Mn prepared by the two-step procedure. Based on the present BET,
II
XRD data, we conclude that, after the grafting of the Mn -Schiff base catalyst on their
surface, a) the ordered structure of pristine materials was retained and b) they are still
porous. A 26-43% decrease of SSA indicates that the pores are filled by the MnII-
complex without blocking the pores.
II
According to the one step procedure, the Mn -L@CMK-3 catalyst has been prepared.
II
Mn -L@CMK-3 is a nonporous material which lost the XRD reflections of the pristine
CMK-3. This behavior, as discussed herein, does not necessarily suggest the damage of
the structural order of CMK-3 and it has been probably originated from the pore filling of
II
CMK-3 with the active Mn -L catalyst.

Comparing the experimental conditions between the one-step and two steps synthetic
processes, both require similar thermal treatment and starting materials. It is obvious that
II
a putative damage of structural order in Mn -L@CMK-3 is unfounded. However, the
raised question why the one step procedure leads to nonporous material is still open. In
II
one step synthesis, the grafting moiety is the entire Mn -L complex, while in the two step
II
one, the initially grafting component is the organic ligand L. If, in Mn -L@CMK-3,
II
Mn -L fills the pores of CMK-3, why, in L@CMK-3, L does not fill the pores
completely, as indicated by SSA, blocking them? With data at hand, an acceptable
II
answer is a possible better mobility and/or affinity of the Mn -L complex within the
pores of CMK-3 vs. the organic ligand L. This could result, in the case of MnII-
L@CMK-3, in completely filling of the pores and the formation of a compact structure.
More studies are needed to prove and control this phenomenon.
The present single-site heterogeneous Mn-catalysts have been evaluated for alkene
epoxidation with H
2
O
2
in presence of the required CH
3
COONH . In these systems also,
4
II
CH
3
COONH
4
plays a key role on the non-heme-Mn -catalysts worked with H
2
2
O ,
II
generating the catalytic efficiency of Mn -catalysts as reported in our previous studies
-
[
32-35, 37, 52, 54]. In details, firstly CH
3
COO abstracts proton from H
2
O
2
promoting its
+, by acting
II
II
coordination to Mn and formation of Mn -OOH species; subsequently, NH
4
II
as a proton-donor to Mn -OOH, accelerates heterolytic O-O cleavage forming the active
IV
Mn =O species which finally catalyses alkene epoxidation [35].
To have an overview of the performance of the developed here L@MCM-41-MnII,
II
II
II
L@SBA-15-Mn , L@CMK-3-Mn and Mn -L@CMK-3 catalysts, it is noted that
II
L@MCM-41-Mn provides the highest TONs values for almost all the evaluated
substrates (Figure 9 A, B). This remarkable catalytic performance is outcompeted by the
II
carbon-based catalyst Mn -L@CMK-3 which presents by far one of the best catalytic
performances of Mn-Schiff base complexes covalently grafted onto CMK-3 matrix with
II
II
H
2
O [31]. When, we direct compare the TONs yielded by L@CMK-3-Mn and Mn -
2
L@CMK-3, we observe higher TONs values in epoxidations catalysed by MnII-
II
L@CMK-3. Mn -L@CMK-3 shows also remarkably higher TOFs (ranged from 65-634
-1
II
-1
h ) vs. the L@CMK-3-Mn catalyst (TOFs 9.8-99.6 h ) and even higher when

II
-1
compared to those achieved by L@MCM-41-Mn (TOFs 2.9-32.5 h ) and L@SBA-15-
II
-1
Mn (TOFs 1.4-29.6 h ) catalysts. Practically, the reactions catalysed by L@CMK-3-
II
II
Mn and Mn -L@CMK-3 were accomplished within 5 h and 1 h respectively and the
II
II
reactions catalysed by L@MCM-41-Mn and L@SBA-15-Mn were accomplished
within 24 h and 21 h respectively. Generally, this trend is in accordance with our
previous studies where carbon-based heterogeneous catalysts showed remarkably higher
TOFs vs. the corresponding silica-based catalysts [35], attributable to the nature of the
II
supports. However, comparing the two CMK-3-based catalysts Mn -L@CMK-3 and
II
II
L@CMK-3-Mn , the observed high rapidity of Mn -L@CMK-3 is probably correlated
with its practically zero porosity. Such effect could be attributable to possible diffusion
limits of reagents and/or products into the porous materials served as catalysts in the
epoxidation reactions studied.
Overall the percent data provide critical clues on the factors that determine the overall
catalytic performance of the hybrids: [i] amount of Mn-coordinated on L vs. Mn-
coordinated on non-specific surface sites, [ii] resilience of the Mn-coordinated on L to
catalytic harassment plays a key role, [iii] porosity profile is key-factor in heterogeneous
catalytic reactions controlling the reaction time.
4
. Conclusions
In this study, we present the development of new heterogeneous catalysts via covalent
attachment of Mn(II)-Shiff base complexes onto mesoporous MCM-41, SBA-15 and
CMK-3 supports. This has been done via two different synthetic protocols: a) a two steps
procedure which includes grafting of the Schiff base ligand L onto the supports and
subsequent metalation of the hybrid materials and b) an one step procedure which allows
II
the covalent grafting of the entire Mn -Schiff base catalyst onto the carbonaceous
II
support. The two steps procedure resulted in the porous L@MCM-41-Mn , L@SBA-
II
II
1
5-Mn and L@CMK-3-Mn single-site Mn-catalysts and the one step procedure
II
provided the non-porous catalyst Mn -L@CMK-3. All the catalysts were evaluated for
alkene epoxidation with H
2
O
2
as oxidant and CH
3
COONH as additive. They are
4
II
effective indicating that the active Mn -Schiff base catalytic component is still functional
after the grafting procedure and selective towards formation of epoxides. The highest

II
TONs, in most of the tested substrates, have been provided by L@MCM-41-Mn and
II
II
Mn -L@CMK-3. The latter catalyst Mn -L@CMK-3 is operative for a second use and
kinetically very fast demonstrating extremely high TOFs. This fast kinetics of MnII-
L@CMK-3 in the catalytic epoxidations studied herein is correlated to its no-porosity.
Further studies are in progress to elucidate, control and capitalise this important
component in heterogeneous catalysis.
Acknowledgment
This research has been co-financed by the European Union (European Social Fund –
ESF) and Greek national funds through the Operational Program "Education and
Lifelong Learning" of the National Strategic Reference Framework (NSRF) - Research
Funding Program: THALIS. Investing in knowledge society through the European Social
Fund.

References
[
1] M.G. Clerici, O.A. Kholdeeva, Liquid Phase Oxidation via Heterogeneous Catalysis:
Organic Synthesis and Industrial Applications, Wiley (2013)
[
[
2] J. M. Thomas, R. Raja, D.W. Lewis, Angew. Chem. Int. Ed. 44 (2005) 6456-6482
3] J. M. Thomas, Design and applications of single-site heterogeneous catalysts:
contributions to green chemistry, clean technology and sustainability. London, UK:
Imperial College Press and World Scientific Press (2012)
[
[
[
[
[
4] D.E. de Vos, M. Dams, B. F. Sols, P. A. Jacobs, Chem. Rev. 102 (2002) 3615-3640
5] M.H. Valkenberg, W.F. Holderich, Catal. Rev. 44 (2002) 321-374
6] A. Corma, Catal. Rev. 46 (2004) 369-417
7] B. S. Lane, K. Burgess, Chem. Rev. 103 (2003) 2457-2474
8] K. Dimos, P. Stathi, M.A. Karakassides, Y. Deligiannakis, Microporous and
Mesoporous Materials 126 (2009) 65-71
[
9] ] D. Zhao, J. Feng, Q. Huo, N. Melosh, G. H. Fredrickson, B. F. Chmelka, G. D.
Stucky, Science 279 (1998) 548-552
[
[
[
10] R. Ryoo, SH. Joo, M. Kruk, M. Jaroniec, Adv Mater 13 (2001) 677–681
11] A. Vinu, T. Mori, K. Ariga, Sci. Tech. Adv. Mater. 7 (2006) 753−771
12] H.J. Shin, R. Ryoo, M. Kruk, M. Jaroniec, Chem. Commun. (2001) 349−350
[
[
13] M. Kruk, M. Jaroniek, R. Ryoo, S.H. Joo, J. Phys. Chem. B. 104 (2000) 7960-7968
14] L. Li, Z. H. Zhu, G. Q. Lu, Z.F. Yan, S.Z. Qiao, Carbon 45 (2007) 11−20
[
15] Q. Sui, J. Huang, Y. Liu, X. Chang, G. Ji, S. Deng, T. Xie, G. Yu, J. Environ. Sci.
2
3(2) (2011) 177−182.
[
16] Z. Wu, P.A. Webley, D. Zhao, Langmuir 26(12) (2010) 10277−10286

[
[
[
17] G. Chandrasekar, W.J. Son, W. Ahn, J.Porous. Mater. 16 (2009) 545−551
18] F. Lufrano, P. Staiti, Energy Fuels 24 (2010) 3313−3320
19] J. Zhang, L.B. Kong, J.J. Cai, Y.C. Luo, L. Kang, Electrochim. Acta 55 (2010)
8
067−8073
[
20] H.M.A. Hassan, E.M.Saad, M.S. Soltan, M.A. Betiha, I.S. Butler, S.I. Mostafa,
Appl. Catal. A. Gen. 488 (2014) 148-159
[
[
21] S. Bhunia, S. Koner, Polyhedron 30 (2011) 1857-1864
22] M. Masteri-Farahani, F. Farzaneh, M. Ghandi, J. Mol. Catal. A. Chem. 248 (2006)
5
3-60
[
23] M. Masteri-Farahani, F. Farzaneh, M. Ghandi, Catal. Commun. 8 (2007) 6-10
[
24] S. Tangestaninejad, M. Moghadam, V. Mirkhani, I. Mohammadpoor-Baltork, K.
Ghani, Catal. Commun. 10 (2009) 853-858
[
[
(
[
25] M.T. Goldani, A. Mohammad, R.Sandaroos, J. Chem.Sci. 126 (2014) 801-805
26] X. G. Zhou, X.Q. Yu, J.S. Huang, S.G. Li, L.S. Li, C.M. Che, Chem. Commun.
1999) 1789-1790
27] Y. Gang, C. Xing, W. Xiaoli, X. Weihong, X. Nanping, Chinese J. Catal. 34 (2013)
1
326-1332
28] P. Oliveira, A. Machado, A.M. Ramos, I.M. Fonseca, F.M. Braz. Fernandes, A.M.
Botelho do Rego, J. Vital, Catal. Commun. 8 (2007) 1366-1372
[
[
29] M. Bagherzadeh, M. Zare, M. Amini, T. Salemnoush, S. Akbayrak, S. Ozkar, J. Mol.
Catal. A. Chem. 395 (2014) 470-480
[
[
30] P. Sharma, A. Lazar, A.P. Singh, Appl. Catal. A Gen. 439-440 (2012) 101-110
31] H. Gaspar, M. Andrade, C. Pereira, A.M. Pereira, S.L.H. Rebelo, J.P. Araujo, J.
Pires, A.P. Carvalho,C. Freire, Catal. Today, 203 (2013) 103-110
[
32] Ch. Vartzouma, E. Evaggellou, Y. Sanakis, N. Hadjiliadis, M. Louloudi, J. Mol.
Catal. A 263 (2007) 77-85