Effects Of Mesoporous Silica Supports And Covalent Attachment Of Cu(ii) And Mn(ii)

Effects of mesoporous silica supports and covalent attachment of Cu(II) and Mn(II) as Schiff base complexes on catalytic activity in oxidation reactions

M. Filipa, I. Georgescub, M. Mureseanuc, G. Munteanua, A.H. Marina, V.M. Neacsua, I. Mateia, G. Ionitaa and V. Parvulescua

a”Ilie Murgulescu” Institute of Physical Chemistry, Romanian Academy, Splaiul Independentei 202, 060021, Bucharest, Romania

b“Babes Bolyai“ University, Faculty of Chemistry and Chemical  Engineering, 11 Arany Janos Street, 400028, Cluj-Napoca, Romania

cUniversity of Craiova, Faculty of Chemistry, A.I. Cuza St, 200144, Craiova, Romania

Abstract

New series of Cu(II) and Mn(II) complexes with Schiff-base ligands derived from 2-furylmethylketone (Met), 2-furaldehyde (Fur) and 2-hydroxyacetopheneone (Hyd) have been synthesized in-situ on SBA-15-NH2, MCM-48-NH2 and MCM-41-NH2 functionalized supports. The hybrid materials were characterized by X-ray diffraction, nitrogen adsorption–desorption, SEM and TEM microscopy, TG analysis, AAS, FTIR and XPS spectroscopies. The optimized structures of immobilized complexes were obtained by using the facilities of GAMESS package programs. The results sustain the typically ordered mesoporous structure of silica supports and its preservation after functionalization and formation of Cu(II) and Mn(II) complexes. The effect of support and ligand type on metal complexes immobilization yield was evidenced. The MCM-41 silica allows the immobilization of the highest ligand amount for all the three tested ligands and, respectively, the highest copper content. Catalytic performances were tested in liquid phase oxidation with hydrogen peroxide of cyclohexene and of different aromatic and aliphatic alcohols (benzyl alcohol, 2-methylpropan-1-ol and 1-buten-3-ol). Comparing the catalytic activity as a function of the mesoporous silica support, the higher performance of the metal complexes supported on MCM-41 was evidenced. A higher catalytic activity was obtained for Mn complexes, especially in oxidation of cyclohexene, in conditions of a significant lower metal/ligand ratio (M/L) and a lower metal content. The best catalytic activity for all tested hybrid materials was obtained in oxidation of cyclohexene, with a 91.5% conversion, 70% selectivity for 2-cyclohexen-1-one, 164.2 h-1 TOF value and SBA-15MetMn as heterogeneous catalyst. Furthermore, the recovery and reuse of the immobilized Mn and Cu complexes were evaluated for the cyclohexene oxidation reaction.

Keywords: Cu(II) Schiff base complex, Mn(II) Schiff base complex, mesoporous silica, cyclohexene oxidation, alcohol oxidation

1. Introduction

Oxidation catalysis is an important research area since it represents the core of a variety of chemical processes for producing bulk and fine chemicals as well as for eliminating pollution. Many transition metal complexes were used as catalyst for the oxidation reactions by using several oxidants, such as oxygen, hydrogen peroxide, and tert-butylhydroperoxide (TBHP) [1-4]. The oxidation of cyclohexene have attracted a deal of attention in recent years, mainly due to the oxidation products of cyclohexene and the derivatives which present the highly reactive carbonyl groups in the cycloaddition reactions [5, 6]. Furthermore, the oxidation of alcohols to the corresponding aldehydes, ketones and carboxylic acids is one of the important reactions in organic chemistry, especially in industrial synthetic chemistry [7-9]. Transition metal Schiff base complexes, as inorganic mimics of enzymes [10-12], have been extensively studied due to their potential use as catalysts in a wide range of oxidation reactions [13-17] and have the advantage to be prepared by simple and cheap methods for industrial applications. The activity of these complexes varied with the type of ligand, coordination sites and metal ion [18]. Cu (II) and Mn(II) are among the metal ions that are often found in the active centers of the metal enzymes involved in the oxygen atom transfer reaction [19]. Although homogeneous catalysts exhibit excellent activity and selectivity, the technical problems encountered in their applications, such as difficulty in product separation and deactivation by self-aggregation of active sites, have slowed their industrial applications. In the search to minimize these disadvantages, the heterogenization of homogeneous catalysts has emerged as a focus of research. Several polymer and inorganic oxide-anchored complexes with catalytic activities as good as those of homogeneous complexes have been developed [20–22]. Physical immobilization of transition metal complexes on the support leads to systems which are susceptible to leaching. More stable catalysts can result from covalent attachment of the reactive groups to a functionalized support [23]. Therefore, covalent anchoring of the Schiff base complexes onto a functionalized siliceous mesoporous material with large pore diameters seems to be promising especially for catalytic applications [24-28]. Ordered mesoporous silica are interesting solid supports due to their uniform and large pores, tunable pore sizes, high-surface areas, defined surface acidity, excellent mechanical stability and high concentration of surface Si–OH groups for the binding of the catalyst active sites [29-33]. In particular, MCM-41 (2d hexagonal p6mm), MCM-48 (3d cubic Ia3d) and SBA-15 (large-pore 2d hexagonal p6mm) mesoporous silicas, are used as supports with a high specific surface area (600-1000 m2g-1) and narrow pore size distribution (pore diameter 3-10 nm) [34-36]. These materials present the advantage of the easy functionalization with organic groups at the pore-wall interface [37, 38], which are used thereafter to immobilize ligands or their complexes with various metal ions. These catalysts combine characteristics of their support, like pore diameter, surface area, ordered porous structure, electrostatic potential, with the electronic and stereochemical properties of the complex [39]. The porous inorganic host is supposed to provide the right steric configuration and orientation of the metal complexes and so the direct access of the substrate molecule to the active site (the metal center) is regulated. Moreover, density of surface hydroxyl groups and their activity varies with the synthesis conditions of mesoporous silica and its porous ordered structure thus determining the degree of functionalization of surface respectively properties of the obtained catalysts. Effect of mesoporous silica supports on covalent attachment of transition metals as Schiff base complexes is still a debated issue. Our aim of this study was to evidenced the effects of mesoporous silica supports and covalent attachment of Cu(II) and Mn(II) as Schiff base complexes on catalytic activity of the obtained hybrid materials in oxidation reactions. Up to now, Cu(II) complexes, obtained through other methods, have been used as catalysts in olefin and alcohol oxidation reactions in homogeneous and heterogeneous medium [40-47] and Mn derivatives which contain chelating N-ligands are among the most active catalysts in hydrocarbons oxidations [20]. Thus, Mn(II) and Mn(III) immobilized complexes were used as efficient and highly selective heterogeneous catalysts in alkene epoxidation [13, 20, 51, 52] or oxidation of alcohols [53]. In this context, the synthesis of new hybrid based on mesoporous silica-Cu(II) and Mn(II) Schiff base complexes are expected to be promising candidates for the oxidation reaction with hydrogen peroxide due to the increase of active sites number, the more facile substrate accessibility and the improvement of diffusional problems.

In this paper we report the immobilization of Cu(II) and Mn(II) complexes with Schiff- base ligands obtained in-situ from 2-furylmethylketone, 2-furaldehyde and 2-hydroxyacetopheneone covalently attached to amino-functionalized SBA-15, MCM-41 and MCM-48 materials. These new hybrid catalysts that we have synthesized were subsequently utilized as catalysts in the oxidation reaction of cyclohexene and of different aromatic and aliphatic alcohols (benzyl alcohol, 1-buten-3-ol, 2-methylpropan-1-ol) with H2O2. Other metal complexes obtained in situ by using a Schiff base ligand derived from an amino-functionalized silica and an aldehyde were used as active catalysts, but none of them in oxidation reactions. For example, a new heterogeneous catalyst for carbon-carbon coupling reaction was obtained by anchoring a Pd(II) complex with a Schiff base ligand derived from amino functionalized MCM-41 silica and 2-thiophenecarboxaldehyde [55]. Concerning the ligands we mention that one of the studied aldehyde, 2-furaldehyde, was chemically immobilized onto silica gel modified with 3-aminoproyltrimethoxysilane and used as an adsorbent for Cu(II) and Ni(II) from aqueous solutions [54].

2. Experimental

2.1. Preparation of functionalized SBA-15, MCM-41 and MCM-48 silicas

SBA-15 material was synthesized as described by Zhao et al [56]. The gel with the molar composition: 1.00 SiO2 : 0.015 P123 : 5 HCl : 140 H2O was hydrothermally treated into a Teflon-lined autoclave at 403K for 24h. The as-synthesized SBA-15 was calcined at 823K, under air flow, during 8 h.

A highly ordered hexagonal siliceous MCM-41 was prepared as described in literature [57]. The resulting gel with the molar composition: 1.00 SiO2 : 0.1 CTABr : 25 H2O : 0.25 NaOH was hydrothermally treated in a Teflon-lined autoclave at 403K for 24h. The solid was filtered, washed with deionized water and dried in air at 353– 373 K for 12 h. The as-synthesized MCM-41 was calcined at 823K under air flow for 6 h in order to remove the template.

MCM-48 mesoporous silica was synthesized as described in literature [58]. The resulting gel with the molar composition: 1.00 SiO2 : 0.175 CTABr : 120 H2O: 0.38 NaOH was hydrothermally treated in a Teflon-lined autoclave at 423K for 15h. The as-synthesized MCM-48 was washed with water until neutral pH, dried in air at 353 K and calcined at 823 K for 8 h in air flow.

The organic-inorganic hybrid materials were obtained by a post-grafting procedure with 3-aminopropyltriethoxysilane (APTES) according to a previously described procedure [59].

2.2. Synthesis of Schiff base grafted onto functionalized silica

The Schiff base were synthesized by the in situ condensation of an aldehyde or a ketone onto the amino functionalized silica supports (MCM-41-NH2, MCM-48-NH2, SBA-15-NH2) as described in literature [60]. 1g amino-functionalized silica (MCM-41-NH2, MCM-48-NH2, SBA-15-NH2) was activated at 393 K for 3 h. The activated support was refluxed with an excess of 2-furylmethylketone, 2-furaldehyde or 2-hydroxyacetopheneone in anhydrous ethanol at 333 K, under N2 atmosphere, for 24 h. The solid was filtered, washed and dried in air at 333K overnight. The pale yellow obtained products were denoted: SBA-15 (MCM-41, MCM-48)-NH2-Fur, SBA-15 (MCM-41, MCM-48)-NH2-Met and SBA-15 (MCM-41, MCM-48)-NH2-Hyd.

2.3. Synthesis of Cu (II) and Mn (II) complexes with Schiff base ligands

The heterogeneous Cu(II) and Mn(II) complexes were obtained by mixing 0.5 g of (Schiff-base) – grafted silica and 10 mmols of Cu (NO3)2.3 H2O or Mn (NO3)2.6 H2O in 25 mL methanol, at room temperature, for 24 h. The solid was filtered, washed and dried overnight.

2.4. Characterization of materials

Small-angle XRD data were acquired on a Bruker AXS D8 diffractometer by using Cu Kα (λ = 0.1541 nm) radiation and Ni filter. The samples were scanned in the range 2θ = 0.4–6.0º. N2 adsorption/desorption isotherms were measured at 77 K with a Micromeritics ASAP 2010. Before measurements, the samples were outgassed at 323 K for 12 h. The specific surface area was calculated by the BET method and the mesopore volume was determined from the isotherm at the end of capillary condensation. The pores size distribution was obtained from the desorption branch using the BJH method and the Harkins-Jura standard isotherm. FTIR spectra of all samples were performed in KBr pellets using a Bruker Alpha spectrometer. Atomic absorption spectroscopy (AAS) measurements were performed on a GBS Avanta spectrometer equipped with multi-element hollow cathode lamps and air-acetylene burner. C, H, and N contents were evaluated by combustion on a Fisons EA1108 elemental analysis apparatus. Thermogravimetric analysis was carried out in a Setaram thermobalance. XPS analysis was performed on a ULVAC-PHI X-ray photoelectron spectroscopy (XPS) system and EPR spectra were recorded on a JEOL FA 100 spectrometer. The chemical microanalyses by scanning electron microscopy (SEM) were performed on a Hitachi S-4500 I scanning microscope and transmission electron microscopy (TEM) characterization was recorded on a JEOL 1200 EX II operated at 100 kV.

2.5. Catalytic Test

The catalytic oxidation reactions were carried out in a 5 mL flask using acetonitrile as solvent and H2O2 (30% aqueous) as oxidant. The amount of catalyst was 30 mg and the tested substrates were: cyclohexene (CH), benzyl alcohol (BzA), 2-methylpropan-1-ol (IzBA), 1-buten-3-ol (1B3A). The molar ratio substrate/solvent/oxidant was 1/3.8/3. The reaction mixture was heated at 303 K for 5 hours with continuous stirring.

The analysis of oxidation products was performed using a GC-MS analysis that was performed on a Thermo Scientific chromatograph equipped with MS DSQ II Thermo and capillary column TR-5MS (30 m x 0.25 ID x 0.25 µm film thickness). The active compound leaching during the reaction was verified by resubmission of filtrates to reaction conditions. Samples were filtered before analysis. The persistence of the catalytic activity was checked in the oxidation of CH for 5 consecutive runs. Copper content in the recovered catalysts was determined by AAS, after each run.

3. Results and Discussion

New heterogeneous oxidation catalyst were obtained and tested for their activity. The supported Cu (II) and Mn (II) complexes with Schiff base ligands were synthesized in three steps: (i) post synthesis functionalization of SBA-15, MCM-41 and MCM-48 mesoporous silica supports with APTES, (ii) condensation of surface NH2 function with 2-furylmethylketone, 2-furaldehyde or 2-hydroxyacetopheneone to forms the Schiff base and (iii) in-situ Cu(II) or Mn(II) coordination with Schiff base ligands to obtain the supported metal complexes (Scheme 1). We mention that the presented structure of Cu(II) and Mn(II) complexes immobilized on mesoporous silica support is the most probable, as resulted from quantum chemical calculations that we present bellow, and is in agreement with the structure of other heterogeneous metal complexes with Schiff base ligands [24].

3.1. Characterization of the mesoporous supports

The characteristics of the obtained silica supports are typically for SBA-15, MCM-41 and MCM-48 mesoporous silica. The small angle XRD patterns, before and after functionalization with amino group, show typically peaks for these materials (Fig. 2).

XRD patterns at small angles of the SBA-15 material (Fig. 2a) emphasize an ordered structure with three diffraction peaks, indexed as (100), (110) and (200) diffraction peaks, associated with the typical two-dimensional hexagonal symmetry (P6mm). For MCM-41 the three peaks (100, 110 and 200) indicate the ordered two-dimensional hexagonal symmetry corresponding to MCM-41 materials (Fig. 2b). Structural analysis of MCM-48 reveals that it has a cubic ordered structure with three-dimensional arrangement characteristic for this type of silica. A slight decreasing of the well resolved peak intensity in the range 0.5-5o after functionalization was evidenced. This indicates a low variation of the inorganic wall structure after the functionalization with amino group. A similar effect can be observed for the porous structure parameters and wall thickness values of all the mesoporous silica supports (Table 1).

The N2 adsorption/desorption isotherms of modified mesoporous molecular sieves confirm that the channels of the support remain accessible. All amino-functionalized materials maintain the characteristics of type IV isotherms and show an uniform pore size distribution in the mesoporous region, similar with the support. However, the pore diameter decreases compared to the pure siliceous material. A pronounced decrease of the BET surface area and of the pore volume of the hybrid materials is evidenced, as well as an increase of the wall thickness, due to the increase of organic group content (Table 1).

SEM images (Fig. 3) highlight the spherical morphology of MCM-41 and MCM-48 supports and rod-like morphology typical of SBA-15 materials.

Detailed structural characterization has been done for MCM-41, MCM-48 and SBA-15 using TEM microscopy (Fig.4). In all these cases, the porous structure was clearly visualized. The image for MCM-41 (Fig. 4A) shows a uniform array of pores with a hexagonally arrangement. The TEM image of cubic MCM-48 (Fig. 4B) shows a highly ordered pore arrangement and confirms the three-dimensional nature of MCM-48 material. TEM micrograph of SBA-15 (Fig. 4C) shows the hexagonal array of uniform channels. TEM results are in good agreement with the XRD and nitrogen adsorption-desorption results discussed above. The effect of functionalization on the support porous structure and morphology was very low.

On the surface of MCM-41, MCM-48 and SBA-15 molecular sieves there are silanol groups that can serve as sites for incorporation of the aminopropyl groups. The literature [63] presents a higher concentration of silanol groups for SBA-15 and MCM-48 (13 mmol/m2) compared to MCM-41 (2 mmol/m2). Considering the values of surface area presented in Table 1, the variation of silanol groups density on the support surface is as follows: Si-OHMCM-41< Si-OHSBA-15< Si-OHMCM-48.

3.2. Structure modeling of the supported metal complexes with Schiff base ligands

The quantum chemical calculations were performed to establish at the beginning if the three molecular formations, 2-furaldehyde (A), 2-furylmethylketone (B) and 2-hydroxyacetopheneone (C), which have functionalized the surface of three types of supports, MCM-41, MCM-48 and SBA-15, could be bidentate ligands. More precisely, we wanted to establish if both oxygen and nitrogen atoms which exist into the structures of these ligands could simultaneously bind copper (or manganese) atoms. Performing DFT – B3LYP/6-31G (d,p) calculations we have shown that the supported Schiff bases on silica mesoporous supports, could be bidentate ligands for Cu(II) or Mn(II) ions. Consequently, using their general formulas we have built the Schiff bases and complexes structures using MOLDEN program [61]. We have to mention that although concentration of silanol groups, surface area and pore diameter of MCM-41, MCM-48 and SBA-15 supports are different, we took into account only a group integrated into the structure of the ligand. We did this approximation because in every case of any of the three supports, the hydrocarbon chain is bound on a silicon atom through the –(CH2)3N= moiety of the Schiff base. Following the first stage of the support functionalization, was found that silicon from APTES is linked to three different oxygen atoms that belong to silanol groups of the silica support. Our attention was finally concentrated to the binding mechanism of the metal ion. We considered that on the surface must be two such ligand molecules in adequate positions allowing to form a square or a tetrahedron of appropriate dimensions, that can bind in its center a copper (or a manganese) ion. Furthermore, the calculated ligand density on the silica surface from TG and elemental analysis sustains this assumption. Having the coordinates of every of these structures we have optimized them using the facilities of GAMESS package programs [62]. In the optimization process we have frozen the group, considering that only the atoms of the bound molecules can change the coordinates. For every surface complex we imagined an adequate initial structure that consists of two bound ligands which bring inside a copper atom (see Fig. 1). Each such geometry was optimized assuming again that the two SiO33- groups, which belong to the support, are rigid. In these optimization processes we have frozen the coordinates of the two groups, allowing to all the other coordinates to be modified. As it can be seen in Fig. 1, starting from the imagined initial structures we could finally obtain enough different optimized structures. The metallic complex changes very much its shape because the ligand molecule is flexible although the SiO33- groups are rigid. As shown in Fig. 1 the metal becomes accessible to interaction with reactant molecules during the catalytic process due to the deformation of the metal complex. Furthermore, this most favorable tetrahedral geometry of the immobilized complex was confirmed by spectroscopic studies (EPR) which are presented in subsection 3.3.

3.3. Characterization of the supported metal complexes with Schiff base ligands

The FTIR spectra for the Schiff base obtained in situ from 2-furylmethylketone, 2-furaldehyde, or 2-hydroxyacetophenone grafted on the amino functionalized silicas (Fig. 5) were registered in the 4000 – 400 cm-1 range and were used to characterize the support and to confirm the complex formation in the solid samples after modification. The FTIR spectrum of the unmodified SBA-15, as well as the spectra of all the modified materials, are dominated by strong bands characteristic of the support matrix, indicating that the support framework remained unchanged.

The FTIR spectra of the free and amino-functionalized supports present bands at about 460, 790, 1080 cm-1 attributed to the vibration modes of the silica host matrix [64]. Symmetrical stretching vibrations of Si-O-Si bonds belonging to ring structures are observed around 790 cm-1. The bands at 460 cm-1 could be assigned to bending vibrations of associated Si-O-Si bonds. The bands at around 1080 cm-1 could be due to antisymmetric stretching vibrations of Si-O-Si, overlapped to Si-O-C, C-O-C and Si-C bond vibrations [65]. The existence of silanol bands at 980 cm-1 could be referred to stretching vibrations of free silanol (Si-OH) groups on the surface of the amorphous solid samples. The stretching vibrations of C-O bonds are also placed in this range [66]. The broad band around 3450 cm-1 corresponds to molecular water hydrogens bonded to each other and to Si-OH groups. The medium intensity bands at 2922 and 2885 cm-1 correspond to asymmetric and symmetric vibration of –CH2- from APTES functionalization agent [67]. In the spectra of Schiff base functionalized silica, the bands at around 1620 cm-1 could be related to bending vibrations of O-H bonds in OH groups, overlapped with the vibration bands characteristics for C=N groups. The condensation reaction of the surface primary amines with the carbonyl functionalities of the aldehydes and ketone can be detected by the appearance of the imine (C=N) bands at 1645 cm-1 for SBA-15- Fur, at 1630 cm-1 for SBA-15-Hyd and at 1616 cm-1 for SBA-15-Met, respectively. The FTIR bands of azomethine group appearing in Schiff bases complexes shifted to lower frequency (10–15 cm−1) confirm the coordination of nitrogen from this group with the metal atoms [68]. Furthermore, the appearance of the new band at around 1385 cm−1 in the spectra of complexes with Schiff bases could be the result of a red shift of the C=N stretching vibration absorption due to coordination to copper ions [69]. For all the immobilized complexes, the vibration bands in the low frequencies range (300-800 cm-1) attributed to M-O and M-N are present but they are overlapped with those characteristic for the silica skeleton.

Thermogravimetric analysis (TGA) of the supported Cu(II) and Mn(II) complexes with Schiff base ligands indicate three major mass losses (Fig. 6). The first two at 1000C and at around 200oC, obtained for all materials, were attributed to the solvent elimination, respectively to the partial decomposition of the metal complexes. The final decomposition of the metal complexes and total organic compound elimination started at ~300oC (Fig. 6A) for the immobilized complexes with 2-furylmethylketone or 2-furaldehyde and at ~400oC for supported Schiff bases obtained with 2-hydroxyaceto-phenone (Fig. 6A).

The insignificantly effect of mesoporous silica support type was evidenced on the shape of TG curves (Fig. 6B). However, the variation of weight loss was significantly influenced by the surface area of the support. The higher weight loss was evidenced for metal complexes immobilized on MCM-48 (Fig. 6B), the support with the highest surface area. The amount of immobilized ligand (L), immobilization yield (Yimm. %) and the metal to ligand molar ratio (M/L) values were obtained by the elemental analysis and are presented in Table 2. The amount of immobilized ligand was determined from TG curves by subtracting the amount of the aminopropyl group previously grafted on the mesoporous silica surface and compared with the elemental analysis (C%, N% and H%). The copper and manganese contents of all samples were determined by AA spectroscopy after their dissolution in 10% HF.

The weight loss due to the aminopropyl moiety was of 11.61% for the SBA-15, 11.11% for MCM-41 and 13.89% for MCM-48 mesopoporos silica, respectively. These values correspond to an aminopropyl density of 3.1 atoms/nm2 for SBA-15 silica, 1.7 atoms/nm2 for MCM-41 and 1.5 atoms/nm2 for MCM-48. Considering the specific surface area and the concentration of silanol groups on the support surface for all three mesoporous silica, we can conclude that the highest degree of silanol participation to the functionalization with amino groups and the most hydrophobic surface was obtained for MCM-41 support. This is in agreement with results presented in Table 2 and Fig. 6A indicating the best immobilization yield for all the metal complexes with Schiff bases on the MCM-41 support. Even if for SBA-15 silica the –NH2 density was greater than for MCM-41, the more hydrophobic surface of the last support was beneficial to the Schiff base ligand immobilization. The elemental analysis confirmed that between 39-89% of the aminopropyl groups grafted on the silica support surface participated in the Schiff base synthesis by condensation with the corresponding aldehyde or ketone. The ligand density varied between 1.9-1.2 atoms/nm2 for SBA-15, between 1.5-0.9 atoms/nm2 for MCM-41 and between 0.8-0.4 atoms /nm2 for MCM-48, respectively. The results presented in Table 2 were obtained for the supported metal complexes after removal by washing of the species that are weakly bound on the surface. In these conditions, we consider that only the unreacted aminopropyl groups and the Schiff base ligand were presented on the silica support surface. Furthermore, not all of the Schiff base ligands have been involved in the complexation reaction with the copper and manganese cations, respectively.

The metal to ligand molar ratio varied between 1/3 to1/6 for copper complexes and between 1/5 to 1/9 for manganese complexes, respectively. The metal ions complexation by the unreacted aminopropyl functions it was also possible. It is clear that the complexation reaction is much favorable for copper ions than for manganese ions (Fig. 7 A, B). From the TG analysis (data not shown) was observed that the Mn(II) complexes immobilized into the tested mesoporous silica supports are a little more stable than the corresponding Cu(II) complexes. Nevertheless, the amount of Mn(II) complexes was smaller than that of the Cu(II) complexes. In Fig. 7B is presented the variation of ligand concentration with silica support type for MetMn, respectively FurMn complexes. For all three type of mesoporous silica, the lower ligand amount was obtained for the Schiff base derived from furaldehyde. Even if for MetMn complex the amount of immobilized ligand was greater, the metal content of all samples was almost the same as for FurMn. It is possible that the uncomplexed ligand to be leached from the support during the catalyst synthesis steps.

Comparing the same ligand but different metals in the complex, initially the ligand amount was the same for the copper and manganese complexes. The greater amount of Cu(II) could stabilize the corresponding complex, which explain the greater amount of ligand determined from elemental and TG analysis.

3.4. XPS studies

XPS spectra of the immobilized complexes and standards obtained by impregnation of mesoporous silica supports with metal nitrate were compared. For these last samples, the highest Cu content was evidenced on MCM-48 support and the lowest on MCM-41 unfunctionalized support (Fig. 8). The highest Mn content was obtained for the impregnated samples into SBA-15 unfunctionalized support and the lowest for the immobilized complexes onto the same support. These results are consistent with the variation of the silanol group concentration on different mesoporous silica support surface (see.3.2). Furthermore, for all the tested supports, the amount of Cu and Mn was grater for the impregnated samples than for those containing immobilized complexes. XPS spectra of the immobilized complexes shown the highest Cu content for MCM-41-Met Cu (Fig. 9) and these results are in agreement with the highest ligand amount (Fig. 6). The highest concentration of Mn complexes was obtained for MCM-48 support. Comparing the amount of immobilized complex correlated with the ligand type, the lower concentration of metal has been observed for Hyd-Cu complexes, results that are in agreement with elemental analysis. Furthermore, considering that the specific form of the support pores was preserved by ligand immobilization on the pore surface, the results of quantum chemical calculations (Fig.1 Cxz-f) showed that the optimized structure of the metal complex with 2-hydroxyacetopheneone ligands is less favorable.

Cu impregnated samples exhibited principle Cu2p3/2 broad profile consisting of two peaks between 930 eV and 938 eV, which could be attributed to Cu2+ species [70]. The XPS deconvoluted spectra (Fig. 8) of Cu supported samples reveal two chemical features of Cu2p3/2 photoelectron line assigned to Cu+ and Cu2+. The amount of Cu+ is increasing from 73.2% (SBA-15-MetCu) to 82.4% (MCM-48-HydCu) with the corresponding decrease of Cu2+ relative concentrations. The presence of Cu+ cation was attributed to an electron transfer from the ligand O and N orbitals to the metal orbitals [71] proving that copper is present as a complex on the silica surface. Moreover, XPS spectra of the immobilized complexes show (Figs. 10, 11) that N1s binding energy, attributed to –N=C, has a little shift to higher binding energy (399.5 eV) when compared to the imine specific band. This positive shift can be the result of electron density decrease around N atom due to metal coordination. A similar effect was revealed by a little shift of O1s binding energy to higher binding energy (532.5 eV) compared to the O1s band of the furan ring.

Fig.12 displays the noisy Mn2p spectra for SBA-15-MetMn and MCM-48-MetMn suggesting a low content of manganese present on the outermost surface layer. The atomic concentrations for SBA-15-MetMn and MCM-48-MetMn samples were found in the range of ~0.5%. The binding energy (642.5 eV) as well as the presence of shake-up characteristic satellite exhibits the fingerprints for MnO in Mn-SBA-15 impregnated sample (Fig. 12). The negative shift of Mn2p binding energy for the immobilized complexes compared with the impregnated sample can be the result of electron density increasing around metal due to an electron transfer from the ligands.

3.5. Catalytic properties

Catalytic activity of Cu(II) and Mn(II) complexes with Schiff-base ligands derived from 2-furylmethylketone (Met), 2-furaldehyde (Fur) and 2-hydroxyacetophenone (Hyd) immobilized into different NH2-functionalized mesoporous silica supports was evaluated for the oxidation of cyclohexene (CH) and aromatic or aliphatic alcohols like benzyl alcohol (BzA), 2-methylpropan-1-ol (IzBA) and 1-buten-3-ol (1B3A). Table 3 and 4 present the results achieved by different catalyst in oxidation of cyclohexene and aromatic or aliphatic alcohol, respectively. Oxidative catalytic conversion of organic compounds has been done with peroxide hydrogen (30% H2O2) and CH3CN as solvent. H2O2 is the preferred oxidant in these systems, since it is highly mobile in the nanopores due to its smaller size. The solvent plays an important and sometimes decisive role in catalytic behavior because it can make different phases uniform, thus promoting mass transportation, and could also change the reaction mechanism by affecting the intermediate species, the catalyst surface properties and reaction pathways. We have chosen for the catalytic tests the CH3CN, which is a polar aprotic solvent, usually used in hydrocarbons oxidation reactions. The catalytic activity was expressed as substrate conversion and also as turnover frequency (TOF) in order to highlight the kinetic effects on the catalytic activity of the different type of mesoporous silica supports. TOF, h−1 is calculated by the expression of moles of substrate converted per mole metal ion per hour. The copper and manganese contents of all samples were determined by AA spectroscopy.

The heterogeneous Cu(II) and Mn(II) complexes did not undergo any color change during the reaction and could be easily separated and reused many times. No oxidation of cyclohexene or alcohols occurred in the absence of the catalyst. A series of blank experiments revealed that each component is essential for an effective catalytic reaction and the system is relatively unaffected by changing the order of mixing. The best activity of the hybrid catalysts was obtained in oxidation of cyclohexene when we have determined the catalytic activity of copper and manganese complexes with Schiff base derived from 2-furaldehyde and 2-furylmethylketone immobilized in the three mesoporous silica supports (Tables 3). Comparing their TOF values (Fig. 13, 14), the higher activity in oxidation of CH, BzA and IzBA was obtained for MetCu complex supported on MCM-41. For this hybrid catalyst, the order of activity was CH>IzBA>BzA>1B3A. A similar effect was evidenced for FurCu complex in oxidation of CH, IzBA and 1B3A. The variation of TOF values for these supported catalysts is consistent with variation of the ligand content on the surface of SBA-15, MCM-48 and MCM-41 silica supports, for the immobilized metal complexes (Fig. 7). These results argue the main effect of silanol concentration on Schiff base formation, complexation and substrate accessibility for the chemical reaction. The same influence of support was observed for MetMn and FurMn complexes (Fig. 13, 14). There are some exceptions for MetCu complex in oxidation of 1B3A and for FurCu complex in oxidation of BzA, when the best activity was obtained for SBA-15 support. In this case the activity can be considered as being strongly influenced by the pore diameter. In conditions of M/L ratio significantly lower for Mn compared to Cu and of a low Mn concentration, both evidenced by AAS and XPS analysis, it was observed a significantly higher catalytic activity for Mn complexes, especially for cyclohexene oxidation. These results confirm that Mn(II) complexes are more active in the tested reactions than the Cu(II) homologues. For Mn(II) complexes, a higher dispersion of the metallic active sites is more important than their concentration.

Aiming the ligand effect on catalytic activity, the results indicate the next general order of activity depending on ligand type: Met>Fur>Hyd either for Cu(II) or Mn(II) complexes. One exception was observed for manganese complex supported on MCM-41 (Fig 15) where the activity of FurMn complex was slightly higher than that of MetMn.

The higher activity in oxidation of the tested alcohols was obtained for the Mn complexes (Table 4). The special interest was for 1 buten-3-ol due to the alcohol/olefin chemoselectivity.

For all the tested catalysts, the relatively high TOF values proved that the mesoporous silica support offer a high substrate accessibility and improvement of the diffusional problems. The most important parameters for a better catalytic activity is a high dispersion of the active site (especially for Mn) and the right steric configuration and orientation of the metal complexes for a direct access of the substrate to the metallic center. For the Cu and Mn complexes with Schiff base that we have tested, the best support was the MCM-41 mesoporous silica.

Among olefins, cyclohexene is one of the most studied substrate in oxidation reactions. In cyclohexene oxidation, two reactions can take place: oxidation in the allylic position and breakdown of the C=C double bond. In the second one, cyclohexene oxide, an interesting monomer for its applications [71] could be obtained, or cyclohexane 1,2 diol after oxide hydrolysis. Studying the CH oxidation by using Mn(II) and Cu(II) complexes immobilized into mesoporous silica supports, both reactions take place but the main followed route was oxidation in allylic position. Table 3 shows a higher selectivity for 2-cyclohexen-one and a very low selectivity for cyclohexene oxide. The presence of 1,2-cyclohexendiol evidence the effect of H2O2 aqueous solution on epoxide hydrolysis.

No significant effect on selectivity was observed for the type of mesoporous support. For all the catalysts the highest selectivity was for 2-cyclohexen-1-one. However, a variation of cyclohexene oxide selectivity was evidenced. The higher selectivity (around 2%) was obtained for MCM-48 support and HydCu complexes.

The main product in oxidation of benzyl alcohol was benzaldehyde and for oxidation of 2 methylpropan-1-ol was 2 methylpropanal. Oxidation of 1-butene-3-ol, an alcohol possessing both hydroxyl an olefinic functional group, led to 1 buten-3-one as main product. Chemoselective oxidation of this unsaturated alcohol can occur by two possible reaction pathways: epoxidation of C=C bound and allylic oxidation of alcohol to aldehyde [72]. Usually, epoxidation is the aim of 1-butene-3-ol oxidation because the epoxide allow many applications, like synthesis of 1,2-epoxybutane-3-ol. Unfortunately, this reaction requires special experimental conditions like higher temperature (100-120oC) and pressure [73]. In conditions of our reactions only 1-butene-3-one was identified.

When compared the oxidative activity of these new catalyst with other heterogeneous metal complexes, both the substrate conversion and TOF values are very good. For example, unsubstituted and tertiary-butyl substituted salicylaldimine complexes of Cu(II) and Co(II) were immobilized on silica supports (MCM-41, SBA-15 and Davisil 710) and tested as catalyst for cyclohexene oxidation using hydrogen peroxide as an oxidant under an oxygen atmosphere [74]. All catalysts were active in the tested reaction and the conversions varied between 41% and 84% depending on the nature of the catalyst. Copper(II)ethylacetonate precursor was immobilized onto organofunctionalized MCM-41 material with N4-(3-(triethoxysilyl)propyl)pyrimidine-2,4,6-triamine compound and was tested in the benzylalcohol oxidation with TBHP as oxidant under solventless conditions, when a high selectivity (97%) to benzaldehyde was achieved for a maximum conversion of 62% [31]. Heterogeneous Mn-nano-catalysts were synthesized and covalently anchored on a modified nanoscale SiO2/Al2O3 and were highly selective catalysts for oxidation of cyclohexene (84% conversion and 95% selectivity for 2-cyclohexen-1-one) and benzyl alcohol (76.1% conversion, 100% selectivity for benzaldehyde), without the need of any solvent and using TBHP as oxidant [75].

3.6. Catalyst integrity and leaching

To test if copper or manganese were leaching out from the solid catalyst during reaction, hot filtration test was done for the cyclohexene oxidation with H2O2, by using the catalyst with the best oxidative activity (SBA-15MetMn and SBA-15MetCu). After the filtration of the reaction mixture at 303 K, the residual activity of the filtrate was further studied in the same reaction conditions. The catalyst filtration was done at the reaction temperature in order to avoid possible re-coordination or precipitation of the metallic ions upon cooling. We fund that after the hot filtration there was no progress of reaction. Furthermore, any metal cation was evidenced by AAS in the reaction products, confirming that was no metal leaching during the catalytic test.

When the same tested catalysts were reused in five consecutive runs, the catalytic performances have not decreased significantly (Table 5) proving that these new catalysts were stable and recyclable. Furthermore, metal leaching was not observed neither in this reusability test.

4. Conclusions

A series of new Cu(II) and Mn(II) complexes with Schiff- base ligands derived from 2-furylmethylketone, 2-furaldehyde and 2-hydroxyacetophenone have been synthesized in-situ into different mesoporous silica supports (MCM-41, SBA-15, MCM-48). Their structure was modeling by using GAMESS package programs and confirmed by the spectroscopic analysis. All three Schiff base derived from Fur, Met and Hyd are bidentate ligands for Cu (Mn) atoms and the L/M ratio is 2/1. Furthermore, the electronic structure of the active metallic sites and their environment due to the ligands were proved by XPS and RPE analysis. The immobilized Cu(II) and Mn(II) complexes were active catalysts for the oxidation of cyclohexene and aromatic and aliphatic alcohols (benzyl alcohol, 2-methylpropan-1-ol, 1-buten-3-ol). These complexes anchored onto mesoporous materials functionalized with Schiff base ligands by in-situ complexation can be easily synthesized and show a great potential due to their tunable large pore size. Furthermore, the increased number of active sites in a right steric configuration, a good dispersion and improvement of diffusional problems, are other advantages that recommend this new heterogeneous oxidation catalysts.

Comparing the effect of silica supports on catalytic activity was evidenced the higher performances of the metal complexes supported on MCM-41. A higher catalytic activity was obtained for Mn complexes, especially in oxidation of cyclohexene in conditions of a significant lower M/L ratio and a lower metal content. The highest selectivity was obtained for aldehydes in oxidation of benzyl alcohol and 2-methylpropan-1-ol, respectively allylic oxidation products for cyclohexene and 1-butene-3-ol.

The obtained results can pave the way for the development of highly effective heterogeneous catalysts, obtained by in-situ immobilization of metal complexes into mesoporous silica matrices, for efficient oxidation of alkene and alcohol substrates.

Acknowledgments: Authors, M.F, G.M and.V.P. are grateful to the Romanian Academy (research programs 2.6.16 IPC) for financial support. The authors would like to thanks INFRANANOCHEM PROJECT POS-CCE O 2.2.1, no. 19/01.03.2009.

References

[1] L. Canali, D.C. Sherrington, Chem. Soc. Rev. 28 (1999) 85–93.

[2] M.R. Maurya, A. Kumar, J. Mol. Catal. A: Chem. 250 (2006) 190–198.

[3] D.E. De Vos, B.F. Sels, P.A. Jacobs, Adv. Synth. Catal. 345 (2003) 457–473.

[4] S. Tangestaninejad, M. Moghadam, V.Mirkhani, I. Mohammadpoor-Baltork, K. Ghani, Inorg. Chem. Commun. 11 (2008) 270–274.

[5] P. Wu, T. Tatsumi, T. Komatsu, T. Yashima, J. Catal. 202 (2001) 245 -255.

[6] N. Sehlotho, T. Nyokong, J. Mol. Catal. A: Chem. 209 (2004) 51-57.

[7] M. M. Dell’Anna, M. Mali, P. Mastrorilli, P. Cotugno, A. Monopoli, J. Mol. Catal. A: Chem. 386 (2014) 114-119.

[8] A.Ourari, D. Aggoun, L. Ouahab, Inorg. Chem. Commun. 33 (2013) 118-124.

[9] C.M.A. Parlett, L.J. Durndell, A. Machado, G. Cibin, D.W. Bruce, N.S. Hondow, K. Wilson, A.F. Lee, Catal. Today 229 (2014) 46-55.

[10] B. Tang, G.-Y. Zhang, Y. Liu, F. Han, Anal. Chim. Acta 459 (2002) 83-91.

[11] X. Wang, G. Wu, W. Wei, Y. Sun, Catal. Lett. 136 (2010) 96–105.

[12] A. M. Abu-Dief, I. M. A. Mohamed, Beni-Suef University Journal of Basic and Applied Sciences 4 (2015) 119-133.

[13] M. Salavati-Niasari, J. Mol. Catal. A: Chem. 283 (2008) 120–128.

[14] W.-J. Zhou, B. Albela, M.-Y. He, L. Bonneviot, Polyhedron 64 (2013) 371-376.

[15] J. Hu, K. Li, W. Li, F. Ma, Y. Guo, Appl. Catal. A: Gen. 364 (2009) 211-220.

[16] X. Wang, G. Wu, J. Li, N. Zhao, W. Wei, Y. Sun, Stud. Surf. Sci. Catal. 170 (2007) 1374-1379.

[17] O. Pouralimardan, A.-C. Chamayou, C. Janiak, H. Hosseini-Monfared, Inorg. Chim. Acta 360 (2007) 1599–1608.

[18] K.C. Gupta, A. K. Sutar, Coord. Chem. Rev. 252 (2008) 1420–1450.

[19] S.T. Prigge, B.A. Eipper, R.E. Mains, L.M. Amzel, Science 304 (2004) 864–867.

[20] S. Alavi, H. Hosseini-Monfared, M. Siczek, J. Mol. Catal. A: Chem. 377 (2013) 16-28. [21] S. Sisodiya, A. Lazar, S. Shylesh, L. Wang, W. R. Thiel, A. Pal Singh, Catal. Commun. 25 (2012) 22-27.

[22] M. Salavati-Niasari, H. Banitaba, J. Mol. Catal. A: Chem. 201 (2003) 43 –54.

[23] A. R. Silva, K. Wilson, J.H. Clark, C. Freire, Micropor. Mesopor. Mater. 91 (2006) 128–138.

[24] U.G. Singh, R.T. Williams, K.R. Hallam, G.C. Allen, J. Solid State Chem. 178 (2005) 3405-3413.

[25] L. Ma, F. Su, X. Zhang, D. Song, Y. Guo, J. Hu, Micropor. Mesopor. Mater. 184 (2014) 37–46.

[26] G. Rezanejade Bardajee, R. Malakooti, I. Abtin, H Atashin, Micropor. Mesopor. Mater. 169 (2013) 67–74.

[27] S. Bhunia, S. Koner, Polyhedron 30 (2011) 1857-1864.

[28] R. J. P. Corriu, E. Lancelle-Beltran, A. Mehdi, C.Reye, S. Brandes, R. Guilard, Chem. Mater. 15 (2003) 3152-3160.

[29] V. Meynen, P. Cool, E.F. Vansant, Micropor. Mesopor. Mater. 125 (2009) 170–223. [30] R. Liu, Y. Zhang, X. Zhao, A. Agarwal, L.J. Mueller, P.Y. Feng, J. Am. Chem. Soc. 132 (2010) 1500–1501.

[31] Y. Perez, R. Ballesteros, M. Fajardo, I. Sierra, I. del Hierro, J. Mol. Catal. A: Chem. 352 (2012) 45–56.

[32] R. Liu, Y. Zhang, P.Y. Feng, J. Am. Chem. Soc. 131 (2009) 15128–15129.

[33] V. Parvulescu, C. Anastasescu, C. Constantin, B.L. Su, Catal. Today 78 (2003) 477-485.

[34] R. Liu, X. Zhao, T. Wu, P.Y. Feng, J. Am. Chem. Soc. 130 (2008) 14418–14419.

[35] A. Y. Khodakov, V. L. Zholobenko, R. Bechara, D. Durand, Micropor. Mesopor. Mater. 79 (2005) 29–39.

[36] S. M. Lopes dos Santos, K. A. Barros Nogueira, M. de Souza Gama, J.D. Ferreira Lima, I. J. da Silva Junior, D.C. Silva de Azevedo, Micropor. Mesopor. Mater. 180 (2013) 284–292.

[37] M. Mureseanu, V. Parvulescu, R. Ene, N. Cioatera, T.D. Pasatoiu, M. Andruh, J. Mat. Sci. 44 (2009) 6795.

[38] R.I. Kureshy, I. Ahmad, N.H. Khan, S.H.R. Abdi, K. Pathak, R.V. Jasra, J. Catal. 238 (2006) 134–141.

[39] E. Paez-Mozo, N. Gabriunas, R. Maggi, D. Acosta, P. Ruiz, B. Delmon, J. Mol. Catal. A: Chem. 91 (1994) 251 –258.

[40] J. I. van der Vlugt, F. Meyer, Top. Organomet. Chem. 22 (2007) 191

[41] V. K. Bansal, R. Kumar, R. Prasad, S. Prasad, J. Mol. Catal. A: Chem. 284 (2008) 69. [42] K. Jahnisch, V. Hessel, H. Lowe, M. Baerns, Angew. Chem., Int. Ed.2004, 43, 406. [43] L. X. Alvarez, M. L. Christ, A. B. Sorokin, Appl. Catal. A: Gen., 325 (2007) 303.

[44] A.-R., Judy-Azar, S. Mohebbi, J. Mol. Catal. A: Chem. 397 (2015) 158–165.

[45] S.M. Islam, A. S. Roy, P. Mondal, S. Paul, N. Salam, Inorg. Chem. Comm. 24 (2012) 170–176.

[46] T. Erkenez, M. Tümer, Arabian Journal of Chemistry (2015)xxx, xxx–xxx.

[47] Y. Yang, Y. Zhang, S. Haob, Q. Kan, Chem. Eng. J. 171 (2011) 1356– 1366.

[48] C. Li, Catal. Rev. Sci. Eng. 46 (2004) 419–492.

[49] Q.H. Zhang, Y. Wang, S. Itsuki, T. Shishido, K. Takehira, Chem. Lett. 9 (2001) 946–947.

[50] P.P. Knops-Gerrits, D.E. De Vos, F. Thibault-Starzyk, P.A. Jacobs, Nature 369 (1994) 543–546.

[51] L. Ma, F. Su, X. Zhang, D. Song, Y. Guo, J. Hu, Micropor. Mesopor. Mater. 184 (2014) 37–46.

[52] M. Ghorbanloo, H. H. Monfared, Ch. Janiak, J. Mol. Catal. A: Chem., 345 (2011) 12– 20.

[53] M. Rong, J. Wang, Y. Shen, J. Han, Catal. Communications 20 (2012) 51–53.

[54] H.Dumrul, A.N.Kursunlu, O.Kocyigit, E.Guler, S.Ertul, Desalinisation 271(2011) 92-99.

[55] S. Bhunia, R. Sen, S. Koner, Inorg. Chim. Acta 363 (2010) 3993–3999.

[56] D. Zhao, J. Feng, Q. Huo, N. Melosh, G.H. Fredrickson, B.F. Chmelka, G.D. Stucky, Science 279 (1998) 548-552.

[57] J.S. Beck, J.C. Vartuli, W.J. Roth, M.E Leonowicz, C.T. Kresge, K.D. Schmitt, C.T.-W. Chu, D.H. Olson, E.W. Sheppard, J. S. McCullen, B.B. Higgins, J.L. Schlenker, J. Am. Chem. Soc. 114 (1992a) 10834-10843.

[58] A. Galarneau, M. Driole, Petitto, B. Chiche, B. Bonelli, M. Armandi, B. Onida, E. Garrone, F. di Renzo, F. Fajula, Micropor. Mesopor. Mat. 83 (2005) 172-180.

[59] D. Brunel, A. Cauvel, F. Fajula, F. di Renzo, Stud. Surf. Sci. Catal. 97 (1995) 173-180.

[60] P. Satya, J. H. Clark, J. Mol. Catal. A: Chem. 215 (2004) 107–111.

[61] G. Schaftenaar, J.H. Noordik, J. Comput. Aided Mol. Design 14 (2000) 123-134.

[62] M.W. Schmidt, K.K. Baldridge, J.A. Boatz, S.T. Elbert, M.S. Gordon, J.H. Jensen, S.Koseki, N. Matsunaga, K.A. Nguyen, S.J. Su, T.L. Windus, J. Comput. Chem. 14 (1993) 1347-1363.

[63] J.M. Gonzalez-Calbert, M.L. Ruiz-Gonzalez, M. Vallet-Regi, in: K.D.M. Harris, P.P. Edwards (Eds.), Turning Points in Solid-State, Materials and Surface Science, Royal Society of Chemistry, Great Britain, 2008, pp 733.

[64] J. George, S. Shylesh, A.P. Singh, Appl. Catal. A. Gen. 290 (2005) 148-158.

[65] N. Rangelova, G. Chernev, S. Nenkova, I.M. Miranda Salvado, Nanoscience and Nanotechnology 11 (2011) 191- 194.

[66] S. Dzwigaj, M.Matsuoka, M.Anpo,M.Che, J. Phys. Chem.B.104 (2000) 6012-6020.

[67] M. Mureseanu, A. Reiss, I. Stefanescu, E. David, V. Parvulescu, G. Renard, V. Hulea, Chemosphere 73 (2008b) 1499-1504.

[68] S. H. Sumrra, Z. H. Chohan, Spectrochim. Acta, Part A 98 (2012) 53–61.

[69] X.Zhu, R.Shen, L.Zhang, Chin. J. Catal. 35(2014) 1716-1726.

[70] R.Figueiredo, A.M.Arias, M.L.Granados, J.L.Fierro, J.Catal. 178 (1998) 146-152.

[71] D.P. Das, K.M. Parida, Catal. Lett. 128 (2009) 111–118.

[72] R. Noyori, M. Aoki, K. Sato, Chem. Commun. 16 (2003) 1977-1986.

[73] A. Wroblewska, J. Wajzberg, A. Fajdek, J. Hazardous Mat. 8 (2008) 1303-1309.

[74] N. Malumbazo, S. F. Mapolie, J. Mol. Catal. A: Chem. 312 (2009) 70–77.

[75] D. Habibi, A. Reza Faraji, Applied Surface Science 276 (2013) 487– 496.

Tables

Tabel 1. Textural and structural properties of modified mesoporous molecular sieves with aminopropyl groups

aHexagonal:; bD 4V/A; cMesopore volume through method BdB; dWall thickness = a0–D.

aCubic: ;bD 4V/A; cMesopore volume through method BdB; d Wall thickness = a0–D.

Table 2. Chemical composition of the metal complexes and their immobilization yield

aDetermined by TGA after subtraction of the aminopropyl moiety previously grafted on the silica support.

Table 3. Catalytic performance of copper and manganese complexes immobilized

into mesoporous support for the oxidation of cyclohexene

*Products formed: 2-cyclohexen-1-one (I), 2-cyclohexen-1-ol (II), 1,2-cyclohexendiol (III) and cyclohexene oxide (IV)

Table 4. Catalytic performance of copper and manganese complexes immobilized

into mesoporous support for the oxidation of alcohols

Table 5 Reusability tests for SBA-15MetMn and SBA-15MetCu in CH oxidation reaction

Figure captions

Fig. 1 The initial (i) and final (f) structures of the copper complexes, formed with ligands A, B and C, before and after optimizing processes (hydrogen- white atom; carbon-brown atom; oxygen-red atom; nitrogen-blue atom; silicon-gray atom; copper-yellow).

Fig. 2. XRD patterns of (a) SBA-15, SBA-15-NH2; (b) MCM-41, MCM-41-NH2; (c) MCM-48, MCM-48-NH2.

Fig. 3. SEM images of A-MCM-41, B-MCM-48, C-SBA-15 mesoporous supports.

Fig. 4. TEM images of A-MCM-41, B-MCM-48, C-SBA-15 mesoporous supports.

Fig. 5. FTIR spectra of the support and of the metal complexes with Schiff base ligands supported on SBA-15 mesoporous silica.

Fig. 6. Effect of ligand (A) and support (B) on mass loss evidenced by TGA analysis.

Fig. 7. Variation of the ligand concentration on support surface for copper complexes (A) and manganese complexes (B).

Fig. 8. Cu2p3/2 XPS superimposed spectra for SBA-15Cu, MCM-41Cu and MCM-48Cu samples.

Fig. 9. Cu2p3/2 XPS deconvoluted spectrum for SBA-15-MetCu, MCM-48-HydCu and MCM-41-MetCu samples.

Fig. 10. N1s XPS deconvoluted spectrum for SBA-15-MetMn sample.

Fig.11. N1s XPS deconvoluted spectrum for MCM-41-MetCu sample.

Fig.12. Mn2p XPS superimposed spectra for SBA-15Mn, SBA-15-MetMn and MCM-48-MetMn samples.

Fig. 13. The variation of TOF values for the supported MetCu(A) and MetMn(B) catalysts in the cyclohexene and alcohols oxidation as a function of the mesoporous silica type.

Fig. 14. The variation of TOF values for the supported FurCu(A) and FurMn(B) catalysts in the cyclohexene and alcohols oxidation as a function of the mesoporous silica type.

Fig. 15. The variation of TOF values for the Mn complexes supported on MCM-41 in the cyclohexene and alcohols oxidation as a function of ligand type.

Figures

Fig. 2.

Fig. 3.

Fig. 4.

Fig. 1.

Fig. 5.

Fig. 6.

Fig .7

Fig. 8.

Fig. 9.

Fig. 10.

Fig.11.

Fig.12.

Fig. 13.

Fig. 14.

Fig. 15.

Scheme caption

Scheme 1. Synthesis of Cu (II) and Mn (II) complexes with Schiff base ligands derived from 2-furaldehyde, 2-furylmethylketone and 2-hydroxyacetopheneone on SBA-15 silica mesoporous support.

Scheme

Scheme 1