Zeolite-encapsulated copper(II) complexes of pyridoxal-based tetradentate ligands for the oxidation of styrene, cyclohexene and methyl phenyl sulfide

FULL PAPER

DOI: 10.1002/ejic.200700797

Zeolite-Encapsulated Copper(II) Complexes of Pyridoxal-Based Tetradentate

Ligands for the Oxidation of Styrene, Cyclohexene and Methyl Phenyl Sulfide

Mannar R. Maurya,*^[a]Baljit Singh,^[a]Pedro Adão,^[b]Fernando Avecilla,^[c]and

João Costa Pessoa^[b]

Keywords: Zeolites / Copper(II) complexes / Heterogeneous catalysis / Oxidation reactions

Interaction of copper(II)-exchanged zeolite-Y with N,NЈ-eth-

version, respectively, using tert-butyl hydroperoxide as oxi-

ylenebis(pyridoxyliminato) (H₂pydx-en, I) and N,NЈ-propyl- dant, where styrene oxide, benzaldehyde, benzoic acid and

enebis(pyridoxyliminato) (H₂pydx-1,3-pn, II) ligands in re-

fluxing methanol leads to the formation of the corresponding

complexes, abbreviated herein as [Cu(pydx-en)]-Y (3) and

[Cu(pydx-1,3-pn)]-Y (4), in the supercages of zeolite-Y. The

neat complexes, [Cu(pydx-en)] (1) and [Cu(pydx-1,3-pn)] (2),

have also been prepared with these ligands. Spectroscopic

studies (IR, UV/Vis and EPR), elemental analyses, thermal

studies, field emission scanning electron micrographs (FE-

SEM) and X-ray diffraction patterns have been used to char-

acterise these complexes. The crystal and molecular struc-

tures of 1 and of 2·CH₃OH, have been determined, confirm-

ing the ONNO binding mode of the ligands. The geometry

around the metal ion is very slightly distorted square-planar

in 1 and distorted square-pyramidal in 2. The encapsulated

complexes catalyse the oxidation, by H₂O₂and tert-butyl hy-

droperoxide, of styrene, cyclohexene and thioanisole ef-

ficiently. Under optimised reaction conditions, the oxidation

of styrene catalysed by 3 and 4 gave 23.6% and 28.0% con-

phenylacetaldehyde are the major products. Better conver-

sions have been obtained using H₂O₂as oxidant. Oxidation

of cyclohexene catalysed by these complexes gave cyclohex-

ene oxide, 2-cyclohexen-1-ol, cyclohexane-1,2-diol and 2-cy-

clohexen-1-one as the major products. A maximum of 90.1%

conversion of cyclohexene with 3 and 83.0% with 4 was ob-

tained under optimised conditions. Similarly, a maximum of

80.3% conversion of methyl phenyl sulfide with 3 and 81.0%

with 4 was observed, where the selectivity of the major prod-

uct methyl phenyl sulfoxide was found to be about 60%.

Tests for the recyclability and heterogeneity of the reactions

have also been carried out, and the results indicate their re-

cyclability. A possible reaction mechanism has been pro-

posed by titrating a methanol solution of 1 and 2 with H₂O₂

to identify the possible intermediates.

Germany, 2007)

heterogeneous catalysts and share many of the advan-

tageous features of homogeneous catalysts.

Introduction

Zeolites find applications in many scientific disciplines as

catalysts and in all types of chemical engineering process

technologies. The exchangeable property of extra-frame-

work cations and suitable cavity size of the zeolites allow

their modification by inclusion of chemically interesting

molecules. A homogeneous catalyst, for example, a metal

complex, may be considered one such interesting molecule

and hence the term zeolite-encapsulated metal complexes.

The large size of the encapsulated homogeneous catalysts

and their rigidity make it difficult for them to escape the

zeolite cages. Zeolite-encapsulated metal complexes having

good catalytic activities possess the advantages of solid

The flexible nature of salen-type [H₂salen = N,NЈ-bis-

(salicylidene)ethane-1,2-diamine] ligands has provided op-

portunities to design various transition-metal complexes in

the nanocavity of zeolites and to develop catalytic processes

for various reactions. Poltowicz et al. have encapsulated a

whole range of metallosalen complexes {e.g., [Fe(salen)],

[Mn(salen)], [Cu(salen)] and Co(salen)]} in zeolite-X to

study their catalytic activity for the oxidation of cyclooc-

tane.^[1]Ratnasamy and co-workers have used copper(II)

and manganese(III) complexes of salen derivatives encapsu-

lated in the cavity of zeolite-X and zeolite-Y for the oxi-

dation of styrene under aerobic conditions using tert-butyl

hydroperoxide as an oxidant. The catalytic efficiency of

these encapsulated complexes was much higher than that of

the neat complexes. Electron-withdrawing substituents such

as Cl, Br and NO₂on the aromatic ring enhance the rate

of oxidation.^[2,3]These complexes also catalyse the oxi-

dation of phenol and p-xylene.^[4,5]Epoxidations of cyclo-

hexene, cyclooctene, 1-hexene and other various types of

alkenes, arenes and cycloalkenes, catalysed by different

[a] Department of Chemistry, Indian Institute of Technology

Roorkee,

Roorkee 247667, India

E-mail: rkmanfcy@iitr.ernet.in

[b] Centro Química Estrutural, Instituto Superior Técnico – TU

Lisbon,

Av. Rovisco Pais, 1049-001 Lisboa, Portugal

[c] Departamento de Química Fundamental, Facultad de Ciencias,

Universidade da Coruña,

Zapateira s/n, 15071 A Coruña, Spain

5720

Eur. J. Inorg. Chem. 2007, 5720–5734

Zeolite-Encapsulated Copper Complexes of Tetradentate Ligands

complexes of salen-type ligands, have also been carried

Reaction of cupric acetate with ligands H₂pydx-en (I)

out.^[6–9][Co(salophen)]-Y [H₂salophen = N,NЈ-bis(salicylid- and H₂pydx-1,3-pn (II) in refluxing methanol yielded [Cu-

ene)benzene-1,2-diamine] and related derivatives catalysed (pydx-en)] (1) and [Cu(pydx-1,3-pn)(MeOH)] (2), respec-

the oxidation of β-isophorone (βIP) to ketoisophorone tively. Complex 1 has previously been isolated by treating I

(KIP) (along with other minor products) in the presence with cupric acetate in aqueous solution.^[16]Encapsulation

of an oxidant at ambient conditions of temperature and of these complexes in the nanocavities of zeolite-Y involved

pressure.^[10]Liquid-phase oxidation of phenol with H₂O₂the exchange of copper(II) ion with Na-Y in water followed

has been reported using copper(II) and oxidovanadium(IV) by reaction of the metal-exchanged zeolite-Y with I or II in

complexes of N,NЈ-bis(salicylidene)propane-1,3-diamine methanol. This method is referred to as the flexible ligand

(H₂sal-1,3-pn) and N,NЈ-bis(salicylidene)diethylenetriamine method. Here, the ligands enter into the cavity of zeolite-Y

(H₂saldien).^[11–13]Copper(II) and oxidovanadium(IV) com- because of their flexible nature and interact with the metal

plexes of N,NЈ-bis(salicylidene)cyclohexane-1,2-diamine ion to give [Cu(pydx-en)]-Y (3) and [Cu(pydx-1,3-pn)]-Y

(H₂sal-dach) catalyse the oxidation of styrene, cyclohexene (4). Finally, extraction of impure samples with methanol

and cyclohexane efficiently.^[14]

using a Soxhlet extractor removed excess free ligand, and

The catalytic potential of metal complexes encapsulated stirring in DMF removed neat metal complex formed on

in the nanocavity of zeolites prompted us to design new the surface of the zeolite, if any. As the crude mass was well

zeolite-Y-encapsulated copper(II) complexes of tetradentate extracted, the metal content found after encapsulation is

ligands I and II (Scheme 1). Catalytic potentials of these only due to the presence of copper(II) complexes in the su-

complexes have been demonstrated by studying the oxi- percages of the zeolite-Y. The diagonal dimension of the

dation of styrene, cyclohexene and methyl phenyl sulfide. complexes is about 12 Å {in [Cu(pydx-en)]} or about 11 Å

Copper(II), nickel(II) and oxidovanadium(IV) complexes of {in [Cu(pydx-1,3-pn)(MeOH)]}, which suggests that they

I have been reported previously.^[15,16]The structure of [Ni- will fit nicely into the supercages of zeolite-Y. The molecu-

(pydx-en)] has been confirmed by single-crystal X-ray dif- lar formulae of the encapsulated complexes are based on

fraction,^[17]while other complexes have been characterised the respective neat complex. The Si/Al ratio of 9.67 of zeo-

by elemental analysis and spectroscopic techniques. We lite-Y taken here corresponds to a unit cell formula

have also prepared neat copper(II) complexes of I and II to Na₁₈(AlO₂)₁₈(SiO₂)₁₇₄. The estimated number of copper

evaluate their catalytic potential and characterised them by atoms per unit cell is 2.5 for [Cu(pydx-en)]-Y (3) and 2.4

single-crystal X-ray diffraction.

for [Cu(pydx-1,3-pn)]-Y (4).

Structure Description of [Cu(pydx-en)] (1) and [Cu(pydx-

1,3-pn)(MeOH)] (2)

Figures 1 and 2 include ORTEP representations of the

molecular structure of [Cu(pydx-en)] (1) and [Cu(pydx-1,3-

pn)(MeOH)] (2), respectively, and Table 1 provides selected

bond lengths and angles of these compounds. The mole-

cules are neutral, as both phenol groups are deprotonated,

and the ligands coordinate to the Cu atom through two

phenolate oxygen (O^–_phenolate) atoms and two imine nitro-

gen (N_imine) atoms. The geometry around the metal ion is

slightly distorted square-planar in 1 and distorted square-

pyramidal in 2. A methanol molecule occupies the fifth co-

ordination position in 2.

Scheme 1.

Results and Discussion

Synthesis and Characterisation of Catalysts

Scheme 2 presents the complexes prepared in this study.

Their structures are based on elemental analysis and vari-

ous spectroscopic studies, as well as on X-ray structure de-

terminations. The presence of very strong peaks corre-

sponding to CuLH⁺and CuLNa⁺in the ESI-MS spectra

(positive mode) of DMSO solutions of 1 and 2, with the

correct isotope ratios, also confirms the integrity of the

complexes in solution.

Figure 1. ORTEP diagram of [Cu(pydx-en)] (1) showing the atom-

labelling scheme. The thermal ellipsoids were drawn at the 30%

probability level.

Scheme 2.

Eur. J. Inorg. Chem. 2007, 5720–5734

www.eurjic.org

5721

M. R. Maurya, B. Singh, P. Adão, F. Avecilla, J. Costa Pessoa

FULL PAPER

Figure 2. ORTEP diagram of [Cu(pydx-1,3-pn)(MeOH)] (2) showing the atom-labelling scheme. The thermal ellipsoids were drawn at the

30% probability level.

Table 1. Selected bond lengths [Å] and angles [°] for [Cu(pydx-en)] (1) and [Cu(pydx-1,3-pn)(MeOH)] (2).

Bond lengths [Å] for [Cu(pydx3-en)] (1)

Bond lengths [Å] for [Cu(pydx-1,3-pn)(MeOH)] (2)

Cu(1)–O(2)

Cu(1)–N(3)

Cu(1)–O(3)

Cu(1)–N(2)

O(3)–C(13)

O(2)–C(6)

N(2)–C(8)

N(2)–C(9)

N(3)–C(11)

N(3)–C(10)

C(6)–C(7)

1.742(6)

1.796(7)

2.341(7)

2.373(9)

1.333(11)

1.380(12)

1.256(12)

1.377(11)

1.434(13)

1.690(9)

1.609(15)

1.338(13)

1.695(15)

1.330(13)

Cu(1)–O(2)

Cu(1)–O(3)

Cu(1)–N(3)

Cu(1)–N(2)

Cu(1)–O(1S)

O(2)–C(6)

N(2)–C(8)

N(2)–C(9)

O(3)–C(14)

N(3)–C(12)

N(3)–C(11)

C(6)–C(7)

1.8965(9)

1.9546(9)

1.9932(11)

1.9935(11)

2.3525(11)

1.2934(15)

1.2962(16)

1.4775(16)

1.2998(15)

1.2905(16)

1.4670(16)

1.4096(17)

1.4443(17)

1.5149(18)

1.5198(19)

1.4494(17)

1.4082(17)

C(7)–C(8)

C(11)–C(12)

C(12)–C(13)

C(7)–C(8)

C(9)–C(10)

C(10)–C(11)

C(12)–C(13)

C(13)–C(14)

Angles [°] for [Cu(pydx-1,3-en)] (1)

Angles [°] for [Cu(pydx-1,3-pn)(MeOH)] (2)

O(2)–Cu(1)–N(3)

O(2)–Cu(1)–O(3)

N(3)–Cu(1)–O(3)

O(2)–Cu(1)–N(2)

N(3)–Cu(1)–N(2)

O(3)–Cu(1)–N(2)

175.2(3)

90.0(3)

94.8(3)

91.2(3)

84.1(3)

178.8(2)

O(2)–Cu(1)–N(3)

O(2)–Cu(1)–O(3)

O(3)–Cu(1)–N(3)

O(2)–Cu(1)–N(2)

N(3)–Cu(1)–N(2)

O(3)–Cu(1)–N(2)

O(2)–Cu(1)–O(1S)

O(3)–Cu(1)–O(1S)

N(3)–Cu(1)–O(1S)

N(2)–Cu(1)–O(1S)

169.92(4)

82.08(4)

87.94(4)

92.29(4)

97.76(4)

166.61(4)

90.18(4)

94.46(5)

89.23(4)

97.71(5)

Lengths [Å] and angles [°] of hydrogen bonds in complexes

O(1)···H(1), 0.82; O(1)···O(3), 2.563(8)

O(3)···H(1), 2.00; O(1)–H(1)–O(3), 125.5

O(4)···H(4), 0.82; O(4)···N(1), 2.582(11)

N(1)···H(4), 2.01; O(4)–H(4)–N(1), 126.5

O(1)···H(1O), 0.69(2); O(1)···O(3), 2.7954(14)

O(3)···H(1O), 2.11(2); O(1)–H(1O)–O(3), 171(2)

O(1S)···H(1S), 0.64(3); O(1S)···N(1), 2.8037(16)

N(1)···H(1S), 2.16(3); O(1S)–H(1S)–N(1), 175(3)

O(4)···H(4O), 0.69(2); O(4)···N(4), 2.7959(15)

N(4)···H(4O), 2.12(2); O(4)–H(4O)–N(4), 166(2)

Deviations from the planarity of the rings

Ring: Cu(1)–O(2)–C(6)–C(7)–C(8)–N(2)

Dev.: 0.0799(52) Å

Ring: Cu(1)–N(2)–C(9)–C(10)–N(3)

Dev.: 0.0976(48) Å

Ring: Cu(1)–O(2)–C(6)–C(7)–C(8)–N(2)

Dev.: 0.0516(07) Å

Ring: Cu(1)–N(2)–C(9)–C(10)–C(11)–N(3)

Dev.: 0.2239(08) Å

Ring: Cu(1)–N(3)–C(11)–C(12)–C(13)–O(3)

Dev.: 0.0334(50) Å

Pyridoxal rings:

Ring: Cu(1)–N(3)–C(12)–C(13)–C(14)–O(3)

Dev.: 0.1928(07) Å

Pyridoxal rings:

Planar, rms 0.0244(63) and 0.0172(60) Å

Torsion angle between rings: 9.58 (0.57)°

Planar, rms 0.0037(08) and 0.0083(09) Å

Torsion angle between rings: 34.27(0.05)°

5722

www.eurjic.org

Eur. J. Inorg. Chem. 2007, 5720–5734

Zeolite-Encapsulated Copper Complexes of Tetradentate Ligands

The planar arrangement Cu(ONNO) is distorted, the

cis/trans angles not being 90° and 180°. The cis angles are

90.0(3), 94.8(3), 91.2(3) and 84.1(3)° in 1, and 82.08(4),

87.94(4), 92.29(4) and 97.76(4)° in 2, and the trans angles

are 175.2(3) and 178.8(2)° in 1, and 169.92(4) and

166.61(4)° in 2. The angles with the apical position of the

O atom of the methanol molecule are 90.18(4), 94.46(5),

89.23(4) and 97.71(5)° in 2.

In complex 1 the ligand forms a set of (6+5+6)-mem-

bered chelate rings, and the deviations from the planarity

are listed in Table 1. The plane formed by the donor atoms

N(2)–N(3)–O(2)–O(3) has a deviation from planarity of

0.0015(30) Å, and if the Cu^IIion is included, 0.0037(30) Å.

In complex 2 the ligand forms a set of (6+6+6)-membered

chelate rings, and the deviations from planarity are also

listed in Table 1. The basal plane of the square pyramid

formed by N(2)–N(3)–O(2)–O(3) atoms has a deviation

from planarity of 0.1063(05) Å, and the Cu^IIion is 0.1020 Å

above the plane. The Cu–O and Cu–N bond lengths

(Table 1) are similar and in the usual range found for these

compounds, 1.9–2.1 Å. The C–O distances are also typical

for this type of ligands,^[15–21]and the C=N bonds are

double bonds, with distances of about 1.3 Å.

In complex 1 the HO(CH₂) groups of the pyridoxal rings

are in a syn position to each other, but in complex 2 the

HO(CH₂) groups of the pyridoxal rings are in an anti con-

formation, similar to other complexes with this type of li-

gand.^[16]The pyridine nitrogen atoms and alcohol group of

HO(CH₂) of the pyridoxal rings are involved in intermo-

lecular H bonds. In complex 2, the OH groups of the coor-

dinated methanol molecule are also involved in H bonds.

The bond lengths and angles of the hydrogen bonds are

also included in Table 1.

Figure 3. Field emission scanning electron micrograph of [Cu-

(pydx-en)]-Y (top) and [Cu(pydx-1,3-pn)]-Y (bottom).

Field Emission Scanning Electron Micrograph and Energy

Dispersive X-ray Analysis Studies

Powder X-ray Diffraction Studies

Figure 3 presents the field emission scanning electron

micrographs of [Cu(pydx-en)]-Y and [Cu(pydx-1,3-pn)]-Y.

It is clear from the micrographs that zeolite-entrapped cop-

per complexes have well-defined crystals, and there is no

indication of the presence of any metal ions or complexes

on the surface. Energy-dispersive X-ray analysis plots sup-

port this conclusion as no copper or nitrogen contents were

noted on the spotted surface in plots for both catalysts. The

average silicon and aluminium percentage obtained on the

spotted surfaces were about 30% and 10%, respectively. An

amount of about 3% sodium suggests the exchange of re-

maining free copper ions by sodium ions during the re-ex-

change process (see Experimental Section). Only a small

amount of carbon (about 5%) but no nitrogen suggests the

presence of a trace amount of solvent (methanol) from

which it was finally washed after Soxhlet extraction. No

morphological changes on the surface upon encapsulation

of the complexes in the cavity are seen because of their poor

loading.

The powder X-ray diffraction patterns of Na-Y, Cu-Y

and encapsulated copper(II) complexes were recorded at 2θ

values between 5 and 50°, and some representative patterns

are presented in Figure 4. Essentially similar diffraction

Figure 4. XRD patterns of Na-Y, Cu-Y and [Cu(pydx-en)]-Y (3).

Eur. J. Inorg. Chem. 2007, 5720–5734

www.eurjic.org

5723

M. R. Maurya, B. Singh, P. Adão, F. Avecilla, J. Costa Pessoa

FULL PAPER

patterns were noticed in encapsulated complexes Cu-Y and hydrogen bond followed by the coordination of the phenol

Na-Y, except zeolite with encapsulated metal complexes has oxygen atom after deprotonation. Existence of this band,

a slightly weaker intensity. These observations indicate that however, in neat complexes hints towards the presence of

the framework of the zeolites do not suffer any significant hydrogen bonding. A sharp band appearing at 1626 cm^–1

structural changes during encapsulation, that is, crystal- (in I) or at 1632 cm^–1(in II) due to the ν(C=N) (azome-

linity of the zeolite-Y is preserved.

thine) shifts to higher wavenumbers in complexes, thereby

suggesting the coordination of the azomethine nitrogen

atom. Several multiple bands of medium intensity covering

the region 2850–3000 cm^–1are observed due to C–H bands.

The spectral patterns of these complexes are similar to

[Cu(pydx-en)] reported in the literature.^[16]

Thermogravimetric Analysis Studies

The decomposition of complexes 3 and 4 occurs in three

overlapping steps. An endothermic loss of trapped water

and methanol occurs between 80 and 200 °C, while exother-

mic removal of intra-zeolite water occurs between 300 and

350 °C. This is followed by endothermic loss of the ligand’s

residue, which continues until the removal of all organic

UV/Vis Spectral Studies

UV/Vis spectral studies of I and its copper(II) complex

mass at about 700 °C has occurred. The low percentage of 1 have been discussed in detail in the literature.^[16]Other

weight loss (about 10%) for this last step indicates the pres- complexes reported here exhibit very similar spectral pat-

ence of an only small amount of metal complexes in the terns. The well-defined band at 590 nm in 1 and a broad

cavities of the zeolite. This is in agreement with the low band covering the region 570–650 nm in 2 have been as-

copper content estimated for these complexes. A weight loss signed to a d–d transition. Such a d–d band in zeolite-

of 6.9%, equivalent to methanol, in 2 occurs up to about encapsulated complexes could not be obtained because of

130 °C. The anhydrous complex is stable up to 260 °C, and their poor loading, while ligand bands appearing in the UV

then it decomposes in two major steps that are complete at region are similar to those observed for neat complexes.

375 °C. The remaining residue of 18.1% slowly stabilises This data is collected in Table 2 and the spectra of encapsu-

with the formation of a final residue of 17.5% (about 17.1% lated complexes are presented in Figure 5.

for CuO) at about 800 °C. A slightly different decomposi-

tion pattern has been obtained for 1, where the decomposi-

tion of the stable complex starts at about 290 °C and occurs

in two overlapping steps. The decomposition is complete at

about 360 °C. The remaining residue of 20.6% further loses

weight up to about 800 °C and stabilises at 18.3%. The cal-

culated residue for CuO of 18.9% agrees with the value ob-

served.

IR Spectral Studies

A partial list of IR spectroscopic data is presented in

Table 2. The intensity of the peaks in encapsulated com-

Figure 5. UV/Vis spectra of [Cu(pydx-en)]-Y and [Cu(pydx-1,3-

plexes is, however, weak because of their low concentration pn)]-Y dispersed in nujol.

in zeolite matrix; the spectra of the encapsulated as well as

neat complexes showed essentially similar bands. Compari-

son of the spectra of these complexes with the respective

ligand provides evidence for the coordinating mode of li-

EPR Studies

gands in complexes. The ligand H₂pydx-en exhibits a me-

The EPR spectra of the neat Cu^IIcomplexes 1 and 2 in

dium-intensity band around 2700 cm^–1due to intramolecu- the polycrystalline state are broad because of dipolar and

lar hydrogen bonding. Absence of this band in the spectra spin–spin exchange interactions. It is clear that they are

of encapsulated complexes indicates the destruction of the characterised by an axial g tensor, but the hyperfine features

Table 2. IR and UV/Vis spectroscopic data of ligand, pure and encapsulated complexes.

Compound

IR [cm^–1]

UV/Vis

ν(OH)

ν(C=N)

Solvent

λ_max[nm]

H₂pydx-en (I)

H₂pydx-1,3-pn (II)

[Cu(pydx-en)] (1)

[Cu(pydx-1,3-pn)(MeOH)] (2)

[Cu(pydx-en)]-Y (3)

[Cu(pydx-1,3-pn)]-Y (4)

3280–3450

2846–3480

ca. 3400

ca. 3450

ca. 3400

ca. 3450

1626

1632

1634

1635

1633

1638

CH₃OH

nujol

216, 252, 335

215, 252, 334

235, 277, 378

233, 273, 377

207, 283, 374

217, 246, 402

nujol

5724

www.eurjic.org

Eur. J. Inorg. Chem. 2007, 5720–5734

Zeolite-Encapsulated Copper Complexes of Tetradentate Ligands

were not resolved, neither at room temperature nor at 77 K, ilar g_ʈand A_ʈparameters and g_ʈ/A_ʈratios were found for

and not much structural information can be obtained. The [Cu^II(salen)] (2.212, 203.7ϫ10^–4cm^–1, 109, respectively)

EPR spectra of the neat Cu^IIcomplexes 1 and 2 in DMF and [Cu^II(sal-1,3-pn)] (2.254, 172.5ϫ10^–4cm^–1, 131, respec-

solutions, and of the catalysts 3 and 4 are given in Figures 6 tively),^[25]in DMF solutions, indicating very similar geome-

and 7.

tries and balance of charge to those of 1 and 2. No super-

hyperfine structure is distinguished.

The EPR spectra of the encapsulated complexes were re-

corded at room temperature and are broader than those for

frozen solutions, suggesting that the complexes located in

the supercages may be present in more than one geometri-

cal conformation. However, the g_ʈand A_ʈparameters and

g_ʈ/A_ʈratios (g_ʈ/A_ʈ= 113 for 3 and 126 for 4) almost coincide

with those obtained in DMF solutions, indicating similar

geometries and balance of charge for the complexes in the

encapsulated and dissolved forms.

Figure 6. EPR spectra of a frozen DMF solution of [Cu(pydx-en)]

(1) and of [Cu(pydx-en)]-Y (3) at room temperature.

Catalytic Activity Studies

The catalytic activities of the zeolite-encapsulated and

neat complexes have been demonstrated by studying the

oxidation of styrene, cyclohexene and methyl phenyl sulfide.

Oxidation of Styrene

The oxidation of styrene, catalysed by [Cu(pydx-en)]-Y

and [Cu(pydx-1,3-pn)]-Y, was carried out using H₂O₂and

tert-butyl hydroperoxide as oxidants to give styrene oxide,

benzaldehyde, benzoic acid and phenylacetaldehyde

(Scheme 3) along with minor amounts of unidentified prod-

ucts. Hulea and Dumitriu^[26]have reported some of these

products while using catalysts TS-1 and MCM-41. All oxi-

dation products as observed here were identified recently

using the polymer-anchored catalyst PS-[VO(sal-ohyba)-

(DMF)] (H₂sal-ohyba = Schiff base derived from salicylal-

dehyde and o-hydroxybenzylamine)^[27]and the zeolite-Y-en-

capsulated complex [VO(sal-dach)]-Y.^[14]

Figure 7. EPR spectra of a frozen DMF solution of [Cu(pydx-1,3-

pn)(MeOH)] (2) and of [Cu(pydx-1,3-pn)]-Y (4) at room tempera-

ture.

The spectra of the encapsulated complexes 3 and 4 are

reasonably resolved, indicating the encapsulation of individ-

ual molecules in the supercages of Na-Y. The inter-

molecular interactions are avoided because of isolation of

the zeolite cavities, indicating that the Cu^IIsites are well

separated in the catalyst, and thereby the spin Hamiltonian

parameters can be estimated with reasonable accuracy;

these are listed in Table 3.

Both neat complexes 1 and 2 have an N₂O₂binding mode

and are monomeric in the solid state. Complex 1 has a

square-planar configuration, and 2 has square-planar con-

figuration with a pyramidal distortion. In DMF solutions

the representation of the A_ʈ, g_ʈvalues fit well with those of

other [Cu^II(salen)] complexes.^[5,22–25]The g_ʈ/A_ʈratio is 112.7

for 1, indicating a remarkable planarity of the square-

planar geometry, as found in the solid state. For 2 this ratio

is 126, but this is mainly due to a decrease in A_ʈ. Very sim- Scheme 3.

Table 3. EPR data for [Cu(pydx-en)] and [Cu(pydx-1,3-pn)(MeOH)] complexes as neat and encapsulated in zeolite-Y.

Complex

g_xx

g_yy

g_zz(ʈ)

A_xx

A_yy

A_zz(ʈ)

(ϫ10⁴cm^–1)

[Cu(pydx-en)] (1)

2.07

2.04

2.03

2.06

2.22

2.24

2.32

2.34

17.39

21.85

28.43

16.88

196.94

178.53

156

[Cu(pydx-1,3-pn)(MeOH)] (2)

[Cu(pydx-en)]-Y (3)

2.08

2.09

–

[Cu(pydx-1,3-pn)]-Y (4)

138

Eur. J. Inorg. Chem. 2007, 5720–5734

www.eurjic.org

5725

M. R. Maurya, B. Singh, P. Adão, F. Avecilla, J. Costa Pessoa

FULL PAPER

In order to achieve the maximum oxidation of styrene,

catalyst 3 was taken as a representative, and the effect of

four different reaction parameters – viz. the amount of

H₂O₂(mol of H₂O₂per mol of styrene), amount of catalyst,

volume of solvent and temperature of the reaction – was

studied.

Three different aqueous 30% H₂O₂/styrene molar ratios

of 1:1, 2:1 and 3:1 were considered while keeping the fixed

amount of styrene (1.04 g, 10 mmol) and catalyst (0.025 g)

in CH₃CN (20 mL) at 80 °C. The periodic analyses of re-

sults up to 6 h are illustrated in Figure 8. Increasing the

H₂O₂/styrene ratio from 1:1 to 2:1 improved the conversion

from 9.1% to 14.8% while a 3:1 ratio showed a further in-

crement of conversion of hardly 1%. Therefore, it is clear

that the 2:1 molar ratio is quite suitable to obtain the opti-

mum styrene conversion of 14.8% in the 6 h reaction time.

Figure 9. Effect of the amount of catalyst on the oxidation of sty-

rene. Reaction conditions: styrene (1.04 g, 10 mmol), 30% H₂O₂

(2.27 g, 20 mmol), CH₃CN (20 mL) and temp. (80 °C).

Figure 10. Effect of the temperature on the oxidation of styrene.

Reaction conditions: styrene (1.04 g, 10 mmol), [Cu(pydx-en)]-Y

(0.025 g), 30% H₂O₂(2.27 g, 20 mmol) and CH₃CN (20 mL).

Figure 8. Effect of H₂O₂on the oxidation of styrene. Reaction con-

ditions: styrene (1.04 g, 10 mmol), [Cu(pydx-en)]-Y (3) (0.025 g),

CH₃CN (20 mL) and temp. (80 °C).

Thus, for the maximum oxidation of 10 mmol of styrene,

the other required conditions as concluded were: [Cu(pydx-

Similarly, for three different amounts – viz. 0.015, 0.025 en)]-Y (0.025 g), H₂O₂(2.27 g, 20 mmol), CH₃CN (10 mL)

and 0.035 g – of [Cu(pydx-en)]-Y at a styrene/aqueous 30% and temperature (80 °C). Under these optimised reaction

H₂O₂ratio of 1:2 under the above reaction conditions, conditions, the catalytic activities of 1–4 were further tested,

0.015 g of catalyst, achieved only 7.8% conversion, whereas and the results obtained are shown in Figure 11. Table 4

0.025 g of catalyst gave a maximum conversion of 14.8% in compares the selectivity of the various products obtained

6 h of reaction time. A further increment of catalyst after 6 h of reaction time, along with the percent conversion

(0.035 g) has shown no improvement (Figure 9). The reason of styrene and turnover frequency. It is clear from Table 4

for reduced activity at higher catalyst dose may be adsorp- and Figure 11 that [Cu(pydx-1,3-pn)]-Y exhibits 37.4%

tion/chemisorption of the two reactants on separate catalyst conversion, which is better than that exhibited by [Cu(pydx-

particles, thereby reducing the chance to interact.

en)]-Y (33.3%). Both catalysts gave mainly four products:

Figure 10 illustrates the oxidation of styrene at three dif- styrene oxide, benzaldehyde, phenylacetaldehyde and ben-

ferent temperatures – viz. 60, 70 and 80 °C –, while keeping zoic acid. The selectivity for benzaldehyde is much higher,

the optimised conditions of 10 mmol of styrene, 20 mmol while the other products are obtained in small amounts.

of H₂O₂and 0.025 g of catalyst in 20 mL of acetonitrile. It

We have also tested the catalytic activity of the neat com-

is evident from the figure that the performance of the cata- plexes 1 and 2 for the oxidation of styrene using an equi-

lyst is poor at 60 °C wherein only 3.9% conversion was ob- molar concentration of metal ion as in their respective zeo-

served. Increasing the reaction temperatures to 70 and lite-encapsulated metal complex. Figure 11 also provides

80 °C increases the conversion to 7.0% and 14.8%, respec- the conversion as a function of time, while Table 4 lists all

tively. Thus, conducting the catalytic reaction at 80 °C is other details. Thus, under the above reaction conditions, the

most suitable to maximise the oxidation of styrene. We have conversion obtained by the neat complexes (18.7% for 1

also observed that reducing the volume of CH₃CN to and 22.3% for 2) is significantly lower than that shown by

10 mL under the above optimised reaction conditions im- the respective encapsulated complexes. Again, all four prod-

proved the conversion considerably, and two more compo- ucts have been obtained with the selectivity order: benzalde-

nents, though in poor yield, have also been obtained.

hyde Ͼ styrene oxide Ͼ phenylacetaldehyde Ͼ benzoic acid.

5726

www.eurjic.org

Eur. J. Inorg. Chem. 2007, 5720–5734

Zeolite-Encapsulated Copper Complexes of Tetradentate Ligands

Figure 12. Catalytic comparison of zeolite-encapsulated metal

complexes and neat complexes for the oxidation of styrene in the

presence of TBHP as oxidant. Reaction conditions: styrene (1.04 g,

10 mmol), catalyst (0.025 g for encapsulated and 0.00133 g in case

of neat complexes), TBHP (2.57 g, 20 mmol), CH₃CN (20 mL) and

temp. (80 °C).

Figure 11. Catalytic comparison of zeolite-encapsulated and neat

complexes for the oxidation of styrene in the presence of H₂O₂as

oxidant. Reaction conditions: styrene (1.04 g, 10 mmol), catalyst

(0.025 g for encapsulated and 0.00133 g in the case of neat com-

plexes), H₂O₂(2.27 g, 20 mmol), CH₃CN (10 mL) and temp.

(80 °C).

Table 5. Product selectivity and percent conversion of styrene with

TBHP after 6 h of reaction time.

Table 4. Product selectivity and percent conversion of styrene after

6 h of reaction time with H₂O₂as oxidant.

[a] TOF [h^–1] (turnover frequency): mol of substrate converted per

mol of metal (in the solid catalyst) per hour.

tion of other products such as phenylacetaldehyde is pos-

sible through isomerisation of styrene oxide, while the for-

mation of benzoic acid from benzaldehyde is straightfor-

ward.

Under the above optimised reaction conditions, the cata-

lytic action of neat as well encapsulated complexes has also

been tested using tert-butyl hydroperoxide (TBHP) as an

oxidant: 70% TBHP (2.57 g, 20 mmol), catalyst (0.025 g for

encapsulated and 0.00133 g, for neat complexes) and sty-

Oxidation of Cyclohexene

The oxidation of cyclohexene catalysed by encapsulated

rene (1.04 g, 10 mmol) were used in CH₃CN (20 mL), and complexes gave cyclohexene oxide, 2-cyclohexen-1-ol, cyclo-

the reaction was carried out at 80 °C. Figure 12 shows the hexane-1,2-diol and 2-cyclohexen-1-one, as shown in

conversion as a function of time, and Table 5 compares the Scheme 4. Recently, we observed all these products while

selectivity data along with the conversion and turnover fre- using catalyst [VO(sal-dach)] encapsulated in zeolite-Y.^[14]

quency after 6 h of reaction time. It is clear that the ob-

The reaction conditions have been optimised considering

tained percentage conversion varies in the range 13.6– 3 as a representative catalyst and varying the amount of

28.0% with the order: [Cu(pydx-1,3-pn)]-Y (28.0%) Ͼ [Cu- catalyst, oxidant, solvent and temperature of the reaction

(pydx-en)]-Y (23.6%) Ͼ [Cu(pydx-1,3-pn)] (16.2%) Ͼ [Cu- mixture. The effect of H₂O₂concentration on the oxidation

(pydx-en)] (13.6%). Thus, the overall conversion of styrene of cyclohexene is shown in Figure 13. Three different H₂O₂/

with TBHP is always lower than with H₂O₂. The selectivity cyclohexene molar ratios of 1:1, 2:1 and 3:1 were used at

of the formation of an important component – styrene ox- a fixed amount of cyclohexene (0.82 g, 10 mmol), catalyst

ide – is better (43.4–70.5%), but two more products, namely (0.025 g), acetonitrile (20 mL) and temperature (75 °C).

benzoic acid and phenylacetaldehyde, have also been ob- From Figure 13 it is clear that a 3:1 molar ratio is the best

tained.

The highest yield of benzaldehyde using H₂O₂as oxidant

of those tested, to obtain a conversion of 38.8% in 6 h.

The amount of catalyst has shown considerable affect on

is possibly due to further oxidation of styrene oxide formed the oxidation of cyclohexene. For a cyclohexene/H₂O₂ratio

in the first step by a nucleophilic attack of H₂O₂on styrene of 1:3, five different amounts of catalyst – viz. 0.015, 0.025,

oxide followed by cleavage of the intermediate hydroper- 0.035, 0.050 and 0.065 g – were used under the above reac-

oxystyrene.^[28]The formation of benzaldehyde may also be tion conditions, and the results obtained are summarised in

facilitated by the direct oxidative cleavage of the styrene Figure 14. The conversion increases with an increase in the

side-chain double bond by a radical mechanism.^[27]Forma- amount of catalyst from 0.015 to 0.050 g (28 to 56.2%), and

Eur. J. Inorg. Chem. 2007, 5720–5734

www.eurjic.org

5727

M. R. Maurya, B. Singh, P. Adão, F. Avecilla, J. Costa Pessoa

FULL PAPER

Scheme 4.

Figure 15. Effect of the temperature on the oxidation of cyclohex-

ene. Reaction conditions: cyclohexene (0.82 g, 10 mmol), [Cu(pydx-

en)]-Y (3) (0.050 g), 30% H₂O₂(3.40 g, 30 mmol) and CH₃CN

(10 mL).

Figure 13. Effect of H₂O₂on the oxidation of cyclohexene. Reac-

tion conditions: cyclohexene (0.82 g, 10 mmol), [Cu(pydx-en)]-Y

(0.025 g), CH₃CN (20 mL) and temp. (75 °C).

Thus, suitable reaction conditions for the maximum oxi-

dation of cyclohexene were optimised as follows: cyclohex-

ene (0.82 g, 10 mmol), H₂O₂(3.40 g, 30 mmol), catalyst

(0.050 g), acetonitrile (10 mL) and a temperature of 75 °C.

Catalyst 4 was also tested under the above reaction condi-

tions, and results along with the selectivity of various reac-

tion products are summarised in Table 6. A maximum of

83.0% conversion has been achieved with 4, which is less

than that obtained by 3 (90.1%) in 6 h of contact time.

However, both catalysts show the highest selectivity

towards the formation of 2-cyclohexen-1-ol, which is fol-

lowed by cyclohexene oxide. The selectivity of the two other

products, 2-cyclohexen-1-one and cyclohexane-1,2-diol, has

no definite trend.

then assumes a lower trend with 0.065 g of catalyst (52.4%).

Thus, an amount of 0.050 g of catalyst is considered ade-

quate to carry out the oxidation of cyclohexene. It was also

noted that by reducing the volume of acetonitrile to 10 mL,

this conversion reached 90.1% with 0.050 g of catalyst un-

der the above reaction conditions.

We have also tested the catalytic activities of encapsu-

lated complexes using tert-butyl hydroperoxide as an oxi-

dant. Thus, catalyst (0.050 g), tert-butyl hydroperoxide

(3.855 g, 30 mmol) and cyclohexene (0.82 g, 10 mmol) were

used in acetonitrile (10 mL), and the reaction was carried

out at 75 °C. The data, also presented in Table 6, show a

poor conversion of cyclohexene (19.5% for 3 and 17.8% for

4) with tert-butyl hydroperoxide in 6 h of contact time.

Here, the major product obtained is cyclohexane-1,2-diol,

while amounts of the other three products are very poor.

The neat complexes gave 45.5% (with 0.0028 g of 1) or

Figure 14. Effect of the amount catalyst on the oxidation of cyclo-

hexene. Reaction conditions: cyclohexene (0.82 g, 10 mmol), 30%

H₂O₂(3.40 g, 30 mmol), CH₃CN (20 mL) and temp.(75 °C).

Figure 15 illustrates the effect of temperature of the reac- 50.7% (with 0.0027 g of 2) conversion using H₂O₂as an

tion medium on the conversion of cyclohexene as a function oxidant (Table 6). Here, the selectivity of cyclohexene oxide

of time. Amongst the three different temperatures (55, 65 is better (33 to 38.7%) in comparison with the encapsulated

and 75 °C) for the fixed amount of cyclohexene (0.82 g, complexes, although the overall selectivity orders of all

10 mmol), aqueous 30% H₂O₂(3.40 g, 30 mmol) and cata- products are the same. However, the turnover rates for en-

lyst (0.050 g) in acetonitrile (10 mL), running the reaction capsulated complexes, their stability towards decomposition

at 75 °C has shown the best result, where 90.1% conversion and easy recovery from the reaction mixture make encapsu-

has been obtained.

lated complexes better catalysts than neat ones.

5728

www.eurjic.org

Eur. J. Inorg. Chem. 2007, 5720–5734

Zeolite-Encapsulated Copper Complexes of Tetradentate Ligands

Table 6. Oxidation of cyclohexene and product selectivity using various catalysts.

The conversion of cyclohexene and selectivity of different

reaction products using [Cu(pydx-1,3-pn)]-Y as catalyst un-

der the optimised reaction conditions have been analysed

as a function of time and are presented in Figure 16. It is

Scheme 5.

clear from the plot that a selectivity of 20.7% of cyclohex-

ene oxide and 35.5% of cyclohexen-1-ol has been obtained

The catalytic potential of these catalysts has also been

at a conversion of 20.2% of cyclohexene in 1 h of reaction

time. The selectivity for these products increases with time

and reaches 37.1% and 51.1%, respectively, with the in-

crease of the conversion of cyclohexene to 83.0% in 6 h.

The other two products, 2-cyclohexen-1-one and cyclohex-

ane-1,2-diol, have 26.8% and 14.7% selectivity in the first

hour, but their selectivities decrease with time and reach 3.5

and 8.2%, respectively. Thus, to obtain a good selectivity

for the important product cyclohexene oxide, it is essential

to run the reaction for at least 6 h under the optimised reac-

tion conditions. After 6 h, only minor changes have been

noted in the selectivities of the different reaction products.

optimised considering the amount of oxidant, solvent and

catalyst. Thus, catalyst 4 was taken as a representative, and

the amount of oxidant was varied at a fixed amount of

methyl phenyl sulfide (1.242 g, 10 mmol) and catalyst

(0.025 g) in acetonitrile (10 mL), and the reaction was car-

ried out at ambient temperature. At an aqueous H₂O₂/sub-

strate ratio of 2:1, a maximum of 69.3% conversion was

achieved in 3 h of reaction time. Decreasing this ratio to 1:1

decreases the conversion of methyl phenyl sulfide consider-

ably, while increasing this ratio to 3:1 has a marginal effect

on the conversion (Figure 17). The only significant influ-

ence when using a 3:1 ratio is the completion of the reaction

in less time. Increasing the volume of solvent to 20 mL un-

der the above reaction conditions decreases the conversion

considerably (to 64.3%).

Figure 16. Conversion of cyclohexene and variation in the selectiv-

ity of various products as a function of time: (a) conversion of

cyclohexene, (b) cyclohexene oxide, (c) 2-cyclohexen-1-ol, (d) 2-cy-

clohexen-1-one and (e) cyclohexane-1,2-diol.

Figure 17. Effect of the amount of H₂O₂on the oxidation of methyl

phenyl sulfide as a function of time. Reaction conditions: methyl

phenyl sulfide (1.242 g, 10 mmol), [Cu(pydx-1,3-pn)]-Y (0.025 g)

and acetonitrile (10 mL).

Oxidation of Methyl Phenyl Sulfide (Thioanisol)

Enzyme sulfoxidases^[29]facilitate sulfide peroxidase ac-

Similarly, four different amounts of catalyst – viz. 0.015,

tivity where the electron-rich sulfur atom of the sulfide un- 0.025, 0.035 and 0.050 g – were considered, while keeping

dergoes electrophilic oxidation by H₂O₂to give the sulfox- methyl phenyl sulfide (1.242 g, 10 mmol) and 30% H₂O₂

ide and, further, sulfone. Such oxidation of methyl phenyl (2.27 g, 20 mmol) in acetonitrile (10 mL). As mentioned in

sulfide was tested with the copper complexes prepared in Figure 18, increasing the catalyst amount from 0.015 to

this work. Oxidation of methyl phenyl sulfide in the pres- 0.025 g increases the conversion from 46.8 to 69.3%. In-

ence of H₂O₂gave two products, namely methyl phenyl sulf- creasing it to 0.035 g improves the conversion to 81.0%,

oxide and methyl phenyl sulfone, as shown in Scheme 5.

while 0.050 g shows no improvement. Therefore, it is clear

Eur. J. Inorg. Chem. 2007, 5720–5734

www.eurjic.org

5729

M. R. Maurya, B. Singh, P. Adão, F. Avecilla, J. Costa Pessoa

FULL PAPER

that 0.035 g of catalyst is adequate to obtain an optimum

methyl phenyl sulfide conversion of 81.0% in 3 h of contact

time.

Figure 19. Oxidation of methyl phenyl sulfide by various catalysts

using H₂O₂as an oxidant.

Table 7. Conversion percentage of methyl phenyl sulfide in 3 h

using H₂O₂as an oxidant and selectivity of sulfoxide and sulfone.

Figure 18. Effect of the amount of catalyst on the oxidation of

methyl phenyl sulfide. Reaction conditions: methyl phenyl sulfide

(1.242 g, 10 mmol), H₂O₂(2.27 g, 20 mmol) and acetonitrile

(10 mL).

Thus, under the optimised reaction conditions, that is,

catalyst (0.035 g), methyl phenyl sulfide (1.242 g, 10 mmol),

30% H₂O₂(2.27, 20 mmol) and acetonitrile (10 mL), cata-

lyst 3 was also tested, and the results of both catalysts are

compared in Figure 19, while Table 7 provides selectivity

details. Under the optimised reaction conditions, 3 and 4

gave 80.3% and 81.0% conversion, respectively. In terms

of the product selectivity, the selectivities of methyl phenyl

sulfoxide and methyl phenyl sulfone are 63.3 and 36.7%,

respectively, for 3 and 53.6 and 46.4%, respectively, for 4.

A blank reaction under similar conditions, that is, methyl

phenyl sulfide (1.242 g, 10 mmol), aqueous 30% H₂O₂

(2.27 g, 20 mmol) and acetonitrile (10 mL), resulted in

35.2% conversion with 68.3% selectivity towards sulfoxide

and 31.7% sulfone. Thus, catalysts 3 and 4 only enhanced

the percent conversion of methyl phenyl sulfide and played

no positive role in enhancing the selectivity of sulfoxide. In

contrast, a positive role in enhancing the selectivity towards

the formation of sulfoxide as well as conversion of methyl

phenyl sulfide has been observed by oxidovanadium(IV)

and copper(II) complexes encapsulated in the cavities of ze-

olite-Y.^[30]The neat complexes also exhibited a good cata-

lytic activity (68.4–69.9% conversion) when 0.0019 g

Figure 20. Oxidation of methyl phenyl sulfide by various catalysts

using tert-butyl hydroperoxide as an oxidant.

Test for Recyclability and Heterogeneity of the Reactions

The recyclability of both the encapsulated complexes has

amounts were used under the above optimised conditions. been tested in all three catalytic reactions. In a typical ex-

However, the selectivity for sulfone was better than for sulf- periment, for example, for styrene, after a contact time of

oxide.

6 h, the reaction mixture was filtered, and after activating

Catalytic activities of these complexes have also been the catalyst by washing with acetonitrile and drying at

tested using TBHP as an oxidant under similar conditions. 120 °C, it was subjected to a further catalytic reaction under

Again, 0.035 g of catalyst has shown the best results at a similar conditions. No appreciable loss in catalytic activity

fixed amount of methyl phenyl sulfide (10 mmol) and suggests that the complex is still present in the cavity of

TBHP (20 mmol) in acetonitrile (10 mL). A maximum of zeolite-Y. The filtrate collected after separating the used cat-

47.1% conversion has been obtained with 3, while 4 gave alyst was placed in the reaction flask and the reaction was

only 18.1% conversion in 3 h of reaction time (Figure 20). continued after adding fresh oxidant for another 4 h. The

Neat complexes 1 and 2 exhibited very poor conversion of gas chromatographic analysis showed no improvement in

4.7 and 4.2%, respectively. Independently of the catalytic conversion, and this confirms that the reaction did not pro-

systems and conversions, the sulfoxide has been obtained ceed upon removal of the solid catalyst. The reaction was,

selectively with TBHP.

therefore, heterogeneous in nature.

5730

www.eurjic.org

Eur. J. Inorg. Chem. 2007, 5720–5734

Zeolite-Encapsulated Copper Complexes of Tetradentate Ligands

Possible Reaction Pathway of the Catalysts

[Cu(pydx-1,3-pn)] are monomeric in the cavity of zeolite,

the facile formation of intermediates [(HOO)-Cu(pydx-en)]

and [(HOO)-Cu(pydx-1,3-pn)], respectively, are expected

because of the vacant available site in them (for example

complex 3 was characterised in the solid state with a coordi-

nated methanol molecule). This intermediate is expected to

transfer the oxygen atoms to the substrates to obtain the

products. Thus, the catalytic performance of the supported

catalyst could be attributed to the facile formation of this

Cu hydroperoxide intermediate species. The broadening of

the band at about 590 nm indicates the merging of the d–d

band with this charge-transfer band because of Cu hydro-

peroxide complex formation. In fact characteristic charge-

transfer bands because of Cu hydroperoxide coordination

are known to appear around 590 nm.^[33]

Attempts to detect Cu hydroperoxide intermediates by

ESI-MS (in the positive and negative modes) failed. The

addition of H₂O₂only increased the intensities of the peaks

corresponding to CuLH⁺and CuLNa⁺(and CuL·2H²⁺for

larger amounts of H₂O₂) in the positive mode, but no peaks

could be found in the negative mode that we could assign

to Cu hydroperoxide intermediate species.

In order to better understand the reaction mechanism,

the neat complexes were treated with H₂O₂and the progress

of the reaction was monitored by electronic absorption

spectroscopy. Thus, titration of a solution of 1 (about

10^–4) in methanol with one-drop portions of 30% H₂O₂

dissolved in methanol resulted in the spectral changes

shown in Figure 21. The intensity of the band at 378 nm

changes marginally, but the band at 277 nm increases con-

siderably, while the band at 235 nm (not shown) continu-

ously gains intensity. With a more concentrated solution

(about 10^–3) now dissolving complex 1 in DMSO, upon

equivalent H₂O₂additions the band appearing at 590 nm

broadens with an increase in the intensity (Figure 21, inset).

A very similar spectral pattern has also been observed for

2 (Figure 22). The data suggest the interaction of copper(II)

complexes with H₂O₂.

Conclusions

Zeolite-Y-encapsulated copper(II) complexes [Cu(pydx-

en)]-Y and [Cu(pydx-1,3-pn)]-Y with N,NЈ-ethylenebis(pyri-

doxyliminato) (H₂pydx-en) and N,NЈ-propylenebis(pyridox-

yliminato) (H₂pydx-1,3-pn) ligands have been prepared.

Neat complexes, [Cu(pydx-en)] and [Cu(pydx-1,3-

pn)(MeOH)], have also been prepared with these ligands

and characterised, and their crystal and molecular struc-

tures were determined, confirming the ONNO binding

mode of the ligands. The encapsulated complexes catalyse

the oxidation of styrene, cyclohexene and thioanisole ef-

ficiently, either using H₂O₂or tert-butyl hydroperoxide as

oxidants. Oxidation of styrene catalysed by [Cu(pydx-en)]-

Y and [Cu(pydx-1,3-pn)]-Y gave 23.6% and 28.0% conver-

sion, respectively, using TBHP as an oxidant. Using H₂O₂

as oxidant the major products were styrene oxide and benz-

aldehyde, while TBHP gave styrene oxide, benzaldehyde,

benzoic acid and phenylacetaldehyde. Under the optimised

conditions the maximum conversions obtained were 90.1%

of cyclohexene, using [Cu(pydx-en)]-Y, and 83.0% using

[Cu(pydx-1,3-pn)]-Y as catalyst; the major products iden-

tified were cyclohexene oxide, 2-cyclohexen-1-ol, cyclohex-

ane-1,2-diol and 2-cyclohexen-1-one. Similarly, maximums

of 80.3% conversion of methyl phenyl sulfide with [Cu-

(pydx-en)]-Y and of 81.0% with [Cu(pydx-1,3-pn)]-Y were

Figure 21. Titration of a methanol solution (about 10^–4) of

[Cu(pydx-en)] (1) with aqueous 30% H₂O₂dissolved in methanol.

The inset displays equivalent titrations, but with higher concentra-

tions of 1 (about 10^–3) dissolved in DMSO.

Figure 22. Titration of a methanol solution (about 10^–4) of

[Cu(pydx-1,3-pn)MeOH] (2) with aqueous 30% H₂O₂dissolved in

methanol. The inset displays equivalent titrations, but with higher observed, where the selectivity for the major product methyl

concentrations of 2 (about 10^–3) dissolved in DMSO.

phenyl sulfoxide was found to be about 60%. Tests for re-

cyclability and heterogeneity of the reactions indicated their

At least three types of intermediates having a copper– good recyclability. Of the alternative suggested mechanisms

oxygen interaction – side-on Cu^III–(µ-η²-peroxido)–Cu^III

,

for catalysis, namely regarding the intermediates specifying

bis(µ-oxido–Cu^III) and Cu^III–O–O–H (copper hydroperox- the copper–oxygen interaction, our data suggest the in-

ide) – have been discussed in the literature during the cata- volvement of Cu^III–O–O–H (copper hydroperoxide) inter-

lytic action.^[31,32]As copper complexes [Cu(pydx-en)] and mediates.

Eur. J. Inorg. Chem. 2007, 5720–5734

www.eurjic.org

5731

M. R. Maurya, B. Singh, P. Adão, F. Avecilla, J. Costa Pessoa

FULL PAPER

X-ray Crystal Structure Determination: Three-dimensional X-ray

data were collected with a Bruker Kappa Apex CCD diffractometer

by the φ-ω scan method. Data were collected at room temperature.

Reflections were measured from a hemisphere of data collected

from frames each covering 0.3° in ω. Of the 13687 reflections mea-

sured for 1 and 20715 for 2, all of which were corrected for Lorentz

and polarization effects, and for absorption by semiempirical meth-

ods based on symmetry-equivalent and repeated reflections, 2139

independent reflections for 1 and 8116 for 2 exceeded the signifi-

cance level |F|/σ(|F|) Ͼ 4.0. Complex scattering factors were taken

from the program package SHELXTL.^[35]The structures were

solved by direct methods and refined by full-matrix least-squares

methods on F². The hydrogen atoms were included in calculated

positions and refined by using a riding mode for 1 and were left to

refine freely for 2. Refinement converged with allowance for ther-

mal anisotropy of all non-hydrogen atoms. Minimum and maxi-

mum final electron density: 2.494 and –2.167°eÅ^–3for 1, 0.997 and

–0.673°eÅ^–3for 2. Crystal data and details on data collection and

refinement are summarised in Table 8. CCDC-655305 (for 1) and

-655304 (for 2) contain the supplementary crystallographic data for

this paper. These data can be obtained free of charge from the

Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/

data_request/cif.

Experimental Section

Materials: Analytical reagent-grade pyridoxal hydrochloride

(Hpydx·HCl, Himedia, Mumbai, India), methyl phenyl sulfide

(Alfa Aeaser, USA), styrene, cyclohexene (Acros Organics, NJ,

USA), cupric nitrate (E. Merck, Mumbai, India) and 30% aqueous

H₂O₂(Qualigens, Mumbai, India) were used as obtained. Y-zeolite

(Si/Al = 10) was obtained from Indian Oil Corporation (R&D),

Faridabad, India. All other chemicals and solvents used were of

AR grade. Ligands H₂pydx-en (I) and H₂pydx-1,2-pn (II) were pre-

pared as described in the literature.^[15]

Physical Methods and Analysis: The copper content was measured

using an inductively coupled plasma spectrometer (ICP; Labtam

8440 Plasmalab) after leaching the metal ions with conc. nitric acid

and diluting with double-distilled water to specific volumes. IR

spectra were recorded as KBr pellets with a Nicolet NEXUS Alig-

ent 1100 series FTIR spectrophotometer, after grinding the sample

with KBr. UV/Vis spectra of zeolite-encapsulated metal complexes

were recorded in Nujol using a Shimadzu 1601 UV/Vis spectropho-

tometer by layering a mull sample to the inside of one of the cu-

vettes and keeping the other one layered with Nujol as reference.

UV/Vis spectra of other ligands and neat complexes were recorded

in methanol. X-ray powder diffractograms of solid catalysts were

recorded using a Bruker AXS D8 advance X-ray powder dif-

fractometer with a Cu-K_αtarget. Thermogravimetric analyses of

pure as well as encapsulated complexes were carried out using a

TG Stanton Redcroft STA 780 instrument. Scanning electron

micrographs (SEM) of zeolite-containing samples were recorded

with a Leo instrument, model 435VP. The samples were dusted on

alumina and coated with a thin film of gold to prevent surface

changing and to protect the surface material from thermal damage

by the electron beam. In all analyses, a uniform thickness of about

0.1 mm was maintained. The energy-dispersive X-ray analysis

(EDAX) plots of encapsulated complexes were recorded with an

FEI Quanta 200 FEG instrument. The ESI-MS data were obtained

with a Varian 500-MS LC ion-trap mass spectrometer operating

either in the positive or in the negative mode. The EPR spectra

were recorded with a Bruker ESP 300E X-band spectrometer. The

spectra were measured at room temperature and at 77 K (with

glasses, made by freezing solutions in liquid nitrogen, or just using

the frozen samples). For the neat complexes we recorded spectra in

DMF and in DMSO. In DMSO apparently there was precipitation

of solids upon freezing, as only broad signals were obtained. The

spin Hamiltonian parameters were obtained by simulation of the

spectra with the computer program of Rockenbauer and Korecz.^[34]

All catalysed reaction products were analysed using a Thermoelec-

tron gas chromatograph fitted with an HP-1 capillary column

(30 mϫ0.25 mmϫ0.25 µm) and FID detector. Styrene, cyclohex-

ene and methyl phenyl sulfide were used as external standards to

quantify their conversion upon catalytic action. To quantify the

conversion of styrene, for example, the calibration of the GC was

carried out using styrene as external standard by injecting three

different but known concentrations of styrene into the GC under

the reaction conditions at which the analyses of catalytic products

were performed and noting the peak area in each case. Finally, the

peak area obtained for the remaining amount of styrene during

catalytic oxidation was correlated with the peak area of the stan-

dard sample. This ultimately gave the percent conversion of sub-

strate. The selectivity of products was then obtained directly from

the peak areas of the respective products. Other external standards

were used similarly. The identities of the products were confirmed

by GC-MS with a Perkin–Elmer Clarus 500 model.

Table 8. Crystal data collection and refinement.

1

2

Empirical formula

C₁₈H₂₀CuN₄O₄

419.92

C₂₀H₂₆CuN₄O₅

465.99

M_r[gmol^–1]

Crystal system

Space group

triclinic

P1

triclinic

P1

¯

T [K]

a [Å]

b [Å]

c [Å]

α [°]

β [°]

293(2)

9.6664(17)

10.0904(19)

10.150(3)

101.418(14)

93.432(14)

116.898(9)

852.7(4)

434

8.0870(2)

9.0492(2)

14.2599(3)

105.8640(10)

90.9030(10)

95.5120(10)

998.20(4)

486

γ [°]

V [Å³]

F(000)

Z

2

D_calcd.[gcm^–3]

1.636

1.315

1.550

1.135

µ [mm^–1]

R_int

0.0918

13687

2139

0.0334

20715

8116

Measured reflections

Observed reflections

Data/restraints/parameters 2975/3/248

10700/0/375

1.029

0.0415

Goodness-of-fit on F²

1.093

0.0918

0.2713

[a]

R₁

[b]

wR₂(all data)

0.1029

[a] R₁= Σ||F_o| – |F_c||/Σ|F_o|. [b] wR₂= {Σ[w(||F_o|²– |F_c|²|)²]|/

Σ[w(F_o⁴)]}^1/2

.

Preparations

Preparation of [Cu(pydx-en)] (1): A stirred solution of H₂pydx-en

(0.717 g, 2 mmol) dissolved in methanol (20 mL) was added to cu-

pric acetate (0.40 g, 2 mmol) dissolved in methanol (30 mL) with

stirring, and the reaction mixture was refluxed in a water bath for

5 h. On reducing the solvent volume to about 20 mL and keeping

the mixture at ambient temperature, a brown solid of [Cu(pydx-

en)] slowly separated within a few hours. This was filtered off,

washed with methanol and dried in vacuo over silica gel. Yield:

78%. Green crystals suitable for X-ray diffraction study were ob-

5732

www.eurjic.org

Eur. J. Inorg. Chem. 2007, 5720–5734

Zeolite-Encapsulated Copper Complexes of Tetradentate Ligands

tained from the filtrate on slow concentration at room temperature.

C₁₈H₂₀CuN₄O₄[Cu(pydx-en)] (419.93): calcd. C 51.48, H 4.80, Cu

15.13, N 13.34; found C 52.0, H 4.6, Cu 15.1, N 13.1.

the UV spectra measured. A solution of H₂O₂was prepared by

dissolving five drops of 30% H₂O₂in methanol (5 mL). One drop

of this H₂O₂solution was added and the UV spectra recorded

again. Several similar additions were made, and after each addition

the spectral changes were recorded similarly. As the solubility of

complexes 1 or 2 is low in methanol, only the UV region could be

covered in this way. For the visible region, the complexes were dis-

solved in DMSO (concentration about 10^–3, 10 mL), and ad-

ditions of the H₂O₂solution were made similarly, now recording

the visible region of the spectra (see insets of Figures 21 and 22).

Preparation of [Cu(pydx-1,3-pn)(MeOH)] (2): [Cu(pydx-1,3-

pn)(MeOH)] was prepared according to essentially the same pro-

cedure as for 1. A green solid was obtained. Green crystals suitable

for X-ray diffraction study were obtained from the filtrate on keep-

ing it at room temperature for slow concentration. Yield: 70%.

C₂₀H₂₆CuN₄O₅[Cu(pydx-1,3-pn)(MeOH)] (466.0): calcd. C 51.55,

H 5.62, Cu 13.64, N 12.02; found C 51.8, H 5.8, Cu 13.8, N 12.2.

Preparation of Cu-Y: Na-Y zeolite (5.0 g) was suspended in distilled

water (300 mL) and after addition of cupric nitrate (2.26 g,

12 mmol), the reaction mixture was heated at 90 °C with stirring

for 24 h. The pale bluish solid was filtered, washed with hot dis-

tilled water until the filtrate was free from any copper(II) ion con-

tent and dried at 150 °C in an air oven for 24 h. Found: Cu 7.60.

Acknowledgments

The authors are thankful to the Council of Scientific and Industrial

Research, New Delhi for financial assistance. The authors also wish

to thank FEDER, Fundação para a Ciência e a Tecnologia and

POCI 2010 (ref. POCI/QUI/55985/2004) for financial support, and

Kamila Koci for the ESI-MS experiments.

Preparation of [Cu(pydx-en)]-Y (3): Cu-Y (3.0 g) and H₂pydx-en

(3.0 g) were mixed in methanol (50 mL), and the reaction mixture

was heated under reflux in an oil bath with stirring for 14 h. The

resulting material was filtered and then Soxhlet-extracted with

methanol to remove unreacted ligand. It was finally treated with

hot DMF while stirring for 1 h, filtered and washed with DMF

followed by hot methanol. The uncomplexed metal ions present in

the zeolite were removed by stirring with double-distilled aqueous

NaCl (0.0.1 , 150 mL) for 10 h. The resulting pale yellow solid

was filtered, and washed with hot distilled water until no precipitate

of AgCl was observed in the filtrate on treating with AgNO₃. Fi-

nally it was dried at 120 °C for several hours. Found: Cu 0.79.

[1] J. Poltowicz, K. Pamin, E. Tabor, J. Haber, A. Adamski, Z.

Sojka, Appl. Catal. A 2006, 299, 235–242.

[2] S. P. Verkey, C. Ratnasamy, P. Ratnasamy, J. Mol. Catal. A

1998, 135, 295–306.

[3] S. P. Verkey, C. Ratnasamy, P. Ratnasamy, Microporous Meso-

porous Mater. 1998, 22, 465–474.

[4] S. P. Verkey, C. Ratnasamy, P. Ratnasamy, Appl. Catal. A 1999,

182, 91–96.

[5] S. Deshpande, D. Srinivas, P. Ratnasamy, J. Catal. 1999, 188,

261–269.

[6] D. Chatterjee, A. Mitra, J. Mol. Catal. A 1999, 144, 363–367.

[7] C. Bowers, P. K. Dutta, J. Catal. 1990, 122, 271–279.

[8] M. S. Niassary, F. Farzaneh, M. Ghandi, L. Turkian, J. Mol.

Catal. A 2000, 157, 183–188.

[9] M. Silva, C. Freire, B. de Castro, J. L. Figueiredo, J. Mol. Ca-

tal. A 2006, 258, 327–333.

[10] T. Joseph, S. B. Halligudi, C. Satyanarayan, D. P. Sawant, S.

Gopinathan, J. Mol. Catal. A 2001, 168, 87–97.

[11] M. R. Maurya, S. J. J. Titinchi, S. Chand, Appl. Catal. A 2002,

228, 177–187.

[12] M. R. Maurya, S. J. J. Titinchi, S. Chand, J. Mol. Catal. A

2003, 201, 119–130.

[13] M. R. Maurya, M. Kumar, S. J. J. Titinchi, H. S. Abbo, S.

Chand, Catal. Lett. 2003, 86, 97–105.

[14] M. R. Maurya, A. K. Chandrakar, S. Chand, J. Mol. Catal. A:

Chem. 2007, 270, 225–235.

[15] I. Correia, J. Costa Pessoa, M. T. Duarte, R. T. Henriques,

M. F. M. Piedade, L. F. Veiros, T. Jackusch, A. Dornyei, T.

Kiss, M. M. C. A. Castro, C. F. G. C. Geraldes, F. Avecilla,

Chem. Eur. J. 2004, 10, 2301–2317.

Preparation of [Cu(pydx-1,3-pn)]-Y (4): The pale yellow [Cu(pydx-

1,3-pn)]-Y was prepared according to essentially the same pro-

cedure as outlined for 3. Found: Cu 0.74.

Catalytic Activity

Oxidation of Styrene: The catalytic oxidation of styrene was carried

out in 50-mL flasks fitted with a water-circulated condenser. In a

typical reaction, an aqueous solution of 30% H₂O₂or 70% tert-

butyl hydroperoxide (TBHP) (20 mmol) and styrene (1.04 g,

10 mmol) was mixed in acetonitrile (20 mL). After the temperature

of the oil bath reached 80 °C, the zeolite-Y-encapsulated catalyst

(0.025 g) was added to the reaction mixture which was continuously

stirred. During the reaction, the products were periodically ana-

lysed using a gas chromatograph by withdrawing small aliquots,

and their identities were confirmed by GC-MS. Various param-

eters, such as amounts of oxidant and catalyst as well as the tem-

perature of the reaction mixture were studied in order to under-

stand their effect on the conversion and patterns of the reaction

products.

[16] I. Correia, A. Dornyei, T. Jakusch, F. Avecilla, T. Kiss, J. Co-

sta Pessoa, Eur. J. Inorg. Chem. 2006, 2819–2830.

[17] I. Correia, A. Dornyei, T. Jakusch, F. Avecilla, T. Kiss, J. Co-

sta Pessoa, Eur. J. Inorg. Chem. 2006, 656–662.

[18] M. Valko, Inorg. Chem. 1993, 32, 4131–4138.

[19] H. Elias, Angew. Chem. Int. Ed. Engl. 1992, 31, 623–625.

[20] D. F. Evans, J. Chem. Soc. 1959, 2003–2005.

[21] E. Sinn, A. B. Blake, W. Haussain, R. Paschke, J. R. Chip-

perfield, Inorg. Chem. 1995, 34, 1125–1129.

Oxidation of Cyclohexene: The catalytic oxidation of cyclohexene

was also carried out using [Cu(pydx-en)]-Y and [Cu(pydx-1,3-pn)]-

Y. Aqueous 30% H₂O₂(2.28 g, 20 mmol), cyclohexene (0.82 g,

10 mmol) and catalyst (0.025 g) were mixed in acetonitrile (20 mL),

and the reaction mixture was heated at 75 °C with continuous stir-

ring in an oil bath. The progress of the reaction was monitored as

mentioned above. The identification of the various products was

confirmed by GC-MS.

[22] J. Peisach, W. E. Blumberg, Arch. Biochem. Biophys. 1974, 165,

Oxidation of Methyl Phenyl Sulfide: Aqueous 30% H₂O₂(2.28 g,

20 mmol) and catalyst (0.025 g) were added to a solution of methyl

phenyl sulfide (1.24 g, 10 mmol) dissolved in acetonitrile (20 mL),

and the reaction mixture was stirred at ambient temperature for

3 h. The formation of the reaction products was monitored by GC.

691–708.

[23] U. Sakaguchi, A. W. Addison, J. Chem. Soc. Dalton Trans.

1979, 600–608.

[24] V. T. Kasumov, F. Koksal, Spectrochim. Acta Part A 2005, 61,

225–231.

[25] K. M. Ananth, M. Kanthimathi, B. U. Nair, Transition Met.

Chem. 2001, 26, 333–338.

[26] V. Hulea, E. Dumitriu, Appl. Catal. A 2004, 277, 99–106.

Measurement of UV/Vis Spectra by Adding H₂O₂: A solution of

complex 1 or 2 in methanol (ca. 10^–4, 10 mL) was prepared and

Eur. J. Inorg. Chem. 2007, 5720–5734

www.eurjic.org

5733

M. R. Maurya, B. Singh, P. Adão, F. Avecilla, J. Costa Pessoa

FULL PAPER

[27] M. R. Maurya, U. Kumar, P. Manikandan, Dalton Trans. 2006,

3561–3575.

[28] M. R. Maurya, U. Kumar, P. Manikandan, Eur. J. Inorg. Chem.

2007, 2303–2314.

[29] A. Butler, M. J. Clague, G. E. Meister, Chem. Rev. 1994, 94,

625–638.

[32] J. P. Klinman, Chem. Rev. 1996, 96, 2541–2561.

[33] K. D. Karlin, J. C. Hayes, Y. Gultneh, R. W. Cruse, J. W.

McKown, J. P. Hutchinson, J. Zubieta, J. Am. Chem. Soc. 1984,

106, 2121–2128.

[34] A. Rockenbauer, L. Korecz, Appl. Magn. Reson. 1996, 10, 29–

43.

[30] M. R. Maurya, A. K. Chandrakar, S. Chand, J. Mol. Catal. A:

Chem. 2007, 263, 227–237.

[31] E. I. Solomon, P. Chen, M. Metz, S.-K. Lee, A. E. Palner, An-

gew. Chem. Int. Ed. 2001, 40, 4570–4590.

[35] G. M. Sheldrick, SHELXTL, release 5.1, Bruker Analytical X-

ray Systems, Madison, WI, 1997.

Received: July 27, 2007

Published Online: November 8, 2007

5734

www.eurjic.org

Eur. J. Inorg. Chem. 2007, 5720–5734

Article Doi

Zeolite-encapsulated copper(II) complexes of pyridoxal-based tetradentate ligands for the oxidation of styrene, cyclohexene and methyl phenyl sulfide

DOI: 10.1002/ejic.200700797

Source and publish data:

Authors:

Article abstract of DOI:10.1002/ejic.200700797

Full text of DOI:10.1002/ejic.200700797

Products guided by the article

Relevant to this article

Hot Product