Synthesis, characterization, reactivity and catalytic activity ofoxidovanadium(IV), oxidovanadium(V) and dioxidovanadium(V) complexes ofbenzimidazole modified ligands†
Mannar R. Maurya,*a Manisha Bisht,a Amit Kumar,b Maxim L. Kuznetsov,b Fernando Avecillac andJoao Costa Pessoa*b
Received 16th February 2011, Accepted 3rd May 2011DOI: 10.1039/c1dt10261c
The reaction between [VIVO(acac)2] and the ONN donor Schiff base obtained by the condensation ofpyridoxal and 2-aminoethylbenzimidazole (Hpydx-aebmz, I) or 2-aminomethylbenzimidazole(Hpydx-ambmz, II) in equimolar amounts results in the formation of [VIVO(acac)(pydx-aebmz)] 1 and[VIVO(acac)(pydx-ambmz)] 2, respectively. The aerobic oxidation of the methanolic solution of 1yielded [VVO2(pydx-aebmz)] 3 and its reaction with aqueous H2O2 gave the oxidoperoxidovanadium(V)complex, [VVO(O2)(pydx-aebmz)] 4. The formation of 4 in solution is also established by titrations ofmethanolic solutions of 1 with H2O2. By titrating solutions of 3 and of 4 with aqueous H2O2 severaldistinct VV-pydx-aebmz species also containing the peroxido ligand are detected. The full geometryoptimization of all species envisaged was done using DFT methods for suitable model complexes. The51V NMR chemical shifts (dV) have also been calculated, the theoretical data being used to supportassignments of the experimental chemical shifts. The 51V hyperfine coupling constants are calculated for1, the obtained values being in good agreement with the experimental EPR data. Reaction between theVIVO2+ exchanged zeolite-Y and Hpydx-aebmz and Hpydx-ambmz in refluxing methanol, followed byaerial oxidation results in the formation of the encapsulated VVO2-complexes, abbreviated herein as[VVO2(pydx-aebmz)]-Y 5 and [VVO2(pydx-ambmz)]-Y 6. The molecular structure of 1, determined bysingle crystal X-ray diffraction, confirms its distorted octahedral geometry with the ONN bindingmode of the tridentate ligand, with one acetylacetonato group remaining bound to the VIVO-centre.Oxidation of styrene is investigated using some of these complexes as catalyst precursors with H2O2 asoxidant. Under optimised reaction conditions for the conversion of styrene in acetonitrile, a maximumof 68% conversion of styrene (with [VVO2(pydx-aebmz)]-Y) and 65% (with [VVO2(pydx-ambmz)]-Y) isachieved in 6 h of reaction time. The selectivity of the various products is similar for both catalysts andfollows the order: benzaldehyde (ca. 55%) > 1-phenylethane-1,2-diol > benzoic acid > styrene oxide >
phenyl acetaldehyde. Speciation of the systems and plausible intermediates involved in the catalyticoxidation processes are established by UV-Vis, EPR, 51V NMR and DFT studies. Both non-radical(Sharpless) and radical mechanisms of the olefin oxidations were theoretically studied, and the radicalpathway was found to be even more favorable than the Sharpless mechanism.
aDepartment of Chemistry, Indian Institute of Technology Roorkee, Roorkee247 667, India. E-mail: [email protected]; Fax: +91 1332 273560;Tel: +91 1332 285327bCentro Quımica Estrutural, Instituto Superior Tecnico, TU Lisbon, AvRovisco Pais, 1049-001 Lisboa, Portugal. E-mail: [email protected];Fax: +351 21 8464455; Tel: +351 218419268cDepartamento de Quımica Fundamental, Universidade da Coruna, Campusde A Zapateira, 15071 A Coruna, Spain† Electronic supplementary information (ESI) available: Additional in-formation on single crystal X-ray diffraction studies of complex 1. 51VNMR spectra of complexes 3 and 4. Calculations of entropies in solution.Total energies, enthalpies, Gibbs free energies (Hartree) and entropies(cal mol-1 K) in gas-phase and CH3CN solution and Cartesian atomiccoordinates (A) of the calculated equilibrium structures. CCDC reference
Introduction
Prospective therapeutic applications of vanadium compounds,particularly in the treatment of Diabetes mellitus1 and cancer,2
and more recently in vitro antiamoebic activity against Entamoebahistolytica3 and anti-trypanosomal activity against Trypanosomacruzi, causative agent of Chagas disease,4 have stimulated thestudy of mixed-ligand vanadium complexes with polydentateligands. The finding of covalent bonds in vanadate-dependenthaloperoxidase enzymes (VHPO) through the Ne of imidazole
number 812058. For ESI and crystallographic data in CIF or otherelectronic format see DOI: 10.1039/c1dt10261c
moieties of histidine residues to vanadate in the active site5
stimulated the design of structural models that contain imidazole,imidazole derivatives, or other N donor ligands coordinated tovanadium.6 These compounds have been extended to functionalsimilarities in that vanadium complexes also model the oxidativehalogenation and sulfoxidation of organic substrates.
Crans et al.7 and Cornman et al.6b used benzimidazole-derivedligands to prepare oxidovanadium(IV) {VIVO} and dioxidovana-dium(V) {VVO2} complexes, providing models for the coordinationof histidine in the enzymes, and some of us also reported structuraland functional models of VHPO using benzimidazole derivedligands,8 finding that these complexes also catalyze the oxidationof styrene by peroxide.
Indeed the addition of an A–B bond across carbon–carbonmultiple bonds is a widely used chemical reaction for thefunctionalization of alkenes, namely their oxidation. Normallya catalyst is needed for such processes and many types of ligandshave been developed which tune the behavior of the metal centres,and a variety of homogeneous catalytic systems have been shownto promote many of these addition reactions. However, in mostcases the solvent required is not environmentally appropriate,or the catalyst is not easily recyclable, or it is de-activated afteruse.
We have been using heterogenized complexes as catalysts forthe functionalization of alkenes. In most cases these catalysts (orcatalyst precursors) have been found to be easily recyclable andmaintain their activity after several cycles of catalytic use.9 Namely,the encapsulation of [VVO2(sal-aebmz)] (Hsal-aebmz = Schiff basederived from salicylaldehyde and 2-aminoethylbenzimidazole)in zeolite-Y, where the zeolite acts as a protein mantle, notonly improved its catalytic activity by yielding higher turn overfrequency, but the catalyst precursor also became recyclable forfurther use.10
In this work we describe the synthesis and characterizationof VIVO- and VVO2-complexes of ligands I and II derived fromthe biogenic carbonyl constituent pyridoxal (one of the formsof vitamin B6; Hpydx) and 2-aminoethylbenzimidazole (aebmz)or 2-aminomethylbenzimidazole (ambmz), Scheme 1. Two of theVVO2-complexes have also been encapsulated in the super cagesof zeolite-Y and their catalytic activities are demonstrated bystudying the liquid phase oxidation of styrene. The use of pyridoxalinstead of salicylaldehyde derivatives features the advantage of thecatalysts being more compatible with the presence of water (andother polar solvents) in the reaction mixtures.11
Scheme 1 Structural formulae of ligands used in this work.
Several aspects related to the mechanisms of the reac-tions are addressed by both examining the speciation ofthese systems by spectroscopic techniques and by theoreticalcalculations.
Experimental section
Materials
VIVOSO4·5H2O, V2O5, o-phenylenediamine (Loba Chemie, Mum-bai, India), b-alanine (Spectrochem, Mumbai, India), glycine,pyridoxal hydrochloride (pydx·HCl) (Himedia, Mumbai, India),acetylacetone (Aldrich, Milwaukee, WI, USA), styrene (AcrosOrganics. New Jersey, USA), and 30% aqueous H2O2 were usedas obtained. Zeolite-Y (Si/Al = ca. 10) was obtained fromIndian Oil Corporation (R&D), Faridabad, India. All otherchemicals and solvents used were of AR grade. [VIVO(acac)2],12 2-aminomethylbenzimidazole dihydrochloride (ambmz·2HCl) and2-aminoethylbenzimidazole dihydrochloride (aebmz·2HCl)13 wereprepared according to the methods reported.
using a Shimadzu 1601 UV-Vis spectrophotometer by layeringthe mull of the sample to the inside of one of the cuvettes whilekeeping the other one layered with Nujol as reference. Spectraof neat complexes were recorded in methanol or DMSO. 1HNMR and 51V NMR spectra were obtained on a Bruker AvanceIII 400 MHz spectrometer with the common parameter settings.NMR spectra were usually recorded in MeOD-d4, DMF-d7 orDMSO-d6, and d (51V) values are referenced relative to neatVVOCl3 as external standard. Magnetic susceptibilities of VIVO-complexes were determined at 298 K using nickel as a standardwith a Vibrating Sample Magnetometer model 155 supplied byPrinceton Applied Research, and diamagnetic corrections weredone with Pascal’s constants.14 EPR spectra were recorded with aBruker ESP 300E X-band spectrometer, and the spin Hamiltonianparameters were obtained by simulation of the spectra with thecomputer program of Rockenbauer and Korecz.15 A ThermoNicolet gas chromatograph with a HP–1 capillary column (30 m ¥0.25 mm ¥ 0.25 mm) was used to analyze the reaction products.The identity of the products was confirmed using a GC-MS modelPerkin-Elmer Clarus 500, by comparing the fragments of eachproduct with the library available.
Preparations
Preparation of Hpydx-aebmz I. Pyridoxal hydrochloride(0.406 g, 0.002 mol) and aebmz·2HCl (0.464 g, 0.002 mol) weredissolved separately in 15 mL of water and neutralized to ca. pH
6.5 by adding an aqueous solution of KOH. The two solutionswere mixed and stirred for 1 h. During this period a yellow solidslowly separated out. This was filtered off, washed thoroughlywith water followed by petroleum ether, and dried in a vacuumdesiccator over silica gel. Yield: 0.56 g (90%). (Found: C, 65.4; H,5.6; N, 18.2%. Calc’d for C17H18N4O2 (310.36) C, 65.73; H, 5.80;N, 18.04%).
Preparation of Hpydx-ambmz II. This was prepared simi-larly following the procedure described above using neutralizedambmz·2HCl (0.436 g, 0.002 mol) and pyridoxal hydrochloride(0.406 g, 0.002 mol). Yield: 0.37 g (69%). (Found: C, 64.4; H, 5.3;N, 18.5%. Calc’d for C16H16N4O2 (296.33) C, 64.79; H, 5.40; N,18.90%).
Preparation of [VIVO(acac)(pydx-aebmz)] 1. A stirred solutionof Hpydx-aebmz (1.55 g, 0.005 mol) in dry methanol (10 mL)was treated with [VIVO(acac)2] (1.33 g, 0.005 mol) dissolved indry methanol (10 mL), and the resulting reaction mixture wasrefluxed for 3 h. Brown crystals of 1 separated out after cooling thereaction mixture to room temperature; these crystals were filteredoff, washed with methanol (2 ¥ 5 mL) and dried in a desiccator oversilica gel. Yield 2.12 g (89%). (Found: C, 55.4; H, 5.2; N, 11.6%.Calc’d for C22H24N4O5V (475.4) C, 55.58; H, 5.09; N, 11.79%.) meff
(293 K) = 1.72 mB.Part of this solid was dissolved in a minimum amount of
methanol and after layering with diethyl ether, the solution waskept in refrigerator at ca. 10 ◦C where dark red crystals of 1appeared in two days. A suitable crystal was picked up directlyfrom the mother liquid and used for the single-crystal X-raydiffraction study.
Preparation of [VIVO(acac)(pydx-ambmz)] 2. The brown[VIVO(acac)(pydx-ambmz)] was prepared from [VIVO(acac)2] andHpydx-ambmz following the procedure outlined for 1. Yield 0.55 g(58%). (Found: C, 54.2; H, 5.1; N, 11.9%. Calc’d for C21H22O5N4V(461.4) C, 54.67; H, 4.81; N, 12.14%.) meff (293 K) = 1.74 mB.
Preparation of [VVO2(pydx-aebmz)] 3. Complex[VIVO(acac)(pydx-aebmz)] 1 (0.950 g, 0.002 mol) was dissolvedin 40 mL methanol while heating. After adding KOH (0.112 g,0.002 mol) and stirring, the solution was filtered and air wasslowly passed through the solution for ca. 3 days with occasionalshaking. During this period, the white solid of [VVO2(pydx-aebmz)] separated out. This was filtered off, washed with hotmethanol and dried in vacuo. Yield 0.390 g (49%). (Found: C,52.5; H, 4.2; N, 14.5%. Calc’d for C17H17O4N4V (392.29) C, 52.05;H, 4.37; N, 14.28%.) Selected IR data (KBr, nmax/cm-1): 1616,1595 (C Nazomethine/Nring), 964, 924 (V O). 1H NMR (DMF-d7,d (ppm)): 9.43 (s, 1H, –NHimadzole), 8.96 (s, 1H, –CH N–), 7.80(s,1H), 7.55 (m, 1H), 7.34 (m, 1H), 7.15(br, 1H) (aromatic); 5.05 (s,1H, –OH), 4.7 (s, 2H, –CH2OH), 4.42 (t, 2H, –CH2CNN), 2.9(t,2H, –CH2), 2.38(t, 3H, –CH3). 51V NMR (DMSO-d6, dV/ppm):-543.
Preparation of [VVO(O2)(pydx-aebmz)] 4. Complex[VIVO(acac)(pydx-aebmz)] (0.951 g, 0.002 mol) was dissolved in40 mL of hot methanol and cooled to ca. 10 ◦C after filtration.A solution of 30% H2O2 (2 mL) in 5 mL of methanol was addeddrop wise under stirring to the above solution within 15 min.Stirring was continued and a yellow solid slowly separated within
45 min. This was filtered off, washed with methanol and driedin vacuum. Yield 0.650 g (75%). (Found: C, 49.9; H, 4.2; N,13.5%. Calc’d for C17H17O5N4V (408.3) C, 50.0; H, 4.20.; N,13.72%.) 51V NMR (DMSO-d7, dV/ppm): -543 (major), -563(minor).
VIVO-Y (oxidovanadium(IV) exchanged zeolite-Y). A filteredsolution of VIVOSO4·5H2O (9.0 g, 0.036 mol) dissolved in 100 mLof distilled water was added to a suspension of Na-Y zeolite(15.0 g) in 900 mL of distilled water and the reaction mixturewas heated at 90 ◦C with stirring for 24 h. The light green solidwas filtered, washed with hot distilled water until filtrate was freefrom any vanadyl ion content and dried at 150 ◦C for 12 h. Yield:14.88 g (99%).% V (ICMPS): 4.6.
[VVO2(pydx-aebmz)]-Y 5. VIVO-Y (3.0 g) and Hpydx-aebmz(2.0 g) were mixed in methanol (50 mL) and the reactionmixture was heated under reflux for 14 h in an oil bath withstirring. The resulting material was filtered and then Soxhletextracted with methanol to remove unreacted ligand. It was finallytreated with hot DMF while stirring for 1 h, filtered, washedwith DMF followed by hot methanol. The solid was furthersuspended in methanol and oxidized by passing air while stirringat room temperature for 24 h. After filtering the non-complexedvanadium(IV/V) ions present in the zeolite were further removedby stirring with aqueous 0.1 M NaCl (150 mL) solution for 10 h.The resulting light yellow solid was filtered, washed with hotdistilled water till no precipitate of AgCl was observed in the filtrateon treating with AgNO3. Finally, it was dried at 120 ◦C for severalhours. Selected IR data (KBr, nmax/cm-1): 1628 (C Nring).% V(ICP-MS): 2.1.
Preparation of [VVO2(pydx-ambmz)]-Y 6. The light yellow[VVO2(pydx-ambmz)]-Y was prepared following the same proce-dure as outlined for [VVO2(pydx-aebmz)]-Y. Selected IR data (KBr,nmax/cm-1): 1625 (C Nring).% V (ICP-MS): 1.9.
X-ray crystal structure determination
Three-dimensional X-ray data for 1 was collected on a BrukerSMART Apex CCD diffractometer at 153(2) K using a graphitemonochromator and Mo-Ka radiation (l = 0.71073 A) by thef-w scan method. Reflections were measured from a hemisphereof data collected of frames each covering 0.3 degrees in w. Ofthe 39550 reflections measured, all of which were corrected forLorentz and polarization effects, and for absorption by semi-empirical methods based on symmetry-equivalent and repeatedreflections, 6009 independent reflections exceeded the significancelevel |F|/s(|F|) > 4.0, respectively. Complex scattering factorswere taken from the program package SHELXTL.16 The structurewas solved by direct method and refined by full-matrix least-squares methods on F2. The non-hydrogen atoms were refinedwith anisotropic thermal parameters. The hydrogen atoms wereleft to refine freely in the other cases except for those attachedto C(1 W) and C(22), which were placed in idealised positionsand refined by using a riding mode. The final difference Fouriermap showed no residual density outside: 1.162 and -0.728 eA-3. Table 1 contains the crystal and structure refinement data16
for 1. CCDC No. 812058 contains the supplementary crystallo-graphic data for this paper. This data can be obtained free of
Table 1 Crystal and structure refinement data16 for [VIVO(acac)(pydx-aebmz)]·MeOH 1
Formula C23H28N4O6VFormula weight 507.43T/K 153.2l/A [Mo, Ka] 0.71073Crystal system MonoclinicSpace group P21/ca/A 11.9158(8)b/A 16.1483(12)c/A 12.7891(9)b (◦) 99.840(4)Z 4Volume/A3 2424.7(3)Dc/g cm-3 1.390m mm-1 0.455Reflections measured 39550Independent reflectionsa 6009R(int) 0.0432Goodness-of-fit on F2 1.163R1
b 0.0443wR2 (all data)b 0.1348
a I > 2s(I). b R1 = R |ıF0ı–ıFcı|/R ıF0ı, wR2 = {R [w(|ıF0ı2–ıFcı2|)2]/R [w(F0
4)]}1/2
charge from the Cambridge Crystallographic Data Centre viawww.ccdc.cam.ac.uk/data_request/cif.
Computational details
The full geometry optimization of all structures envisaged was car-ried out at the DFT level of theory using B3P8617 functional withthe help of the Gaussian-0318 program package. This functionalwas found to be appropriate for the theoretical studies of structuralparameters and 51V NMR chemical shifts of V complexes withSchiff bases.19 No symmetry operations were applied for anyof the structures calculated. The geometry optimization wascarried out using a relativistic Stuttgart pseudopotential thatdescribed 10 core electrons and the appropriate contracted basisset (8s7p6d1f)/[6s5p3d1f]20 for the vanadium atom and the 6-31G(d) basis set for other atoms. The Hessian matrix wascalculated analytically for all optimized structures in order toprove the location of correct minima (no imaginary frequencies)or transition state (one imaginary frequency) and to estimate thethermodynamic parameters, the latter being calculated at 25 ◦C.The nature of transition states was investigated by the analysis ofvectors associated with the imaginary frequency.
Most of the DFT calculations were done for the Schiff basemodel ligand III instead of the pydx-aebmz- ligand I. Thenumbering of model complexes bearing III corresponds to thereal species with addition of the symbol “ ¢ ”. Various possibleisomers of the calculated structures were considered and only themost stable ones were included to the discussion of the reactionmechanism.
For the mechanistic part, total energies corrected for sol-vent effects (Es) were estimated at the single-point calcula-tions on the basis of gas-phase geometries at the CPCM-B3P86/6-311+G(d,p)//gas-B3P86/6-31G(d) level of theory us-ing the polarizable continuum model21 in the CPCM version22
with acetonitrile as solvent. The UAKS model was appliedfor the molecular cavity. The entropic term in solutions(Ss) was calculated according to the procedure described byWertz,23 and Cooper and Ziegler24 (see Supporting Informa-tion for details†). The enthalpies and Gibbs free energiesin solution (H s and Gs) were estimated using the followingequations
H s = Es(6-311+G(d,p)) + Hg(6-31G(d)) - Eg(6-31G(d))
Gs = H s - TSs
where Es, Eg and Hg are the total energies in solution and ingas phase and gas-phase enthalpy calculated at the correspondinglevel.
Magnetic shielding was calculated for the equilibrium ge-ometries using the GIAO25 method at the CPCM-B3P86/6-311+G(2d,p)//gas-B3P86/6-31G(d) level including the solventeffects with methanol as solvent. 51V Chemical shifts (dV
calc) wereestimated relative to VVOCl3 (s of -2914 calculated at the samelevel of theory). For all peroxo-complexes the empirical correctionof -42 ppm was introduced.19
The 51V hyperfine coupling constants in the VIVO-complex1¢, model compound of 1, were estimated at the single-pointcalculations using the BH and HLYP functional and 6-311G*basis set for all atoms on the basis of the equilibrium ge-ometry obtained at the B3P86/6-31G*(V-ECP) level of the-ory. The anisotropic 51V hyperfine coupling constants Ax, Ay,and Az were estimated as the sum of the isotropic Fermicontact term and corresponding dipolar hyperfine interactionterm.26
Catalytic activity studies
The catalytic potential of the compounds prepared was tested,choosing the oxidation of styrene as a model reaction. Thus,[VVO2(pydx-aebmz)]-Y and [VVO2(pydx-ambmz)]-Y, as well as 1and 2 were used as catalyst precursors. All catalytic reactionswere carried out in a 50 mL flask fitted with a water circulatedcondenser. Typically, styrene (0.50 g, 0.005 mol) and aqueous 30%H2O2 (1.70 g, 0.015 mol) were taken in 10 mL of acetonitrile andthe temperature of the reaction mixture was maintained at 75 ◦Cwith continuous stirring. The reaction was considered to beginwhen the appropriate catalyst precursor (0.015 g) was added tothe reaction mixture.
The effects of various parameters, such as amounts of oxidantand catalyst as well as the temperature of the reaction mixture werestudied to check their effect on the conversion and selectivity of thereaction products. During the reaction the reagents and productsformed were analyzed by gas chromatography, withdrawing smallaliquots at chosen time intervals. Their identity was confirmed byGC-MS.
The reaction between equimolar amounts of [VIVO(acac)2] andHpydx-aebmz I or Hpydx-ambmz II (cf. Scheme 1) in dry, reflux-ing methanol yielded the brown oxidovanadium(IV) complexes[VIVO(acac)(pydx-aebmz)] 1 and [VIVO(acac)(pydx-ambmz)] 2,respectively. The synthetic procedures can be outlined by eqn (1)and (2)
The aerobic oxidation of 1 in methanol yielded the dioxidovana-dium(V) compound [VVO2(pydx-aebmz)] 3 (eqn (3)). A smallamount (c.a. 0.2 equiv.) of KOH assisted this oxidation.
The paramagnetic VIVO-complexes 1 and 2 exhibit normalmagnetic moment (1.72 mB to 1.74 mB) as expected for magneticallydilute d1 systems while all other compounds are diamagnetic. Allcomplexes are soluble in methanol, ethanol, DMSO and DMF.Scheme 2 presents the structural formulae and binding modesproposed for these complexes, based on their spectroscopic char-acterization (IR, electronic, EPR, 1H and 51V NMR), elementalanalyses and single crystal X-ray analysis of 1. The coordinationof the ligands involves their monoanionic (ONN-) form.
Scheme 2 Structural formulae of the complexes prepared in this work.
Encapsulation of [VVO2(pydx-aebmz)] and [VVO2(pydx-ambmz)] in the nano-cavities of zeolite-Y involved (i) the exchangeof VIVO2+ with Na+ of Na-Y in water to form zeolite-VIVO-Yspecies, followed by reaction of the metal-exchanged zeolite-Ywith I or II in methanol, and (ii) the slow oxidation of the initiallyformed VIVO-species by air when these samples are suspended
in MeOH to give [VVO2(pydx-aebmz)]-Y 5 and [VVO2(pydx-ambmz)]-Y 6. The remaining uncomplexed metal ions in thezeolite were removed by exchanging with aqueous 0.1 M NaClsolution. Extraction of impure samples with methanol using aSoxhlet extractor removed excess free ligand and stirring in DMFremoved the neat metal complex formed on the surface of thezeolite, if any. The metal content found after encapsulation for5 and 6 is mostly due to the presence of VVO2-complexes inthe cavities of the zeolite-Y. Notwithstanding, small amounts of[VIVO(OH)(pydx-aebmz)]n-Y 7 or [VIVO(OH)(pydx-ambmz)]m-Y8, respectively, are detected by EPR.
These encapsulated complexes were additionally characterisedby thermogravimetric patterns, Field-Emission-Scanning Elec-tron Micrograph (FE-SEM), Energy Dispersive X-Ray analy-sis (EDX) and powder X-ray Diffraction (XRD) Studies. Thebinding modes for these encapsulated complexes are proposedby comparison with the corresponding homogeneous modelcomplex 3.
Characterization of compounds
Structural description of [VIVO(acac)(pydx-aebmz)]·MeOH 1.Crystals suitable for X-ray diffraction studies were obtainedfor [VIVO(acac)(pydx-aebmz)]·MeOH 1. The asymmetric unitcontains one molecule of [VIVO(acac)(pydx-aebmz)] and onemolecule of methanol. Fig. 1 shows an ORTEP diagram withthe atom-labeling scheme, and selected bond lengths and anglesare given in Table 2. The [VIVO(acac)(pydx-aebmz)] is neutral withN(3) and O(5) protonated and the vanadium atom in the oxidationstate IV. The pydx-aebmz- ligand binds to the VIVO2+ moiety bymeans of one phenolate-O-, one imine-N and one benzoimidazole-N atoms forming a distorted octahedral coordination polyhedron.The distortion is reflected in the coordination distances andangles: there is one short V O bond [1.6204(12) A], which istrans to the acac-O(3) atom, which forms a longer V–O bond[2.1484(12) A] than the other acac-O(4) atom [1.9841(11) A].The metal ion is located 0.2743 A above the equatorial planedefined by the phenolate-O, O(2), acac-O, O(4), imine-N (N2)and benzoimidazole-N, N(4), with a deviation from planarity of0.0280(6) A. Atom N(1) is involved in a short intermolecular Hbond with O(1 W) of neighboring methanol molecule, while atomsN(3) and O(5) are involved in one with O(1 W) of a symmetry-related molecule (see Table S1 of ESI†).
Fig. 1 Molecular structure of [VIVO(acac)(pydx-aebmz)]·MeOH 1, show-ing the atomic numbering scheme. The ORTEP plot is at 30% probabilitylevel.
Thermogravimetric analysis. Thermogravimetric analysis isvery useful for encapsulated complexes to estimate their thermalstability. The thermal decomposition of [VVO2(pydx-aebmz)]-Yand [VVO2(pydx-ambmz)]-Y proceeds in two major steps. Anendothermic weight loss of ca. 16% occurs in the temperaturerange 100 to 300 ◦C, which is possibly due to the removal ofintrazeolitic water. The second step of endothermic weight lossstarts immediately after the first step and consists of several substeps which continue until ca. 650 ◦C to constant weight. Atotal weight loss of ca. 11% in [VVO2(pydx-aebmz)]-Y and 8% in
[VVO2(pydx-ambmz)]-Y in the temperature range of 300–650 ◦Cdue to the slow decomposition of vanadium complexes suggeststhe insertion of only a small amount of metal complex in the cavityof the zeolite-Y. This is in agreement with the low percentageof metal content obtained for encapsulated complexes. In factthe analytical results of %V obtained by ICPMS are: [VO2(sal-aebmz)]-Y: 2.1%; [VO2(sal-ambmz)]-Y: 1.9%.
Field-Emission-Scanning Electron Micrograph and Energy Dis-persive X-Ray analysis studies. The field emission scanningelectron micrographs of [VVO2(pydx-aebmz)]-Y and [VVO2(pydx-ambmz)]-Y along with their energy dispersive X-ray analysis(EDX) profiles are presented in Fig. 2. Accurate information onthe morphological changes in terms of exact orientation of ligandscoordinated to the metal ion was not possible due to poor loadingof the metal complexes. However, it is clear from the micrographsthat the vanadium complexes entrapped zeolites have well definedcrystals and there is no indication of the presence of any metalions or complexes on the surface. Energy dispersive X-ray analysisplots, evaluated semi-quantitatively, support this conclusion asno vanadium or nitrogen contents were detected on the spottedsurface for both catalysts. Only a small amount of carbon (ca.5%) but no nitrogen on the spotted image of SEM suggests thepresence of trace amount of solvent (methanol) from which itwas finally washed after Soxhlet extraction. An amount of ca.3% sodium suggests the exchange of remaining free Vanadiumions by sodium ions during re-exchange process (see experimentalsection). The average Si and Al percentage obtained were ca. 35%and ca. 12%, respectively.
Fig. 2 Field-Emission-Scanning Electron Micrographs (left) and Energy Dispersive X-ray analysis profiles (right) of [VVO2(pydx-aebmz)]-Y (above) and[VVO2(pydx-ambmz)]-Y (below).
Powder X-ray diffraction studies. The powder X-ray diffrac-tion patterns of Na-Y, VIVO-Y and of encapsulated oxidovana-dium(IV) complexes [VVO2(pydx-aebmz)]-Y and [VVO2(pydx-ambmz)]-Y were recorded at 2q values between 5 and 70◦, andthe obtained patterns are presented in Fig. 3. Apart from aslightly weaker intensity of the zeolites having encapsulated metalcomplexes, essentially similar diffraction patterns in encapsulatedcomplexes, VIVO-Y and Na-Y, were obtained. These observationsindicate that the framework of the zeolite did not undergo anysignificant structural change during incorporation of the catalysts,i.e., crystallinity of the zeolite-Y is preserved during encapsulation.
Fig. 3 XRD patterns of powdered samples of Na-Y, VIVO-Y,[VVO2(pydx-aebmz)]-Y and [VVO2(pydx-ambmz)]-Y.
IR spectral studies. The IR spectrum of Hpydx-aebmz exhibitstwo sharp bands at 1606 and 1636 cm-1 due to n(C N) (azome-thine/ring stretch). These bands undergo a bathochromic effectin complexes, indicating coordination of the azomethine/ringnitrogen to the vanadium. The presence of several mediumintensity bands between 2500 and 2700 cm-1 in the ligand as wellas in most of the complexes suggests the existence of hydrogenbonding between NH of benzimidazole and other electronegativeatoms. The coordination of the phenolic oxygen could not beascertained unequivocally from IR data but by comparisonwith the structurally characterized complex [VVO2(sal-aebmz)],8 amonobasic tridentate ONN binding of the ligand is also expectedfor the present complexes. Table 3 summarizes IR data for thecompounds discussed in this work.
The VIVO-compounds [VIVO(acac)(pydx-aebmz)] 1 and[VIVO(acac)(pydx-ambmz)] 2 display one sharp band at 950and 925 cm-1, respectively, due to the n(V O) mode (seeTable 3). There are two such bands at 924 and 964 cm-1
in [VVO2(pydx-aebmz)] 3, corresponding to the ns(O V O)and nas(O V O) modes, respectively. The peroxido complex[VVO(O2)(pydx–aebmz)] 4 exhibits three IR active vibration modesassociated with the peroxido moiety {V(O2)3+} at 878, 755 and614 cm-1, and these are assigned to the O–O intra-stretch (n1), theantisymmetric V(O2) stretch (n3), and the symmetric V(O2) stretch(n2), respectively. The presence of these bands confirms the usualh2-coordination of the peroxido group. In addition, the sharp bandappearing at 962 cm-1 is assigned to the n(V O) stretch.27,28
Peaks due to the zeolite-Na-Y normally dominate the spectraof these materials. Some of the major zeolite framework absorp-tions appear as strong bands around 1140, 1040, 960, 785 and740 cm-1.28,29 Therefore, it was not possible to clearly identify bandsdue to the cis-VVO2 moiety in zeolite encapsulated vanadiumcomplexes 1 and 2.
Electronic spectral studies. Table 3 also includes electronicspectral data of ligands and complexes and Fig. 4 provides spectraof encapsulated complexes. The UV spectra of the ligands showfive sharp absorption maxima. The first two bands are mostlyassigned due to j→j* and p→p*, while the last band is assignableto n→p* transitions. Other intra-ligand bands are appearing dueto the presence of the benzimidazole residue.8 The electronicspectral data of neat as well as encapsulated complexes exhibit lmctbands at ca. 375–400 nm. In addition, they also exhibit two or threeUV bands. The lower energy and less intense bands appearing at541 and 790 nm {in [VIVO(acac)(pydx-aebmz)] 1}, or 525 and805 nm {in [VIVO(acac)(pydx-ambmz)] 2}, are assigned to d→d
Fig. 4 Electronic spectra of encapsulated complexes: (a)[VVO2(pydx-aebmz)]-Y (5) and [VVO2(pydx-ambmz)]-Y (6) recorded inNujol (see experimental section).
Table 3 IR and electronic spectral data of ligand and complexes
transitions. As VV-complexes have 3d0 configuration, d→d bandsare not expected.
51V NMR studies. The 51V NMR spectra of [VVO2(pydx-aebmz)] 3 and [VVO(O2)(pydx-aebmz)] 4 (ca. 4 mM) dissolved inDMSO-d6 were recorded to establish their behavior in solutionand their speciation upon modifying the composition of thesolutions (see Fig. S1). The assignment of resonances was done byconsidering both the experiments described below and the resultsof DFT calculations carried out for the corresponding complexesinvolving the model ligand III (see experimental section).
The 51V NMR spectrum of [VVO2(pydx-aebmz)] 3 inDMSO exhibits a resonance at -543 ppm, this beingwithin the values expected for VVO2-complexes containinga O/N donor set.6a,6b,30 It also agrees with the value of-542 ppm reported for [VVO2(SALIMH)] {SALIMH = [4-(2-salicylideneaminato)ethyl]imidazole).6b The value calculated forthe model complex 3¢, considering MeOH as solvent, is dV
cal = -507ppm; for the DFT optimized structure of 3¢·MeOH, where MeOHis also coordinated to VV, but very weakly, dV
cal = -522 ppm (seealso below). The 1H NMR spectrum (vide supra) also supports themonobasic tridentate ONN behavior of the ligand as formulatedin Scheme 2. A trigonal bipyramidal geometry distorted towardssquare pyramidal was confirmed for the closely related complex[VVO2(sal-aebmz)] (Hsal-aebmz = Schiff base derived from sali-cylaldehyde and 2-aminoethylbenzimidazole) by single crystal X-ray diffraction.8 Thus, a similar trigonal bipyrimidal geometryis suggested for 3 as well as for the corresponding encapsulatedVVO2-complexes 5 and 6.
The 51V NMR spectrum of [VVO(O2)(pydx-aebmz)] 4 in DMSOshows a resonance at -543 ppm assigned to complex 3, and aminor peak at -563 ppm (assigned to 4). Thus, upon dissolving4 most of the peroxido ligand is lost, complex 3 being the mostimportant VV-containing species in solution.
EPR studies. The EPR spectra of “frozen” solutions (77 K) of1 and 2 in MeOH/DMSO are depicted in Fig. 5. The spectra arenearly axial and the hyperfine features are well resolved in both theparallel and perpendicular regions. The spectra were simulated15
and the obtained spin Hamiltonian parameters gz and Az agreewell with the values estimated using the additivity relationship pro-posed by Wuthrich31 and Chasteen,32 with estimated accuracy of± 1.5 ¥ 10-4 cm-1, including the contribution of Az(benzimidazole){considered as equal to Az(imidazole) predicted by Smith et al.33}.
The gz, Az obtained, 1.951, 162.0 ¥ 10-4 cm-1 for 1 and 1.951,163.3 ¥ 10-4 cm-1 for 2, respectively (Table 4), agree well with
Fig. 5 First derivative EPR spectra of frozen methanolic so-lutions (at 77 K) of (a), [VIVO(acac)(pydx-aebmz)] 1 and (b)[VIVO(acac)(pydx-ambmz)] 2.
those reported by Smith et al.33 for [VIVO(SALIMH)(acac)], 1.960,162.3 ¥ 10-4 cm-1, for which a single crystal X-ray diffractionstudy also confirmed the binding set (OPhen, Nimine, Nimidazole,Oacac)equatorial(Oacac)axial. The quantum-chemical calculations of thehyperfine coupling constants for [VIVO(acac)(L¢)] (1¢) with themodel ligand L¢ (L¢ = III) reproduce well the experimental data.Indeed, the calculated |Az|, |Ay|, and |Ax| values are 162.2 ¥10-4, 63.5 ¥ 10-4, and 60.2 ¥ 10-4 cm-1; the hyperfine tensorhaving an axial symmetry, the direction of Az coinciding withthe V O bond. The perpendicular tensors are directed betweenthe equatorial donor atoms.
The EPR spectra of 1 and 2 each show a minor sec-ondary species with Az values of ~164 ¥ 10-4 cm-1 (for 1) and~165 ¥ 10-4 cm-1 (for 2), which probably correspond to the hy-drolytic products: [VIVO(pydx-aebmz)(MeOH)] and [VIVO(pydx-ambmz)(MeOH)], respectively. In fact, the difference in thecontributions to Az assumed for Oacac (41.6 ¥ 10-4 cm-1)34 and forOMeOH (~45.6 ¥ 10-4 cm-1),32 and the probably distinct contributionof Az(benzimidazole), is compatible with the difference in the Az
values observed for 1 and 2 and for their hydrolytic products.Hydrated zeolite-VIVO-Y was previously reported and its EPR
spectrum recorded.35 It was concluded that the VIVO2+ ions arelargely located on type III sites in the large cavities bound tohydroxyl or O atoms, the spin Hamiltonian parameters being:gz = 1.938 and Az = 178 ¥ 10-4 cm-1. Upon dehydration theseparameters change, but upon re-hydration the original parameterswere restored. We prepared zeolite-VIVO-Y and recorded its EPRspectrum (Fig. 6). The spectra presents rather broad lines due todipolar interactions and the spin Hamiltonian parameters cannot
Table 4 Spin Hamiltonian parameters of several VIVO-complexes obtained from the experimental EPR spectra recorded at 77 K, and calculated forcomplex 1¢
Fig. 6 First derivative EPR spectra of (a) [VIVO-Y], (b) [VIVO(OH)(pydx-aebmz)]-Y 7 solid; (c) [VIVO(OH)(pydx-ambmz)]-Y 8 solid; at 77 K (see text).
be determined accurately being gz ~ 1.939 and Az ~ 176 ¥ 10-4 cm-1,in good agreement with the data previously reported.35 This is notsurprising as the %V found was relatively high, ca. 4.6.
The change of the EPR spectrum after introduction of a ligandand its similarity with the corresponding neat complex is goodevidence of the encapsulation of the VIVO(ligand) species inside thepores of zeolite-Y. As mentioned, encapsulation of [VVO2(pydx-aebmz)] and [VVO2(pydx-ambmz)] in zeolite-Y involved the reac-tion of zeolite-VIVO-Y species with I or II in methanol. Here,ligands enter into the nano-cavities of zeolite-Y due to theirflexible nature and interact with VIVO2+ to give VIVO-complexesthat we tentatively formulate as [VIVO(OH)(pydx-aebmz)]n-Y 7and [VIVO(OH)(pydx-ambmz)]m-Y 8, respectively, with n, m = 1(or 2), where the OH donor is supplied by the surface groups ofthe cavity walls. It is probable that, besides this binding, hydrogenbond interactions are established between the zeolite-OH groupsand the –CH2OH and NH moieties of the ligands. The EPR spectrarecorded for 7 and 8 at 77 K also show relatively broad lines, butdiffer from the spectrum of zeolite-VIVO-Y used as precursor (Fig.6), confirming the change in environment of the VIVO-centre. Thespectra are compatible with the formulation as [VIVO(OH)(pydx-aebmz)]n-Y 7 and [VIVO(OH)(pydx-ambmz)]m-Y 8, respectively.Indeed the spin Hamiltonian parameters obtained, gz ~1.944, Az =165 ¥ 10-4 cm-1 (7), and gz ~1.946, Az = 167 ¥ 10-4cm-1 (8), arein the range estimated for binding modes (OHY, OPhenolate, Nimine,Nbenzimidazole)equatorial(Nbenzimidazole)axial. No signal was recorded in thefield range corresponding to g = 4.
Reactivity of [VIVO(acac)(pydx-aebmz)] 1 and[VVO2(pydx-aebmz)] 3 with H2O2: possible speciation in solution
It is known that several VVO2-complexes react with H2O2 togive the corresponding VVO(O2)-complexes, and it is normallyassumed that the corresponding hydroperoxido-complex is theactive catalyst which mediates oxygenation, including the oxida-tion of sulfides to sulfoxides and sulfones and the epoxidationof alkenes and allylic alcohols.36 As monitored by UV-Vis and51V NMR, addition of H2O2 to solutions of [VVO2(pydx-aebmz)]in DMF yield the oxidoperoxido species. Fig. 7 presents thespectral changes observed. Thus, the dropwise addition of a 30%
Fig. 7 Spectral changes observed during titration of ca. 10-3 M solution of[VVO2(pydx-aebmz)] in DMF with aqueous 30% H2O2 mixed with a smallamount of DMF. A band at ca. 400 nm emerges and progressively increasesits intensity, this indicating the formation of [VVO(O2)(pydx-aebmz)].
H2O2 solution diluted in DMF to a ca. 10-3 molar solution of[VVO2(pydx-aebmz)] in DMF results in the partial reduction ofthe intensity of the 340 nm peak and the appearance of a band atca. 400 nm. These changes indicate the interaction of [VVO2(pydx–aebmz)] with H2O2 in DMF, the band at ca. 400 nm correspondingto a peroxido-to-vanadium charge transfer transition.37 The finalspectrum is similar to the one obtained for freshly prepared[VVO(O2)(pydx-aebmz)] 4.
Treatment of [VIVO(acac)(pydx-aebmz)] 1 with H2O2 at ca. 10 ◦Cyielded the VVO-complex [VVO(O2)(pydx-aebmz)] 4. Changes inthe absorption spectra during titration of a methanolic solution of1 with H2O2 dissolved in MeOH are shown in Fig. 8, and resemblethose observed for similar titrations of methanolic solutions of[VIVO(acac)(sal-dmen)] with H2O2
19 (Hsal-dmen = Schiff basederived from salicylaldehyde and N,N-dimethylethylene diamine).The d–d bands appearing at 541 and 790 nm and the band at402 nm gradually disappear, while a band at 337 nm increases itsintensity. Other UV bands also change intensity and/or their lmax.
Progressive addition of H2O2 (2, 4, 5, 6 and 7 equivalents ofH2O2) to a 5 mM methanolic solution of 1 led to the continuousdecrease of the intensity of the EPR signal without changing itspattern, as shown in Fig. 9. Reduction of intensity to ca. 1/8th of
Fig. 8 UV-Vis spectral changes observed during titration of [VIVO(acac)(pydx-aebmz)] with H2O2. (a) The spectra were recorded after successiveadditions of one drop portions of H2O2 (1 ¥ 10-4 mmol of 30% H2O2 in 10 mL of methanol) to 50 mL of ca. 10-3 M solution of [VIVO(acac)(pydx-aebmz)]in methanol. (b) The equivalent titration, but with lower concentration of a [VIVO(acac)(pydx-aebmz)] solution (ca. 10-4 M).
Fig. 9 (a) First derivative EPR spectrum of a methanolic solution of [VIVO(acac)(pydx-aebmz)] (1) (ca. 4 mM); (b) after addition of a total of 2.0 equiv.(30%) H2O2 (total) to the solution of (a); (c) after addition of 4.0 equiv. (30%) H2O2 (total) to the solution of (b); (d) after addition of 5.0 equiv. (30%)H2O2 (total) to the solution of (c); (e) after addition of 7.0 equiv. (30%) H2O2 (total) to the solution of (d).
that of the initial solution was observed after addition of 7 equiv. ofH2O2. These results indicate that the VIV-centre in 1 is oxidized byH2O2, but that the remaining [VIVO(acac)(pydx-aebmz)] preservesits integrity, no other VIV-complexes being detected by EPR.
After addition of 1.0 or 2.0 equiv. of H2O2 to the methanolicsolution of 1 (Fig. 10), by 51V NMR spectroscopy three signals, atdV = -490, -543 and -563 ppm are detected. We tentatively assignthe dV = -563 ppm peak to [VVO(O)2(pydx-aetbmz)]. Indeed thedV
calc(3¢) = -507 ppm, dVcalc(3¢·MeOH) = -522 ppm and dV
calc (4¢) =-550 ppm, supporting this assignment. The resonance at dV = -490ppm is assigned to protonation of complex 3, yielding a speciesdesignated by CI. This protonation can occur either at one of theOoxido moieties or at one of the ligand donor atoms, probably theNbenzimidazole. Several isomeric structures were calculated for CI andare included as Supplementary Information†, with dV
calc in therange of -440 to -539 ppm (see Scheme 3). The protonation atthe Nbenzimidazole yielding CI or CIa·MeOH is therefore a plausibleassignment for the -490 ppm peak.
Upon further additions of aqueous H2O2 (Fig. 10 (A)), severalother resonances progressively appear at -602, -575, -644 and-725 ppm, all these being assigned to peroxido-VV complexes.
Such assignments are supported by the theoretically calculatedvalues of the 51V NMR chemical shifts for the correspondingmodel species with ligand III (see Scheme 3). Moreover, eitheradding styrene to these solutions (e.g. Fig. 10B (f)) or leavingthem standing in contact with air, the peroxido-VV complexesprogressively yield mainly the peak at -543 ppm correspond-ing to complex 3, demonstrating the reversibility of theseprocesses.
The 51V NMR spectrum of a 5 mM solution of [VVO(O2)(pydx-aebmz)] 4 in DMSO (Fig. 10B), shows a major peak at -543 ppmdue to complex 3, and a minor peak at -563 ppm (assigned to4). Upon successive additions of portions of a 30% H2O2 aqueoussolution, the relative intensity of the -543 ppm peak decreases andseveral new peaks appear: after addition of ca. 2 equiv. of H2O2, apeak at -602 ppm emerges, and upon further additions of portionsof the 30% H2O2 solution, new peaks appear at -644 and -725 ppmand increase intensity at the expense of the resonances at -543 ppm(and -602 ppm). Again upon addition of styrene the peaks at -602,-644 and -725 ppm gradually disappear (the last to disappearis the one at -602 ppm). The tentative assignment of 51V NMRpeaks is indicated in Scheme 3, and is based both on the theoretical
Fig. 10 (A) 51V NMR spectra of (a) A methanolic solution of [VIVO(acac)(pydx-aebmz)] 1 (ca. 4 mM) the pH is ca. 8.0; (b) addition of 1.0 equiv. of 30%H2O2 to the solution of (a); (c) after addition of 2.0 equiv. 30% H2O2 (total) to the solution of (b); (d) after addition of 3.0 equiv. of 30% H2O2 (total) tothe solution of (c) the pH is ca. 5.0; (e) after addition of 4.0 equiv. of 30% H2O2 (total) to the solution of (d) the pH is ca. 4.5; (f) after addition of 5.0equiv. of 30% H2O2 (total) to the solution of (e) the pH is ca. 4.0; (g) after addition of 7.0 equiv. 30% H2O2 (total) to the solution of (f) the pH is ca. 4.0;(h) solution of (g) after 5 h; (i) solution of (h) after 24 h. (B) 51V NMR spectra of (a) A methanolic solution of [VVO(O)2(pydx-aebmz)] 4 (ca. 4mM) thepH is ca. 7.5; (b) addition of 2.0 equiv. (30%) H2O2 (total) to the solution of (a); (c) after addition of 5.0 equiv. (30%) H2O2 (total) to the solution of (b)the pH is ca. 5.0; (d) solution of (c) after 1 h; (e) after addition of 7.0 equiv. (30%) H2O2 (total) to the solution of (d) the pH is ca. 4.0–4.5; (f) solution of(e) after addition of 10 equiv. methyl phenyl sulfide; (g) solution of (f) after 4 h.
calculations of dV resonances and on the expected changes of theconcentrations of H2O2 and H+ upon adding the aqueous H2O2
solution. These results indicate that during the catalytic reactions,namely the oxidation of styrene by peroxide, intermediate peroxidospecies indeed form,38 and protonated complexes CII (and possiblyalso CI) are expected to have significant concentrations in solution.
Catalytic activities
Oxidation of styrene. To test the catalytic potential of thecomplexes prepared, the oxidation of styrene was chosen as amodel reaction. Thus, the oxidation of styrene, catalyzed by[VVO2(pydx-aebmz)]-Y and [VVO2(pydx-aebmz)]-Y was carriedout using aqueous 30% H2O2 as an oxidant. It gave styrene oxide,benzaldehyde, 1-phenylethane-1,2-diol, benzoic acid and pheny-lacetaldehyde along with only minor amounts of unidentifiedproducts. Scheme 4 represents the formation of all these products.These products of styrene oxidation were previously observed byothers as well.7,39–42
To achieve reaction conditions for the maximum oxidativeconversion of styrene [VVO2(pydx-aebmz)]-Y was taken as arepresentative catalyst and after a few preliminary tests concerningthe reaction time and temperature, several different parameters,viz. amount of oxidant, catalyst and effect of H2O2 amount were
varied. The effect of the amount/concentration of H2O2 on theoxidation of styrene is illustrated in Fig. 11. At a styrene to30% H2O2 molar ratio of 1 : 1, a maximum of 33% conversion
Fig. 11 Effect of amount of H2O2 concentration (H2O2: styrene molarratio) on oxidation of styrene. Reaction conditions: styrene (0.50 g, 0.005mol), aqueous 30% H2O2 (0.005, 0.010 or 0.015 mol), acetonitrile (10 mL)and catalyst precursor (0.015 g) at ca. 75 ◦C with continuous stirring. Thereaction took place during 6 h counted after addition of the catalyst.
Scheme 3 Summary of speciation of V-pydx-aebmz species in solution. The solvent is methanol unless specified. The experimental dV are indicated as(E: xxx ppm), and the calculated dV
calc values as (T: xxx ppm) for the model compounds 3¢, CI¢ and CII¢ with ligand L¢ = III instead of pydx-aebmz.
Scheme 4 Various products obtained upon oxidation of styrene.
was achieved in 6 h of contact time in the conditions specifiedin the Figure caption. Increasing the ratio to 1 : 2 improved the
conversion to ca. 58%, while at a 1 : 3 ratio the conversion was ca.68%. Further increments of H2O2 did not improve the conversionfurther.
Similarly, for three different amounts (viz. 0.005, 0.015, and0.020 g) of catalyst [VVO2(pydx-aebmz)]-Y and a H2O2 to styrenemolar ratio of 3 : 1 under above reaction conditions, 0.005 g gaveonly 30% oxidative conversion while 0.015 and 0.020 g have shownnearly identical results with ca. 68% conversion in 6 h of contacttime; Fig. S2†. Thus, 0.015 g catalyst may be considered adequateto run the reaction under the above conditions.
Considering these selected “optimized” reaction conditions, thecatalyst precursor [VVO2(pydx-ambmz)]-Y was also tested underthe same conditions. Thus, for 5 mmol of styrene, 15 mmol of30% H2O2 and 0.015 g of [VVO2(pydx-ambmz)]-Y were taken in10 mL of CH3CN and the reaction was carried out at 75 ◦C.A maximum of ca. 65% conversion was achieved after 6 h ofreaction time, which is marginally less than that observed for[VVO2(pydx-ambmz)]-Y; Fig. S3†. The turn over rate calculatedfor these catalysts is ca. 55 h-1). A conversion of ca. 2.9% wasobtained in the blank reaction, and ca. 4.8% for VO-Y. Thecatalytic activity of a similar catalyst [VVO2(sal-ambmz)]-Y was
a SO: Styrene oxide, bza: Benzaldehyde, phed: 1-phenylethane-1,2-diol, bzac: Benzoic acid, phaa: Phenyl acetaldehyde.
ca. 97%.9 Furthermore ca. 80% conversion of styrene was reportedwhen using the polymer supported catalyst PS-K[VVO2(sal-oaba)](H2sal-oaba = Schiff base derived from salicylaldehyde and o-aminobenzyl alcohol).43 The selectivity of styrene oxide in all thesecatalysts is always low, but indeed the experimental conditionswere not developed to optimize the formation of this product.
Table 5 presents a comparative report on the conversion ofstyrene and selectivity of the various products of the reaction.The selectivity profile for the various products is similar for bothcomplexes and follows the order: benzaldehyde >1-phenylethane-1,2-diol > benzoic acid > styrene oxide > phenyl acetaldehyde.Aspects related to the mechanism of reaction are discussedbelow. The high yield of benzaldehyde may possibly be due tothe further oxidation of styrene oxide formed in the first stepby a nucleophilic attack of H2O2 on styrene oxide followed bycleavage of the intermediate hydroperoxystyrene.41 The formationof benzaldehyde may also result from direct oxidative cleavageof the styrene side chain double bond via a radical mechanism.The high amount of water present in H2O2 is partly responsiblefor the possible hydrolysis of styrene oxide to 1-phenylethane-1,2-diol. Other products, e.g. benzoic acid formation through furtheroxidation of benzaldehyde, only form in low amount in thesereactions. Similarly the formation of phenylacetaldehyde throughisomerisation of styrene oxide is quite low in all cases.
Tests of recycle ability and heterogeneity of the zeolite-Y encap-sulated catalytic reactions. The recycle ability of encapsulatedcomplexes was examined up to two cycles considering the oxida-tion of styrene. After a contact time of 6 h the reaction mixture wasfiltered, the catalyst was washed with acetonitrile and dried at ca.120 ◦C. It was subjected to further catalytic reaction under similarconditions. No appreciable loss in catalytic activity was observed,this indicating that most of the complex is still present in the cavityof the zeolite-Y. In another experiment, after ca. 6 h of reactiontime the filtrate collected after separating the used catalyst wasplaced into the reaction flask and the reaction was continued afteradding fresh oxidant for another 4 h. The gas chromatographicanalysis showed no improvement in conversion and this confirmsthat the reaction did not proceed upon removal of the solidcatalyst. The oxidation of styrene is, therefore, heterogeneous innature.
Theoretical mechanistic studies
For the oxidation of hydrocarbons with transition metal complexesas catalysts in the presence of H2O2 two general groups ofmechanisms are known: the radical and non-radical ones, bothinvolving metal-peroxido intermediates. The radical mechanism,developed by Shul’pin for the V-catalysts44 and theoretically
Table 6 Energetic characteristics (in kcal mol-1) of the calculatedreactionsa
a The negative values of DH sπ for some reactions are accounted for by the
formation, on the first step of the processes, of molecular van der Waalscomplexes with total energies lower than the sum of the total energies ofthe separate reactants.
studied by Bell et al.45 as well as by Pombeiro, Shul’pin et al.,46
is based on the formation of the free HOO∑ and HO∑ radicals,the latter directly oxidizing hydrocarbons giving alkylperoxideswhich then undergo decomposition to the final oxidation products.Among the non-radical pathways, the Sharpless mechanism isusually considered as the most favorable one for the epoxidationof olefins.47,48 In the present work, in order to better understandthe mechanism of the oxidation of alkenes relevant here, quantum-chemical calculations of both (i) the non-radical Sharpless routeand (ii) the radical pathway have been undertaken for the modelsystem 3¢/H2O2/CH3CN + C2H4, where 3¢ is the model complex[VVO2L¢] with L¢ = III.
Non-radical Sharpless mechanism. This mechanism startsfrom the slightly endoergonic formation (by 5.0 kcal mol-1) ofan H2O2 adduct [VO2L¢(H2O2)] (TI) with the catalyst precursormolecule (Scheme 5, Table 6). The H-transfer from the coordinatedH2O2 to one of the oxo-ligands affords the hydroxo-hydroperoxocomplex TIIa. As was shown previously,45,49 the H-transfers inthe V or Re peroxo complexes are facilitated by a water moleculewhich provides the formation of stable six-membered transitionstates (TS) instead of the less stable four-membered TSs. In thepresent case, the reaction TI → TIIa occurs via TS1 (Fig. 12) withthe activation barrier 8.7 kcal mol-1 (in terms of DGs
π).Complex TIIa then converts to the peroxo complex 4¢. From the
search of the potential energy surface, two possible routes werefound for this step. The first one includes the V–N bond cleavageto give TIIb, which is by 0.4 kcal mol-1 more stable than TIIa,the H-transfer from the hydroperoxo ligand to the hydroxo ligand
Fig. 12 Equilibrium structures of the calculated transition states.
Scheme 5 Sharpless mechanism of the olefin epoxidation with complex3¢.
via TS2 to form TSIIIa and then 4¢ upon the H2O elimination.The second route involves the isomerization of TIIa to TIIc, theH-transfer through TS3 leading to TIIIb, and the water liberationaffording 4¢. The second route appears to be more favorable thanthe first one, see the left part of Fig. 13.
Finally, complex 4¢ reacts with the olefin (ethylene) in theoxygen transfer process via the Sharpless-type TS (TS4) with theformation of epoxide and regeneration of the catalyst. The DGs
π
value for this step is 33.2 kcal mol-1 that is similar to the activationbarrier for the catalysts based on the V-salan complexes (ca.32 kcal mol-1).48 The apparent activation energy of the Sharplessmechanism relative to complex 3¢ is 27.7 kcal mol-1 (Fig. 13).
Radical mechanism. The radical pathway also starts from theformation of TI and TIIa followed by the homolytic cleavage ofthe V–OOH bond to give the HOO∑ radical and the VIV complexTIV (Scheme 6). The calculated DH s values of the reaction TIIa →
Scheme 6 Radical mechanism of the HOO∑ and HO∑ formation catalyzedby 3¢.
TIV + HOO∑ (e.g., the adiabatic V–OOH bond enthalpy) is onlyslightly positive (5.6 kcal mol-1). However, the corresponding DGs
value is negative (-3.9 kcal mol-1) due to a favorable entropic factorindicating that the elimination of HOO∑ from TIIa is spontaneous(Table 6).
In the next step, the addition of another H2O2 molecule to TIVleads to complex TV, which, upon the H-transfer via TS5 andsimultaneous liberation of H2O, converts to TVI. The latter speciesundergoes an elimination of the HO∑ radical upon the O–OH bondcleavage through TS6 with the regeneration of the catalyst 3¢. TheHO∑ radicals react with hydrocarbons in the presence of molecularoxygen according to known reactions.44,51
The inspection of the energy profile (Fig. 13) demonstrates thatthe rate limiting step of the formation of HO∑ radical is the last stepof the O–O(H) bond cleavage in TVI, although the previous stepof the H-transfer in TV is only by 0.8 kcal mol-1 less energetic.
Fig. 13 Energy profiles of the catalytic cycles for the Sharpless mechanism of ethylene epoxidation (left) and the radical mechanism of HO∑ formation(right) with complex 3¢. Only V-bearing species, radicals and epoxide are indicated.50
The apparent activation energy of this mechanism relative to 3¢is 22.4 kcal mol-1. This value is clearly lower than the activationbarrier of the Sharpless route (27.7 kcal mol-1) indicating that theradical pathway is even more favorable than the non-radical one.This conclusion is consistent with the experimental observations(Table 5) indicating that the principal product of the styrene oxida-tion in the present system 3/H2O2/MeCN–H2O is benzaldehyde–the product of the C–C bond cleavage which usually forms underfree radical conditions.51
Conclusions
Compounds [VIVO(acac)(pydx-aebmz)] 1 and [VIVO(acac)(pydx-ambmz)] 2 where the ligands are the ONN donor Schiffbases obtained by the condensation of pyridoxal and 2-aminoethylbenzimidazole or 2-aminomethylbenzimidazole weresynthesized. A distorted octahedral geometry with the ONNequatorial binding mode of the tridentate ligand was confirmedfor 1 by single-crystal X-ray diffraction. Aerobic oxidation ofthe methanolic solution of 1 yielded [VVO2(pydx-aebmz)] 3 andits reaction with aqueous H2O2 gave [VVO(O2)(pydx-aebmz)]4. The encapsulated complexes [VVO2(pydx-aebmz)]-Y 5 and[VVO2(pydx-ambmz)]-Y 6 were also obtained.
Oxidation of styrene in acetonitrile using H2O2 as oxidantunder optimised reaction conditions for maximum conversionafter 6 h of reaction time gave conversions of styrene of 68%(with 5) and 65% (with 6), benzaldehyde being the main productformed (ca. 55%), and 1-phenylethane-1,2-diol > benzoic acid >
styrene oxide > phenyl acetaldehyde (order of selectivity). Theencapsulated complexes are recyclable at least up to two cycles ofcatalytic oxidation.
The radical and non-radical (Sharpless-type) mechanism ofepoxidation were studied by DFT methods for the model com-pound 3¢ and ethene as substrate. The radical pathway was foundto be even more favorable compared to the non-radical route,
accounting for the formation of benzaldehyde as the main productof the styrene oxidation.
The speciation of the system VV-pydx-aebmz upon additionof aqueous H2O2, and consequent change in pH was studied byUV-Vis and 51V NMR spectroscopy. Several distinct species form,and yield different dV values, namely [VVO2(pydx-aebmz)] and[VVO(O2)(pydx-aebmz)] as well as their protonated complexes,and assignments were proposed also with the support of DFTcalculations. The formation of peroxido-complexes is relevant forthe catalytic reactions. The fact that upon addition of H2O2 to theVV-pydx-aebmz complexes several types of peroxido-complexesform may partly explain the relatively low selectivity obtained.
Acknowledgements
MRM and MB thank the Department of Science and Technology,the Government of India, New Delhi for financial support ofthe work. JCP and AK thank the Portuguese NMR Network(IST-UTL Center), and the POCI 2010, FEDER, Fundacao paraa Ciencia e Tecnologia, SFRH/BPD/34835/2007. M.L.K. isgrateful to the IST and FCT for a research contract within theCiencia 2007 scientific programme.
References
1 (a) H. Sakurai, Y. Yoshikawa and H. Yasui, Chem. Soc. Rev., 2008, 37,2383–2392; (b) Y. Shechter, I. Goldwaser, M. Mironchik, M. Fridkinand D. Gefel, Coord. Chem. Rev., 2003, 237, 3–11; (c) K. H. Thompson,J. H. McNeill and C. Orvig, Chem. Rev., 1999, 99, 2561–2572; (d) K.H. Thompson and C. Orvig, Dalton Trans., 2006, 761–764; (e) K. H.Thompson, J. Lichter, C LeBel, M. C. Scaife, J. H. McNeil and C.Orvig, J. Inorg. Biochem., 2009, 103, 554–558; (f) D. C. Crans, J. Inorg.Biochem., 2000, 80, 123–131; (g) M. Yamaguchi, K. Wakasugi, R. Saito,Y. Adachi, Y. Yoshikawa, H. Sakurai and A. Katoh, J. Inorg. Biochem.,2006, 100, 260–269; (h) D. Rehder, J. Costa Pessoa, C. F. G. C. Geraldes,M. M. C. A. Castro, T. Kabanos, T. Kiss, B. Meier, G. Micera, L.Pettersson, M. Rangel, A. Salifoglou, I. Turel and D. Wang, JBIC, J.Biol. Inorg. Chem., 2002, 7, 384–396.
2 (a) A. M. Evangelou, Crit. Rev. Oncol. Hematol., 2002, 42, 249–265;(b) A. Y. Louie and T. J. Meade, Chem. Rev., 1999, 99, 2711–2734; (c) T.
Chakraborty, A. Chartterjee, M. G. Saralaya and M. Chartterjee, J.Biol. Inorg. Chem., 2006, 11, 855–866; (d) A. Morinville, D. Maysingerand A. Shaver, TIPS, 1998, 19, 452–460; (e) A. Papaioannou, M.Manos, S. Karkabounas, R. Liasko, V. Kalfakakou, I. Correia, A.Enangelou, J. Costa Pessoa and T. Kabanos, J. Inorg. Biochem., 2004,98, 959–968; (f) R. K. Narla, Y. Dong, D. Klis and F. M. Uckun, Clin.Can. Res., 2001, 1094–1101.
3 (a) M. R. Maurya, S. Khurana, Shailendra, A. Azam, W. Zhang and D.Rehder, Eur. J. Inorg. Chem., 2003, 1966–1973; (b) M. R. Maurya, A.Kumar, A. R. Bhat, A. Azam, C. Bader and D. Rehder, Inorg. Chem.,2006, 45, 1260–1269; (c) M. R. Maurya, A. Kumar, M. Abid and A.Azam, Inorg. Chim. Acta, 2006, 359, 2439–2447; (d) M. R. Maurya, S.Agarwal, M. Abid, A. Azam, C. Bader, M. Ebel and D. Rehder, DaltonTrans., 2006, 937–947; (e) M. R. Maurya, A. A. Khan, A. Azam, S.Ranjan, N. Mondal, A. Kumar and J. Costa Pessoa, Eur. J. Inorg.Chem., 2009, 5377–5390; (f) M. R. Maurya, A. A. Khan, A. Azam, S.Ranjan, N. Mondal, A. Kumar, F. Avecilla and J. Costa Pessoa, DaltonTrans., 2010, 39, 1345–1360.
4 J. Benıtez, L. Guggeri, I. Tomaz, G. Arrambide, M. Navarro, J. CostaPessoa, B. Garat and D. Gambino, J. Inorg. Biochem., 2009, 103, 609–616.
5 (a) A. Messerschmidt and R. Wever, Proc. Natl. Acad. Sci. U. S. A.,1996, 93, 392–396; (b) M. Weyand, H. J. Hecht, M. Kiesz, M. F. Liaud,H. Vilter and D. Schomburg, J. Mol. Biol., 1999, 293, 595–611; (c) M. I.Isupov, A. R. Dalby, A. Brindley, Y. Izumi, T. Tanabe, G. N. Murshudovand J. A. Littlechild, J. Mol. Biol., 2000, 299, 1035–1049.
6 (a) C. R. Cornman, J. Kampf, M. S. Lah and V. L Pecoraro, Inorg.Chem., 1992, 31, 2035–2043; (b) C. R. Cornman, G. J. Colpas, J. D.Hoeschele, J. Kampf and V. L. Pecoraro, J. Am. Chem. Soc., 1992,114, 9925–9933; (c) C. R. Cornman, J. Kampf and V. L. Pecoraro,Inorg. Chem., 1992, 31, 1981–1983; (d) V. Vergopoulos, W. Priebsch,M. Fritzsche and D. Rehder, Inorg. Chem., 1993, 32, 1844–1849; (e) L.J. Calviou, J. M. Arber, D. Collison, C. D. Garner and W. Clegg, J.Chem. Soc., Chem. Commun., 1992, 654–656; (f) A. D. Keramidas, S.M. Miller, O. P. Anderson and D. C. Crans, J. Am. Chem. Soc., 1997,119, 8901–8915; (g) S. K. Dutta, S. B. Kumar, S. Bhattacharyya, E.R. T. Tiekink and M. Chaudhury, Inorg. Chem., 1997, 36, 4954–4960;(h) S. K. Dutta, S. Samanta, S. B. Kumar, O. H. Han, P. Burckel,A. A. Pinkerton and M. Chaudhury, Inorg. Chem., 1999, 38, 1982–1988; (i) S. K. Dutta, S. Samanta, S. Mukhopadhyay, P. Burckel, A. A.Pinkerton and M. Chaudhury, Inorg. Chem., 2002, 41, 2946–2952; (j) S.Samanta, D. Ghosh, S. Mukhopadhyay, A. Endo, T. J. R. Weakley andM. Chaudhury, Inorg. Chem., 2003, 42, 1508–1517.
7 D. C. Crans, A. D. Keramidas, S. S. Amin, O. P. Anderson and S. M.Miller, J. Chem. Soc., Dalton Trans., 1997, 2799–2812.
8 M. R. Maurya, A. Kumar, M. Ebel and D. Rehder, Inorg. Chem., 2006,45, 5924–5937.
9 M. R. Maurya and J. Costa Pessoa, J. Organomet. Chem., 2011, 696,244–254.
10 M. R. Maurya, A. K. Chandrakar and S. Chand, J. Mol. Catal. A:Chem., 2007, 263, 227–237.
11 (a) I. Correia, J. Costa Pessoa, M. Teresa Duarte, R. T Henriques, M.F. M. Piedade, L. F. Veiros, T. Jakusch, T. Kiss, A. Dornyei, M. M. C.A. Castro, C. F. G. C. Geraldes and F. Avecilla, Chem.–Eur. J., 2004,10, 2301–2317; (b) I. Correia, A. Dorney, F. Avecilla, T. Kiss and J.Costa Pessoa, Eur. J. Inorg. Chem., 2006, 656–662.
12 R. A. Rowe and M. M. Jones, Inorg. Synth., 1957, 5, 113–116.13 D. C. Crans, A. D. Keramidas, S. S. Amin, O. P. Anderson and S. S.
Miller, J. Chem. Soc., Dalton Trans., 1997, 2799–2812.14 R. L. Dutta and A. Syamal, Elements of Magnetochemistry, Affiliated
East-West Press, New Delhi, 1993, 2nd edn, p. 8.15 A. Rockenbauer and L. Korecz, Appl. Magn. Reson., 1996, 10, 29–43.16 G. M. Sheldrick, SHELXL-97: An Integrated System for Solving
and Refining Crystal Structures from Diffraction Data (Revision 5.1),University of Gottingen, Germany, 1997.
17 (a) A. D. Becke, J. Chem. Phys., 1993, 98, 5648–5652; (b) J. P. Perdew,Phys. Rev., 1986, B33, 8822–8824.
18 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb,J. R. Cheeseman, J. A. Montgomery, Jr., T. Vreven, K. N. Kudin, J. C.Burant, J. M. Millam, S. S. Iyengar, J. Tomasi, V. Barone, B. Mennucci,M. Cossi, G. Scalmani, N. Rega, G. A. Petersson, H. Nakatsuji, M.Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T.Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J. E.Knox, H. P. Hratchian, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo,R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi,
C. Pomelli, J. W. Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth,P. Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D.Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K.Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S.Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz,I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y.Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W.Chen, M. W. Wong, C. Gonzalez and J. A. Pople, Gaussian 03, revisionD.01, Gaussian, Inc., Wallingford CT, 2004.
19 M. R. Maurya, A. Arya, A. Kumar, M. L. Kuznetsov, F. Avecilla andJ. Costa Pessoa, Inorg. Chem., 2010, 49, 6586–6600.
20 M. Dolg, U. Wedig, H. Stoll and H. Preuss, J. Chem. Phys., 1987, 86,866–872.
21 J. Tomasi and M. Persico, Chem. Rev., 1994, 94, 2027–2094.22 V. Barone and M. Cossi, J. Phys. Chem. A, 1998, 102, 1995–2001.23 D. H. Wertz, J. Am. Chem. Soc., 1980, 102, 5316–5322.24 J. Cooper and T. Ziegler, Inorg. Chem., 2002, 41, 6614–6622.25 (a) R. Ditchfield, Mol. Phys., 1974, 27, 789; (b) K. Wolinski, J. F. Hinton
and P. Pulay, J. Am. Chem. Soc., 1990, 112, 8251–8260.26 S. Gorelsky, G. Micera and E. Garribba, Chem. Eur. J., 2010, 16, 8167–
8180.27 K. Nakamoto, Infrared and Raman Spectra of Inorganic Compounds,
Part B: Applications in Coordination, Organometallic, and BioinorganicChemistry, 5th Ed., John Wiley & Sons, New York, 1997.
28 M. R. Maurya, A. Kumar and J. Costa Pessoa, Coord. Chem. Rev.,2011, DOI: 10.1016/j.ccr.2011.01.050.
29 (a) T. Joseph, D. Srinivas, C. S. Gopinath and S. B. Halligudi, Catal.Lett., 2002, 83, 209–214; (b) A. P. A. Marques, E. R. Dockal, F. C.Skrobot and I. L. V. Rosa, Inorg. Chem. Commun., 2007, 10, 255–261.
30 D. Rehder, Bioinorganic Vanadium Chemistry, John Wiley & Sons, NewYork, 2008.
31 K. Wuthrich, Helv. Chim. Acta, 1965, 48, 1012–1017.32 N. D. Chasteen, in, J. Reuben (Ed.), Biological Magnetic Resonance,
Plenum, New York, 1981, 53.33 T. S. Smith, II, C. A. Root, J. W. Kamp, P. G. Rasmussen and V. L.
Pecoraro, J. Am. Chem. Soc., 2000, 122, 767–775.34 E. Garribba, G. Micera and D. Sanna, Inorg. Chim. Acta, 2006, 359,
4470–4476.35 G. Martini, M. F. Ottaviani and G. L. Seravalli, J. Phys. Chem., 1975,
79, 1716–1720.36 (a) A. G. J. Ligtenbarg, R. Hage and B. L. Feringa, Coord. Chem. Rev.,
2003, 237, 89–101; (b) A. Butler, M. J. Clague and G. E. Meister, Chem.Rev., 1994, 94, 625–638.
37 B. J. Hamstra, G. J. Colpas and V. L. Pecoraro, Inorg. Chem., 1998, 37,949–955.
38 S. Smith II and V. L. Pecoraro, Inorg. Chem., 2002, 41, 6754–6760.39 K. Srinivasan, P. Michaud and J. K. Kochi, J. Am. Chem. Soc., 1986,
108, 2309–2320.40 R. Irie, Y. Ito and T. Katsuki, Synlett, 1991, 265–266.41 V. Hulea and E. Dumitriu, Appl. Catal., A, 2004, 277, 99–106.42 S. B. Kumar, S. P. Mirajkar, G. C. G. Pais, P. Kumar and R. Kumar, J.
Catal., 1995, 156, 163–166.43 M. R. Maurya, A. K. Chandrakar and S. Chand, J. Mol. Catal. A:
Chem., 2007, 274, 192–201.44 (a) G. B. Shul’pin, Y. N. Kozlov, G. V. Nizova, G. Suss-Fink, S. Stanislas,
A. Kitaygorodskiy and V. S. Kulikova, J. Chem. Soc., Perkin Trans. 2,2001, 1351–1371; (b) Y. N. Kozlov, V. B. Romakh, A. Kitaygorodskiy,P. Buglyo, G. Suss-Fink and G. B. Shul’pin, J. Phys. Chem. A, 2007,111, 7736–7752.
45 R. Z. Khaliullin, A. T. Bell and M. Head-Gordon, J. Phys. Chem. B,2005, 109, 17984–17992.
46 M. V. Kirillova, M. L. Kuznetsov, V. B. Romakh, L. S. Shul’pina, J. J.R. Frausto da Silva, A. J. L. Pombeiro and G. B. Shul’pin, J. Catal.,2009, 267, 140–157.
47 D. V. Deubel, G. Frenking, P. Gisdakis, W. A. Herrmann, N. Roschand J. Sundermeyer, Acc. Chem. Res., 2004, 37, 645–652.
48 M. L. Kuznetsov and J. Costa Pessoa, Dalton Trans., 2009, 5460–5468.49 M. L. Kuznetsov and A. J. L. Pombeiro, Inorg. Chem., 2009, 48, 307–
318.50 The energy of TIIa is higher than the energy of TS1 due to the
formation, on the final step of the reaction, of a molecular Van DerWaals complex with the total energy lower than that of TIIa. ComplexTIIa then is formed upon decomposition of the van der Waals complex.
51 G. B. Shul’pin, Y. Ishii, S. Sakaguchi and T. Iwahama, Russ. Chem.Bull., 1999, 48, 887–890.
