Dioxotungsten(VI) complexes with isoniazid-related hydrazones as (pre)catalysts for olefin epoxidation: solvent and ligand substituent effects

Abstract

The mononuclear dioxotungsten(VI) complexes [WO₂(L^3OMe)(D)] (1a and 1b), [WO₂(L^4OMe)(D)] (2a and 2b) and [WO₂(L^H)(D)] (3a and 3b) (D = EtOH (1a–3a) or MeOH (1b–3b); L^3OMe = 3-methoxy-2-oxybenzaldehyde isonicotinoyl hydrazonato, L^4OMe = 4-methoxy-2-oxybenzaldehyde isonicotinoyl hydrazonato, L^H = 2-oxybenzaldehyde isonicotinoyl hydrazonato) were synthesized by the reaction of [WO₂(acac)₂]·0.5C₆H₅Me with the respective isoniazid-related hydrazone. The compounds were characterized by microanalysis, FT-IR and NMR spectroscopy, thermogravimetric analysis, and powder X-ray diffraction method. The crystal and molecular structures of 1a, 1b, 3a and [WO₂(acac)₂]·0.5C₆H₅Me were determined by single crystal X-ray diffraction. The structures of 1a, 1b, 3a are mononuclear and form hydrogen bonded centrosymmetric dimers. In all three complexes, the dimers are also held together by π⋯π interactions between aromatic rings. The catalytic performances (activity and selectivity) of 1a–3a and 1b–3b towards alkene epoxidation by tert-butyl hydroperoxide (TBHP) were investigated under different conditions.

---

Introduction

The chemistry of hydrazones receives ongoing attention in various fields. Thus, isoniazid-related aroylhydrazones and their coordination compounds are of particular interest because of their biological activities¹ and their structural diversity.² Although various metallosupramolecular assemblies² as well as mononuclear³ dioxomolybdenum(VI) complexes with isoniazid-based aroylhydrazone derivatives are known, related dioxotungsten(VI) complexes are rare. To the best of our knowledge, only the dioxotungsten(VI) compounds [WO₂(NIH)]_n and [WO₂(VIH)]₄ (where NIH²⁻ = 2-oxy-1-naphthaldehyde isonicotinoyl hydrazonato and VIH²⁻ = 3-methoxy-2-oxybenzaldehyde isonicotinoyl hydrazonato) were structurally characterized by the powder X-ray diffraction method (PXRD).^2a

Compounds with active metal centres play an important role as functional materials in diverse fields.⁴ In this context, metal complexes with coordinatively unsaturated metal ions as well as those ligated by weakly coordinating ligands are of particular interest. In addition to having other functions, they also act as catalysts for the epoxidation of alkenes, an important process of wide chemical and biochemical significance. A variety of complexes has been used in this way and those containing the {MO₂}²⁺ unit are known as highly selective catalysts.^5–7

The catalytic conversion of olefins to epoxides may involve oxidants such as tert-butylhydroperoxide (TBHP),^3b,5a–5c,7 O₂,⁸ or H₂O₂,^5d,5e usually in organic solvents. In accordance with the requirements of “green” chemistry, these reactions have also been performed under organic solvent-free conditions. However, contrary to dioxomolybdenum(VI),^{3b,5a,5c,6,7b–7d} investigations of solvent-free epoxidations involving dioxotungsten(VI) complexes are quite scarce. Examples are [WO₂(L′)(solv)] (solv = alcohol, L′ = Schiff base)^3b,9 [WO₂L] (L = aminobisphenolato ligand)¹⁰ and [WO₂(salen)] (salen = bis(salicylidenediamine)) complexes.¹¹ We have previously compared the catalytic activity of dioxomolybdenum(VI) and dioxotungsten(VI) complexes with ligands derived from 4-hydroxy-benzhydrazone.¹² It was shown that the Mo^VI complexes are more efficient catalysts for cyclooctene epoxidation by tert-butyl hydroperoxide (TBHP) under organic solvent-free conditions.

Considering these features, we have decided to investigate the influence of the coordinated ligands (solvent (D) and isoniazid-related hydrazones (L^R)) on the solid-state structure and catalytic activity of the [WO₂(L^R)(D)] complexes. To achieve this aim, the investigation included a systematic variation of: (i) the tridentate ligand (2-hydroxybenzaldehyde isonicotinoyl hydrazone, H₂L^H, and two methoxy-substituted derivatives, i.e. 3-methoxy-2-hydroxybenzaldehyde isonicotinoyl hydrazone, H₂L^3OMe, and 4-methoxy-2-hydroxybenzaldehyde isonicotinoyl hydrazone, H₂L^4OMe, see Scheme 1), and (ii) the solvent D (MeOH and EtOH). The mononuclear W^VI complexes [WO₂(L^R)(EtOH)] (1a–3a) and [WO₂(L^R)(MeOH)] (1b–3b), where R = 3OMe, 4OMe or H, have been investigated as (pre)catalysts for the epoxidation of cyclooctene in the presence of organic solvent as well as in its absence, using TBHP (in decane or in water) as an oxidant.

Scheme 1 The isoniazid-based hydrazones H₂L^R (H₂L^3OMe, H₂L^4OMe and H₂L^H). The substituent is shown in red colour.

Results and discussion

Synthesis of bis(acetylacetonato)dioxotungsten(VI) toluene solvate (2/1), [WO₂(acac)₂]·0.5C₆H₅Me

The synthesis of the starting tungsten precursor [WO₂(acac)₂]·0.5C₆H₅Me was performed from freshly prepared WO₂Cl₂ and an excess of dry acetylacetone in toluene. Crystals suitable for X-ray structure analysis were obtained directly from the reaction mixture. The product can be recrystallized from a toluene/acetylacetone mixture. The previously reported pseudopolymorph was identified as [WO₂(acac)₂].¹³ Surprisingly, the crystal and molecular structure of this compound has not been reported previously.

Synthesis of dioxotungsten(VI) complexes chelated by ONO tridentate ligands

Compound [WO₂(acac)₂]·0.5C₆H₅Me was used as a precursor for the synthesis of dioxotungsten(VI) complexes of the type [WO₂(L^R)(D)], Scheme 2, with isoniazid-related hydrazone ligands. The products were obtained by the ligand exchange strategy using the corresponding aroylhydrazone ligand in ethanol (1a–3a) or methanol (1b–3b). The reactions were performed under a dry argon atmosphere.

Scheme 2 Mononuclear [WO₂(L^R)(D)] complexes (D = MeOH or EtOH). The R′ substituent is shown in red colour in Scheme 1.

Synthesis in the presence of a small amount of DMF proved to be effective for producing products in crystalline form. Good quality single crystals of 1a, 1b and 3a suitable for X-ray diffraction analysis were obtained in this way. Otherwise, in the absence of DMF, yellow powdered solids were obtained.

Thermal analysis

The thermal stability of the complexes was investigated under an oxygen atmosphere. The mass loss corresponding to the first step in the TG curves of 1a–3a, 1b–3b was related to the loss of the coordinated ethanol or methanol molecule. This process occurred in the range 135–186 °C (1a), 159–210 °C (1b), 141–197 °C (2a), 154–193 °C (2b), 190–234 °C (3a) and 175–237 °C (2b). These temperatures are generally much higher than those required for alcohol removal from similar dioxomolybdenum(VI) complexes, for instance 110–120 °C for EtOH loss from a salicylidene-2-aminophenolato derivative,¹⁴ indicating that the alcohol donor forms a stronger bond with W relative to Mo. The same trend was observed in a previous report.¹² All remaining unsolvated products started to decompose at 298 °C for 1a and 1b, 308 °C for 2a and 2b, and 295 °C for 3a and 3b. The final residue was identified as WO₃.

The first step in the thermogravimetric curve of [WO₂(acac)₂]·0.5C₆H₅Me was related to the loss of toluene (65–104 °C) and was followed, upon further heating, by significant weight loss in the range 128–449 °C due to ligand decomposition.

Crystal and molecular structure of [WO₂(acac)₂]·0.5C₆H₅Me

In [WO₂(acac)₂]·0.5C₆H₅Me the tungsten atom exhibits the distorted octahedral coordination sphere. It is coordinated by two cis oxo oxygen atoms and by four oxygen atoms from the two acetylacetonate ligands (Fig. 1). The asymmetric unit consists of the complex molecule and one half of the toluene molecule of crystallization (Fig. S1, see ESI†).

Fig. 1 ORTEP drawing of [WO₂(acac)₂]·0.5C₆H₅Me with the atom-labelling scheme (displacement ellipsoids of non-hydrogen atoms are drawn at the 50% probability level). Toluene molecule is omitted for clarity.

The Cambridge Structural Database¹⁵ was searched for structures with the XO₂ unit (X was any element with the coordination number of 6: two oxygen atoms from two dioxo groups and four oxygen atoms from two acac ligands, including substituted acac ligands). The search gave only 9 hits (6 Mo complexes, 2 U compounds and a W complex). Since the data for the W complex (having 1,3-phenyl substituted acacetylacetonate) include only the unit cell and lack the atomic coordinates, a comparison of bonds and angles with the present structure is not possible. The bond distances and angles around the W atom and within the chelate rings are given in Table S1, see ESI.† There are no classical hydrogen bonds in the crystal structure, only a weak C–H⋯O interaction between the toluene molecule and an oxo oxygen atom (d(C15⋯O5[−1/2−x, 1/2 + y, 1/2−z]) = 3.337(12) Å).

Crystal and molecular structure of [WO₂(L^R)(D)]

In all compounds, the tungsten atoms exhibit the distorted octahedral coordination sphere. The ligand is coordinated in the dianionic form to the cis-{WO₂}²⁺ core via the ONO donor atoms. The remaining sixth coordination site is occupied by the oxygen atom from the solvent molecule (ethanol in 1a and 3a, methanol in 1b). The ORTEP plots of the molecular structures of 1a, 1b and 3a are given in the ESI (Fig. S2, see ESI†).

The distance from the W atom to the O donor atom of the solvent molecule in 1a, 1b and 3a represents the longest bond within the distorted octahedron (Table 1). The C7–N1 bond of the L^3OMe ligand in complexes 1a and 1b is 1.275(4) and 1.284(5) Å, respectively and is not significantly different than in the free ligands (1.2791(18) Å in H₂L^3OMe and 1.276(2) Å in H₂L^3OMe·H₂O).¹⁶ The N2–C1 bond length is shortened in the complexes 1a and 1b (1.290(5) Å in 1a and 1.270(5) Å in 1b) in comparison to the free ligands (1.3552(18) Å and 1.339(2) Å in H₂L^3OMe and H₂L^3OMe·H₂O, respectively), while the N1–N2 bond is lengthened (1.401(4) Å in 1a and 1.395(3) Å in 1b) in comparison to the free ligands (1.3791(19) Å and 1.3724(15) Å in H₂L^3OMe and H₂L^3OMe·H₂O, respectively), due to electron delocalization.

Table 1 Selected bond lengths (Å) and angles (°) for compounds 1a, 1b and 3a

	1a	1b	3a
W–O1	2.001(2)	2.000(3)	1.997(2)
W–O2	1.922(2)	1.927(3)	1.927(3)
W–O3	1.706(2)	1.704(3)	1.704(2)
W–O4	1.718(2)	1.721(3)	1.719(2)
W–O5	2.292(2)	2.327(3)	2.306(3)
W–N1	2.249(2)	2.217(3)	2.248(3)
C7–N1	1.275(4)	1.284(5)	1.282(4)
N1–N2	1.401(4)	1.395(4)	1.399(4)
N2–C1	1.290(5)	1.270(5)	1.291(4)
O1–W–O2	149.17(9)	150.99(11)	148.58(9)
O1–W–O3	97.60(10)	96.44(14)	97.96(11)
O1–W–O4	97.71(10)	98.32(13)	98.93(11)
O1–W–O5	78.35(9)	80.63(11)	79.48(9)
O1–W–N1	71.81(9)	72.09(11)	71.91(9)
O2–W–O3	98.56(10)	98.56(14)	99.16(11)
O2–W–O4	103.40(10)	101.59(13)	102.06(11)
O2–W–O5	80.73(9)	80.73(12)	78.98(10)
O2–W–N1	81.42(9)	81.72(11)	81.17(9)
O3–W–O4	104.48(11)	104.90(14)	104.37(12)
O3–W–O5	168.71(10)	171.06(12)	169.69(11)
O3–W–N1	92.41(10)	96.10(12)	92.58(10)
O4–W–O5	86.59(9)	83.92(12)	85.92(10)
O4–W–N1	161.35(10)	157.88(13)	161.87(10)
O5–W–N1	76.32(8)	74.97(11)	77.13(9)

---

The C7–N1 bond length in 3a is 1.282(4) Å and is not significantly different than in the free H₂L^H ligand, 1.2756(16) Å.¹⁷ The N2–C1 bond is shortened in 3a (1.291(4) Å) in comparison to the free H₂L^H ligand (1.3558(17) Å), whereas the N1–N2 bond is lengthened in the complexes (1.399(4) Å) in comparison to the free H₂L^H ligand (1.3699(15) Å) due to the electron delocalization.

In 1a, 1b and 3a the ligand is not planar with the largest deviation from planarity being that of the phenyl and the pyridyl moieties in 3a (φ = 8.94(12)°, Table S2, see ESI†). The largest dihedral angle between the planes of the five- and six-membered chelate rings is also observed in 3a (ψ = 16.87(16)°, Table S2†).

A common feature of 1a, 1b and 3a is the association of the molecules by hydrogen bonds (O5–H50⋯N3) into centrosymmetric dimers, graph set R₂²(18), Fig. 2. The shortest hydrogen bond distance is in complex 1b, d(O5⋯N3[−x, −y, 1−z]) = 2.692(4) Å (Table S3, see ESI†).

Fig. 2 Discrete dimer formed through intermolecular O5–H50⋯N3[−x, −y, 1−z] hydrogen bonds in 1b. Hydrogen bonds are shown by dotted lines.

In all three complexes (1a, 1b and 3a), the dimers are also held together by π⋯π interactions between the two pyridyl rings. Neighboring dimer molecules are further connected by π⋯π interactions between: the five-membered chelate rings and phenyl rings in 1a, between the five-membered chelate rings and pyridyl rings in 1b, and between the two phenyl rings in 3a (Fig. 3 and Table S4, see ESI†).

Fig. 3 Partial structural motif of endless chains in 1a, 1b and 3a. Hydrogen-bonds are shown by blue dotted lines. Green dashed lines indicate π⋯π interactions (values are given in Å). The centroids of the five-membered chelate rings are shown as yellow spheres, the centroids of pyridyl rings as green spheres and the centroids of the phenyl rings as red spheres.

Complex 1b is isostructural with the dioxomolybdenum(VI) complex [MoO₂(L^3OMe)(MeOH)] (CSD code KANYIJ¹⁸ and KANYIJ01).¹⁹ The powder X-ray diffraction patterns of 2a and 2b indicate that they are isostructural (Fig. S3†). The dioxotungsten(VI) complex 3a is isostructural with the corresponding dioxomolybdenum(VI) complexes, [MoO₂(L^H)(EtOH)] (CSD code UVAQOY)^2b and [MoO₂(L^H)(MeOH)] (CSD code PAGCEF,²⁰ PAGCEF01).²¹ PXRD of 3a and 3b indicate that they are also isostructural (Fig. S3†).

IR spectroscopy

The IR spectroscopic analysis confirms the doubly deprotonated state of the ONO ligands for all [WO₂(L^R)(D)] complexes (Scheme 2). The bands found in the IR spectra of H₂L^R, characteristic for the N–H and C=O groups (at ca. 3150 cm⁻¹ and 1645 cm⁻¹, respectively) are absent in the IR spectra of the tungsten complexes, suggesting tautomerism (C=N–NH–(C=O)– → = N–NC=(C–OH)–), deprotonation and coordination through the oxygen atom. This is additionally supported by the presence of a new band at ca. 1315 cm⁻¹ assigned to the stretching vibrations of the C–O bond.²² Two stretching vibration bands belonging to the C=N_imine and C–O_phenolic bonds appeared in the spectra of the complexes at ca. 1600 cm⁻¹ and 1340 cm⁻¹.²² They indicate that coordination of the ligand L²⁻ to the cis-{WO₂}²⁺ core occurs through the ONO donor atoms (Scheme 2). A band around 1050 cm⁻¹, seen in the IR spectra of 1a–3a and 1b–3b, is assigned to the C–O stretching vibration of the coordinated alcohol molecule.²³

The presence of the {WO₂}²⁺ core in 1a–3a and 1b–3b was identified by the appearance of the strong absorption bands found in the range 959–949 cm⁻¹ and 920–896 cm⁻¹, characteristic for ν_asym(WO₂) and for the antisymmetric combination of W=O and W–O_EtOH (or W–O_MeOH) stretchings, respectively.²⁴ The corresponding symmetric stretching bands ν_sym(WO₂) having significantly lower intensities appeared in the same region but they either overlap with the asymmetric ones or appear as shoulders.

UV-Vis spectroscopy

The electronic absorption spectrum of the ligands exhibit several absorption bands with distinct maxima at 310 and 225 nm (H₂L^3OMe), at 338 and 219 nm (H₂L^4OMe), and at 294 and 325 nm (H₂L^H). These transitions are assigned to the intra-ligand charge transfer transitions.^25,26 Electronic spectra of the dioxotungsten(VI) complexes 1a–3a and 1b–3b recorded in methanol exhibited N(pπ)–W(dπ) LMCT and O(pπ)–W(dπ) LMCT bands in the 410–404 nm and 337–309 nm regions.²⁷ Each complex displays additional bands in the higher energy region characteristic for intraligand transitions.

NMR spectroscopy

The structures of all compounds in solution have been confirmed by one- and two-dimensional NMR techniques. The proton and carbon resonances were unambiguously assigned by using ¹H, APT, COSY, HSQC and HMBC experiments and are displayed in the ESI (Tables S5 and S6, Scheme S1, see ESI†). In the ¹H NMR spectra of the ligand molecules the most deshielded protons were those belonging to the NH and OH groups resonating at around 12 and 11 ppm, respectively (Fig. S4–S6, see ESI†). The signals were somewhat broadened indicating their involvement in hydrogen bonding interactions. In the complexes 1a, 2a and 3a those signals disappeared reflecting their deprotonation upon coordination of the ligands to tungsten.

Deprotonation was accompanied by shielding and deshielding of the neighbouring hydrogen and carbon atoms (see Tables S1 and S2†), as a result of the electron redistribution upon complexation. Most of the neighbouring carbon atoms were deshielded, i.e. they underwent upfield shifts upon complexation up to 10.34 ppm as observed for carbon C-1 in 1a. The atoms C-4, C-12, C-14 and C-15 were also deshielded but to a smaller extent, while C-5 experienced a shielding effect up to 3.15 ppm as observed in 1a. These findings are in accordance with the crystal structures obtained for these compounds.

Catalytic studies

The catalytic reactions were performed under different experimental conditions in order to better understand the role of solvent that delivers TBHP, water (procedure A) and decane (procedure C) as well as the role of added organic solvent (herein MeCN) with aqueous TBHP (procedure B) (Scheme 3).

Scheme 3 The different conditions used for the epoxidation.

As seen previously,^6a TBHP in water without catalyst does react poorly with cyclooctene under organic solvent free conditions, as well as in the presence of organic solvent.²⁸

Organic solvent-free epoxidation (catalytic procedure A)

All prepared mononuclear tungsten complexes [WO₂(L^R)(D)] were employed as (pre)catalysts using aqueous TBHP as oxidant, no additional organic solvent, and a 0.25 mol% catalyst charge. After the TBHP addition at 80 °C, the tungsten complexes remained partially undissolved at the beginning of the reaction, but dissolved completely as the reactions progressed. Under these conditions, it is known that TBHP is mostly transferred to the organic phase.^6b The substrate conversion and the desired epoxide formation were monitored by analyzing only the organic layer. The analyses of the aqueous layer were not carried out because the desired epoxide is confined in the organic layer and only diol by-product may be partially lost in the aqueous layer. The obtained results are compiled in Table 2 and the kinetic profiles are presented in Fig. 4. The complexes with coordinated ethanol (1a–3a) are slightly more active than those with methanol (1b–3b). The profiles show similar evolution, with good activity in the first 20 minutes followed by a slowdown. The two (pre)catalysts 3a and 2a, however, seemed to regain activity after 150 min.

Table 2 Relevant catalytic results for the cyclooctene epoxidation by aqueous TBHP^a

Cat.	Conversion^b/%	Selectivity^c/%	TOF20 min^d/h^−1	TON^e
a Reaction conditions: time, 5 h; temperature, 80 °C; [W]/cyclooctene/TBHP molar ratio: 0.25/100/200.b Calculated after 5 h.c Formed epoxide per converted olefin after 5 h.d n(cyclooctene transformed)/n(catalyst)/time at 20 minutes.e n(cyclooctene transformed)/n(catalyst at 5 h).
1a	20	3	73	63
1b	17	12	171	70
2a	28	33	81	117
2b	23	8	111	93
3a	31	11	204	129
3b	22	6	168	90

---

Fig. 4 Kinetic monitoring of cis-cyclooctene epoxidation with TBHP–H₂O in the presence of tungsten complexes: 1a – black circle, 1b – white circle, 2a – black triangle, 2b – white triangle, 3a – black square, 3b – white square. Experimental conditions are in Table 1.

The cyclooctene conversion for all tested compounds after 5 h is between 17 and 31%, following the order 3a > 2a > 2b ≈ 3b > 1a > 1b. The cycloctene oxide selectivity is rather poor, varying from 8 to 33%, certainly implying that the organic solvent-free conditions and water from TBHP promote the cyclooctene epoxide ring opening into the corresponding cyclooctanediol.⁶ The highest selectivity was observed for 2a and the lowest one for 1a. The OMe substituent position has a marked influence on the reaction selectivity. Poor results in terms of selectivity and conversion could also be caused by the inhibition of the tungsten catalyst in water, as stated elsewhere with other tungsten-containing complexes.^29–31 The highest turnover frequency (TOF) is observed for 3a.

We previously reported results of epoxidation catalysis with [WO₂(L^R*)(EtOH)] complexes (where L^R* are the corresponding 4-hydroxy-benzhydrazide-related ligands).¹² When comparing with the isoniazide performance, it can be concluded that the ligand hydrazone moiety does not influence the activity whereas the aldehyde moiety does, the 4OMe derivative always giving higher conversion than the corresponding 3OMe derivative (Tables S7 and Scheme S2, see ESI†). This could be related to a steric impediment of the 3OMe substituent toward the approach and coordination of TBHP to the metal centre. As discussed earlier for the catalysts with molybdenum complexes having a similar coordination sphere, the catalytic cycle probably starts from the pentacoordinated mononuclear compound [WO₂(L^R)]. The sixth coordination site is available for the TBHP coordination and activation through the assistance of an O–H⋯O hydrogen bond.^6b

Epoxidation of cyclooctene by aqueous TBHP in MeCN (catalytic procedure B)

Acetonitrile is a commonly used solvent for metal-catalyzed epoxidation.³⁰ Addition of MeCN to the reaction mixture positively affected the 1–3 catalyst activity, see Table 3. In all cases, significant improvement of cyclooctene conversion was observed (35–63%) after only 20 min of the reaction progress. Like under organic solvent-free conditions, the ethanol-stabilized complexes were more active than those containing methanol. The higher conversion for the complexes containing (L^4OMe)²⁻ and (L^3OMe)²⁻ ligands than for those having the (L^H)²⁻ ligand is more pronounced in MeCN. After 20 min, the conversion reached again a plateau, implying possible catalyst deactivation. All catalysts were poorly selective towards the corresponding epoxide, which clearly indicates that the addition of organic solvent does not promote epoxide formation. The acetonitrile effect seems complex, since many parameters affect the catalyst action: polarity, reactant and product solubility, diffusion effects, solvent coordination to the metal center, side reactions, etc.^32,33 It has been suggested that MeCN addition to alcohol-stabilized mononuclear tungsten complexes leads to oligomerization.¹² Certain polynuclear complexes were found more active than the corresponding alcohol-stabilized mononuclear ones,^5a,5c possibly explaining the better conversion in presence of MeCN with the tested catalysts.

Table 3 Comparison of the catalytic parameters (conversion and selectivity) for the cyclooctene epoxidation by TBHP (procedures A, B and C)^a

a Reaction conditions: time, 5 h; temperature, 80 °C; [W]/cyclooctene/TBHP molar ratio: 0.25/100/200 for all compounds.
Procedure	A	B	C
Solvent	—	MeCN	—
TBHP in	Water	Water	Decane

---

Catalyst	Conv. (selec.) %	Conv. (selec.) %	Conv. (selec.) %
1a	20 (3)	66 (1)	20 (31)
1b	17 (12)	51 (2)	22 (12)
2a	28 (33)	78 (3)	20 (27)
2b	23 (8)	63 (2)	25 (19)
3a	31 (11)	47 (2)	29 (31)
3b	22 (6)	35 (1)	23 (13)

---

Epoxidation of cyclooctene by TBHP/decane under organic solvent-free conditions (catalytic procedure C)

The use of TBHP in decane did not result in any dramatic change of the conversion parameters compared to procedure A, yielding very similar results for all catalysts. This contrasts with the positive contribution of decane observed with other complexes.³⁰ The solvent that delivers TBHP to the cyclooctene solution (water or decane) has no influence on the conversion rate. On the other hand, the epoxide selectivity was the same or better than using aqueous TBHP. This has certainly to be linked to the protective hydrophobic effect of decane towards the epoxide opening with water.³⁴

One blank reaction was performed using TBHP in decane and no solvent. After 5 h cyclooctene conversion of 17% was observed with selectivity of 49% towards epoxide. The catalysts have a very low conversion when using 0.25 mol% catalyst charge. However, when adding the W complexes, the lower selectivity showed that another transformation occurred in addition to the epoxidation, but the nature of the other compounds was not determined.

Conclusions

Replacement of the acetylacetonate ligands in [WO₂(acac)₂] by the corresponding aroylhydrazone ligand H₂L^R in ethanol or methanol results in mononuclear complexes [WO₂(L^R)(EtOH)] (1a–3a) and [WO₂(L^R)(MeOH)] (1b–3b), respectively. Reactions in the presence of a small amount of DMF are effective in producing crystals suitable for single crystal X-ray diffraction experiment.

The association of the molecules by hydrogen bonds into centrosymmetric dimers is a common feature of 1a, 1b and 3a. Furthermore, the dimers are held together by π⋯π interactions involving the pyridyl rings.

All the dioxotungsten(VI) complexes led to greater cyclooctene conversion by oxidation with TBHP in an acetonitrile solution than in an organic solvent-free process but with very poor selectivity towards the desired epoxide. Under organic solvent-free conditions, use of TBHP in water or decane has no impact to the cyclooctene conversion. Nevertheless, the use of TBHP in decane has a positive impact on the selectivity in comparison to the use of aqueous TBHP.

In general, the catalytic behaviour of the complexes depends on the aroylhydrazone ligand, more precisely on the presence and position of the substituent present in the salicylaldehyde moiety. Compounds [WO₂(L^4OMe)(D)] achieved higher conversions than the corresponding [WO₂(L^3OMe)(D)], indicating the importance of the ring substituent position. The complexes coordinated with ethanol showed higher conversions than those with methanol.

Experimental section

Preparative part

Ligands H₂L^3OMe, H₂L^4OMe and H₂L^H were prepared by Schiff base condensation according to the procedure described in the literature.³⁵ The tungsten complex WO₂Cl₂ was prepared according to literature procedure.³⁶ Toluene and acetylacetone were dried by refluxing over P₂O₅ and then distilled. Diethyl-ether was dried over sodium wire and distilled. Methanol and ethanol were dried using magnesium turnings and iodine and then distilled. For the catalytic investigations cis-cyclooctene (98% Aldrich), cyclooctene oxide (Aldrich), cis-1,2-cyclooctanediol (99% Aldrich), TBHP (70% in water, Aldrich), H₂O₂ (35% in water, Aldrich) and organic solvents (acetonitrile, diethyl ether from ACROS) were commercially available and used as received.

Synthesis of [WO₂(acac)₂]·0.5C₆H₅Me

A suspension of WO₂Cl₂ (2 g, 7 mmol) and 20 mL of acetylacetone in 120 mL of toluene was refluxed under a dry argon atmosphere for 9 h and left overnight. The obtained mixture was refluxed additionally for 2 hours and filtered in a glove box while still hot to remove insoluble impurities. The solvent was evaporated by vacuum distillation to approximately 5–10 mL and the resulting solution was then kept overnight at 10 °C to give a white solid. The obtained precipitate was filtered and washed with dry diethyl-ether. Yield: 1.9 g; 59%. Anal. calcd for C_13.5H₁₈O₆W (460.12): C, 35.2; H, 3.9. Found: C, 35.4; H, 3.8%. TG: calc. for C₆H₅CH₃ 10.01%, found 9.85% calc. for WO₃ 50.4%, found 50.35%. Selected IR data (cm⁻¹): 1558, 1515, 954, 937, 909 cm⁻¹.

Synthesis of dioxotungsten(VI) complexes with hydrazone ligands

A mixture of [WO₂(acac)₂]·0.5C₆H₅Me (0.22 mmol) and H₂L^R (0.22 mmol) and 20 μL DMF in dry methanol or ethanol (35 mL) was refluxed for five hours under a dry argon atmosphere. The obtained solution was cooled down to room temperature and then concentrated under vacuum to one third of its volume. The resulting solution was left at room temperature for a few days. The obtained yellow precipitate was filtered, rinsed with cold alcohol and dried in a desiccator up to the constant weight.

[WO₂(L^3OMe)(EtOH)] (1a). Yield: 0.05 g; 43%. Anal. calcd for C₁₆H₁₇N₃O₆W (531.17): C, 36.2; H, 3.2; N, 7.9. Found: C, 35.9; H, 3.2; N, 7.6%. TG: calc. for EtOH 8.7%, found 8.35% calc. for WO₃ 43.6%, found 43.9%. Selected IR data (cm⁻¹): 1622 (C=N)_py, 1600 (C=N), 1341 (C–O_phenolate), 1315 (C–O), 1194 (C–O_EtOH), 959, 948 (WO₂), 920, 910 (O=W–O_EtOH). UV-Vis (methanol): λ/nm (ε/dm³ mol⁻¹ cm⁻¹): 225 (42 500), 309 (27 350) and 410 (16 314).

[WO₂(L^4OMe)(EtOH)] (2a). Yield: 0.05 g, 43%. Calc. for C₁₆H₁₇N₃O₆W (531.14): C, 36.2; H, 3.2; N, 7.9. Found: C, 36.0; H, 3.0; N, 7.6%. TG: calc. for EtOH 8.7%, found 8.4%, calc. for WO₃ 43.65%, found 43.1%. IR data (cm⁻¹): 1616 (C=N)_py, 1600 (C=N), 1334 (C–O_phenolate), 1315 (C–O_enolate), 1199 (C–O_EtOH), 949, 943 (WO₂), 896 (O=W–O_EtOH). UV-Vis (methanol): λ/nm (ε/dm³ mol⁻¹ cm⁻¹): 217 (38 300), 337 (24 800) and 407 (6414).

[WO₂(L^H)(EtOH)] (3a). Yield: 0.06 g, 55%. Calc. for C₁₅H₁₅N₃O₅W (501.14): C, 35.95; H, 3.0; N, 8.4. Found: C, 35.65; H, 2.8; N, 8.2%. TG: calc. for EtOH 9.2%, found 8.9%, calc. for WO₃ 46.3%, found 45.95%. Selected IR data (cm⁻¹): 1621 (C=N)_py, 1600 (C=N), 1339 (C–O_phenolate), 1317 (C–O), 954 (WO₂), 913, 905 (O=W–O_EtOH). UV-Vis (methanol): λ/nm (ε/dm³ mol⁻¹ cm⁻¹): 216 (27 500), 294 (15 650), 333 (12 950) and 404 (1507).

[WO₂(L^3OMe)(MeOH)] (1b). Yield: 0.07 g; 62%. Anal. calcd for C₁₅H₁₇N₃O₆W (517.142): C, 34.8; H, 2.9; N, 8.1. Found: C, 35.9; H, 3.2; N, 7.6%. TG: calc. for MeOH 6.2%, found 6.2%, calc. for WO₃ 44.8%, found 43.45%. Selected IR data (cm⁻¹): 1621 (C=N)_py, 1600 (C=N), 1345 (C–O_phenolate), 1315 (C–O), 1194 (C–O_EtOH), 949 (WO₂), 920, 910 (O=W–O_EtOH). UV-Vis (methanol): λ/nm (ε/dm³ mol⁻¹ cm⁻¹): 225 (42 500), 309 (27 350) and 410 (16 314).

[WO₂(L^4OMe)(MeOH)] (2b). Yield: 0.06 g, 53%. Calc. for C₁₅H₁₅N₃O₆W (517.14): C, 34.8; H, 2.9; N, 8.1. Found: C, 38.5; H, 3.1; N, 7.9%. TG: calc. for MeOH 6.2%, found 6.4%, calc. for WO₃ 44.8%, found 44.4%. IR data (cm⁻¹): 1615 (C=N)_py, 1600 (C=N), 1334 (C–O_phenolate), 1314 (C–O), 1199 (C–O_EtOH), 949, 943 (WO₂), 896 (O=W–O_EtOH). UV-Vis (methanol): λ/nm (ε/dm³ mol⁻¹ cm⁻¹): 217 (38 300), 337 (24 800) and 407 (6414).

[WO₂(L^H)(MeOH)] (3b). Yield: 0.08 g, 75%. Calc. for C₁₄H₁₃N₃O₅W (487.1): C, 34.5; H, 2.7; N, 8.6. Found: C, 34.7; H, 2.8; N, 8.7%. TG: calc. for MeOH 6.6%, found 6.3%, calc. for WO₃ 47.6%, found 47.4%. Selected IR data (cm⁻¹): 1622 (C=N)_py, 1602 (C=N), 1340 (C–O_phenolate), 1317 (C–O), 955 (WO_{2_asym}), 914, 906 (WO_{2_sym}). UV-Vis (methanol): λ/nm (ε/dm³ mol⁻¹ cm⁻¹): 216 (27 500), 294 (15 650), 333 (12 950) and 404 (1507).

General procedure for the epoxidation of cyclooctene by aqueous TBHP

Procedure A. A mixture of cyclooctene (2.76 mL, 20 mmol), acetophenone (as an internal reference) and W (pre)catalyst 1a–3a and 1b–3b (0.05 mmol) was stirred and heated up to 80 °C before addition of aqueous TBHP (70% w/w, 5.48 mL, 40 mmol). The mixture is initially an emulsion, but two phases become clearly visible as the reaction progresses, a colourless aqueous one and a yellowish organic one. The reaction was monitored for 5 h with withdrawal and analysis of organic phase aliquots (0.1 mL) at required times. Each withdrawn sample was mixed with 2 mL of Et₂O, treated with a small quantity of MnO₂ and then filtered through silica and analyzed by GC.

Procedure B. The procedure was performed in the same manner as procedure A with addition of 2 mL of MeCN before the addition of aqueous TBHP.

Procedure C. The same as procedure A with using TBHP in decane (40 mmol, 6.5 mL) as oxidizing agent.

Physical methods. Elemental analyses were provided by the Analytical Services Laboratory of the Ruđer Bošković Institute, Zagreb. Thermogravimetric (TG) analysis was carried out with a Mettler-Toledo TGA/SDTA851e thermobalance using aluminum crucibles. All experiments were recorded in a dynamic atmosphere with a flow rate of 200 cm³ min⁻¹. Heating rates of 5 K min⁻¹ were used for all investigations. Fourier Transform Infrared spectra (FT-IR) were recorded in KBr pellets with a Perkin-Elmer 502 spectrophotometer. Spectra were recorded in the spectral range between 4500 and 450 cm⁻¹. Electronic absorption spectra were recorded at 25 °C on a Cary 100 UV-Vis Spectrophotometer.

All NMR spectra were recorded on Bruker Avance III HD 400 spectrometer operating at 400 MHz equipped with a broadband observed (BBO) Prodigy cryoprobe and z-gradient accessory. Compounds were dissolved in DMSO-d₆ and measured in 5 mm NMR tubes at 298 K with TMS as an internal standard. The sample concentration was 10 mg mL⁻¹.

Proton spectra with spectral width of 6002 Hz and a digital resolution of 0.36 Hz per point were measured with 64 and 128 scans. In the gCOSY experiment, 2048 points in the f2 dimension and 128 increments in the f1 dimension were used. For each increment, 16 scans and the spectral width of 6010 Hz were applied. Digital resolution was 5.87 and 93.78 Hz per point in f2 and f1 dimensions, respectively. The gHSQC and gHMBC spectra were acquired with 16 and 32 scans, respectively. Spectral width was 6009 Hz in f2 and 20 124 Hz in f1 dimension for both experiments. 1k data points were applied in the time domains and for each data set 256 increments were collected. The resulting digital resolution was 5.86 Hz per point in f2 dimension and 314.4 Hz per point in f1 dimension.

Catalytic reactions were followed by gas chromatography on an Agilent 6890A chromatograph equipped with FID detector, a HP5-MS capillary column (30 m × 0.25 mm × 0.25 μ) and automatic sampling, or on a Fisons GC 8000 chromatograph equipped with FID detector and with a SPB-5 capillary column (30 m × 0.32 mm × 0.25 μ). The GC parameters were quantified with authentic samples of the reactants and products. The conversion of cis-cyclooctene and the formation of cyclooctene oxide were calculated from calibration curves (r² = 0.999) relatively to an internal standard.

X-ray crystallography. Powder diffraction. The powder X-ray diffraction data were collected by the Panalytical X'Change powder diffractometer in the Bragg–Brentano geometry using CuK_α radiation. The sample was contained on a Si sample holder. Patterns were collected in the range of 2θ = 5–40° with the step size of 0.03° and at 1.5 s per step. The data were collected and visualized using the X'Pert programs Suite.³⁷

X-ray crystallography. Single crystal diffraction. The X-ray single-crystal structure data obtained for [WO₂(acac)₂]·0.5C₆H₅Me, 1a, 1b and 3a are tabulated in Table 4. The single crystal X-ray diffraction data were collected by ω-scans on an Oxford Diffraction Xcalibur 3 CCD diffractometer with graphite-monochromated Mo Kα radiation (λ = 0.71073 Å). Each single crystal was glued onto a glass fibber and the data were collected at room temperature. The data reduction was performed using the CrysAlis software package.³⁸ Solution, refinement and analysis of the structures were done using the programs integrated in the WinGX system.³⁹

Table 4 Crystallographic data for compounds [WO₂(C₅H₇O₂)₂]·0.5C₆H₅Me, 1a, 1b and 3a

	[WO2(acac)2]·0.5C6H5Me	1a	1b	3a
a R = Σ∣∣Fo∣ − ∣Fc∣∣/Σ∣Fo∣.b wR = [Σ(Fo^2 − Fc^2)^2/Σw(Fo^2)^2]^1/2.c S = Σ[w(Fo^2 − Fc^2)^2/(Nobs − Nparam)]^1/2
Empirical formula	C13.5H18O6W	C16H17N3O6W	C15H15N3O6W	C15H15N3O5W
Mr	460.12	531.17	517.14	501.14
Crystal colour, habit	Colourless, plate	Yellow, plate	Yellow, plate	Yellow, plate
Crystal size (mm^3)	0.56 × 0.36 × 0.18	0.56 × 0.38 × 0.18	0.48 × 0.34 × 0.18	0.56 × 0.20 × 0.12
Crystal system	Monoclinic	Triclinic	Monoclinic	Monoclinic
Space group	P21/n	P1̄	P21/n	P21/n
unit cell parameters
a (Å)	7.3569(4)	7.2563(4)	6.7820(3)	8.1744(2)
b (Å)	16.2269(7)	10.4696(5)	30.3344(16)	12.5034(2)
c (Å)	13.5020(11)	12.6154(5)	7.9111(5)	15.3736(3)
α (°)	90	113.702(4)	90	90
β (°)	95.619(6)	96.023(4)	93.335(5)	97.032(2)
γ (°)	90	97.909(4)	90	90
V (Å^3)	1604.12(17)	855.54(8)	1624.78(15)	1559.48(6)
Z	4	2	4	4
Dcalc (g cm^−3)	1.905	2.062	2.114	2.135
Temperature (K)	295	295	295	295
μ (mm^−1)	7.220	6.791	7.148	7.439
F(000)	884	512	992	960
Number of unique data	3639	5682	3835	3735
Number of data [Fo ≥ 4σ(Fo)]	2646	5228	3053	2805
Number of parameters	172	236	227	221
R1^a, [Fo ≥ 4σ(Fo)]	0.0327	0.0269	0.0251	0.0235
wR2^b	0.0841	0.0585	0.0579	0.0438
Goodness of fit on F^2, S^c	1.00	1.00	1.10	0.94
Min. and max. electron density (e Å^−3)	−2.47, 0.76	−1.35, 1.67	−1.80, 1.33	−0.61, 0.86

---

The structures were solved by using the SHELXS program and the Patterson method. The refinement procedure was performed by the full-matrix least-squares method based on F² against all reflections using SHELXL.⁴⁰

In [WO₂(acac)₂]·0.5C₆H₅Me the non-hydrogen atoms were refined anisotropically with exception of the toluene solvent molecule atoms which were refined isotropically. The toluene solvent molecule is situated near the centre of symmetry and is disordered over the two centrosymmetrically related positions. Restraints on the C–C distances and angles, and planarity (FLAT) were used for the toluene molecule. The centre of symmetry is within the toluene molecule and is located between atoms C12 and its centrosymmetrically related pair C12[−x, −y, −z] so placement of H-atoms at C12[−x, −y, −z] and the neighbouring C15 was difficult and their positions had to be fixed (Fig. S1†). All other hydrogen atoms were placed in calculated positions and refined using the riding model.

The non-hydrogen atoms in 1a, 1b and 3a were refined anisotropically. The hydrogen atoms in 1a, 1b and 3a were placed in calculated positions and refined using the riding model, exceptions were hydrogen atoms of the coordinated solvent molecule hydroxyl group.

Acknowledgements

Financial support for this research was provided by Ministry of Science and Technology of the Republic of Croatia. We acknowledge the Centre National de la Recherche Scientifique (CNRS) for financial support, the Université Paul Sabatier and its Institut Universitaire Technologique for the facilities. All authors want to thank Prof. R. Poli for suggestions and advices.

Notes and references

1. (a) M. C. Rodríguez-Argüelles, S. Mosquera-Vázquez, P. Tourón-Touceda, J. Sanmartín-Matalobos, A. M. García-Deibe, M. Belicchi Ferrari, G. Pelosi, C. Pelizzi and F. Zani, J. Inorg. Biochem., 2007, 101, 138; (b) D. S. Kalinowski, P. C. Sharpe, P. V. Bernhardt and D. R. Richardson, J. Med. Chem., 2008, 51, 331.
2. (a) V. Vrdoljak, B. Prugovečki, D. Matković-Čalogović, R. Dreos, P. Siega and C. Tavagnacco, Cryst. Growth Des., 2010, 10, 1373; (b) V. Vrdoljak, B. Prugovečki, D. Matković-Čalogović, J. Pisk, R. Dreos and P. Siega, Cryst. Growth Des., 2011, 11, 1244; (c) V. Vrdoljak, B. Prugovečki, D. Matković-Čalogović, T. Hrenar, R. Dreos and P. Siega, Cryst. Growth Des., 2013, 13, 3773; (d) W.-X. Xu, Y.-M. Yuan and W. H. Li, J. Coord. Chem., 2013, 66, 2726; (e) M. R. Maurya, L. Rana and F. Avecilla, Inorg. Chim. Acta, 2015, 429, 138.
3. (a) V. Vrdoljak, B. Prugovečki, I. Pulić, M. Cigler, D. Sviben, J. Parlov Vuković, P. Novak, D. Matković-Čalogović and M. Cindrić, New J. Chem., 2015, 39, 7322; (b) J. Pisk, B. Prugovečki, D. Matković-Čalogović, T. Jednačak, P. Novak, D. Agustin and V. Vrdoljak, RSC Adv., 2014, 4, 39000; (c) S.-B. Ding and W.-H. Li, J. Coord. Chem., 2013, 66, 2023; (d) M. Nandy, S. Shit, C. Razzoli, G. Pilet and S. Mitra, Polyhedron, 2015, 88, 63.
4. A. Dolbecq, P. Mialane, L. Lisnard, J. Marrot and F. Sécheresse, Chem.–Eur. J., 2003, 9, 2914.
5. (a) J. Pisk, D. Agustin, V. Vrdoljak and R. Poli, Adv. Synth. Catal., 2011, 353, 2910; (b) M. E. Judmaier, C. Holzer, M. Volpe and N. C. Mösch-Zanetti, Inorg. Chem., 2012, 51, 9956; (c) J. Pisk, B. Prugovečki, D. Matković-Čalogović, R. Poli, D. Agustin and V. Vrdoljak, Polyhedron, 2012, 33, 441; (d) M. Bagherzadeh, M. Amini, H. Parastar, M. Jalali-Heravi, A. Ellern and L. K. Woo, Inorg. Chem. Commun., 2012, 20, 86; (e) M. Bagherzadeh, M. Zare, V. Amani, A. Ellern and L. K. Woo, Polyhedron, 2013, 55, 223.
6. (a) B. Guérin, D. Mesquita-Fernandes, J.-C. Daran, D. Agustin and R. Poli, New J. Chem., 2013, 37, 3466; (b) J. Morlot, N. Uyttebroeck, D. Agustin and R. Poli, ChemCatChem, 2013, 5, 601.
7. (a) A. Rezaeifard, I. Sheikhshoaie, N. Monadi and M. Alipour, Polyhedron, 2010, 29, 2703; (b) M. Loubidi and D. Agustin, C. R. Chim., 2014, 17, 549; (c) W. Wang, T. Vanderbeeken, D. Agustin and R. Poli, Catal. Commun., 2015, 63, 26; (d) W. Wang, T. Guerrero, S. R. Merecias, H. García-Ortega, R. Santillan, J.-C. Daran, N. Farfán, D. Agustin and R. Poli, Inorg. Chim. Acta, 2015, 431, 176.
8. M. A. Katkar, S. N. Rao and H. D. Juneja, RSC Adv., 2012, 2, 8071.
9. X. Zhou, J. Zhao, A. M. Santos and F. E. Kühn, Z. Naturforsch., B: J. Chem. Sci., 2004, 59, 1223.
10. A. Dupé, M. K. Hossain, J. A. Schachner, F. Belaj, A. Lehtonen, E. Nordlander and N. C. Mösch-Zanetti, Eur. J. Inorg. Chem., 2015, 3572.
11. D. D. Agarwal, R. Rastogi, V. K. Gupta and S. Singh, Indian J. Chem., Sect. A: Inorg., Bio-inorg., Phys., Theor. Anal. Chem., 1999, 38, 369.
12. V. Vrdoljak, J. Pisk, D. Agustin, P. Novak, J. Parlov Vuković and D. Matković-Čalogović, New J. Chem., 2014, 38, 6176.
13. (a) S.-B. Yu and R. H. Holm, Inorg. Chem., 1989, 28, 4385; (b) W. A. Herrmann, J. Fridgen, G. M. Lobmaier and M. Spiegler, New J. Chem., 1999, 23, 5.
14. D. Agustin, C. Bibal, B. Neveux, J.-C. Daran and R. Poli, Z. Anorg. Allg. Chem., 2009, 635, 2120.
15. Cambridge Structural Database V5.37, Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge, England, 2015.
16. (a) F. Zhi and R. Wang, Acta Crystallogr., Sect. E: Struct. Rep. Online, 2010, 66, o892; (b) S.-J. Peng and H.-Y. Hou, Acta Crystallogr., Sect. E: Struct. Rep. Online, 2008, 64, o1996.
17. Y.-J. Xu, S. Zhao and S. Bi, Acta Crystallogr., Sect. E: Struct. Rep. Online, 2007, 63, o4633.
18. Y. Lei and C. Fu, Synth. React. Inorg., Met.-Org., Nano-Met. Chem., 2011, 41, 704.
19. S. Alghool and C. Slebodnick, Polyhedron, 2014, 67, 11.
20. Y.-L. Zhai, X.-X. Xu and Y. Wang, Polyhedron, 1992, 11, 415.
21. V. S. Sergienko, V. L. Abramenko, L. K. Minacheva, M. A. Porai-Koshits and V. G. Sakharova, Koord. Khim., 1993, 19, 28.
22. M. R. Maurya, S. Agarwal, C. Bader and D. Rehder, Eur. J. Inorg. Chem., 2005, 147.
23. M. R. Maurya, S. Gopinathan, C. Gopinathan and R. C. Maurya, Polyhedron, 1993, 12, 159.
24. (a) Y.-L. Wong, J.-F. Ma, W.-F. Law, Y. Yan, W.-T. Wong, Z.-Y. Zhang, T. C. W. Mak and D. K. P. Ng, Eur. J. Inorg. Chem., 1999, 313; (b) Y.-L. Wong, Y. Yan, E. S. H. Chan, Q. Yang, T. C. W. Mak and D. K. P. Ng, J. Chem. Soc., Dalton Trans., 1998, 3057.
25. K. Chichak, U. Jacquenard and N. R. Branda, Eur. J. Inorg. Chem., 2002, 357.
26. P. Krishnamoorthy, P. Satyadevi, P. T. Mutiah and N. Dharmaraj, RSC Adv., 2012, 2, 12190.
27. S. Gupta, A. Kumar Barik, S. Pal, A. Hazra, S. Roy, R. J. Butcher and S. Kumar Kar, Polyhedron, 2007, 26, 133.
28. (a) M. Lashanizadegan, S. Shayegan and M. Sarkheil, J. Serb. Chem. Soc., 2016, 81, 153; (b) M. Ghorbanloo, R. Bikas and G. Małecki, Inorg. Chim. Acta, 2016, 445, 8; (c) C. I. Fernandes, M. D. Carvalho, L. P. Ferreira, C. D. Nunes and P. D. Vaz, J. Organomet. Chem., 2014, 760, 2.
29. M. Ciclosi, C. Dinoi, L. Gonsalvi, M. Peruzzini, E. Manoury and R. Poli, Organometallics, 2008, 27, 2281.
30. C. Dinoi, M. Ciclosi, E. Manoury, L. Maron, L. Perrin and R. Poli, Chem.–Eur. J., 2010, 16, 9572.
31. P. Sőzen-Aktas, E. Manoury, F. Demirhan and R. Poli, Eur. J. Inorg. Chem., 2013, 2728.
32. A. Corma, P. Esteve and A. Martínez, J. Catal., 1996, 161, 11.
33. Z. Opre, T. Mallat and A. Baiker, J. Catal., 2007, 245, 482.
34. S. Bonollo, D. Lanari and L. Vaccaro, Eur. J. Org. Chem., 2011, 2587.
35. (a) J.-F. Lu, Acta Crystallogr., Sect. E: Struct. Rep. Online, 2008, 64, o2032; (b) H. H. Rassem, A. Salhin, B. Bin Salleh, M. M. Rosli and H.-K. Fun, Acta Crystallogr., Sect. E: Struct. Rep. Online, 2012, 68, o2279; (c) A. Salhin, A. A. Tameem, B. Saad, S.-L. Ng and H.-K. Fun, Acta Crystallogr., Sect. E: Struct. Rep. Online, 2007, 63, o2880.
36. W. C. Gibson, T. P. Kee and A. Shaw, Polyhedron, 1988, 7, 579.
37. X'Pert Software Suite, Version 1.3e, Panalytical B. V., Almelo, The Netherlands, 2001.
38. CrysAlisPro, Version 1.171.37.33, Agilent Technologies, release 27-03-2014 CrysAlis171.NET.
39. L. J. Farrugia, J. Appl. Crystallogr., 1999, 32, 837.
40. G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr., 2008, 64, 112.

---

Footnote

† Electronic supplementary information (ESI) available: (1) Spectral data, (2) XRPD patterns, (3) tables and (4) figures for compounds. Crystallographic data sets for the structures 1a, 1b, 3a, and [WO₂(acac)₂]·0.5C₆H₅Me. CCDC 1454679–1454682. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c6ra05067k

---