Mononuclear/Binuclear [V^IVO]/[V^VO₂] Complexes Derived from 1,3-Diaminoguanidine and Their Catalytic Application for the Oxidation of Benzoin via Oxygen Atom Transfer

Abstract

Ligands H₄sal-dag (I) and H₄Brsal-dag (II) derived from 1,3-diaminoguanidine and salicylaldehyde or 5-bromosalicylaldehyde react with one or 2 mol equivalent of vanadium precursor to give two different series of vanadium complexes. Thus, complexes [V^IVO(H₂sal-dag) (H₂O)] (1) and [V^IVO(H₂Brsal-dag) (H₂O)] (2) were isolated by the reaction of an equimolar ratio of these ligands with [V^IVO(acac)₂] in MeOH. In the presence of K⁺/Cs⁺ ion and using aerially oxidized [V^IVO(acac)₂], the above reaction gave complexes [K(H₂O){V^VO₂(H₂sal-dag)}]₂ (3), [Cs(H₂O){V^VO₂(H₂sal-dag)}]₂ (4), [K(H₂O){VO₂(H₂Brsal-dag)}]₂ (5), and [Cs(H₂O){V^VO₂(H₂Brsal-dag)}]₂ (6), which could also be isolated by direct aerial oxidation of complexes 1 and 2 in MeOH in the presence of K⁺/Cs⁺ ion. Complexes [(H₂O)V^IVO(Hsal-dag)V^VO₂] (7) and [(H₂O)V^IVO(HBrsal-dag)V^VO₂] (8) were isolated upon increasing the ligand-to-vanadium precursor molar ratio to 1:2 under an air atmosphere. When I and II were reacted with aerially oxidized [V^IVO(acac)₂] in a 1:2 molar ratio in MeOH in the presence of K⁺/Cs⁺ ion, they formed [K(H₂O)₅{(V^VO₂)₂(Hsal-dag)}]₂ (9), [Cs(H₂O)₂{(V^VO₂)₂(Hsal-dag)}]₂ (10), [K₂(H₂O)₄{(V^VO₂)₂(Brsal-dag)}]₂ (11), and [Cs₂(H₂O)₄{(V^VO₂)₂(Brsal-dag)}]₂ (12). The structures of complexes 3, 4, 5, and 9 determined by single-crystal X-ray diffraction study confirm the mono-, bi-, tri-, and tetra-anionic behaviors of the ligands. All complexes were found to be an effective catalyst for the oxidation of benzoin to benzil via oxygen atom transfer (OAT) between DMSO and benzoin. Under aerobic condition, this oxidation also proceeds effectively in the absence of DMSO. Electron paramagnetic resonance and ⁵¹V NMR studies demonstrated the active role of a stable V(IV) intermediate during OAT between DMSO and benzoin.

Introduction

Robenzidene, popularly known as “Robenidine,” is 1,3-diaminoguanidine-derived N′, 2-bis(€-4-chlorobenzylidene)hydrazine-1-carboximidhydrazide hydrochloride and was possibly the first compound developed as a poultry anticoccidial agent in the early 1970s. This was later extended to the whole range of apicomplexan parasites causing the disease coccidiosis in other smaller cattle such as dog, cat, sheep, goat, and so forth.¹⁻⁷ Very recently, Krollenbrock et al. have prepared Robenidine analogues having different substituents on the benzaldehyde moiety and tested them as potential antimalarial drugs in vitro as well as in vivo against Plasmodium falciparum, including multidrug-resistant strains.⁸ Over the years, modified Robenidine analogues have also been prepared by several groups and screened for anti-inflammatory activities.⁹ Other areas where such compounds find applications are colorimetric determination of fluoride,¹⁰ chemo sensor for Hg(II) ions,¹¹ fluorescent detection of Zn(II) ions,¹¹ and fluorometric determination of microgram level of gallium in biological tissues.¹²

However, the first report on the metal complexes of 1,3-diaminoguanidine-derived ligand (R = H, Scheme 1) appeared in 2008 where binuclear complexes of Cu(II), Ni(II), Mn(II), and V(IV) were isolated, but the structure of only the Cu(II) complex was confirmed by the single-crystal X-ray study (SC-XRD); structures of others were ventured only on the basis of the corresponding (thio)carbohydrazide-derived ligands.¹³ Tang et al. reported Ru(II) complexes of ligands presented in Scheme 1 (R = H, 5-Cl, 3,5-Br₂) where ligands were coordinated to Ru(II) as monoanionic NN bidentate, leaving azomethine nitrogens and hydroxyl groups free from coordination.¹⁴ Mononuclear cis-[Mo^VIO₂] as well as binuclear bis{cis-[Mo^VIO₂]} complexes of the ligand with R = 2,4-di-^tertBu of Scheme 1 were isolated in 2020, and their structures were confirmed by SC-XRD and examined for catalytic potential toward epoxidation of cyclohexene.¹⁵

Scheme 1. 1,3-Diaminoguanidine Derived Ligands (I and II) Used in This Study and Ligand III Earlier Used for Vanadium Complexes.

Among various transition metals, the coordination chemistry of vanadium has been inspired by its role in biological systems.¹⁶ The presence of vanadate moiety at the active sites of haloperoxidase enzymes and its ability to catalyze oxidation and oxidative bromination of organic molecules by hydrogen peroxide¹⁷ attracted the attention of researchers to search for its structural as well as functional models.^18,19 In addition to the functional model, vanadium complexes have extensively been explored in other catalytic reactions²⁰ as well, like peroxidase mimicking,²¹ sulphoxidation,²² bromination,²³ oxidation of alkanes,²⁴ cyclohexane,²⁵ phenols and alcohols,²⁶ oxidative coupling of arenols,²⁷ epoxidation of olefins,^26c,28 hydroxylation of aromatic hydrocarbons,²⁸ multicomponent reactions,²⁹ and so forth.

Since ligands I and II (Scheme 1) derived from 1,3-diaminoguanidine undergo several tautomeric structures, they show very flexible coordination behaviors. In fact, monobasic N̂N bidentate,¹⁴ tribasic hexadentate,¹³ and tetrabasic hexadentate¹⁵ behaviors of these ligands are well established. In principle, these ligands have five titrable protons, but so far only four of them have been found to be deprotonable in metal complexes;¹⁵ therefore, we have also considered them as tetrabasic hexadentate in our study. Thus, compared to vanadium complexes of closely related ligands, 1,5-bis(substituted 2-hydroxybenzaldehyde)carbohydrazone (III),^30,31 where these ligands are restricted to provide two different ONN and ONO binding sites in a combined triply deprotonated form, ligands I and II, with flexible coordination behaviors, may present different vanadium coordination chemistries. Our interest in vanadium complexes of polynucleating ligands¹⁹ prompted us to consider ligands I and II and undertake systematic investigation on their coordination behavior toward vanadium precursor(s). We report herein the isolation of various mononuclear/binuclear [V^IVO]/[V^VO₂] complexes of I and II and their characterization and reactivity. It is expected that the electronegative character of the N-donor atom (of amine residue) of one pocket of these ligands may help in recognizing individual vanadium centers by nuclear magnetic resonance (NMR) (¹H and ⁵¹V) study if unsymmetry exists in binuclear complexes.

Further, benzil is an important substrate in the classic benzil-benzilic rearrangement for the synthesis of biologically active compounds A-Norpreganes.³² Benzil has also been used as a starting material for the synthesis of important heterocyclic compounds.³³⁻³⁵ A number of methods are reported in the literature for the oxidation of α-hydroxycarbonyl compounds using different metal-based catalysts including Mn³⁶ and Cr.³⁷ However, these methods involve the formation of environmentally toxic and hazardous side products. Noble metal catalysts including ruthenium, gold, palladium, and platinum have also been reported in the literature.³⁸⁻⁴⁶ However, these noble metal-based catalysts are very expensive, and therefore, it is worth developing non-noble transition-metal-based catalysts like vanadium for effective oxidation of α-hydroxycarbonyl compounds. However, it is to be noted that not all oxidovanadium(IV/V) complexes are nontoxic unless they are verified.⁴⁷ There is very limited literature on the aerobic oxidation of α-hydroxycarbonyl compounds including benzoin by oxidovanadium(IV/V) based compounds.⁴⁸⁻⁵¹ We have reported that dioxidomolybdenum(VI) complexes catalyzed the oxidation of benzoin to benzil using the oxygen atom transfer (OAT) reaction between benzoin and dimethyl sulfoxide (DMSO).⁵² However, such an OAT reaction catalyzed by oxido/dioxido-vanadium(IV/V) complexes has not been explored. Therefore, herein we also focus on the catalytic oxidation of benzoin to benzil using the OAT reaction between benzoin and DMSO under aerobic/inert conditions. Role of the active catalytic intermediate in this OAT reaction has also been identified.

Experimental Section

Materials and Methods

Analytical reagent grade salicylaldehyde, acetylacetone, 1,3-diaminoguanidine monohydrochloride, DMSO-d₆ (Sigma-Aldrich, USA), and V₂O₅ (Himedia, India) were used as procured. Other chemicals and solvents used in this study were procured from standard sources. Precursor [V^IVO(acac)₂]⁵³ was prepared following the literature method.

All measurements were made after drying the ligands and complexes at 100 °C. The CHN analysis of all complexes were carried out on an elementar model Vario-EL-III element analyzer. Infrared (IR) spectra were recorded with KBr pellets on a Nicolet 1100 FT-IR spectrometer. Electronic spectra of the ligands and complexes were recorded in DMSO on a Shimadzu 2600 UV–vis spectrophotometer. ¹H, ¹³C, and ⁵¹V NMR spectra were recorded in DMSO-d₆ on a Jeol 500 MHz spectrometer. The electron paramagnetic resonance (EPR) spectra of the complexes were recorded in DMSO at 100 K on a Bruker Biospin EMXmicro A200-9.5/12/S/W spectrometer. High-performance liquid chromatography (HPLC) was performed on a Shimadzu LC-2010A HT instrument in the low-pressure gradient mode with a flow rate of 0.5 mL min^–1 and an injection volume of 15 μL.

Synthesis of Ligands

The method reported in the literature^10,13 was used to prepare 1,3-bis(2-hydroxyphenylmethylideneamino)guanidine monohydrochloride (H₄sal-dag·HCl) (I), and 1,3-bis(5-bromo-2-hydroxyphenylmethylideneamino)guanidine monohydrochloride (H₄Brsal-dag·HCl) (II). Characterization data (¹H and ¹³C NMR) presented in the Supporting Information match well with the literature values.

Synthesis of Complexes

[V^IVO(H₂sal-dag) (H₂O)] (1)

To a methanolic solution (30 mL) of [V^IVO(acac)₂] (0.265 g, 1.00 mmol) was added H₄sal-dag (I) (0.334 g, 1.00 mmol) as solid with shaking, and the resulting reaction mixture was refluxed for ca. 10 h. The reaction mixture was allowed to cool at ambient temperature where a dark brown solid precipitated out within 24 h. The brown solid was filtered, washed with methanol, and dried under vacuum. Yield: 0.260 g (68.4%). C₁₅H₁₅N₅O₄V (MW = 380.26) calcd C, 47.38; H, 3.98; N, 18.42. Found: C, 47.07; H, 4.12; N, 18.24%. UV–vis (DMSO) [λ_max, nm (ε, litre mol^–1 cm^–1)]: 648 (0.98 × 10^–2), 400 (4.2 × 10²), 336 (1.25 × 10³), 276 (1.46 × 10³), 258 (1.31 × 10³). IR (KBr, ν_max/cm^–1): 1632, 1606 (C=N), 996 (V=O). ¹H NMR (500 MHz, DMSO-d₆): δ/ppm = 11.91 (br, 1H, OH), 10.09 (s, 1H, −C=NH), 9.65 (s, 1H, Ph–CH=N−), 8.70 (s, 1H, Ph–CH=N−), 8.01 (d, 1H, aromatic), 7.65 (s, 2H, aromatic), 7.54 (d, 1H, aromatic), 7.43 (s, 1H, aromatic), 7.25 (s, 1H, aromatic), 6.94 (s, 1H, aromatic), 6.83 (b, 2H, =CNH).

[V^IVO(H₂Brsal-dag) (H₂O)] (2)

The brown complex 2 was prepared from [V^IVO(acac)₂] (0.265 g, 1.00 mmol) and H₄Brsal-dag (II) (0.419 g, 1.00 mmol) by adopting the procedure outlined for 1. Yield = 0.370 g (68.8%). C₁₅H₁₃Br₂N₅O₄V (MW = 538.05) calcd C, 33.48; H, 2.44; N, 13.02. Found: C, 33.92; H, 2.29; N, 13.28%. UV–vis (DMSO) [λ_max, nm (ε, litre mol^–1 cm^–1)]: 647 (1.23 × 10^–2), 405 (5.85 × 10²), 350 (1.26 × 10³), 260 (1.79 × 10³). IR (KBr, ν_max/cm^–1): 1639, 1613 (C=N), 970 (V=O). ¹H NMR (500 MHz, DMSO-d₆): δ/ppm = 12.05 (br, 1H, OH), 10.32 (s, 1H, −C=NH), 9.51 (s, 1H, Ph–CH=N−), 8.59 (s, 1H, Ph–CH=N−), 8.20 (s, 1H, aromatic), 7.72 (s, 1H, aromatic), 7.62 (s, 1H, aromatic), 7.46 (s, 1H, aromatic), 7.33 (s, 1H, aromatic), 6.85 (s, 1H, aromatic), 6.74 (s, 1H, =CNH).

[K(H₂O){V^VO₂(H₂sal-dag)}]₂ (3)

A solution of [V^IVO(acac)₂] (0.265 g, 1.00 mmol) prepared in EtOH (30 mL) was kept open for aerial oxidation at ambient temperature with slow stirring for 2 days. After adding H₄sal-dag (I) (0.334 g, 1.00 mmol) as solid, a solution of K₂CO₃ (0.85 g, 1.50 mmol) (prepared in 8 mL EtOH +2 mL H₂O) was added to the reaction mixture. The pH of the resulting reaction mixture was ∼9 after complete addition of potassium carbonate. Thereafter, the reaction mixture was stirred at room temperature for 2 days whereupon a transparent orange-red solution was obtained. The solvent volume was reduced to ca. 10 mL, and the flask was kept at room temperature where an orange-yellow solid precipitated out within 24 h. This was filtered, washed with cold EtOH and dried under vacuum. Keeping the filtrate at room temperature, orange-yellow crystals of 3 suitable for single-crystal X-ray study were formed after ca. 4 days. Yield 0.315 g (71.4%). C₃₀H₃₀K₂N₁₀O₉V₂ (MW = 854.71) calcd C, 42.16; H, 3.54; N, 16.39. Found: C, 42.63; H, 3.65; N, 16.68%. UV–vis (DMSO) [λ_max, nm (ε, litre mol^–1 cm^–1)]: 435 (2.12 × 10³), 360 (3.16 × 10³), 266 (2.98 × 10³). IR (KBr, ν_max/cm^–1): 1637, 1621 (C=N), 928, 910 (O=V=O). ¹H NMR (500 MHz, DMSO-d₆): δ/ppm = 10.75 (br, 1H, OH), 9.16 (s, 1H, Ph–CH=N−), 8.41 (s, 1H, Ph–CH=N−), 7.59 (d, 1H, aromatic), 7.36 (d, 1H, aromatic), 7.20 (t, 1H, aromatic), 7.15 (t, 1H, aromatic), 6.87 (t, 2H, aromatic), 6.69 (m, 2H, aromatic), 5.84 (br, 2H, =CNH₂).

[Cs(H₂O){V^VO₂(H₂sal-dag)}]₂ (4)

The reddish–orange complex 4 was prepared similarly from [V^IVO(acac)₂] (0.265 g, 1.00 mmol) and H₄sal-dag (I) (0.334 g, 1.00 mmol) in MeOH in the presence of Cs₂CO₃ solution (0.488 g, 1.50 mmol in 8 mL MeOH +2 mL H₂O), as mentioned for 3. Crystals suitable for single-crystal X-ray study were obtained from the filtrate. Yield 0.360 g (68.1%). C₃₀H₃₀Cs₂N₁₀O₁₀V₂ (MW = 1058.32) calcd C, 34.05; H, 2.86; N, 13.24. Found: C, 34.33; H, 2.71; N, 13.03%. UV–vis (DMSO) [λ_max, nm (ε, litre mol^–1 cm^–1)]: 435 (2.4 × 10³), 359 (3.46 × 10³), 264 (3.54 × 10³). IR (KBr, ν_max/cm^–1): 1634, 1621 (C=N), 930, 911, 885 (O=V=O). ¹H NMR (500 MHz, DMSO-d₆): δ/ppm = 10.74 (s, 1H, OH), 9.15 (s, 1H, Ph–CH=N−), 8.40 (s, 1H, Ph–CH=N−), 7.59 (d, 1H, aromatic), 7.36 (d,1H, aromatic), 7.20 (t, 1H, aromatic), 7.15 (t, 1H, aromatic), 6.86 (m, 2H, aromatic), 6.70 (m, 2H, aromatic) 5.83 (s, 2H, =CNH₂).

[K(H₂O){V^VO₂(H₂Brsal-dag)}]₂ (5)

Using the procedure outlined for 3, maroon-red complex 5 was prepared from [V^IVO(acac)₂] (0.265 g, 1.00 mmol) and H₄Brsal-dag (II) (0.419 g, 1.00 mmol) in MeOH in the presence of K₂CO₃ solution (0.85 g, 1.50 mmol in 8 mL MeOH +2 mL H₂O). Crystals suitable for single-crystal X-ray study were obtained from the filtrate. Yield 0.335 g (53.6%). C₃₁H₃₀Br₄K₂N₁₀O₁₁V₂ (MW = 1218.34) calcd C, 30.56; H, 2.48; N, 11.50. Found: C, 30.73; H, 2.57; N, 11.28%. UV–vis (DMSO) [λ_max, nm (ε, litre mol^–1 cm^–1)]: 444 (3.42 × 10³), 369 (4.28 × 10³), 267 (4.76 × 10³). IR (KBr, ν_max/cm^–1): 1631, 1611 (C=N), 949, 885 (O=V=O). ¹H NMR (500 MHz, DMSO-d₆): δ/ppm = 10.47 (br, 1H, OH), 9.08 (s, 1H, Ph–CH=N−), 8.38 (s, 1H, Ph–CH=N−), 7.92 (d, 1H, aromatic), 7.57 (d, 1H, aromatic), 7.27–7.22 (m, 2H, aromatic), 6.83 (d, 1H, aromatic), 6.64 (d, 1H, aromatic), 6.14 (br, 2H, =CNH₂), 3.37 (br, 2H, coordinated H₂O).

[Cs(H₂O){V^VO₂(H₂Brsal-dag)}]₂ (6)

The yellowish-orange complex 6 was prepared in MeOH taking [V^IVO(acac)₂] (0.265 g, 1.00 mmol) and H₄Brsal-dag (II) (0.419 g, 1.00 mmol) in the presence of Cs₂CO₃ solution (0.488 g, 1.50 mmol in 8 mL EtOH +2 mL H₂O) according to the procedure mentioned for 3. Yield 0.400 g (58.2%). C₃₀H₂₆Br₄Cs₂N₁₀O₁₀V₂ (MW = 1373.91) calcd C, 26.23; H, 1.91; N, 10.20. Found C, 26.42; H, 1.84; N, 10.38%. UV–vis (DMSO) [λ_max, nm (ε, litre mol^–1 cm^–1)]: 445 (3.26 × 10³), 369 (3.72 × 10³), 265 (4.5 × 10³). IR (KBr, ν_max/cm^–1): 1643, 1613 (C=N), 917, 864 (O=V=O). ¹H NMR (500 MHz, DMSO-d₆): δ/ppm = 10.39 (s, 1H, OH), 9.08 (s, 1H, Ph–CH=N−), 8.38 (s, 1H, Ph–CH=N−), 7.92 (s, 1H, aromatic), 7.57 (s, 1H, aromatic), 7.26 (m, 2H, aromatic), 6.84 (d, 1H, aromatic), 6.63 (d, 1H, aromatic), 6.10 (s, 2H, =CNH₂).

[(H₂O)V^IVO(Hsal-dag)V^VO₂] (7)

Precursor [V^IVO(acac)₂] (0.663 g, 2.50 mmol) and H₄sal-dag (I) (0.334 g, 1.00 mmol) were dissolved together in MeOH (30 mL), and the resulting reaction mixture was refluxed for ca. 10 h. Thereafter, the flask was left open for slow oxidation of the product at room temperature for ca. 2 days with slow stirring where a brown solid precipitated. The solid was filtered, washed with methanol, and dried under vacuum. Yield 0.300 g (64.9%). C₁₅H₁₄N₅O₆V₂ (MW = 462.19) calcd C, 38.98; H, 3.05; N, 15.15. Found C, 39.32; H, 2.97; N, 15.28%. UV–vis (DMSO) [λ_max, nm (ε, litre mol^–1 cm^–1)]: 648 (1.23 × 10^–2), 400 (5.3 × 10²), 341 (9.92 × 10²), 283 (1.04 × 10³), 257 (1.60 × 10³). IR (KBr, ν_max/cm^–1): 1603 (C=N), 961 (V=O), 916, 898 (O=V=O). ¹H NMR (500 MHz, DMSO-d₆): δ/ppm = 10.08 (s, 1H, −C=NH), 9.65 (s, 1H, Ph–CH=N−), 8.70 (s, 1H, Ph–CH=N−), 8.31 (s, 2H, Aromatic), 7.65 (s, 1H, aromatic), 7.52 (s, 1H, aromatic), 7.42 (s, 1H, aromatic), 7.25 (s, 1H, aromatic), 6.84 (b, 3H, aromatic and =CNH).

[(H₂O)V^IVO(HBrsal-dag)V^VO₂] (8)

Brown complex 8 was prepared from [V^IVO(acac)₂] (0.663 g, 2.50 mmol) and H₄Brsal-dag (II) (0.419 g, 1.00 mmol) in MeOH (30 mL) employing the procedure outlined for 7. Yield 0.390 g (62.9%). C₁₅H₁₂Br₂N₅O₆V₂ (MW = 619.98) calcd C, 29.06; H, 1.95; N, 11.30. Found: C, 29.48; H, 1.72; N, 11.43%. UV–vis (DMSO) [λ_max, nm (ε, litre mol^–1 cm^–1)]: 650 (1.43 × 10^–2), 408 (16.38 × 10²), 351 (1.52 × 10³), 266 (2.13 × 10³). IR (KBr, ν_max/cm^–1): 1612 (C=N), 955 (V=O), 919, 880 (O=V=O). ¹H NMR (paramagnetic) (500 MHz, DMSO-d₆): δ/ppm = 10.37 (br, 1H, −C=NH), 9.58 (s, 1H, Ph–CH=N−), 8.65 (s, 1H, Ph–CH=N−), 8.28 (s, 1H, aromatic), 7.79 (s, 2H, aromatic), 7.51 (s, 1H, aromatic), 6.89 (s, 1H, aromatic), 6.79 (s, 2H, aromatic and =CNH).

[K(H₂O)₅{(V^VO₂)₂(Hsal-dag)}]₂ (9)

An oxidized solution of [V^IVO(acac)₂] (0.663 g, 2.50 mmol) in ethanol (30 mL) was prepared as mentioned for 3, and to this was added H₄sal-dag (I) (0.334 g, 1.00 mmol) as solid followed by KOH solution (0.140 g, 2.50 mmol in 8 mL EtOH +2 mL H₂O) slowly with stirring. This stirring was continued for ca. 3 days where most of the ligand was dissolved and an orange-yellow solution was obtained. This was filtered, and the solvent volume was reduced to half. After keeping the solution overnight, the yellowish-orange solid was precipitated which was filtered, washed with cold EtOH, and dried under vacuum. The filtrate left after slow evaporation gave yellow-orange crystals suitable for single-crystal X-ray study. Combined yield 0.320 g (54.3%). C₃₀H₄₄K₂N₁₀O₂₂V₄ (MW = 1178.7) calcd C, 30.57; H, 3.76; N, 11.88. Found: C, 30.82; H, 3.61; N, 12.08%. UV–vis (DMSO) [λ_max, nm (ε, litre mol^–1 cm^–1)]: 434 (2.64 × 10³), 360 (3.8 × 10³), 263 (3.98 × 10³). IR (KBr, ν_max/cm^–1): 1609 (C=N), 937, 911, 889 (ν_symO=V=O). ¹H NMR (500 MHz, DMSO-d₆): δ/ppm = 11.56 (br, 1H, −C=NH), 9.67 (s, 1H, Ph–CH=N−), 8.62 (s, 1H, Ph–CH=N−), 7.57 (s, 1H, aromatic), 7.50 (d, 1H, aromatic), 7.40 (t, 1H, aromatic), 7.34 (d, 1H, aromatic), 6.90–6.80 (m, 5H, aromatic and =CNH), 3.17 (s, 1H, coordinated H₂O).

[Cs(H₂O)₂{(V^VO₂)₂(Hsal-dag)}]₂ (10)

Complex 10 was prepared in EtOH from [V^IVO(acac)₂] (0.663 g, 2.50 mmol) and H₄sal-dag (I) (0.334 g, 1.00 mmol) in the presence of Cs₂CO₃ solution (0.815 g, 2.50 mmol in 8 mL of EtOH +2 mL of H₂O) using the procedure presented for 9. A yellowish-orange solid precipitate was filtered, washed with cold EtOH, and dried under vacuum. Yield 0.580 g (94.9%). C₃₀H₃₂Cs₂N₁₀O₁₆V₄ (MW = 1258.22) calcd C, 28.64; H, 2.56; N, 11.13 Found: C, 28.22; H, 2.71; N, 11.28%. UV–vis (DMSO) [λ_max, nm (ε, litre mol^–1 cm^–1)]: 463 (4.54 × 10²), 370 (2.1 × 10³), 280 (2.06 × 10³), 258 (4.1 × 10³). IR (KBr, ν_max/cm^–1): 1610 (C=N), 937, 927, 917, 894 (O=V=O). ¹H NMR (500 MHz, DMSO-d₆): δ/ppm = 11.55 (s, 1H, −C=NH), 9.67 (s, 1H, Ph–CH=N−), 8.63 (s, 1H, Ph–CH=N−), 7.57 (d, 1H, aromatic), 7.52 (d, 1H, aromatic), 7.42 (t, 1H, aromatic), 7.35 (t, 1H, aromatic), 6.90 (t, 1H, aromatic), 6.84 (t, 4H, aromatic and =CNH).

[K₂(H₂O)₄{(V^VO₂)₂(Brsal-dag)}]₂ (11)

Orange-yellow complex 11 was prepared in EtOH from [V^IVO(acac)₂] (0.663 g, 2.50 mmol) and H₄Brsal-dag (II) (0.419 g, 1.00 mmol) in the presence of Cs₂CO₃ solution (0.815 g, 2.50 mmol in 8 mL of EtOH +2 mL of H₂O) using the procedure presented for 9. Yield 0.390 g (47.4%). C₃₀H₃₄Br₄K₄N₁₀O₂₀V₄ (MW = 1534.43) calcd C, 23.48; H, 2.23; N, 9.13. Found: C, 23.72; H, 2.17; N, 9.21%. UV–vis (DMSO) [λ_max, nm (ε, litre mol^–1 cm^–1)]: 457 (2.86 × 10³), 382 (2.94 × 10³), 279 (5.5 × 10³), 258 (5.52 × 10³). IR (KBr, ν_max/cm^–1): 1616 (C=N), 925, 860 (O=V=O). ¹H NMR (500 MHz, DMSO-d₆): δ/ppm = 9.37 (s, 1H, Ph–CH=N−), 8.47 (s, 1H, Ph–CH=N−), 7.65 (s, 1H, aromatic), 7.59 (s, 1H, aromatic), 7.37 (t, 2H, aromatic), 6.74 (d, 2H, aromatic), 6.47 (b, 1H, =CNH), 3.17 (s, 1H, coordinated H₂O).

[Cs₂(H₂O)₄{(V^VO₂)₂(Brsal-dag)}]₂ (12)

Orange-yellow complex 12 was prepared in EtOH from [V^IVO(acac)₂] (0.663 g, 2.50 mmol) and H₄Brsal-dag (II) (0.419 g, 1.00 mmol) in the presence of Cs₂CO₃ solution (0.815 g, 2.50 mmol in 8 mL of EtOH +2 mL of H₂O) following the procedure outlines for 9. Yield 0.772 g (84.0%). C₃₀H₃₄Br₄Cs₄N₁₀O₂₀V₄ (MW = 1909.66) calcd C, 18.87; H, 1.79; N, 7.33. Found: C, 19.06; H, 1.66; N, 7.45%. UV–vis (DMSO) [λ_max, nm (ε, litre mol^–1 cm^–1)]: 458 (4.14 × 10³), 382 (3.06 × 10³), 298 (4.66 × 10³), 257 (5.04 × 10³). IR (KBr, ν_max/cm^–1): 1607 (C=N), 923, 867 (O=V=O). ¹H NMR (500 MHz, DMSO-d₆): δ/ppm = 9.10 (s, 1H, Ph–CH=N−), 8.30 (s, 1H, Ph–CH=N−), 7.46 (d, 2H, aromatic), 7.24 (m, 2H, aromatic), 6.67 (m, 2H, aromatic), 5.70 (b, 1H, =CNH).

X-ray Crystal Structure Determination

Three-dimensional X-ray data were collected on a Bruker Kappa Apex CCD diffractometer at 293 K for compounds 3, 4, 5, and 9 by the ϕ−ω scan method. Reflections were measured from a hemisphere of data collected from frames, each of them covering 0.3° in ω. A total of 58215 for 3, 13654 for 4, 34841 for 5, and 11861 for 9 reflections measured were corrected for Lorentz and polarization effects and for absorption by multiscan methods based on symmetry-equivalent and repeated reflections. Of the total, 3841 for 3, 4120 for 4, 4921 for 5, and 4290 for 9 independent reflections exceeded the significance level (|F|/σ|F|) > 4.0. After data collection, a multiscan absorption correction (SADABS)⁵⁴ was applied in each case. The structures of compounds 3, 4, and 5 were solved by direct methods and refined by full matrix least-squares on F² data using Olex2,⁵⁵ and the structure of 9 was solved and refined using SHELX suite of programs.⁵⁶ Hydrogen atoms were located in a difference Fourier map and left to refine freely in compound 4, except for O(5), which was located in a difference Fourier map and fixed to the oxygen atom. In the other three compounds, hydrogen atoms were included in calculated positions and refined in the riding mode, except for C(3), O(4), C(4), C(5), C(7), C(9), C(14), and C(15) in compound 3, O(1M), C(2), N(3), O(4), C(5), C(11), C(13), and C(14) in compound 5, and O(3W) and N(5) in compound 9, which were located in a difference Fourier map and left to refine freely. Refinements were done with allowance for thermal anisotropy of all non-hydrogen atoms. Compound 3 presents a disorder on the oxygen atom of the half-coordinated ethanol molecule and compound 9 on one water molecule. These disorders have been refined, and two atomic sites have been observed and refined for the atoms implied with the anisotropic atomic displacement parameters. More specifically, these disorders were refined using 17 and 12 restraints in compound 3 and compound 9, respectively (SADI, SIMU, and DELU restraints were used). The site occupancy factors were 0.50738 for O(1SA) in compound 3 and 0.82932 for O(1WA) in compound 9. In compound 5, the hydrogen atoms of water molecule present in the crystal packing were not located. A final difference Fourier map showed no residual electronic density: 0.65 and −0.43 e Å^–3 for 3. For the others, the final difference Fourier map showed a residual electronic density: 1.85 and −1.44 e Å^–3 for 4, 1.87 and −1.46 e Å^–3 for 5, and 1.651 and −0.688 e Å^–3 for 9, next to the metal atoms. Weighting schemes of w = 1/[σ²(F_o²) + (0.089468 P)² + 2.042964 P] for 3, 1/[σ²(F_o²) + (0.026258 P)² + 2.582623 P] for 4, 1/[σ²(F_o²) + (0.055665 P)² + 7.259597 P] for 5, and 1/[σ²(F_o²) + (0.072100 P)² + 2.567600 P] for 9, where P = (|F_o|² + 2|F_c|²)/3, were used in the latter stages of refinement. Further details of the crystal structure determination are given in Table 1.

Table 1. Crystal Data and Structure Refinement for 3, 4, 5, and 9.

	3	4	5	9
formula	C16H17N5O4.5KV	C15H14N5O5VCs	C16H17Br2KN5O6V	C15H22KN5O11V2
formula weight	441.384	528.156	625.19	589.36
T, K	293(2)	293(2)	293(2)	293(2)
wavelength, Å	0.71073	0.71073	0.71073	0.71073
crystal system	orthorhombic	triclinic	monoclinic	triclinic
space group	Pbca	P1̅	P21/c	P1̅
a/Å	11.0214(5)	6.9094(4)	8.0685(4)	6.9544(13)
b/Å	14.4994(7)	11.2611(6)	29.1227(15)	11.256(2)
c/Å	25.0218(13)	11.6773(6)	9.5344(5)	15.632(3)
α/deg	90	95.021(1)	90	71.482(6)
β/deg	90	96.263(1)	109.2720(10)	89.174(5)
γ/deg	90	99.426(1)	90	84.835(5)
V/Å3	3998.6(3)	885.80(8)	2114.81(19)	1155.5(4)
Z	8	2	4	2
F000	1813.0	514.4	1233.0	600
Dcalc/g cm–3	1.466	1.980	1.964	1.694
μ/mm–1	0.738	2.624	4.489	1.054
2θ/(deg)	4.92 to 56.64	6.02 to 56.72	4.74 to 68.76	6.46 to 56.70
Rint	0.0532	0.0546	0.0507	0.0485
crystal size/mm3	0.30 × 0.20 × 0.10	0.3 × 0.2 × 0.1	0.3 × 0.2 × 0.1	0.21 × 0.16 × 0.08
goodness-of-fit on F2	1.062	1.051	1.048	1.024
R1[I > 2σ(I)]a	0.0439	0.0340	0.0426	0.0596
wR2(all data)b	0.1561	0.0842	0.1161	0.1679
largest differences peak and hole (eÅ–3)	0.65 and −0.43	1.85 and −1.44	1.87 and −1.47	1.649 and −0.692

^aR₁ = Σ||F_o| – |F_c||/Σ|F_o|.

^bwR₂ = {Σ[w(||F_o|² – |F_c|²|)²]|/Σ[w(F_o²)²]}^1/2.

Results and Discussion

Synthesis of Complexes and Their Stability

Ligands H₄sal-dag (I) and H₄Brsal-dag (II) (c.f.Scheme 1) have competence to react under aerobic/anaerobic conditions with 1 or 2 mol equivalent of the vanadium precursor in the presence or absence of K⁺/Cs⁺ ion to give 12 different types of complexes (Scheme 2). Thus, the reaction of equimolar ratio of these ligands with [V^IVO(acac)₂] in MeOH under an inert atmosphere resulted in the formation of complexes [V^IVO(H₂sal-dag) (H₂O)] (1) and [V^IVO(H₂Brsal-dag) (H₂O)] (2). Increasing this ratio to 1:2 (L/V precursor) in the presence of air, these ligands provide complexes [(H₂O)V^IVO(Hsal-dag)V^VO₂] (7) and [(H₂O)V^IVO(HBrsal-dag)V^VO₂] (8), where one of the V ions is oxidized to oxidation state 5 and generate [V^VO₂] moiety. Direct aerial oxidation of complexes 1 and 2 in the presence of K⁺/Cs⁺ ion or the reactions of ligands I and II with aerially oxidized [V^IVO(acac)₂] in a 1:1 molar ratio in MeOH in the presence of K⁺/Cs⁺ ion give complexes [K(H₂O){V^VO₂(H₂sal-dag)}]₂ (3), [Cs(H₂O){V^VO₂(H₂sal-dag)}]₂ (4), [K(H₂O){VO₂(H₂Brsal-dag)}]₂ (5), and [Cs(H₂O){V^VO₂(H₂Brsal-dag)}]₂ (6). Similarly, carrying out the reactions of I and II with aerially oxidized [V^IVO(acac)₂] in a 1:2 molar ratio in MeOH in the presence of K⁺/Cs⁺ ion form [K(H₂O)₅{(V^VO₂)₂(Hsal-dag)}]₂ (9), [Cs(H₂O)₂{(V^VO₂)₂(Hsal-dag)}]₂ (10), [K₂(H₂O)₄{(V^VO₂)₂(Brsal-dag)}]₂ (11), and [Cs₂(H₂O)₄{(V^VO₂)₂(Brsal-dag)}]₂ (12). Whole reaction schemes and structures of different complexes (based on different studies) are summarized in Scheme 2. These complexes were characterized by elemental and thermal analyses, spectroscopic (IR, UV–visible, ¹H, ¹³C, and ⁵¹V NMR), and EPR (for [V^IVO] complexes). Additionally, complexes 3, 4, 5, and 9 were confirmed by single-crystal X-ray diffraction (SC-XRD) study.

Scheme 2. Synthetic Routes to Prepare Different Types of Vanadium Complexes.

(a) Reaction Conditions: (i) [V^IVO(acac)₂] (M/L = 1:1)/MeOH, Δ, (ii) [V^IVO(acac)₂] (M/L = 1:1)/MeOH, K₂CO₃ or Cs₂CO₃(1.5 equiv), RT, (iii) [V^IVO(acac)₂] (M/L = 2.5:1)/MeOH, Δ, O₂/RT, (iv) [V^IVO(acac)₂] (M/L = 2.5:1)/EtOH, KOH or Cs₂CO₃ (2.5 equiv), RT, and (v) [V^IVO(acac)₂] (M/L = 2.5:1)/EtOH, KOH or Cs₂CO₃ (2.5 equiv), RT. (b) Structures of the complexes are based on different studies. For complexes 3–6 and 9–11, only the anionic part of the structures is shown.

All these complexes are stable for a long period at room temperature but show a mass loss corresponding to the coordinated solvent molecule(s) (MeOH, EtOH, and H₂O) over a temperature range of 100–280 °C. All complexes decompose on further heating in two or more exothermic steps with the formation of V₂O₅ and MVO₃ (M⁺ = K⁺ or Cs⁺) at higher temperature (300–800 °C). The observed and calculated values of V₂O₅ and MVO₃ residues are presented in Table S1 [for thermogravimetric analysis (TGA) profiles, see Figure S1], and they confirm the expected composition of complexes.

Structure Descriptions

Figures S2–S5 depict the ORTEP representations of the asymmetric units of the compounds 3, 4, 5, and 9, respectively. Figures 1–4 show that these compounds form dimeric aggregates; such dimerization has already been observed in other published works.²⁹ The vanadium metal centers are five-coordinated by two terminal oxygen atoms, which act as bridging with alkali metal ions, and three donor atoms of the ligand, two nitrogen atoms and one oxygen atom. The distances of the vanadium center to the terminal oxygen atoms are similar [V(1)–O(1), 1.6209(19) Å in 3, 1.649(2) Å in 4 and 1.653(3) Å in 9; V(1)–O(2), 1.6389(18) Å in 3, 1.632(3) Å in 4, 1.679(2) Å in 5, and 1.629(3) Å in 9], except in the case of the bond length of V(1)–O(1) in compound 5, which is a little longer, 1.853(2) Å. In 3 and 5, the oxygen atom of the phenol coordination of the ligand also forms a bridge with the K⁺ ion. The distances to water molecules are Cs(1)–O(5), 3.077(3) Å in 4, K(1)-O(1W), 2.944(6) Å, through the symmetry transformation 1 – x, −y, 1 – z, and 3.420(6) Å and through the symmetry transformation x + 1, y, and z in 5. The K to water molecules distances are between 2.680(3) and 2.922(14) Å in compound 9. Table 2 presents the list of the main bond lengths and angles.

Figure 1.

ORTEP of the dimeric aggregate present in the crystal packing of compound 3, [K(EtOH)_0.5{VO₂(H₂sal-dag)}]₂. Non-hydrogen atoms were drawn using 50% probability ellipsoids. Hydrogen atoms are omitted for clarity.

Figure 4.

ORTEP of the dimeric aggregate present in the crystal packing of compound 9, [K(H₂O)₅{(V^VO₂)₂(Hsal-dag)}]₂. Non-hydrogen atoms were drawn using 50% probability ellipsoids. Hydrogen atoms are omitted for clarity.

Table 2. Bond Lengths [Å] and Angles [°] for the Compounds 3, 4, 5, and 9.

compounds	3	4	5	9
Bond Lengths
V(1)–O(1)	1.6209(19)	1.649(2)	1.853(2)		1.653(3)
V(1)–O(2)	1.6389(18)	1.632(3)	1.679(2)		1.629(3)
V(1)–O(3)	1.9356(17)	1.932(2)	1.7867(19)		1.891(3)
V(1)–N(2)	2.033(2)	2.025(3)	1.967(2)	V(1)–N(1)	2.213(3)
V(1)–N(5)	2.1339(19)	2.138(3)	2.052(2)		1.999(3)
C(8)–N(2)	1.355(3)	1.369(4)	1.283(3)	V(2)–O(4)	1.614(3)
C(8)–N(3)	1.353(3)	1.361(4)	1.334(3)	V(2)–O(5)	1.621(3)
C(8)–N(4)	1.320(3)	1.315(4)	1.360(4)	V(2)–O(6)	1.922(3)
				V(2)–N(2)	2.058(3)
				V(2)–N(4)	2.179(3)
				N(2)–C(8)	1.339(5)
				N(3)–C(8)	1.342(5)
V(1)–Cs(1)		3.8434(6)		N(5)–C(8)	1.318(5)
V(1)–K(1)	3.6219(7)		3.1975(7)
Angles
O(2)-V(1)-O(1)	107.99(10)	107.23(12)	113.52(10)		107.28(15)
O(3)-V(1)-O(1)	99.87(9)	93.92(11)	93.94(11)		97.83(13)
O(2)-V(1)-O(3)	95.13(8)	101.66(11)	107.80(8)		106.04(14)
O(1)-V(1)-N(2)	98.72(9)	94.13(11)	115.14(9)	O(1)-V(1)-N(1)	153.85(14)
O(2)-V(1)-N(2)	94.33(8)	100.15(12)	78.05(9)	O(2)-V(1)-N(1)	97.93(13)
O(3)-V(1)-N(2)	155.38(8)	153.31(11)	45.68(10)	O(3)-V(1)-N(1)	81.39(12)
O(1)-V(1)-N(5)	118.42(9)	135.74(12)	107.34(10)		92.36(14)
O(2)-V(1)-N(5)	133.10(8)	116.62(12)	138.68(11)		109.19(14)
O(3)-V(1)-N(5)	83.82(7)	83.31(10)	74.46(8)		138.34(13)

Figure 3.

ORTEP of the dimeric aggregate present in the crystal packing of compound 5, [K(H₂O) (MeOH){VO₂(H₂Brsal-dag)}]₂. Non-hydrogen atoms were drawn using 50% probability ellipsoids. Hydrogen atoms are omitted for clarity.

The geometric parameter τ = (β – α)/60, where β and α are the two largest L–M–L angles, can be used to describe the polyhedron defined by five-coordinated atoms. This parameter has a value of 1 if the structure is pentagonal bipyramidal and 0 if it is square pyramidal. For 3 (α_O3–V1–N2 = 155.38°, β_O2–V1–N5 = 133.10°), for 4 (α_O3–V1–N2 = 153.31°, β_O1–V1–N5 = 135.74°), for 5 (α_O3–V1–N2 = 145.68°, β_O2–V1–N5 = 138.68°), and for 9 [α_O1–V1–N1 = 153.85°, β_O3–V1–N5 = 138.34°) for V(1) and (α_O6–V2–N2 = 152.10°, β_O4–V2–N4 = 135.37°) for V(2)], the τ values of 0.37 in 3, 0.29 in 4, 0.12 in 5, and 0.26 for V(1) and 0.27 for V(2) in 9 indicate an intermediate position between the two geometries but closer to that of a square pyramidal polyhedron, especially in the case of compound 5.⁵⁷

Dimeric aggregate in compound 4 (Figure 2) shows lateral η²-cation−π interactions of Cs⁺ ions and carbon atoms of phenol groups in dashed black lines. The distances to carbon atoms are 3.821(4) Å to C(14) and 3.845(3) Å to C(15). In addition, hydrogen bonds and electrostatic interactions are present in the crystal packing. The ethanol molecule coordinated to the metal center prevents the formation of hydrogen bonds in compound 3. Table S2 presents the hydrogen bonds that are formed in compounds 4, 5, and 9. Dihedral angles between the phenyl rings [C(1)-C(2)-C(3)-C(4)-C(5)-C(6) and C(10)-C(11)-C(12)-C(13)-C(14)-C(15)], 10.50(16)° in compound 3, 15.58(6)° in 4, 38.22(13)° in 5, and 16.34(9)° in 9, show that the ligands are basically flat, except in the case of compound 5, which has a more distorted structure.

Figure 2.

ORTEP of the dimeric aggregate present in the crystal packing of compound 4, [Cs(H₂O){VO₂(H₂sal-dag)}]₂. Non-hydrogen atoms were drawn using 50% probability ellipsoids. Hydrogen atoms are omitted for clarity.

While the interaction of the K⁺/Cs⁺ cation in present complexes with two vanadium centers of two independent molecules of the complex forces these complexes to exit as dimer (3 and 4) or tetramer (dimers of dimes; 5 and 9), binuclear complex of ligand III, where NH₄⁺ is the counterion, does not facilitate such dimerization because NH₄⁺ cation cannot bring such two molecules together through the interaction. However, clustering in two anions occurs through weak interaction of the two sets of oxido groups on different V of one anion with two V of the second one. In other binuclear complexes of III, such clustering of two molecules (i.e., dimer of dimer) occurs through two V-(μ-O)-V bridges.^30,31

IR and UV–Visible Spectral Studies

Three main IR peaks at 3000–3450, 2900–3250, and 1651–1676 cm^–1 of ligands are worth to consider which are due to ν(O–H), ν(NH), and ν(C=N) stretches, respectively. In binuclear complexes 7–12, there is a sharp band at 1603–1616 cm^–1, while in mononuclear complexes, there are at least two such bands in the 1611–1643 cm^–1 region. The electronic current redistribution due to coordination of one of the azomethine nitrogen possibly pushes the azomethine nitrogen to appear at a lower wave number, justifying the coordination of azomethine nitrogen to V. The ν(O–H) and ν(NH) stretching region is a bit complicated in complexes, and no conclusive interpretation could be made for the phenolate O and other nitrogen(s) coordinated to V. The (O=V=O) group of complexes displays two/three bands at 911–949 and 860–917 cm^–1. Such two bands for the (O=V=O) group approve the presence of cis-[VO₂] structure in these complexes.^19a More than two bands in the ionic (Cs⁺ and K⁺ containing) complexes possibly suggest the weak interaction of such oxygen to Cs⁺ and K⁺ ions, thereby lowering some of these bands. Complexes 1, 2, 7, and 8 exhibit a sharp foot print at 955–996 cm^–1 due to the ν(V=O) stretch.

Mono- as well as binuclear complexes display one band at 400–458 nm in the visible region due to ligand-to-metal charge-transfer transition arising due to the transfer of electron density from the filled p-orbital of coordinated oxygen atoms to a vacant d-orbital of vanadium. In [V^IVO] complexes, the LMCT band was observed at ca. 400 nm (for ligands’ spectra see Figure S6). At higher concentration, a weak shoulder band centered around 650 nm has also been observed which is assigned to d–d transition (Figure 5). Further details are presented in Table S3.

Figure 5.

(a) UV–vis spectra of [V^IVO] complexes recorded in DMSO: 1 (conc. 1.25 × 10^–3 M), 2 (conc. 9.29 × 10^–4 M), 7 (conc. 1.08 × 10^–3 M), and 8 (conc. 8.06 × 10^–4 M). Inset shows corresponding spectra in the 500–800 nm range at higher concentration. 1 (conc. 1.64 × 10^–2 M), 2 (conc. 1.16 × 10^–2 M), 7 (conc. 1.35 × 10^–2 M), and 8 (conc. 1.01 × 10^–2 M). (b) UV–vis spectra of mononuclear cis-[V^VO₂] complexes recorded in DMSO: 3 (conc. 2.06 × 10^–3 M), 4 (conc. 18.90 × 10^–4 M), 5 (conc. 16.0 × 10^–4 M), and 6 (conc. 14.56 × 10^–4 M). (c) UV–vis spectra of binuclear cis-[V^VO₂] complexes recorded in DMSO: 9 (conc. 16.96 × 10^–4 M), 10 (conc. 16.36 × 10^–4 M), 11 (conc. 12.14 × 10^–4 M), and 12 (conc. 10.88 × 10^–4 M).

EPR Spectral Study

The X-band first derivative EPR spectra of complexes 1, 2, 7, and 8 were recorded in DMSO at 100 K; the EPR spectra of complexes 2 and 8 are presented in Figure 6 (for others see Figure S7). The spin Hamiltonian parameters (Table 3) were calculated for all complexes from the spectra obtained and are well within the range for normal [V^IVO] octahedral complexes having O and N coordination.⁵⁸ Considering the weak interaction of DMSO to vanadium in these complexes, a cis-form (related to axial oxygen and equatorial DMSO) seems to be more reasonable based on the high value of [A_z] (Table 3).⁵⁸ In the EPR spectra of mononuclear as well as binuclear [V^IVO] complexes, we observed eight lines without any noticeable broadening, and this confirm the presence of V(IV) center in these complexes. EPR patterns of binuclear complexes 7 and 8 further confirm that the unpaired electron of the V(IV) center is localized and not mixed with the V(V) center.

Figure 6.

X-band EPR spectrum of 2 (a) and 8 (b) recorded in DMSO at 100 K.

Table 3. Spin Hamiltonian Parameters Obtained from the EPR Spectra of Complexes 1, 2, 7, and 8 Recorded in DMSO at 100 K.

compounds	gx, gy	|Ax|, |Ay| (×10–4 cm–1)	gz	|Az| (×10–4 cm–1)
[VIVO(H2sal-dag) (H2O)] (1)	1.991	58.9	1.942	171.9
[VIVO(H2Brsal-dag) (H2O)] (2)	1.992	63.1	1.950	170.2
[(H2O)VIVO(Hsal-dag)VVO2] (7)	1.985	67.3	1.947	178.2
[(H2O)VIVO(HBrsal-dag)VVO2] (8)	1.993	67.3	1.946	175.4

¹H NMR Study

Experimental section contains ¹H NMR spectral data of ligands and complexes recorded in DMSO-d₆; spectra of H₄sal-dag·HCl (I), [K(H₂O){V^VO₂(H₂sal-dag)}]₂ (3), and [K(H₂O)₅{(V^VO₂)₂(Hsal-dag)}]₂(9) as a representative is presented in Figure 7 (for others, see Figures S8–S18). The position of the peaks, particularly assigned to NH/OH protons, was confirmed by D₂O exchange (Figure S19–S24). Thus, two protons of the hydrazide residue (=C–NH−) generally appear as a singlet at ca. 8.77 ppm. One of the two nitrogens of the hydrazide residue coordinates to vanadium after losing the proton, and the remaining =C–NH– proton still resonates as a singlet in complexes 1, 2, and 7–12 but relatively at upfield (6.47–6.84 ppm) possibly due to electronic current readjustment. However, in complexes 3–6, this proton migrates to −C=NH which converts it to −C–NH₂ and appears at ca. 6.00 ppm along with the vanishing of ca. 8.77 ppm signal. It is to be noted that the −C=NH·HCl group in the amine fragment of ligands exhibits a broad singlet at 12.16–12.30 ppm equivalent to two protons which disappears upon D₂O exchange, indicating their acidic character. The proton of −C=NH group, after losing HCl and coordinating to vanadium, appears at ca. 11.55 ppm (e.g., in complexes 9 and 10), while when it is only free from HCl, it appears at relatively upfield (10.09–10.32 ppm e.g. in 1 and 2) compared to those of complexes 9 and 10. Absence of this signal in complexes 11 and 12 hints toward complete deprotonation and tetra basic behavior of ligands to balance the charges of two K⁺/Cs⁺ ions per molecule (see Figures S13 and S14). Both ligands show a singlet at 10.30 ppm (in I) and 10.72 ppm (in II) equivalent to two protons due to phenolic OH. Absence of this signal in binuclear complexes (7–12) is in agreement with the involvement of both phenolic oxygens in coordination with vanadium. However, the existence of a phenolic signal integrating to one proton in 1–6 indicates the involvement of only one of the phenolic oxygens in coordination with vanadium after proton replacement. Similarly, the azomethine protons of ligands resonate at 8.38 (s, 2H in I) and at 8.57 (s, 2H in II), and the appearance of two such resonances of one proton each with considerable shift to down field in complexes 7–12 and the downfield shift of at least one resonance in others (i.e., in 1–6) signify that both azomethine nitrogens are coordinated to vanadium in binuclear complexes while at least one azomethine nitrogen in mononuclear complexes.

Figure 7.

¹H NMR spectra of H₄sal-dag·HCl (I), [K(H₂O){V^VO₂(H₂sal-dag)}]₂ (3), and [K(H₂O)₅{(V^VO₂)₂(Hsal-dag)}]₂ (9). Water in the formula of complexes is not shown for clarity.

¹³C NMR Study

Comparing the ¹³C NMR spectra of ligands with the corresponding complexes provides useful information. Due to symmetry in the structure of I and II, the number of ¹³C signals is relatively less than the total number of carbons, while complexes generally show signals for all carbons (Table S4). Figure 8 reproduces the spectra of H₄sal-dag·HCl (I), [K(H₂O){V^VO₂(H₂sal-dag)}]₂ (3), and [K(H₂O)₅{(V^VO₂)₂(Hsal-dag)}]₂ (9). (For other spectra, see Figures S25–S31). Coordination of metal ion to the ligands causes significant downfield shift Δδ = [δ(complex) – δ(ligand)] in the signal of those carbon atoms that are close to the coordination functionalities like phenolate O-atom(s) (i.e., C1/C1′) and two types of nitrogen atoms (i.e., C7/C7′and C8). In the case of mononuclear complexes, signals related to C1′ and C7′ are least affected; therefore, two signals for such carbons with significant difference in their positions are observed in their spectra. In other cases, normally one signal is observed for each set of carbons (C1/C1′or C7/C7′), and if there are two signals, their positions are very close to each other. Peaks due to carbons of phenyl ring also appear at the expected δ values in the spectra of ligands and complexes.

Figure 8.

¹³C NMR spectra of H₄sal-dag·HCl (I) and its vanadium complexes [K(H₂O){V^VO₂(H₂sal-dag)}]₂ (3) and [K(H₂O)₅{(V^VO₂)₂(Hsal-dag)}]₂ (9).

⁵¹V NMR Study

The ⁵¹V NMR spectra of all cis-[V^VO₂] complexes were measured in DMSO-d₆, and the data are summarized in Table 4. Figure 9 presents the ⁵¹V NMR spectra of complexes 3–6 and 9–12. The mononuclear complexes show one sharp peak at ca. −447 ppm. On the other hand, binuclear complexes 9–11 exhibit two sharp signals of almost equal intensity at much upfield that appear at a very narrow range of −559 to −562 and −565 to −572 ppm. A significant upfield shift in binuclear complexes compared to the mononuclear ones is possibly due to more electronegative character of N-donor atoms (of amine residue) relative to O-donor atoms of phenol.⁵⁹ While all ⁵¹V NMR chemical shifts are consistent with the earlier reported δ_V values for cis-[V^VO₂] complexes,⁶⁰ the presence of two signals in binuclear complexes suggests a slight asymmetry in the two vanadium centers even after delocalization of partial electronegative character of N-donor atoms of the amine residue. Accordingly, the signal at the highest upfield region may be assigned to that vanadium which is coordinated to the N-donor atom of the amine residue. These interpretations are also consistent with the crystal structures (vide supra) of particularly mononuclear complexes where coordination functionalities of other than the amine residue site of the ligands are involved in coordination to vanadium. Complex 12, however, shows only one signal at −559 ppm, showing better symmetry in the two vanadium centers in this complex. Further, the tetrabasic nature of the ligand in 11 and 12 also pushes them to be more symmetric. Thus, when we move from complex 9 to 12 through 10 and 11, the difference between two signals (in ppm) follows the trend 11(9) ≈ 11(10) > 6 (11) > 0 (12), signifying better symmetry in dianionic complexes having a bromo substituent at the ligand.

Table 4. ⁵¹V NMR Data of Vanadium Complexes Recorded in DMSO-d₆.

sr. no	compound	δ (ppm)	δ (ppm) after addition of H2O2
1	[VIVO(H2sal-dag) (H2O)] (1)		–568, −579
2	[VIVO(H2Brsal-dag) (H2O)] (2)		–568, −579
3	[(H2O)VIVO(Hsal-dag)VVO2] (7)		–567, −578
4	[[(H2O)VIVO(HBrsal-dag)VVO2] (8)		–568, −579
5	[K(H2O){VVO2(H2sal-dag)}]2 (3)	–548
6	[Cs(H2O){VVO2(H2sal-dag)}]2 (4)	–547	–547, −561, −570
7	[K(H2O){VO2(H2Brsal-dag)}]2 (5)	–547
8	[Cs(H2O){VVO2(H2Brsal-dag)}]2 (6)	–547	–547, −561, −570
9	[K(H2O)5{(VVO2)2(Hsal-dag)}]2 (9)	–561, −572	–570, −582
10	[Cs(H2O)2{(VVO2)2(Hsal-dag)}]2 (10)	–562, −573	–570, −582
11	[K2(H2O)4{(VVO2)2(Brsal-dag)}]2 (11)	–559, −565	–563, −566, −570, −583
12	[Cs2(H2O)4{(VVO2)2(Brsal-dag)}]2 (12)	–555	–564, −576

Figure 9.

⁵¹V NMR spectra of mononuclear (3–6) and binuclear cis-[V^VO₂] complexes (9–12) recorded in DMSO-d₆.

Reactivity of cis-[V^VO₂] Complexes (3–6 and 9–12) with H₂O₂: a ⁵¹V-NMR Study

The reactivity of some selected mono- and binuclear cis-[V^VO₂] complexes with H₂O₂ was also studied by recording the changes in the ⁵¹V NMR spectra of these complexes in DMSO-d₆. Addition of 10 equivalents of H₂O₂ to ca. 1.0 mM DMSO-d₆ solution of 4 resulted in no change, while 50 equivalents of H₂O₂ caused the appearance of two additional signals at −561 and −570 ppm along with the original one at −547 ppm [Figure 10a]. The additional peaks are well within the range of the peroxidovanadium(V) complexes.^61,62 Almost similar changes were observed for 6 under similar conditions (Figure S32). Similarly, complex 9 does not show any change when 10 equivalents of H₂O₂ was added to it, while addition of 50 equivalents of H₂O₂ resulted in the disappearance of the −561 ppm signal and appearance of one additional signal at −582 ppm. A slight shift in the position of −572 ppm signal to −570 ppm was also noted (Figure 10b). This clearly demonstrates the conversion of one of the dioxido groups to the corresponding oxidoperoxido group. Other binuclear complexes have similar observations (Figures S33 and S34). Interestingly, [Cs₂(H₂O)₂{V^VO₂}₂(Brsal-dag)]₂ (12) under the above conditions (i.e., in the presence of 50 equivalents of H₂O₂) shows two distinct peaks generated at −564 and −576 ppm, while the original −555 ppm peak completely vanished (Figure S35).

Figure 10.

⁵¹V NMR spectra of [Cs(H₂O){V^VO₂(H₂sal-dag)}]₂ (4) (a) and [K(H₂O)₅{(V^VO₂)₂(Hsal-dag)}]₂(9) (b) recorded in DMSO-d₆ and after addition of 30% aq. H₂O₂. Equivalents of H₂O₂ are shown along with the particular spectrum.

The oxidovanadium(IV) complexes 1, 2, 7, and 8 are stable in air as well as in the solvent and therefore are not suitable for direct ⁵¹V NMR measurement. Therefore, ⁵¹V NMR chemical shift of these complexes was measured after adding different amounts of 30% aqueous H₂O₂ to ca. 1.0 mM DMSO-d₆ solutions of these complexes. Thus, no NMR signal could be observed even after adding 20 equivalents of H₂O₂; however, upon addition of 50 equivalents of H₂O₂, these complexes show two peaks in the region −567 to −579 ppm with a major peak at −568 ppm (Figure 11). Compared to the ⁵¹V NMR resonance(s) in the corresponding [V^VO₂] complexes, these signals are much in upfield, and therefore, we tentatively assign these signals due to the corresponding oxidoperoxidovanadium(V) complexes.^21a,61

Figure 11.

⁵¹V NMR spectra of [V^IVO] complexes 1, 2, 7, and 8 recorded in DMSO-d₆ after addition of 50 equivalents of 30% aq. H₂O₂ to ca. 1.0 mM DMSO-d₆ solutions of these complexes.

Catalytic Oxidation of Benzoin to Benzil via OAT in the Presence of DMSO

Catalytic oxidation of benzoin was studied considering complex [K(H₂O){V^VO₂(H₂sal-dag)}]₂ (3) as a representative catalyst (Scheme 3). Initially, various parameters like the amount of complex 3, the volume of solvent (MeCN), the volume of DMSO, and the temperature of the reaction were optimized for the oxidation of 1.06 g (5 mmol) of benzoin. Finally, for the fixed amount of benzoin, that is, 1.06 g (5 mmol), DMSO (1 mL) and complex 3 (3.0 mg, 3.5 × 10^–3 mmol) in 10 mL MeCN were found to be the most ideal ones to carry out the reaction at 80 °C under aerobic condition to get a good yield of benzil. Thus, the decrease in the concentration of benzoin with time was monitored under above reaction conditions aerobically by HPLC using MeCN–water–trifluoroacetic acid mixture (60:40:0.02) as the eluent. The results depicted in Figure 12 show that the conversion of benzoin increases with the elapse of time and reaches 87% along with 100% selectivity toward benzil in 20 h of reaction time.

Scheme 3. Oxidation of Benzoin to Benzil via OAT Catalyzed by Vanadium Complexes.

Figure 12.

Conversion of benzoin to benzil as a function of time. Reaction conditions: benzoin (1.06 g, 5 mmol), DMSO (1 mL), Catalyst 3 (3.0 mg, 3.5 × 10^–3 mmol) and MeCN (10 mL) at 80 °C under aerobic condition. Best of three sets of reactions is considered in each case. The bars also represent the standard deviation of the mean of conversion.

Catalytic oxidation of benzoin was also performed under an argon atmosphere in the presence as well as absence of DMSO. The obtained conversions are much lower compared to the one carried out under aerobic conditions (49 vs 87% in the presence of DMSO and 41.3 vs 84.6% in the absence) (Table 5). This suggests that complex 3 can also be used as an effective catalyst for the aerobic oxidation of benzoin to benzil. Additionally, this also validates the role of DMSO in the oxidation of benzoin to benzil via OAT as ca. 8% oxidation of benzoin increased in the presence of DMSO. Interestingly, addition of 30% aqueous H₂O₂ does not help in improving the oxidation of benzoin (Table 5, entries 3, 9, and 10). A blank reaction, that is, in the absence of the catalyst, gave only 22% conversion (Table 5, entry 4). Only 3.1% conversion was noted when the reaction was conducted at room temperature under above reaction conditions, while no conversion was observed when the reaction was carried out at room temperature in the absence of the catalyst as well as DMSO.

Table 5. Catalytic Oxidation of Benzoin (1.06 g, 5 mmol) to Benzil Selectively in 20 h of Reaction Time by 3 under Different Reaction Conditions in 10 mL of MeCN.

entry no.	catalyst 3 (mg, mmol)	DMSO (mL)	medium	temp (°C)	% conversion of benzoin	TOF (h–1)
1	3.0, 3.5 × 10–3	1	aerobic	80	87	35
2	3.0, 3.5 × 10–3		aerobic	80	84.6	34
3a	3.0, 3.5 × 10–3	1	aerobic	80	86.5	35
4		1	aerobic	80	22
5	3.0, 3.5 × 10–3	1	aerobic	RT	3.1	1
6	3.0, 3.5 × 10–3		aerobic	RT	no reaction
7	3.0, 3.5 × 10–3	1	argon	80	49	20
8	3.0, 3.5 × 10–3		argon	80	41.3	17
9a	3.0, 3.5 × 10–3		argon	80	17.7	7
10a	3.0, 3.5 × 10–3		aerial	80	27.3	11

^aReactions at entries 3, 9, and 10 are in the presence of 30% aqueous H₂O₂ (1.7 g, 15 mmol).

Keeping in mind the impact of the catalyst and DMSO on the oxidation of benzoin under aerobic condition, catalytic potentials of other vanadium complexes were also checked considering the same mmol of the complexes per metal center, and the results are summarized in Table 6. In general, dioxidovanadium(V) complexes show better activity than oxidovanadium(IV) complexes. However, in spite of lower conversion obtained by oxidovanadium(IV) complexes, the influence of DMSO is much more visible in these complexes (entry 1 vs 2 of Table 6). Within dioxidovanadium(V) complexes, slightly lower conversion was experienced by the complexes that have bromo bearing on the para position of the benzene ring. Since OAT reaction involves the attack of lone pairs of sulfur on the vanadium center, the presence of the electron-donating group on the benzene ring does not favour such an attack. Though the catalytic system considered here is different from other vanadium complex-catalyzed oxidation of alcohols, these catalysts seem no better than those reported in the literature for the oxidation of alcohols.^26c A possible reason may be due to their different behaviors in solution.

Table 6. Catalytic Oxidation of Benzoin (1.06 g, 5 mmol) to Benzil with 100% Selectivity in 20 h of Reaction Time by Oxido/Dioxido–Vanadium(IV/V) Complexes (1–12).

s. no	catalyst	catalysta(mg, mmol)	DMSO (mL)	medium or atmosphere	T (°C)	% conversion	TOF (h–1)
1	1	2.6, 7.0 × 10–3	1	aerobic	80	59.4	24
2	1	2.6, 7.0 × 10–3		aerobic	80	25.7	10
3	2	3.8, 7.0 × 10–3	1	aerobic	80	65.7	26
4	4	3.7, 3.5 × 10–3	1	aerobic	80	73.7	30
5	5	4.3, 3.5 × 10–3	1	aerobic	80	56.7	28
6	6	4.8, 3.5 × 10–3	1	aerobic	80	34.7	17
7	7	1.6, 3.5 × 10–3	1	aerobic	80	68.8	55
8	8	2.2, 3.5 × 10–3	1	aerobic	80	70	56
9	9	2.1, 1.75 × 10–3	1	aerobic	80	89.6	72
10	10	2.2, 1.75 × 10–3	1	aerobic	80	69.2	55
11	11	2.7, 1.75 × 10–3	1	aerobic	80	75.8	61
12	12	3.4, 1.75 × 10–3	1	aerobic	80	45.2	36

^aFor all complexes, the catalyst amount is considered per V center.

Mechanistic Study

The role of catalyst in the OAT pathway between DMSO and benzoin is in order by the fact that only 22% conversion of benzoin could be achieved without the complex, while this conversion reached 87% in the presence of catalyst 3. It follows that DMSO partially oxidizes benzoin to benzil and itself get reduced to dimethyl sulfide. In the presence of catalyst 3 (i.e. cis-[V^VO₂] complex), dimethyl sulfide reoxidizes to DMSO and the cis-[V^VO₂] species reduces to a stable [V^IVO] species in an irreversible manner. This process continues until all dioxidovanadium(V) species is converted to a stable oxidovanadium(IV) species. The oxidovanadium(IV) species further participates in the catalytic cycle, possibly as depicted in Scheme 4. The highly unstable V^II species possibly quickly oxidizes to a stable [V^IVO] in the presence of air/DMSO (see later part of this section). The formation of dimethyl sulfide was confirmed by recording the GC–MS spectrum of the reaction mixture (m/z = 62 for DMS) (Figure S36). Since the reaction takes longer time to acquire equilibrium, the sustainability and stability of the reduced [V^IVO] species for such a long period were ascertained by monitoring the UV–visible spectral changes of the reaction mixture [i.e., benzoin (5 mmol), DMSO (1 mL), and 3 (10 mg) in 10 mL of MeCN] at 80 °C for ca. 24 h. It is clear from the spectral changes that the reaction starts within a few minutes, but a clear broad band generates at ca. 647 nm due to d–d transition in ca. 6 h of reaction time (Figure 13a. This is further supported by the fact that a sharp peak appearing at −547 ppm in the ⁵¹V NMR spectrum [Figure 13b] of complex 3 vanishes after ca. 6 h when a mixture of 3 (4.0 mg), benzoin (100 mg), and DMSO-d₆ (0.6 mL) was heated at 80 °C.

Scheme 4. Possible Reaction Pathway Involving Oxygen Transfer between Benzoin and DMSO Catalyzed by Oxidovanadium(IV) Complex.

Figure 13.

(a) UV–visible spectra of pure catalyst 3 (spectrum 1) and time-dependent spectral changes observed in the OAT reaction between benzoin and DMSO assisted by dioxidovanadium(V) complex 3 in MeCN after 6 h of reaction time: immediately after 6 h at 80 °C (2), after cooling at RT for 30 s (3), after cooling at RT for 60 s (4), after cooling at RT for 5 min (5), after cooling at RT for 6 h (6), after cooling at RT for 16 h (7), and after cooling at RT for 24 h (8). (b) ⁵¹V NMR spectra of the reaction containing benzoin (100 mg) and catalyst 3 (5.0 mg) in 0.6 mL of DMSO-d₆: (i) at time zero, (ii) after 6 h of reaction time at 80 °C, and (iii) after keeping the reaction mixture at RT for 24 h.

The EPR spectrum of the same reaction mixture (i.e., after 6 h of reaction time at 80 °C) in DMSO at 100 K exhibits eight-line spectrum [Figure 14a] {[g_x, g_y = 1.994, |A_x|, |A_y| = 55.9 × 10^–4 cm^–1; g_z = 1.961, |A_z| = 160.4 × 10^–4 cm^–1]}, suggesting the generation of a stable oxidovanadium(IV) species which participates in the catalytic cycle. We have also recorded the EPR spectrum of the same reaction mixture after keeping it at RT for 3 days, and the EPR exhibited a similar eight-line spectrum (Figure S37) confirming the formation of a stable oxovanadium(IV) species, that is, the active species involved in the catalytic cycle. It is to be noted that the EPR spectrum of a reaction mixture containing benzoin (1.06 g, 5 mmol), DMSO (1 mL), acetonitrile (10 mL). and oxidovanadium(IV) complex 1 (5.0 mg) was also recorded after heating the reaction mixture at 80 °C for 6 h, and there was no change in the EPR pattern {[g_x, g_y = 1.993, |A_x|, |A_y| = 58.2 × 10^–4 cm^–1], but we were unable to calculate the g_z and |A_z|] values because the last resonances at high magnetic field (after 3600–3700 Gauss) are too weak to be observed (Figure 14b), what was obtained with the pure complex 1 in DMSO. However, this further stresses the involvement of V(IV) complex in the possible catalytic cycle (Scheme 4).

Figure 14.

(a) EPR spectrum of the reaction mixture containing benzoin (500 mg), catalyst 3 (4.0 mg), and 1.0 mL of DMSO in 10 mL of MeCN after refluxing the reaction mixture at 80 °C for ca. 6 h. (b) EPR spectrum of the reaction mixture containing benzoin (1.06 g, 5 mmol), complex 1 (5.0 mg), and DMSO (1 mL) in 10 mL of acetonitrile after refluxing the reaction mixture at 80 °C for ca. 6 h. Both spectra were recorded at 100 K.

This mechanism is also supported by the results obtained when the catalytic oxidation of benzoin catalyzed by V^IVO(H₂sal-dag) (H₂O)] (1) was performed in the presence and absence of DMSO. The reaction in which DMSO was added gave 59.4% conversion, whereas only 25.7% conversion was achieved when no DMSO was added to the reaction mixture. Thus, DMSO facilitates the oxidation of benzoin to benzil by transferring its oxygen. Running the reaction under aerobic condition further improves the oxidation and thus also approves the role of dioxygen in the catalytic oxidation of benzoin to benzil.

Kinetic Study

We have also studied the kinetics of the decrease of the concentration of benzoin with time for catalysts 1, 3, 7, and 9. The rate equation (eq 1) for the OAT reaction can be written as

1

Assuming that the concentration of the catalyst remains largely unaffected during the reaction, eq 1 can be rewritten as eq 2-4.

2

where

3

4

The plot of 1/[benzoin]_t versus time is shown in Figure 15 which suggests that the reaction follows a second-order kinetics with respect to benzoin. The rate equation for a second-order reaction is shown in eq 5.

5

Figure 15.

Variation of the concentration of benzoin with time for the oxidation of benzoin to benzil in the presence of DMSO by vanadium complexes (1, 3, 7, and 9). Reaction conditions: benzoin (5 mmol. 1.06 g), DMSO (1 mL), MeCN (10 mL), and catalysts 1 (2.6 mg, 7.0 × 10^–3 mmol), 3 (3.0 mg, 3.5 × 10^–3 mmol), 7 (1.6 mg, 3.5 × 10^–3 mmol), and 9 (4.2 mg, 1.75 × 10^–3 mmol) at 80 °C.

The slope of the line of the plot of 1/[benzoin]_t versus time gave the corresponding rate constants for a second-order reaction (Table 7). Thus, the overall rate equation (eq 5) can be written as

6

Table 7. Kinetic Data for the OAT Reaction between Benzoin and DMSO in MeCN at 80 °C under Aerobic Conditions.

s. no	catalyst	K1 ± SEa(mol–1 L h–1)	Kobs ± SEa(mol–1 L h–1)
1.	[VIVO(H2sal-dag) (H2O)] (1)	1.04 × 10–4 ± 1.28 × 10–4
2.	[K(H2O){VVO2(H2sal-dag)}]2 (3)	7.01 × 10–4 ± 6.52 × 10–5	3.50 × 10–4 ± 6.58× 10–5
3.	[(H2O)VIVO(Hsal-dag)VVO2] (7)	1.11 × 10–4 ± 5.94 × 10–5
4.	[K(H2O)5{(VVO2)2(Hsal-dag)}]2 (9)	1.69 × 10–3 ± 8.13 × 10–5

^aStandard error.

Using the eq 4, the order of the reaction with respect to the catalyst can also be determined by calculating k₁ using different concentrations of, for example, catalyst 3. The plot of lnk₁ versus ln[catalyst] is shown in Figure 16 and the order of the reaction with respect to the catalyst was found to be 0.72 with k_obs = 3.50 × 10^–4. Therefore, the overall rate equation for the OAT reaction may be represented as

Figure 16.

Plot of lnk₁ vs ln[catalyst] for the OAT reaction between benzoin and DMSO catalyzed by 3.

Conclusions

In this paper, we have studied the coordination behavior of two polynucleating biaminoguanidine-derived Schiff base ligands H₄sal-dag (I) and H₄Brsal-dag (II) toward vanadium metal under different reaction conditions. These ligands resulted in complexes [V^IVO(H₂sal-dag) (H₂O)] (1), [V^IVO(H₂Brsal-dag) (H₂O)] (2), [K(H₂O){V^VO₂(H₂sal-dag)}]₂ (3), [Cs(H₂O){V^VO₂(H₂sal-dag)}]₂ (4), [K(H₂O){VO₂(H₂Brsal-dag)}]₂ (5), [Cs(H₂O){V^VO₂(H₂Brsal-dag)}]₂ (6), [(H₂O)V^IVO(Hsal-dag)V^VO₂] (7), [(H₂O)V^IVO(HBrsal-dag)V^VO₂] (8), [K(H₂O)₅{(V^VO₂)₂(Hsal-dag)}]₂(9), [Cs(H₂O)₂{(V^VO₂)₂(Hsal-dag)}]₂ (10), [K₂(H₂O)₄ {(V^VO₂)₂(Brsal-dag)}]₂ (11), and [Cs₂(H₂O)₄{(V^VO₂)₂(Brsal-dag)}]₂ (12) which were characterized by various techniques. They behave as dibasic tridentate ONN in 1–6 where only one vanadium(IV or V) ion is coordinated. In complexes 7 and 8, these ligands behave as binucleating and accommodate two heterovalent vanadium(IV and V) ions. With the accommodation of two homovalent vanadium(V) ions, these ligands have tribasic (ONN)₂ nature in 9 and 10, while bis(dibasic ONN) in 11 and 12. Interaction of the K⁺/Cs⁺ cation in complexes with two vanadium centers of two independent molecules of the complex forces 3 and 4 to exit as dimers and 5 and 9 as tetramers (dimers of dimers), but this dimerization is entirely different form the vanadium complexes of ligand III reported in the literature³¹ However, it is believed that such dimerization should not exist in solution as bond distances associated with the K⁺/Cs⁺ cation with vanadium are relatively long. Evidenced by NMR and SC-XRD, the coordination functionalities (ONN) of the nonamine residue site of the ligands are involved in the coordination to vanadium in mononuclear V(IV) and V(V) complexes. ⁵¹V NMR study confirms slight nonsymmetric nature of ligands in 9–11, while 12 is symmetric in nature. The difference between two signals (in ppm) follows the order 11(9) ≈ 11(10) > 6 (11) > 0 (12). While all V(IV) complexes give eight-line EPR spectra, the oxidation of V(IV) as well as V(V) complexes, upon reaction with H₂O₂, changes them to the corresponding peroxidovanadium(V) complexes as confirmed by ⁵¹V NMR studies. Thus, these ligands behave as mono-, bi-, tri-, and tetra-anionic toward vanadium under different reaction conditions. Such diverse behavior of these ligands is rarely seen in other metal ions.

A limited study has been reported on the catalytic potential of oxidovanadium(IV/V) complexes in the oxidation of benzoin to benzyl and that too under aerobic conditions. The vanadium complexes reported here are effective catalysts for the oxidation of benzoin to benzil via OAT between DMSO and benzoin. Due to the OAT between DMSO and benzoin, the reaction also works under anaerobic conditions. Independent of the catalysts used, a stable oxidovanadium(IV) intermediate is the proposed active species involved in the catalytic cycle which is based on the fact that the intermediate compound shows eight-line EPR spectrum very similar to those exhibited by oxidovanadium(IV) complexes 1, 2, 7, and 8. This is further supported by the fact that 3 loses its ⁵¹V NMR signal during the catalytic reaction and remain ⁵¹V NMR silent even after several hours of catalytic reaction. Such identification of the V(IV) catalytic intermediate during the OAT reaction between DMSO and benzoin has not been observed earlier.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.2c06732.

• These data can be obtained free of charge via http://www.ccdc.cam.ac.uk/conts/retrieving.html or from the Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: (+44) 1223 336 033; or e-mail: deposit@ccdc.cam.ac.uk (PDF)

• Crystallographic data for 3 (CIF)

• Crystallographic data for 4 (CIF)

• Crystallographic data for 5 (CIF)

• Crystallographic data for 9 (CIF)

The authors declare no competing financial interest.