Peroxidative Oxidation of Alkanes and Alcohols under Mild Conditions by Di- and Tetranuclear Copper (II) Complexes of Bis (2-Hydroxybenzylidene) Isophthalohydrazide

Bis(2-hydroxybenzylidene)isophthalohydrazide (H4L) has been used to synthesize the dinuclear [Cu2(1κNO2:2κN′O′2-H2L)(NO3)2(H2O)2] (1) and the tetranuclear [Cu4(μ-1κNO2:2κN′O2-H2L)2(μ-NO3)2(H2O)4]·2C2H5OH (2) complexes. The solvent plays an important role in determining the ligand behaviour in the syntheses of the complexes. An ethanol-acetonitrile mixture of solvents favours partials enolization in the case of 2. Both complexes have been characterized by elemental analysis, infrared radiation (IR), single crystal X-ray crystallography and electrochemical methods. The variable temperature magnetic susceptibility measurements of 2 show strong antiferromagnetic coupling between the central nitrato-bridged Cu (II) ions. The catalytic activity of both 1 and 2 has been screened toward the solvent-free microwave-assisted oxidation of alcohols and the peroxidative oxidation of alkanes under mild conditions. Complex 1 exhibits the highest activity for both oxidation reactions, leading selectively to a maximum product yield of 99% (for the 1-phenylethanol oxidation after 1 h without any additive) and 13% (for the cyclohexane oxidation to cyclohexyl hydroperoxide, cyclohexanol and cyclohexanone after 3 h).


Introduction
The oxidation of alcohols to carbonyl-containing compounds [1][2][3] is one of the fundamental reactions in organic synthesis [4,5], with a great interest due to the application of carbonyl compounds in research and industrial manufacturing, e.g., in the production of new materials [6] and energy sources [7]. In view of their central role in synthetic chemistry and expected further applications, these reactions continue to attract great attention in order to develop environmentally benign processes [8,9] disclosing new efficient catalysts [10,11], substrates or oxidation agents which could successfully be used in the near future and make a difference in terms of efficiency, selectivity, economy and/or sustainability of the processes. The metal-catalyzed aerobic and peroxidative oxidations of alcohols, in particular of benzylic alcohols, are typical model reactions due to their importance and generality; inexpensive primary oxidants such as O 2 , H 2 O 2 or tert-butylhydroperoxide (TBHP) and simple procedures are usually explored [1][2][3][4][5]. The accelerating effect of microwave (MW) irradiation in the synthesis of ketones from secondary alcohols with TBHP as an oxidant has been largely reported [12][13][14][15][16][17][18]; this technology is a useful alternative source in organic synthesis, with an environmentally friendly nature.
It has been shown that copper (II) complexes are useful catalysts towards the MW assisted oxidation of alcohols and also the functionalization of inert alkanes into valuable organic products, using readily available and cheap oxidants [22,[25][26][27][28]; however, such applications are still limited and the subject requires further exploration. Moreover, aroylhydrazone complexes are potentially important as oxidation catalyst and some of them also showed interesting magnetic properties [29][30][31][32].
Continuing our research on the syntheses of aroylhydrazone metal complexes and their application in catalysis, herein we report the syntheses and characterization of new dinuclear and tetranuclear Cu(II) complexes derived from the aroylhydrazone Schiff base bis (2-hydroxybenzylidene) isophthalohydrazide (H 4 L), their electrochemical behaviour and magnetic properties (2), as well as their catalytic activity in the solvent-free MW assisted peroxidative oxidation of alcohols and also in the oxidation of alkanes under mild conditions, towards the development of environmentally friendly catalytic systems.

Syntheses and Characterizations
The aroylhydrazone Schiff base bis(2-hydroxybenzylidene) isophthalohydrazide (H 4 L) has two potential tridentate coordination pockets which can bind two metal centres simultaneously. It has previously been observed [32][33][34] that the aroylhydrazone can act as ligand in two tautomeric (keto or enol) forms. At room temperature, deprotonation of the phenolic -OH group from the aroylhydrazone in the presence of a copper (II) source, in methanol, generates the H 2 L 2− species and leads to the stable dinuclear complex 1 (Scheme 1). When the reaction is carried out in a 2:1 acetonitrile-ethanol solvent mixture, part of the aroylhydrazone remains in the keto form, the remaining undergoing enolization and deprotonation during complexation. As a result, a HL 3− species is formed in solution and provides the tetranuclear complex 2 (Scheme 1). This compound 2 can be considered as a dimer of 1, where the two central Cu (II) ions are bridged by two nitrate ions and one of the phenoxido groups of the organic ligand. The IR spectra of complexes 1 and 2 contain all the characteristic bands of the corresponding coordinated tridentate anionic ligand viz., 3476, 3024, 1608, 1254 and 1159 cm −

General Description of the Crystal Structures
X-ray low quality (low diffracting) crystals of 1 and 2 were obtained upon slow evaporation of a methanolic (for 1) or an acetonitrile-ethanol solution mixture (for 2), at room temperature. Crystallographic data are summarized in Table 1, representative plots are displayed in Figures 1 and  2, and selected dimensions are presented in Table 2. Cif files for Compound 1 and Compound 2 are available as supporting information for this paper.
The organic ligand in 1 and 2 is almost planar, behaves as a hexadentate chelator and acts as a NO2 donor for each metal cation in 1 and for the outer ones in 2, while for the inner metals it behaves as an Ophenolate-bridging species as well. The copper cations exhibit square pyramidal [Cu1: τ5 = 0.08 (1) and 0.06 (2)], distorted penta-coordinate (Cu2 in 1, τ5 = 0.38) or distorted octahedral geometries (Cu2 in 2; quadratic elongation = 1.078, angle variance = 77.60 º 2 ). The relative orientation of one of the nitrate ligands of Cu1 in 1 caused a measured Cu-Onitrate bond distance of 2.69(1) Å which is similar to those of 2.680 (7) and 2.709 (5) Å in 2, well below the sum on the van der Waals radii of copper and oxygen (1.40 and 1.52 Å, respectively). Therefore, on this basis, the geometry of that cation in 1 may be thought as octahedral, although with a great distortion (quadratic elongation = 1.125, angle variance = 272.62 º 2 ).
Compound 2 can be considered as a dimer of 1, where dimerization took place at the level of the metals bearing nitrates, leading to a central four-membered dicopper metalacycle. This process affected not only the dimensions involving the central copper cations (Cu(nitrate), see Table 2), but also the outer ones with water ligands bound to copper (Cu(water), see Table 2), which is probably related to the observed change of geometry (see above the values of τ5 descriptor for such metals). Thus, from 1 to 2, the Cu-Ophenolate increased in both types of metals, but the Cu-Oketone increased only in the metal with water ligands, and decreased slightly in the other. Concerning the Cu-Owater lengths, they differ markedly in 2 with the apical much longer than the equatorial one; in 1 the dissimilarity is not so significant, but still the longer distance pertains to the water molecule more distant from the least-square plane of the molecule. An influence has also been perceived in the Cu···Cu lengths; in the molecule of 2 the distance between the Cu(nitrate) and the Cu(water) is more than 0.100 Å shorter than in 1, but the shortest intermolecular distance between the metal cations is higher in the former. The effect on selected O-Cu-O and O-Cu-N angles in 1 and 2 (Table 2) is most probably related to the aforementioned differences in geometry.
Moreover, the enolate forms are more strongly coordinated to the metal centres with shorter M-O bond distances than the corresponding ones in the keto forms, as observed in other cases [32][33][34]

General Description of the Crystal Structures
X-ray low quality (low diffracting) crystals of 1 and 2 were obtained upon slow evaporation of a methanolic (for 1) or an acetonitrile-ethanol solution mixture (for 2), at room temperature. Crystallographic data are summarized in Table 1, representative plots are displayed in Figures 1  and 2, and selected dimensions are presented in Table 2. Cif files for Compound 1 and Compound 2 are available as Supporting Information for this paper.
The organic ligand in 1 and 2 is almost planar, behaves as a hexadentate chelator and acts as a NO 2 donor for each metal cation in 1 and for the outer ones in 2, while for the inner metals it behaves as an O phenolate -bridging species as well. The copper cations exhibit square pyramidal [Cu1: τ 5 = 0.08 (1) and 0.06 (2)], distorted penta-coordinate (Cu2 in 1, τ 5 = 0.38) or distorted octahedral geometries (Cu2 in 2; quadratic elongation = 1.078, angle variance = 77.60

General Description of the Crystal Structures
X-ray low quality (low diffracting) crystals of 1 and 2 were obtained upon slow evaporation of a methanolic (for 1) or an acetonitrile-ethanol solution mixture (for 2), at room temperature. Crystallographic data are summarized in Table 1, representative plots are displayed in Figures 1 and  2, and selected dimensions are presented in Table 2. Cif files for Compound 1 and Compound 2 are available as supporting information for this paper.
The organic ligand in 1 and 2 is almost planar, behaves as a hexadentate chelator and acts as a NO2 donor for each metal cation in 1 and for the outer ones in 2, while for the inner metals it behaves as an Ophenolate-bridging species as well. The copper cations exhibit square pyramidal [Cu1: τ5 = 0.08 (1) and 0.06 (2)], distorted penta-coordinate (Cu2 in 1, τ5 = 0.38) or distorted octahedral geometries (Cu2 in 2; quadratic elongation = 1.078, angle variance = 77.60 º 2 ). The relative orientation of one of the nitrate ligands of Cu1 in 1 caused a measured Cu-Onitrate bond distance of 2.69(1) Å which is similar to those of 2.680 (7) and 2.709 (5) Å in 2, well below the sum on the van der Waals radii of copper and oxygen (1.40 and 1.52 Å, respectively). Therefore, on this basis, the geometry of that cation in 1 may be thought as octahedral, although with a great distortion (quadratic elongation = 1.125, angle variance = 272.62 º 2 ).
Compound 2 can be considered as a dimer of 1, where dimerization took place at the level of the metals bearing nitrates, leading to a central four-membered dicopper metalacycle. This process affected not only the dimensions involving the central copper cations (Cu(nitrate), see Table 2), but also the outer ones with water ligands bound to copper (Cu(water), see Table 2), which is probably related to the observed change of geometry (see above the values of τ5 descriptor for such metals). Thus, from 1 to 2, the Cu-Ophenolate increased in both types of metals, but the Cu-Oketone increased only in the metal with water ligands, and decreased slightly in the other. Concerning the Cu-Owater lengths, they differ markedly in 2 with the apical much longer than the equatorial one; in 1 the dissimilarity is not so significant, but still the longer distance pertains to the water molecule more distant from the least-square plane of the molecule. An influence has also been perceived in the Cu···Cu lengths; in the molecule of 2 the distance between the Cu(nitrate) and the Cu(water) is more than 0.100 Å shorter than in 1, but the shortest intermolecular distance between the metal cations is higher in the former. The effect on selected O-Cu-O and O-Cu-N angles in 1 and 2 ( Table 2) is most probably related to the aforementioned differences in geometry.
Moreover, the enolate forms are more strongly coordinated to the metal centres with shorter M-O bond distances than the corresponding ones in the keto forms, as observed in other cases [32][33][34].
). The relative orientation of one of the nitrate ligands of Cu1 in 1 caused a measured Cu-O nitrate bond distance of 2.69(1) Å which is similar to those of 2.680 (7) and 2.709 (5) Å in 2, well below the sum on the van der Waals radii of copper and oxygen (1.40 and 1.52 Å, respectively). Therefore, on this basis, the geometry of that cation in 1 may be thought as octahedral, although with a great distortion (quadratic elongation = 1.125, angle variance = 272.62 of the Crystal Structures (low diffracting) crystals of 1 and 2 were obtained upon slow evaporation of a an acetonitrile-ethanol solution mixture (for 2), at room temperature. are summarized in Table 1, representative plots are displayed in Figures 1 and ions are presented in Table 2. Cif files for Compound 1 and Compound 2 are information for this paper. d in 1 and 2 is almost planar, behaves as a hexadentate chelator and acts as a tal cation in 1 and for the outer ones in 2, while for the inner metals it behaves species as well. The copper cations exhibit square pyramidal [Cu1: τ5 = 0.08 rted penta-coordinate (Cu2 in 1, τ5 = 0.38) or distorted octahedral geometries ongation = 1.078, angle variance = 77.60 º 2 ). The relative orientation of one of Cu1 in 1 caused a measured Cu-Onitrate bond distance of 2.69(1) Å which is 0 (7) and 2.709 (5) Å in 2, well below the sum on the van der Waals radii of .40 and 1.52 Å, respectively). Therefore, on this basis, the geometry of that ught as octahedral, although with a great distortion (quadratic elongation = 272.62 º 2 ). be considered as a dimer of 1, where dimerization took place at the level of the s, leading to a central four-membered dicopper metalacycle. This process imensions involving the central copper cations (Cu(nitrate), see Table 2), but also ter ligands bound to copper (Cu(water), see Table 2), which is probably related e of geometry (see above the values of τ5 descriptor for such metals). Thus, enolate increased in both types of metals, but the Cu-Oketone increased only in the ds, and decreased slightly in the other. Concerning the Cu-Owater lengths, they ith the apical much longer than the equatorial one; in 1 the dissimilarity is not l the longer distance pertains to the water molecule more distant from the he molecule. An influence has also been perceived in the Cu···Cu lengths; in istance between the Cu(nitrate) and the Cu(water) is more than 0.100 Å shorter than termolecular distance between the metal cations is higher in the former. The u-O and O-Cu-N angles in 1 and 2 ( Table 2) is most probably related to the nces in geometry. olate forms are more strongly coordinated to the metal centres with shorter than the corresponding ones in the keto forms, as observed in other cases ). Compound 2 can be considered as a dimer of 1, where dimerization took place at the level of the metals bearing nitrates, leading to a central four-membered dicopper metalacycle. This process affected not only the dimensions involving the central copper cations (Cu (nitrate) , see Table 2), but also the outer ones with water ligands bound to copper (Cu (water) , see Table 2), which is probably related to the observed change of geometry (see above the values of τ 5 descriptor for such metals). Thus, from 1 to 2, the Cu-O phenolate increased in both types of metals, but the Cu-O ketone increased only in the metal with water ligands, and decreased slightly in the other. Concerning the Cu-O water lengths, they differ markedly in 2 with the apical much longer than the equatorial one; in 1 the dissimilarity is not so significant, but still the longer distance pertains to the water molecule more distant from the least-square plane of the molecule. An influence has also been perceived in the Cu···Cu lengths; in the molecule of 2 the distance between the Cu (nitrate) and the Cu (water) is more than 0.100 Å shorter than in 1, but the shortest intermolecular distance between the metal cations is higher in the former. The effect on selected O-Cu-O and O-Cu-N angles in 1 and 2 ( Table 2) is most probably related to the aforementioned differences in geometry.

Electrochemical Properties
Cyclic voltammograms of 1 and 2 exhibit a first single-electron (per metal atom) irreversible reduction process (wave I red ) at I E p red −0.27 or −0.43 V vs. SCE, for 1 or 2, respectively, followed, at a lower potential, by a second single-electron (per metal atom) reduction (wave II red ) at II E p red = −1.12 or −1.08 V vs. SCE, for 1 or 2, respectively, which are believed to correspond to the Cu II → Cu I (wave I red ) and Cu I → Cu 0 (wave II red ) cathodic processes, in this order. Upon scan reversal following the first reduction wave, an irreversible oxidation at I E p ox −0.05 or 0.02 V vs. SCE, for 1 or 2, respectively, was detected and assigned to the oxidation of a novel Cu(I) species. By reversing the direction of the potential scan after the formation of wave II red , an adsorption wave was formed at II E p ox 0.15 or 0.27 V vs. SCE for 1 or 2, respectively.
Expectedly, no genuine anodic waves have been detected for any of the complexes. Bis(2-hydroxybenzylidene) isophthalohydrazide (H 4 L) was not redox active under our experimental conditions, and hence the cyclic voltammetric waves of 1 and 2 can be due to metal-based electron transfer processes.

Magnetic Properties
The dc magnetic susceptibilities of complex 2 have been determined under 2000 Oe from 2 K to 300 K. As shown in Figure 3, the χ M T value at 300 K is 1.48 cm 3 K mol −1 , slightly smaller than the expected value for four noninteracting Cu 2+ ions (g = 2.0). The χ M T product decreases with the decreased temperture until about 100 K, then drops slowly to about 10 K; below this temperature, χ M T further decreases. These results indicate that there are antiferromagnetic interactions between copper(II) ions.
A linear tetranuclear copper (II) magnetic coupling model was adopted to analyze the magnetic interaction [35]. Owing to the crystallographic inversion symmetry, four exchange coupling constants J 1 , J 2 , J 3 and J 4 were used in the Hamiltonian, H = −2J 1 (S 1 S 2 +S 1 S 2 ) − 2J 2 S 2 S 2 − 2J 3 (S 1 S 2 +S 1 S 2 ) − 2J 4 (S 1 S 1 ) [35]; they represent the magnetic interactions between Cu1 and Cu2 (or Cu1 i and Cu2 i ), Cu2 and Cu2 i , Cu1 and Cu2 i (or Cu1 i and Cu2), and between Cu1 and Cu1 i , respectively. Given the high number of variables, fitting the magnetic data using all four coupling constants led to unreasonable results. Since the distance between Cu1 and Cu2 i (or Cu1 i and Cu2) and the distance between Cu1 and Cu1 i are large enough, J 3 and J 4 could be considered as zero for a simplified model, and better results were thus achieved. The best fitting gave g = 2.02, 2 ). The large negative value of J 2 reveals that there is a strong antiferromagnetic exchange between the Cu (II) ions mediated by the O phenolate and NO 3 bridges.
magnetic interaction [35]. Owing to the crystallographic inversion symmetry, four exchange coupling constants J1, J2, J3 and J4 were used in the Hamiltonian, H = −2J1(S1S2+S1′S2′) − 2J2 S2S2′ − 2J3(S1S2′+S1′S2) − 2J4(S1S1′) [35]; they represent the magnetic interactions between Cu1 and Cu2 (or Cu1 i and Cu2 i ), Cu2 and Cu2 i , Cu1 and Cu2 i (or Cu1 i and Cu2), and between Cu1 and Cu1 i , respectively. Given the high number of variables, fitting the magnetic data using all four coupling constants led to unreasonable results. Since the distance between Cu1 and Cu2 i (or Cu1 i and Cu2) and the distance between Cu1 and Cu1 i are large enough, J3 and J4 could be considered as zero for a simplified model, and better results were thus achieved. The best fitting gave g = 2.02, J1 = -5.  Complexes 1 and 2 were tested as catalyst precursors for the homogeneous oxidation of secondary alcohols [1-phenylethanol (model substrate), cyclohexanol, 2-hexanol and 3-hexanol] to the corresponding ketones following our previously developed procedure [18,31,[36][37][38][39] using tert-butylhydroperoxide (t-BuOOH, Equation (2), used as aq. 70%) as oxidizing agent, under typical conditions of 80-120 °C, microwave (MW) irradiation (5-20 W), 0.5-3 h reaction time and in the absence of any added solvent (Scheme 2 for 1-phenylethanol and cyclohexanol oxidation). Results are summarized in Tables 3 and 4.  Tables 3 and 4. Under typical reaction conditions (120 °C and 0.5 h reaction time) ketone yields up to 37% (TOF = 264 h −1 ) are obtained for the oxidation of 1-phenylethanol (Table 3, entry 3) by the 1/TBHP/MW system (catalyst/substrate molar ratio of 0.2%) and in the absence of any additive. For a longer reaction time of 1 and 3 h, and for the same catalytic system, the oxidation by TBHP of 1-phenyletanol leads to 75 and 95% of acetophenone, respectively (  Under typical reaction conditions (120 • C and 0.5 h reaction time) ketone yields up to 37% (TOF = 264 h −1 ) are obtained for the oxidation of 1-phenylethanol (Table 3, entry 3) by the 1/TBHP/MW system (catalyst/substrate molar ratio of 0.2%) and in the absence of any additive. For a longer reaction time of 1 and 3 h, and for the same catalytic system, the oxidation by TBHP of 1-phenyletanol leads to 75 and 95% of acetophenone, respectively (Table 3, entries 4 and 5). The acetophenone yield value for 1 h reaction (Table 3, entry 4) is significantly higher than that obtained (17%, Table 3, entry 15) without MW assistance, for the same reaction time. Reactions performed at 120 • C with 20 W of MW irradiation and in the presence of complex 2 (2/TBHP/MW) resulted only in 28, 42 and 71% of acetophenone after 0.5, 1 and 3 h of oxidation reaction, respectively (Table 3, entries 16-18, respectively). The addition of the heteroaromatic 2-pyrazynecarboxylic acid (Hpca) or of trifluoroacetic acid (TFA) to the 1/TBHP/MW system [n (acid)/n (catalyst 1) = 10], slightly improved the yield from 11 to 14% (Table 3, entry 6) or to 18% (Table 3, entry 9), respectively, when the reaction was performed at 80 • C for 0.5 h.
Attempts to perform microwave-assisted oxidation of secondary alcohols in the presence of 1 or 2, in the absence of any additive and at room temperature, failed. Performing the reaction at 80 • C did not allow yields beyond 27% in 3 h (with catalyst 1), but at 120 • C the product was obtained in 95 (with 1) or 71% yield (with 2), in the same reaction time (Table 3, entries 5 and 20). The accelerating effect of the increase in temperature was also observed in the presence of the tested additives ( Figure 4). Thus, for example, in the presence of the TEMPO radical, one observes a change in the conversion of 1-phenylethanol from 22 to 88% and in the TOF value from 168 to 1112 h −1 ( Table 3, entries 12 and 13) when going from 80 or 120 • C, in 0.5 h. This feature also concerns the microwave power since at 80 • C the power went up to 10 W in the first 10 s but then, after reaching the desired temperature, it stabilized at ca. 5 W, while for the temperature of 120 • C, ca. 40 W were reached in the first 10 s followed by stabilisation at 10-15 W for the remaining time.  Complexes 1 and 2 were also tested towards the oxidation of aliphatic alcohols, namely cyclohexanol and the linear 2-and 3-hexanols. As expected, they were less reactive than 1-phenylethanol (benzylic alcohol derivative), leading to moderate yields in the 15-31% range (in 0.5 h reaction at 120 °C, without any additive (Table 4,    Complexes 1 and 2 were also tested towards the oxidation of aliphatic alcohols, namely cyclohexanol and the linear 2-and 3-hexanols. As expected, they were less reactive than 1-phenylethanol (benzylic alcohol derivative), leading to moderate yields in the 15-31% range (in 0.5 h reaction at 120 • C, without any additive (Table 4, entries 1, 4, 7, 10, 13 and 16). The efficiency of the oxidation of those alcohols could also be enhanced by using TEMPO radical and the yields increased to 29-58% in 0.5 h (Table 4, entries 2, 5, 8, 11, 14 and 17). Extending the reaction time to 1 h and in the presence of the same promoter, the conversions raised to 87% for cyclohexanol (Table 4, entry 3), 71% for 2-hexanol (Table 4, entry 6) and 59% for 3-hexanol (Table 4, entry 9) with 1 as catalyst. The position (2 or 3) of the OH group in the aliphatic chain of the linear alcohols (2-hexanol and 3-hexanol) appeared not to influence significantly the efficiency of the system, in view of the obtained yields (compare, e.g., entries 4 and 7 or 6 and 9, Table 4, for complex 1). Blank tests were performed for all aliphatic alcohols in the absence of any catalyst and residual conversions (up to 5%) were recorded. The catalytic mechanism may proceed through the metal-assisted generation of t-BuOO and t-BuO radicals [40][41][42][43][44][45][46][47][48], upon oxidation or reduction of t-BuOOH by a Cu II or Cu I centre (Equations (1) and (2)), respectively, and is summed up in Equations (1)- (6).

Peroxidative Oxidation of Cyclohexane
Compounds 1 and 2 were also tested as catalyst precursors in the oxidation of cyclohexane by H2O2 (Equation (2), used as 50% aqueous solution) at 50 °C in MeCN/H2O medium (Scheme 3). The reaction was monitored by Gas Chromatography (GC) to determine the amount of cyclohexanol and cyclohexanone formed, typically after treatment with PPh3 (to reduce cyclohexyl hydroperoxide to cyclohexanol) [29,30,[49][50][51]. The accumulation of oxygenated products (cyclohexanol and cyclohexanone) in the cyclohexane oxidation catalysed by 1 and 2, in the absence and in the presence of TFA, is given in Table 5. Both complexes were active for this reaction in the absence of any additive ( Figure 5) with approximately 7 and 5% (for 1 and 2, respectively) of total product yield after ca. 0.5 h (Table 5, entries 3 and 15, for 1 and 2, respectively). Further increase of the reaction time to 2 h slightly Scheme 3. Oxidation of cyclohexane to cyclohexyl hydroperoxide, cyclohexanol and cyclohexanone.
The accumulation of oxygenated products (cyclohexanol and cyclohexanone) in the cyclohexane oxidation catalysed by 1 and 2, in the absence and in the presence of TFA, is given in Table 5. Both complexes were active for this reaction in the absence of any additive ( Figure 5) with approximately 7 and 5% (for 1 and 2, respectively) of total product yield after ca. 0.5 h (Table 5, entries 3 and 15, for 1 and 2, respectively). Further increase of the reaction time to 2 h slightly improved the total yield of the products in the presence of 1 and almost had not effect when 2 was the catalyst. The presence of TFA improved the catalytic performance of both copper compounds, more markedly for 1 with the total yield of cyclohexanol and cyclohexanone achieving ca. 13% in 1 h (Table 5, entry 11). The promoting effect of an acid co-catalyst was already observed for other Cu-catalysed systems in the oxidative transformation of alkanes and can be related to (i) its involvement in proton transfer steps, (ii) catalyst activation by unsaturation of the Cu (II) centres upon ligand protonation or to the (iii) facilitation of the formation of peroxo complexes [27,[52][53][54][55][56].
The activity exhibited by compounds 1 and 2, even in the absence of TFA, is higher than that shown, e.g., by [Cu(OTf) 2 (Py 2 S 2 )] (Py 2 S 2 = 1,6-bis(2 -pyridyl)-2,5-dithiahexane) (with 4.3% overall yield) [57] and is comparable to those of the complexes bearing azathia macrocycles, e.g., [Cu(OTf) 2 (L 3 )] (L 3 = mixed 14-membered N 2 S 2 azathia macrocycle) or [Cu(OTf)(L 4 )(H 2 O)](OTf) (L 4 = nine-membered NS 2 macrocyclic ligand with a 2-methylpyridyl pendant arm) (overall yield of ca. 8%) [58].  The formation of cyclohexyl hydroperoxide CyOOH was confirmed by GC-MS methods and accounts for a free radical reaction mechanism, which conceivably involves the presence of oxygen-centered radicals, HOO • and HO•, very reactive species, formed upon reaction of the catalyst with hydrogen peroxide (Equations (7) and (8)) [50,[59][60][61][62] and enhanced by the presence of TEMPO radical. The HO• radical abstracts hydrogen from cyclohexane CyH to produce cyclohexyl radical Cy• (Equation (9)), which is then trapped by dioxygen to give CyOO• radical (Equation (10)). The latter may react with the oxidant to form CyOOH (Equation (11)). Metal-assisted decomposition of CyOOH to CyO• and CyOO• (Equations (12) and (13)) would then lead to cyclohexanol (CyOH) and cyclohexanone (Cy-H=O) products (Equations (14) and (15) The coordination of the oxidant to the metal centre is dependent on the Lewis acidity of the latter. The more Lewis acid metal centre should be easier to reduce. To correlate the electrochemical and catalytic behaviours, the reduction potentials of both complexes were determined by cyclic voltammetry (see above). In fact, complex 1, the most active one in both oxidation reactions, is easier to reduce than 2 (reduction at a less negative potential, i.e., I Ep red -0.27 V versus −0.43 V vs. SCE). The formation of cyclohexyl hydroperoxide CyOOH was confirmed by GC-MS methods and accounts for a free radical reaction mechanism, which conceivably involves the presence of oxygen-centered radicals, HOO • and HO•, very reactive species, formed upon reaction of the catalyst with hydrogen peroxide (Equations (7) and (8)) [50,[59][60][61][62] and enhanced by the presence of TEMPO radical. The HO• radical abstracts hydrogen from cyclohexane CyH to produce cyclohexyl radical Cy• (Equation (9)), which is then trapped by dioxygen to give CyOO• radical (Equation (10)). The latter may react with the oxidant to form CyOOH (Equation (11)). Metal-assisted decomposition of CyOOH to CyO• and CyOO• (Equations (12) and (13)) would then lead to cyclohexanol (CyOH) and cyclohexanone (Cy-H=O) products (Equations (14) and (15) The coordination of the oxidant to the metal centre is dependent on the Lewis acidity of the latter. The more Lewis acid metal centre should be easier to reduce. To correlate the electrochemical and catalytic behaviours, the reduction potentials of both complexes were determined by cyclic voltammetry (see above). In fact, complex 1, the most active one in both oxidation reactions, is easier to reduce than 2 (reduction at a less negative potential, i.e., I E p red −0.27 V versus −0.43 V vs. SCE).

General Materials and Equipment
All synthetic work was performed in air. Commercially available reagents and solvents were used as received, without further purification or drying. Cu(NO 3 ) 2 ·5H 2 O was used as a metal source for the synthesis of the complexes.
C, H, and N elemental analyses were carried out by the Microanalytical Service of Instituto Superior Técnico. Infrared spectra (4000-400 cm -1 ) were recorded on a Bruker Vertex 70 instrument in KBr pellets; wavenumbers are in cm -1 . The 1 H NMR spectra were recorded at room temperature on a Bruker Avance II + 400.13 MHz (UltraShieldTM Magnet, Rheinstetten, Germany) spectrometer. Tetramethylsilane was used as the internal reference and the chemical shifts are reported in ppm. Mass spectra were run in a Varian 500-MS LC Ion Trap Mass Spectrometer equipped with an electrospray (ESI) ion source. For electrospray ionization, the drying gas and flow rate were optimized according to the particular sample with 35 p.s.i. nebulizer pressure. Scanning was performed from m/z 100 to 1200 in methanol solution. The compounds were observed in the positive mode (capillary voltage = 80-105 V). The catalytic tests were performed under microwave (MW) irradiation using a focused Anton Paar Monowave 300 microwave fitted with a rotational system and an IR temperature detector, using a 10 mL capacity reaction tube with a 13 mm internal diameter. Gas chromatographic (GC) measurements were carried in a FISONS Instrument GC 8000 series gas chromatograph with a capillary DB-WAX column (30 m × 0.32 mm), a FID detector, helium as the carrier gas and using the Jasco-Borwin v.1.50 software. The magnetic susceptibility measurements were carried out on polycrystalline samples with a Quantum Design MPMS-XL5 SQUID magnetometer in the temperature range of 2-300 K and at an applied field of 2000 Oe. Diamagnetic corrections were estimated from Pascal's constants for all constituent atoms [63]. The electrochemical experiments were performed in an EG&G PAR 273A potentiostat/galvanostat connected to a personal computer through a GPIB interface.

X-Ray Measurements
Good quality single crystals suitable for X-ray diffraction of 1 and 2 were immersed in cryo-oil, mounted in Nylon loops and measured at a temperature of 296 K. Intensity data were collected using a Bruker AXS PHOTON 100 diffractometer with graphite monochromated Mo-Kα (λ 0.71073) radiation. Data were collected using omega scans of 0.5 • per frame and full spheres of data were obtained. Cell parameters were retrieved using Bruker SMART [66] software and refined using Bruker SAINT [66] on all the observed reflections. Absorption corrections were applied using SADABS [66]. Structures were solved by direct methods by using SIR97 [67] and refined with SHELXL2014 [68]. Calculations were performed using WinGX version 2014.1 [69]. All non-hydrogen atoms were refined anisotropically. The H-atoms bonded to carbons were included in the model at geometrically calculated positions and refined using a riding model. U iso (H) values were defined as 1.2U eq of the parent aromatic and methylene groups and 1.5U eq of the parent methyl ones. The other hydrogen atoms (O-H and N-H) were located in the difference Fourier synthesis and refined, in some cases with the help of distance and angle restraints, their isotropic thermal parameter set at 1.5 times the average thermal parameter of the parent oxygen or nitrogen atom. The structure of 1 appears to have three water molecules per unit cell, but their hydrogen atoms could not be ascertained and the Calc-OH routine of WinGX led to unreasonable locations. The possibility of disordered solvent molecules in a void was taken into consideration and Platon/Squeeze [70] was used to establish a void of 158 Å 3 with 56 electrons; removing such content in the void did not lead to any improved solution, the reason why such strategy was not followed. Least square refinements with anisotropic thermal motion parameters for all the non-hydrogen atoms and isotropic for the remaining atoms were employed.

Electrochemical Studies
Cyclic voltammetry (CV) and controlled-potential electrolyses (CPE) were undertaken in 0.2 M [ n Bu 4 N][BF 4 ]/DMSO electrolyte solutions, saturated with N 2 before each run, at room temperature. They were performed in a three-electrode cell equipped with a Luggin capillary connected to a silver wire pseudo-reference electrode. Platinum disc (for CV; d = 1 mm) or gauze (CPE) working electrodes were used, as well as platinum wire (CV) or gauze (CPE) counter electrodes. In CPE experiments, the working and counter compartments were separated by a sintered glass frit and the studies were monitored regularly by CV. The redox potentials were measured using ferrocene as internal standard, and their values are quoted relative to the SCE by using the [Fe(η 5 -C 5 H 5 ) 2 ] 0/+ redox couple (E OX 1/2 = 0.44 V vs. SCE for DMSO) [71][72][73][74][75][76].

Typical Procedures and Product Analysis for Catalysis
The microwave-assisted (MW) solvent-free peroxidative catalytic oxidation of 1-phenyethanol was performed in a focused Anton Paar Monowave 300 reactor using a 10 mL capacity cylindric Pyrex tube with a 13 mm internal diameter, fitted with a rotational system and an IR temperature detector. The tube was charged with the alcohol (2.5 mmol), 5 µmol catalyst precursor 1 or 2 (0.2 mol% vs. substrate) and a 70% aqueous solution of t-BuOOH (5 mmol), sealed, placed in the microwave reactor and the system left stirring under irradiation (5 or 20 W) at 80 or 120 • C for 0.5-3 h. After cooling to room temperature, 150 µL of benzaldehyde (internal standard) and 2.5 mL of MeCN (to extract the substrate and the organic products from the reaction mixture) were added. The obtained mixture was stirred for 10 min and then a sample (1 µL) was taken from the organic phase and analysed by GC.
The peroxidative oxidations of cyclohexane were performed in round bottom flasks with vigorous stirring and using MeCN as solvent (up to 5.0 mL total volume), under air. The catalyst precursors 1 and 2 and trifluoroacetic acid (TFA, optional), in the form of a stock solution in acetonitrile, were mixed, cyclohexane (0.25 mL, 2.3 mmol) was included and the reaction started when hydrogen peroxide was added (50% in H 2 O, 0.68 mL, 11 mmol). The concentrations of the reactants were thus as follows: catalyst precursor (5 × 10 −4 mol L −1 ), TFA (0.005 mol L −1 ), substrate (0.46 mol L −1 ) and H 2 O 2 (2.2 mol L −1 ). The reaction mixture was stirred at 50 • C for 2 h and at 5, 15, 30, 45, 60 and 120 min, an aliquot was taken and analysed by GC using nitromethane (50 µL) as internal standard. Before the GC analysis an excess of triphenylphosphine was added, in order to reduce the formed cyclohexyl hydroperoxide to the corresponding alcohol and hydrogen peroxide to water, following a method developed by Shul'pin [49][50][51]. Blank experiments were performed and confirmed that no product of cyclohexane oxidation was obtained unless the metal catalyst was used.
For the GC measurements, the temperature of injection was 240 • C. The initial temperature of the column was maintained at 100 • C (oxidation of cyclohexane) or 120 • C (oxidation of alcohol) for 1 min, then raised 10 • C/min up to 180 • C (oxidation of cyclohexane) or 200 • C (oxidation of alcohol), and held at this temperature for 1 min. Attribution of peaks was made by comparison with chromatograms of genuine samples and, in some cases, by GC-MS analyses using a Perkin Elmer Clarus 600 C instrument (He as the carrier gas), equipped with a 30 m × 0.22 mm × 25 µm BPX5 (SGE) capillary column.

Conclusions
The Both complexes act as catalyst precursors towards the solvent-free microwave-assisted oxidation of alcohols and the peroxidative oxidation of alkanes under mild conditions. The highest activity and selectivity with a maximum product yield of 99% (for the 1-phenylethanol oxidation after 1 h without any additive) were observed with 1. In the case of oxidation of cyclohexane, 1 also exhibits a better activity (a maximum product yield of 13% after 3h) than 2.
The catalytic oxidations under microwave assisted and solvent-free conditions have significance for the development of environmentally friendly catalytic systems and deserve further investigations. Our study was an attempt to contribute to this aim.