Vanadium-Substituted Wells-Dawson Heteropolyacid as Catalyst for Liquid Phase Oxidation of 1,4-Dihydropyridine Derivative

H7P2W17VO62.25H2O Wells-Dawson heteropolyacid was prepared, characterized and evaluated as catalyst in a 1,4-dihydropyridine oxidation reaction. The optimal conditions are the following: 1 mmol% of HWDV, a molar ratio 1,4-DHP:H2O2 (1:10) at reflux of acetonitrile. A conversion of 98.6 % is achieved in only 120 min of reaction.


Introduction
Catalysis by heteropolyacids (HPAs) and related compounds is a field of increasing importance worldwide.Numerous developments are being carried out in basic research as well as in fine chemistry processes.It has been proved that the introduction of vanadium into the Keggin framework is beneficial for redox catalysis, shifting its activity from acid to redox-dominated [1].This can be explained through the energy gap between the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO), which is correlated with the reduction potential of the HPAs.It has been demonstrated that the more reducible HPA has the smaller energy gap between the HOMO and the LUMO.It has also been reported that vanadium-containing HPAs show excellent redox properties because the substitution with vanadium stabilizes the LUMOs [2].In the last years we have prepared many Wells-Dawson heteropolycompounds such as K 6 P 2 W 18 O 62 Á10H 2 O and H 6 P 2 W 18 O 62 Á24H 2 O [3,4].Wells-Dawson structure known as a-isomer possesses two identical ''half units'', XM 9 O 31, of the central atom surrounded by nine octahedral units linked through oxygen atoms and related by the symmetry perpendicular to the trigonal axis.The use of the Wells-Dawson HPAs in the synthesis of diverse compounds such as flavones [5], coumarins [6], 1,4-dihydropiridines [7], pyridines [8] and crotonates [9], among others, has already been reported by us.
On the other hand, pyridines are present in the important niacin and B6 vitamins, and also in highly toxic alkaloids such as nicotine [17].They are important as anti-inflammatory, antiasthmatic [18], antidepressant [21], antitubercular and antibacterial agents [17].There also are examples of pyridines that act as potent HIV protease inhibitor [22], and some pyridine derivates and their metal complexes are important building blocks for the construction of chemosensors, self-organized assemblies, or photoactive molecular devices [23].
From this, several procedures have been developed in order to study the oxidative dehydrogenation of 1,4-DHPs.Due to the need for working with non-contaminating technologies, new procedures include some environmentally compatible oxidative agents such as H 2 O 2 [24,25], t-BuOOH [26] and CO(NH 2 ) 2 ÁH 2 O 2 [27], among others.The use of alternative energy sources such as microwaves has also been reported [28].On the other hand, a mixedaddenda vanadomolybdophosphate heteropolyacid, H 6 PMo 9 V 3 O 40 , was used by Heravi et al. in the oxidation of 1,4-DHPs [29].
Continuing with our studies on the synthesis of heteropolyacids and their use as effective, reusable and stable solid catalysts, in the present work we synthesized and characterized the V substituted Wells-Dawson heteropolyacid, H 7 P 2 W 17 VO 62 Á25H 2 O (HWDV).We performed the oxidation of a 1,4-DHP using this solid as homogeneous catalyst in acetonitrile media, with H 2 O 2 as oxidant agent.In addition, hydrogen peroxide is the technologically green oxidant by election [30] in agreement with its accessibility and because it is cheap and environmentally friendly since water is the generated by-product.The unsubstituted WD was synthesized by the etherate method described in detail in a previous paper [4].

Catalyst Characterization
Fresh solid acid samples were characterized by FTIR in Bruker IFS 66 equipment, and by 31 P MAS-NMR measurements in Bruker Avance II-300 equipment, using 85 % H 3 PO 4 as an external standard under the following operation conditions: pulse width of p/2 31 P: 5 ls; repetition time ( 31 P): 60 s; rotation speed: 7 kHz.Besides, the morphology of the compounds was analyzed by scanning electron microscopy (SEM) with Philips SEM 505 microscope.Thermogravimetric analyses (TGA) were carried out in Shimadzu TGA-50H equipment.Samples were heated in air from room temperature to 800 °C at a rate of 10 °C min -1 .X-ray diffraction (XRD) spectra for 2h values between 5°and 45°were recorded in Philips PW 1732 equipment.The following operating conditions were used: source voltage, 40 kV; source current, 20 mA; goniometer speed, Dh = 2 min -1 ; chart speed, 2 cm min -1 ; slit, 1/0.1/ 1°.Besides, potentiometric titration measurements were made in Metrohm 794 Basic Titrino equipment, according with the following procedure: each solid catalyst sample was suspended in acetonitrile, and the system was stirred at room temperature over 60 min.Then, the acetonitrile catalyst solution was titrated with n-butylamine solution (0.025 N).

Catalytic Tests
Catalytic tests in order to analyze the effect of catalyst, amount of oxidant agent, reaction temperature and amount of catalyst on reaction conversion and the selectivity to the desired oxidation product (2,6-dimethyl-4phenyl-3,5-pyridinedicarboxylic acid, dimethyl ester), were made.With that purpose, the general reaction presented in Scheme 1 was performed.Also, decomposition of H 2 O 2 was evaluated through iodimetric method.The starting 1,4-DHP was prepared according to Sathicq et al. [7].The general procedure used is the following: a mixture of 1,4-dihydropyridine, H 2 O 2 and HWDV as catalyst in acetonitrile as reaction solvent was stirred for the appropriate amount of time.Previously, the reaction solvent (acetonitrile) was selected in order to assume a homogeneous work system.Reactions were monitored by gas chromatography (GC), with Shimadzu GC-2014 equipment according to the following conditions; detector temperature, 320 °C; injector temperature, 200 °C; split ratio, 1:10; pressure, 60 kPa; purge flow, 3.0 ml min -1 ; and using a flame ionization detector (FID).The conversion was defined as the ratio of the oxidation product (a pyridine derivative) to the initial concentration of reactant (1,4-DHP).The product was identified by comparison of physical data (mp, TLC, NMR) with those reported [31]. 13C NMR and 1 H NMR spectra were recorded at room temperature on Bruker AC 200 using tetramethylsilane (TMS) as internal standard.

Kinetic Parameters Determination for 1,4-Dihydropyridines Oxidation
The general procedure for the oxidation of 1,4-DHPs was devised from the results of the experiments mentioned above.The initial rate method was used to measure the kinetic parameters of the reaction.The 1,4-DHP apparent reaction order was determined by varying its concentration between 0.0083 and 0.05 mol L -1 , maintaining all the other variables (0.25 mol L -1 of H 2 O 2 , 3.72 g L -1 of catalyst, acetonitrile volume (5 mL) and 354 K) constant.
On the other hand, the reaction temperature was varied between 323 and 354 K to obtain the specific reaction rate at different temperatures and the apparent activation energy.
3 Results and Discussion It is known that when the two half-anions are identicals, the symmetric species show one signal in the 31 P MAS-NMR spectrum, and when the species are unsymmetrical two 31 P signals are observed [33].In a previous work, we have reported that WD sample has two equivalent phosphorus atoms and consequently, it shows a mainly peak at-12.7 ppm that corresponds to the a-isomer [4].In Fig. 2a the mainly expected peak is present, and also it can be observed the presence of two small signals at -12.2 and 11.5 ppm that corresponds to the presence of the Wells-Dawson b-isomer as impurity [34].Figure 2b shows the 31 P-MAS-NMR spectrum of the monosubstituted V compound (HWDV).It can be observed that HWDV sample shows two peaks with a chemical shift of -12.4 and -11.2 ppm.The signal at -12.4 ppm corresponds to the unperturbed P atom of the WD structure [3] and the other one at -11.2 ppm corresponds to vanadium incorporation into a polar position of the PW 9 half unit, in agreement with previous literature reports [35,36].These results indicate that the V atom has been successfully included in the Wells-Dawson structure.
Furthermore, Fig. 3 shows the infrared spectroscopic measurements for the lacunar heteropolyoxoanion, WD and HWDV catalysts.In the case of the lacunar heteropolyoxoanion, the typical signals expected for this structure were found: 740, 805, 880, 905, 940, 985, 1,022 and 1,084 cm -1 [4,37].For WD and HWDV samples, Fig. 3 shows the typical signals expected for the Wells-Dawson structure: PO 4 tetrahedron (1,089 cm -1 ), W = O (950 cm -1 ) and W-O-W (912 and 779 cm -1 ) bonds [3,4].Therefore, it is possible to assume that the V-substituted catalyst keeps the corresponding Wells-Dawson structure.A widening of the PO 4 band for HWDV, probably due to the loss of symmetry of one of the tetrahedral PO 4 , is observed [32].
X-ray diffraction results, 31 P-MAS-NMR and FTIR measurements confirm that HWDV keeps the Wells-Dawson structure after vanadium substitution.
On the other hand, the number of water molecules involved in WD and HWDV structures was determined by TGA.The results indicated that HWDV and WD possess 25 and 24H 2 O hydration molecules, respectively.
Figure 5 depicts a SEM micrograph of fresh vanadium substituted and unsubstitued WD.Both acid samples show a smooth morphology, in contrast to the rough texture obtained for other heteropolyacids with Keggin structure [3].
The acidic properties of the solid acid catalysts were evaluated by potentiometric titration with n-butylamine.The maximum acid strength of superficial sites is indicated by the initial electrode potential (E).According to this, the acid strength of the superficial sites could be classified by using the following range: very strong site: E [ 100 mV; strong site: 0 \ E \ 100 mV; weak site: -100 \ E \ 0 mV; and Fig. 3 FTIR spectra of fresh lacunar heteropolyoxoanion, WD and HWDV catalysts Fig. 4 XRD pattern of fresh HWDV very weak site: E \ -100 mV [38].The area under each curve in the graph of potential versus equivalents of nbutylamine solution per gram of sample gives us information about the number of acidic sites in the catalyst.In addition, by plotting the potential derivative obtained versus nbutylamine solution volume, it is possible to determine the number of different acidic sites.
Figure 6 shows the obtained results by potentiometric titration for fresh vanadium substituted and unsubstitued WD.It can be observed that both acids have very strong acidic sites (E [ 100 mV), although as can be seen, HWDV has more acidic sites than WD, and they are a little more acid than WD sites.Figure 7 shows that, WD presents one peak corresponding to one type of acidic site; meanwhile, HWDV has two different acidic sites types.

Catalytic Tests
Physical data of the oxidized product were analyzed (mp, TLC, NMR).A comparison between reported and experimental mp, 13 C NMR and 1 H NMR was done.Experimental data spectra are shown below.As can be seen from these and TLC results that are not reported here, the desired product was obtained.2,6-Dimethyl-4-phenyl-3,5-pyridinedicarboxylic acid, dimethyl ester, mp: 135-137 °C (ethanol) (lit mp: 136-138 °C [27]), 13  In order to study the effect of the catalyst, the comparison between the use of 1 mmol% of WD, HWDV and no catalyst was made (Fig. 8).It was found that for HWDV system decomposition of H 2 O 2 is 18 % in a reaction time of 240 min (Fig. 8).In the WD catalyst and no catalyst systems, decomposition of H 2 O 2 is even a bit lower.As we can see, the conversion is higher using HWDV catalyst, with a minimal H 2 O 2 decomposition.This catalyst has a bigger number of stronger acidic sites, and according to Shaabani et al. [39] they can make redox potential of H 2 O 2 stronger than in neutral or weak acidic media.HWDV also has a vanadium atom that is beneficial for redox catalysis [1].
A plausible mechanism in acidic media is proposed in Fig. 9, similar to what was proposed by Sharbatdaran et al. [40].A peroxo intermediate is formed when H 2 O 2 coordinates with V or W atom of the HWDV catalyst.After  Furthermore, the optimal ratio between 1,4-DHP and H 2 O 2 was selected by carrying out some experiences with different quantities of the oxidizing agent: 1:0, 1:1, 1:5, 1:10 and 1:25 (molar ratio 1,4-DHP: H 2 O 2 , Fig. 10).There is a big difference when lower relationships than 1:10 are used, but there is no significant difference by using larger quantities of oxidant agent.So, we consider that a ratio 1:10 is appropriate to achieve excellent conversion values.
The effect of reaction temperature on reactant conversion was tested in a range from 323 to 354 K (Fig. 11).As is shown, conversions are 11.9, 47.3, 77.1 and 98.6 % at 240 min of reaction, for a temperature of 323, 333, 343 and 354 K, respectively.At 354 K a conversion of 98.6 % is achieved in only 120 min of reaction.According to this, the most convenient reaction temperature is at reflux of the selected reaction solvent (354 K).
The last parameter tested was the amount of catalyst (Fig. 12) by using 0.5, 1 and 3 mmol % of HWDV.The There is no significant difference between 1 and 3 mmol%: at 120 min of reaction both reaction systems achieve the same conversion as what is obtained at 240 min.Therefore, 1 mmol% is the optimal amount for our system.The optimal procedure is the following: 1 mmol% of HWDV, a ratio 1,4-DHP:H 2 O 2 (1:10) at reflux of acetonitrile (0.025 M).

Kinetic Parameters Determination for 1,4-Dihydropyridines Oxidation
The dependence of the reaction rate on the 1,4-DHP concentration was determined by fixing H 2 O 2 and HWDV concentration (0.25 mol L -1 and 3.72 g L -1 , respectively), and the reaction temperature (354 K).The 1,4-DHP oxidation conversion versus the reaction time for a 1,4-DHP concentration range of 0.0083-0.05mol L -1 was measured.Figure 13 shows the relation between initial reaction rate and the 1,4-DHP concentration initially added.Each initial reaction rate was obtained by plotting the 1,4-DHP concentration versus time, and extrapolating to the initial condition.Figure 13 displays a linear relation between the initial reaction rates calculated for each experiment and the initial 1,4-DHP concentration, indicating a pseudo-first order with respect to substrate concentration.
The graph obtained by plotting Ln [1,4-DHP] versus reaction time for each temperature is shown in Fig. 14.The corresponding k values were obtained from these fittings.The linearity of the data corroborates the reaction pseudo-first order shown before.From the typical Arrhenius plot that is shown in Fig. 15, it is possible to calculate the apparent activation energy (Ea).In the case of the reaction we studied, Ea has a value of 95.20 kJ mol -1 .

Conclusions
We have synthesized and characterized Wells-Dawson heteropolyacid with a V substitution (HWDV).By comparing different characterization results with those from another known WD, it is possible to conclude that de synthesized material maintains the Wells-Dawson structure.Moreover, the V substitution was performed successfully according to the comparison between HWDV characterization results and those from the literature.Then, the performances of both catalysts (WD and HWDV) were tested in a desired oxidation reaction of a 1,4-DHP.As HWDV shows better conversion results, we selected this catalyst and looked for an optimal procedure to oxidize a 1,4-DHP by using an excess of H 2 O 2 in acetonitrile.A conversion of 98.6 % is achieved in only 120 min of reaction.It must be noted that along all reactions no secondary products were observed.
Once a general procedure was devised, we wanted to establish some kinetic parameters.As a result we found a pseudo-first order with respect to the 1,4-DHP and the apparent activation energy was calculated; a value of 95.2 kJ mol -1 was obtained.
Acknowledgments The authors thank to CONICET, Agencia Nacional de Promocio ´n Cientı ´fica y Tecnolo ´gica (Argentina), Comisio ´n de Investigaciones Cientı ´ficas (CIC), and Universidad Nacional de La Plata for financial support.AGS, GPR, GTB and HJT are members of CONICET.

3. 1 Fig. 2
Figure 1 shows the 31 P MAS-NMR results of unsubstituted WD and the monosubstituted V compound (HWDV) synthesized in this work.Wells-Dawson structure known as aisomer possesses two identical ''half units'', XM 9 O 31 , of the central atom surrounded by nine octahedral units linked through oxygen atoms and related by the symmetry perpendicular to the trigonal axis.Figure 2 shows the polyhedral representation of the Wells-Dawson structure of