Kinetics and mechanism of permanganate oxidation of nalidixic acid in aqueous alkaline medium

Article history: Received on: 06/09/2016 Revised on: 24/09/2016 Accepted on: 16/10/2016 Available online: 31/01/2017 The kinetics and mechanism of oxidation of nalidixic acid (NA) by permanganate ion in alkaline medium have been studied at 40 ± 1 o C. The Stoichiometry was observed to be 2:1 in terms of mole ratio of permanganate ion and nalidixic acid consumed. The reaction shows first order with respect to oxidant and fractional order in both the substrate and alkali concentration. The oxidation reaction proceeds via an alkali permanganate species that forms a complex with nalidixic acid and the complex then decomposes to give the product. The effects of added products and ionic strength have also been investigated. The main products identified were hydroxylated NA and Mn(VI). A mechanism was proposed on the basis of experimental results. Investigation of the reaction at different temperature allowed the determination of the activation parameters with respect to the slow step of the proposed mechanism.


INTRODUCTION
Potassium permanganate widely used as oxidizing agent play vital role in the kinetics of number of organic and biological active compounds (Fatiadi, 1987;Ladbury and Cullis, 1958;William, 1958;Banerji, 1988;Baljeet and Kothari, 1997).Oxidation reactions by Potassium permanganate are of considerable academic and technological importance because of variable oxidation states.Permanganate is one such powerful multi-electron oxidant which can exist in various oxidation states, among which +7 is its highest oxidation state, which occurs in the Oxo compounds like MnO 4 -, Mn 2 O 7 , MnO 3 F. Out of which MnO 4 -is the most commonly used well known oxidant species to carry out kinetic studies in acidic, neutral and alkaline media.Oxidation by permanganate ion find extensive applications in organic syntheses (Fatiadi, 1987;Stewart and .Wiberg, 1965;Freeman, 1976;Lee, 1980;Lee and Tranhanovsky, 1982;Simandi et al., 1983;Lee et al., 1987) especially since the introduction of phase transfer catalysis (Lee, 1980;Lee and Tranhanovsky, 1982;Lee et al., 1987) which permits the use of solvents like methylene chloride and benzene.Kinetic studies are vital sources of mechanistic information on these reactions, as validated by result stating to unsaturated acids in both aqueous (Fatiadi, 1987;Stewart and Wiberg, 1965;Freeman, 1976;Lee, 1980;Lee and Tranhanovsky, 1982;Simandi et al., 1983;Lee et al., 1987;Wiberg et al. 1973) and non-aqueous media (Wiberg et al. 1973).As is known, in aqueous alkaline medium the permanganate ion oxidizes a number of organic compounds, which are not, or only very slowly, attacked in acidic or neutral medium (Ladbury and Cullis, 1958;William, 1958), (Drummond and Waters, 1935).The mechanism of oxidation depends on the nature of the substrate and pH of the reaction mixture (Stewart et al. 1997).In strongly alkaline medium, the stable reduction product (Simandi et al., 1985;Timmanagoudar et al., 1997;Nadimpalli et al., 1993) of permanganate is manganate ion, MnO 4 2- . MnO 2 appears only after long time, i.e., after the complete consumption of MnO 4 -.
No mechanistic information is available to discriminate between a direct one-electron reduction to Mn(VI) and a mechanism in which a hypomanganate ion formed in a twoelectron reduction followed by its rapid re-oxidation (Panari et al., 1998;Bohn et al., 1992).
The manganese chemistry involved in these multistep redox reactions is a significant source of information as the manganese intermediates are relatively easy to identify when they have sufficiently long life time and oxidation states of the intermediates permit useful deductions as to the possible reaction mechanisms including the nature of intermediates.
Fluoroquinolones are broad-spectrum antibacterial agents used to treat the bacterial infections in human beings.Pharmaceuticals, of which antibacterial groups are important, have been identified as evolving environmental contaminants (Johnson et al., 2003).A major fraction of fluoroquinolones pass into the domestic sewage due to partial metabolism in the human body.This represents the main route for entry of such pharmaceutical compounds into natural aquatic environment.In this perception, transformations of fluoroquinolone antibacterial agents in suitable water treatment process definitely play a major role (Halling-Sorensen et al., 1998).Nalidixic acid (NA) with molecular formula C 12 H 12 N 2 O 3 (1-ethydm-3, 4-dihydro-7-methyl-4-oxo-1, 8naphthyridine-3-carboxylic acid) (Figure 1) is the first synthesized antimicrobial quinolone.NA is an ionizable, non-biodegradable photosensitive molecule (Mascolo et al., 2010;Ge et al., 2010) with a carboxylic acid function having a pKa of 5.95 (Ross and Riley, 1990).NA is an antibacterial drug still widely used for urinary tract infections (Barlow, 1963).Its two major metabolites are 7hydroxynalidixic acid (HNA), which exhibits antibacterial properties equal to NA (Mcchesney et al., 1964;Moore et al., 1965;Portmann et al., 1966) and 7-carboxynalidixic acid (CNA), which is inactive.Permanganate has been widely used for the water and wastewater treatment from last five decades (Hicks, 1976).The oxidation of nalidixic acid by permanganate was studied to investigate the kinetics and mechanism.

Chemicals
All chemicals used were of analytical grade and doubly distilled water was used throughout this study.Standard solution of nalidixic acid (KORES India Limited) was prepared by dissolving calculated quantity of pure drug in double distilled water.Permanganate solution was obtained by dissolving potassium permanganate (BDH Analar) in water and standardized by titrating against oxalic acid (Vogel, 1967).Freshly prepared & standardized permanganate solutions were always used in kinetics experiments.The Mn(II) solution was made by dissolving manganese sulphate (BDH) in water.NaOH (BDH) and NaNO 3 (MERCK) were used to provide required alkalinity and ionic strength respectively.

Kinetic Measurements
All kinetic measurements were conducted under pseudofirst-order conditions, where the concentration of nalidixic acid was much greater than permanganate ion concentration at constant temperature 40 ± 0.1 o C unless otherwise stated.The reaction was initiated by mixing thermostated solution of permanganate and nalidixic acid; in addition to that required quantities of NaOH and NaNO 3 are added to provide required alkalinity and ionic strength of reaction.The progress of the reaction was followed spectrophotometrically at 525nm.The application of Beer's law to permanganate at 525nm had been verified.The molar absorptivity index of permanganate was found to be ε = 2260 ±50 dm 3 mol -1 cm -1 as a function of time (compared to the literature, ε = 2200, Simandi et al., 1985).The kinetics reactions were followed more than 85 % completion of the reaction.The pseudo-first-order rate constants k obs were calculated from the plots of the logarithm of absorbance versus time, which were linear.The values of k obs were reproducible within ± 5%.

Stoichiometry and Product Analysis
Different sets of concentration of reactants in 0.5 mol dm -3 of OH -ion and at constant ionic strength, 0.5mol dm -3 , were kept over 24 hours at 40 o C in a closed container.When LC/MS analysis of the reaction indicated the presence of a product with molecular ion of m/z 248 corresponds to 1-ethyl-2hydroxy-1, 4-dihydro-7-methyl-4-oxo-1, 8-naphthyridine-3carboxylic acid (Figure 2).The molecular ion of nalidixic acid is m/z 232.2.The IR spectroscopy shows a broad peak at 3382.39 cm -1 which is due to -OH stretching (Figure 3) and the remaining peaks are of the parent compound.

RESULTS
The reaction orders were determined from the slopes of log k obs versus log [concentration] plots by different concentration of nalidixic acid, permanganate and alkali in turn, keeping all other concentration and conditions constant.

Effect of Concentration of Manganese(VII)
The oxidant permanganate [MnO 4 -] concentration varied from 1×10 -4 to 7×10 -4 mol dm -3 , and all other concentrations and conditions were constant.The plot of log absorbance versus time was linear (Figure 4) indicating that the reaction is first order with respect to [KMnO 4 ].The observed pseudo first order rate constant (k obs ) were independent of the concentration of KMnO 4 .

Effect of Concentration of Nalidixic acid
The effect of concentration variation of nalidixic acid on the rate of reaction was studied in the range 2 × 10 -3 to 10 × 10 - 3 mol dm -3 at constant concentration of permanganate, alkali and ionic strength at 35 o , 40 o , 45 o C respectively.The rate of reaction increases with increasing concentration of nalidixic acid (Table 1).A plot of log k obs versus log [NA] was linear with a slope of 0.52, thus indicating a fractional-order dependence on nalidixic acid concentration.This was confirmed by the plot of 1/k obs versus 1/ [NA] (Figure 5) which was also linear with a positive intercept.

Effect of Concentration of Alkali
The effect of concentration variation of alkali on the rate of reaction was studied in the concentration range 2.0 × 10 -1 to 10 × 10 -1 mol dm -3 at fixed concentration of permanganate, nalidixic acid and ionic strength at three temperatures viz.35 o , 40 o , 45 o C respectively.
Pseudo first-order rate constant (k obs) was found to be increased with increase in [OH -] (Table 1).A plot of log k obs versus log [OH -] was linear with a fractional slope of 0.56.This was confirmed by the plot of 1/k obs versus 1/ [OH -] (Figure 6) which was also linear with a positive intercept.

Effect of Ionic Strength and Dielectric Constant
At constant concentration of reactants and other conditions constant, the ionic strength was varied by varying concentration of sodium nitrate from 0.75 -1.75 mol dm -3 .Ionic strength had negligible effect on the rate of reaction.The effect of the dielectric constant (D) was studied by varying the t-butanolwater content (v/v) in the reaction mixture with all other conditions being maintained constant.The rate of reaction increases with increasing t-butanol volume.The plot of log k obs versus 1/D was linear with positive slope (Figure 7).

Effect of Added Products
The manganate ion concentration was varied from 4.0 × 10 -5 to 4.0 × 10 -4 mol dm -3 at constant concentrations of permanganate, nalidixic acid, alkali, and ionic strength.It was found that initially added manganate ion had no effect on the rate of reaction.

Tests for Free Radical
The reaction mixture (10ml) in which known quantity (2ml) of acrylonitrile has been added and kept in an inert atmosphere for 5 hours then diluted with methanol, white precipitate was formed, indicating the intervention of free radicals in the reaction.

DISCUSSION
Permanganate ion is a strong oxidant in an aqueous alkaline media.Since it shows various oxidation states, the stoichiometric results and the pH of reaction medium play a significant role.Under the present experimental conditions at pH > 12, the reduction product of Mn(VII) is stable and further reduction of Mn(VI) might be stopped (Simandi et al., 1985;Timmanagoudar et al., 1987).However, prolong standing, green Mn(VI) is reduced to Mn(IV) under experimental conditions.The permanganate shows various oxidation states, such as Mn(VII), Mn(V), and Mn(VI) in the alkaline medium.The colour of the reaction mixture changes from violet Mn(VII) to dark green Mn(VI) through blue Mn(IV) were observed.It is plausible that blue colour originated from the violet of permanganate and the green from manganate, excluding the accumulation of hypomanganate.It is clear from Figure 8 that the concentration of MnO 4 -decreases at wavelength 526 nm, while increases at 610 and 460 nm are due to Mn(VI).As the reaction proceeds, a yellow turbidity slowly develops, and after keeping for a long time the solution decolourises and forms a brown precipitate.This suggests that the initial products might have undergone further oxidation resulting in a lower oxidation state of manganese.
The less than unit order with respect to NA may be due to the complex formation between the [MnO 4 •OH] 2− and NA before the rate determining step.A plot of 1/k obs versus 1/ [NA] (Figure 5) shows an intercept in agreement with complex formation.Further evidence for complex formation was obtained from the UV-VIS spectra of reaction mixtures.Two isosbestic points were observed for this reaction (Figure 8), indicating the presence of an equilibrium before the slow step of the mechanism (Chang, 1981;Sathyanarayana, 2001).Within the complex one electron is transferred from nalidixic acid to Mn(VII).The breaking of this complex (C) is assigned as the slowest step, leading to the formation of an NA radical intermediate and Mn(VI).The radical intermediate reacts with another Mn(VII) species, [MnO 4 •OH] 2− , to give the final products (Scheme 2).The effect of ionic strength and dielectric constant on the rate explains qualitatively the involvement of a neutral molecule in the reaction.From the above mechanism the following rate law, eqn.( 1) -( 8) can be derived. - ] f are total and free concentration of Mn (VII) respectively.Total concentration of [OH -] is given by: - In view of low concentration of MnO 4 -and nalidixic acid used, above equation can be written as: …5 Substituting equation ( 2), ( 4) and (5) in equation ( 1) and omitting "t" and "f" subscripts Equation ( 7) can be rearranged as According to Eqn (8) the plot of 1/k obs versus 1/ [NA] (Figure 5) is linear with positive intercept and slope at three different temperatures.The rate constant k, of the slow step, (Scheme 2) was obtained from the intercept of the plots 1/k obs versus 1/ [NA] (Table 2).The energy of activation was determined by the plot of log k versus 1/T from which activation parameters were calculated (Table 2).The equilibrium constant (K 1 ) and the equilibrium constant of complex (K 2 ) in Scheme 2 were calculated from the intercept and slope of the plot 1/ k obs versus 1/ [OH -] (Figure 6) (Table 2).
The value of K 1 is in good agreement with earlier work (Thabaj et al., 2007) at 40 o C. Van't Hoff's plots of log K 1 versus 1/T and log K 2 versus 1/T gave the values of enthalpy of reaction ΔH, entropy of reaction ΔS and free energy of reaction ΔG, calculated for the first, and second equilibrium steps (Table 2).
The values of ΔH ≠ and ΔS ≠ are both favourable for electron transfer process (Farokhi and Nandibewoor, 2004).The value of ΔS ≠ within the range of radical reaction has been ascribed (Walling, 1957) to the nature of electron pairing and unpairing process.The negative value of ΔS ≠ indicates that complex is more ordered than the reactants (Rangappa et al., 2001;Bugarcic et al., 2006).The observed modest enthalpy of activation and a relatively low value of the entropy of activation as well as a higher rate constant of the slow step indicate that the oxidation probably occurs via inner-sphere mechanism (Farokhi and Nandibewoor, 2003).

CONCLUSION
It is interesting that the oxidant species [MnO 4 -] requires pH > 12, below which the system becomes disturbed and the reaction proceeds further to give a reduced oxidation product as manganese(IV), which slowly develops a yellow turbidity.Hence, the role of pH in the reaction medium is crucial.The oxidant, manganese(VII), exists in alkali media as alkali-permanganate species [MnO 4 . OH] 2-, which takes part in the chemical reaction.Chemical oxidation using Mn(VII) has been widely used for treatment of pollutants in drinking water and waste water applications.The proposed mechanism is consistent with product, mechanism and kinetic studies.

Table 1 :
Effects of variation of [MnO4 -], [NA] and [OH -] on the oxidation of nalidixic acid by alkaline permanganate at 40 °C and I = 0.5 mol dm -3 .

Fig. 3 :
Fig. 3: FT-IR spectra of the product of oxidation of Nalidixic acid by permanganate.

Fig. 7 :
Fig. 7: Effect of dielectric constant on the oxidation of nalidixic acid by alkaline permanganate at 40 o C.

Table 2 :
Activation and thermodynamic quantities for the oxidation of nalidixic acid by alkaline permanganate.