Introduction

Platinum(IV) complexes find applications in cancer chemotherapy and treatment of various other diseases and as gel components for several types of medical implants (breast implants, joint replacement prosthetics, artificial lumbar disks, vascular access ports, etc.). Ascorbic acid (vitamin C), being an important biochemical compound, is also suspected to play some role in the treatment of various tumors. Hence, the interactions of platinum(IV) with ascorbic acid appear to be worthy of investigation. L-ascorbic acid in recent years has been extensively studied as a reductant both for free metal ions and for metal complexes [1–12]. These reactions are categorized into three main groups, namely outer-sphere, inner-sphere, and mixed outer-and-inner-sphere electron-transfer reactions.

Trace metal ion catalysis in the oxidation of ascorbic acid, for example by peroxodiphosphate [13] in acetate buffers and peroxo-bound chromium(V) [14] in acid medium, is another important category of the reactions of ascorbic acid. The title reaction has been studied previously [15], but the proposed reaction mechanism is not complete as it lacks the effect of chloride ion concentration on the rate. This has prompted us to undertake a study of the kinetics of oxidation of ascorbic acid by platinum(IV) in aqueous acid medium, so that a comprehensive mechanism accounting for the effect of chloride ion can be delineated.

L-ascorbic acid can exhibit one-equivalent or two-equivalent nature in oxidations. Further, characteristic hydrogen ion dependence is reported in a large number of oxidations of ascorbic acid by metal ion oxidants. We were interested to see if such hydrogen ion dependence would also be observed in the present system.

Experimental

Materials and methods

L-ascorbic acid (Merck) (AnalaR grade) was used as received, and its aqueous solution was standardized iodimetrically [16]. A stock solution of platinum(IV) was prepared by dissolving the requisite quantity of hexachloroplatinic acid (Acros) in 0.3 mol dm3 HCl and standardized iodometrically. Both solutions were kept in bottles painted black from the outside to suppress photodecomposition. Other reagents were either of AnalaR or guaranteed reagent grade and were employed as supplied. Triply distilled water was used to prepare the solutions; the second and third distillations were from alkaline permanganate and edta, respectively, in an all-glass still.

Kinetic procedure

The reactions were conducted in glass-stoppered Erlenmeyer flasks immersed in a water-bath thermostated at ±0.1 °C unless stated otherwise. The reactions were initiated by adding Pt(IV) solution to temperature pre-equilibrated reaction mixtures containing all the other reagents. The time of initiation was recorded when the solution of platinum(IV) from the pipette was half-released into the reaction mixture. Aliquots (5 cm3) of the reaction mixture were withdrawn at different time intervals and then discharged into ice-chilled cerium(IV) sulfate solution of known concentration. The excess cerium(IV) was estimated by titrating against iron(II) sulfate solution employing ferroin as an indicator. Further experiments showed no difference whether the reaction was initiated by adding Pt(IV) or ascorbic acid. Initial rates (k i , mol dm−3 s−1) were computed [17] by employing the plane mirror method. Second-order plots were made for comparable concentrations of the reactants. Triplicate rate measurements were reproducible to within ±5 %.

Stoichiometry

The stoichiometry of the reaction was determined by conducting a set of reactions with an excess concentration of Pt(IV) over that of ascorbic acid, or vice versa, in a thermostated water-bath maintained at (30.0 ± 0.1) °C for ca. 6 h. The excess ascorbic acid was estimated iodimetrically in ice-cold solution. The results corresponded to the stoichiometry of the reaction shown in Eq. (1);

$$ {\text{Pt}}\left( {\text{IV}} \right) + {\text{H}}_{ 2} {\text{A}} \to {\text{Pt}}\left( {\text{II}} \right) + 2 {\text{H}}^{ + } + {\text{A}} $$
(1)

where H2A and A represent ascorbic acid and dehydroascorbic acid, respectively (Scheme 1); the latter was also detected qualitatively [18]. Dehydroascorbic acid being of triketo structure has also been reported [19, 20] in other oxidation reactions of ascorbic acid.

Scheme 1
scheme 1

Structural representation of Ascorbic acid and De-hydro ascorbic acid

Full size image

The stoichiometry of the reaction was further confirmed by studying the kinetics of a few reactions using stoichiometric concentrations of the reactants under identical experimental conditions. Second-order plots of 1/[H2A]t− versus time were made, and the second-order rate constants calculated from the gradients of the resulting straight lines were in agreement with the rate constants calculated from initial rates and second-order plots made for comparable concentrations of the reactants (Table 1).

Table 1 Initial rates (k i , mol dm−3 s−1) and second-order rate constants (k, dm3 mol−1 s−1) in the reaction of Pt(IV) and ascorbic acid in aqueous acid medium ([H+] = 0.5 mol dm−3; 30 °C)
Full size table

Results

The order with respect to the oxidant was determined by varying the concentration of Pt(IV) at fixed concentrations of other reactants at (30.0 ± 0.1) °C. Initial rates (k i , mol dm−3 s−1) were computed by employing the plane mirror method [17]. A plot of initial rate versus concentration of Pt(IV) yielded a straight line passing through the origin, conforming that the reaction is first order with respect to the oxidant. Similarly, initial rates for varying concentrations of ascorbic acid at constant concentrations of other reactants were calculated. The plot of initial rate against the concentration of ascorbic acid also yielded a straight line passing through the origin, conforming to first-order dependence with respect to the substrate. Second-order rate constants were evaluated from second-order plots of log[Pt(IV)]t versus [H2A]t or log[H2A]t versus [Pt(IV)]t (Fig. 1) for comparable concentrations of the reactants, and found to be in close agreement with those calculated from initial rates and stoichiometric plots (Table 1).

Fig. 1
figure 1

Second-order plots in the reaction of Pt(IV) and ascorbic acid (H2A). [Pt(IV) = (1) 4.0 × 10−4 (2) 7.0 × 10−4 (3) 8.0 × 10−4 (4) 1.0 × 10−3 (5) 2.0 × 10−4 mol dm−3]; [H2A] = 5.0 × 10−4  mol dm−3; 30 °C

Full size image

The concentration of sodium chloride was varied in the range of (0.05–0.5) mol dm−3, keeping fixed concentrations of other reactants. Initial rates were found to decrease with increasing concentration of chloride ion. Hydrogen ion concentration was varied from 0.1 to 0.5 mol dm−3 at fixed concentrations of other reactants at I = 1.0 mol dm−3 (Ionic strength (I) was kept constant by employing lithium perchlorate). The rate decreased with increasing hydrogen ion concentration.

The effect of ionic strength was studied by employing lithium perchlorate; the rate increased with increasing ionic strength, consistent with an interaction between like-charged species of the reactants (Fig. 2). The effect of temperature on the rate of the reaction was studied at constant concentrations of all reactants. The energy and entropy of activation were calculated as (48.24 ± 0.49) kJ mol−1 and (−92.31 ± 1.65) JK−1 mol−1, respectively, by employing Eyring’s equation [21].

Fig. 2
figure 2

A plot of slope versus [H+]. I = 1.0 mol dm−3; 30 °C

Full size image

Discussion

If one takes into account the percentage distribution of ascorbic acid species, namely H2A, HA−, and A2− considering the following equilibria (2) and (3), HA− appears to be the predominant species of ascorbic acid under the experimental conditions used in this study.

$$ {\text{H}}_{ 2} {\text{A}}\overset {{\text{K}}_{ 1} } \rightleftharpoons {\text{HA}}^{ - } + {\text{H}}^{ + } $$
(2)
$$ {\text{HA}}^{ - } \overset {{\text{K}}_{ 2} } \rightleftharpoons {\text{A}}^{ 2- } + {\text{H}}^{ + } $$
(3)

where the values of pK1 and pK2 for steps (2) and (3) are 4.03 and 11.3, respectively.

In view of the decelerating effect of chloride ion, the speciation of platinum(IV) appears to be governed by equilibrium (4). This effect of chloride ion also has been observed previously [15].

The effect of ionic strength on the rate has been studied by employing lithium perchlorate; contrary to the effect of chloride ion, the rate increases with increasing ionic strength under these conditions. This pattern of reactivity can be explained by considering a mechanism consisting of step (2) above and steps (4–6). It is worth mentioning that the effect of chloride ion was assumed to be equivalent to the effect of ionic strength in an earlier study [15], whereas the present study shows that this is an oversimplification, due to the observed rate increase when using lithium perchlorate to increase the ionic strength. Such an effect requires an interaction of like-charged species of reactants. Therefore, step (6) in the proposed mechanism is predominant. This revised mechanism is more reasonable; in that, it accounts for the effects of both chloride ion and ionic strength on the rate.

$$ {\text{PtCl}}_{6}^{2 - } \overset {K_{ 2} } \rightleftharpoons {\text{PtCl}}_{5}^{ - } + {\text{Cl}}^{ - } $$
(4)
$$ {\text{PtCl}}_{5}^{ - } + {\text{H}}_{ 2} {\text{A}}\mathop{\longrightarrow}\limits^{k}{\text{Products}} $$
(5)
$$ {\text{PtCl}}_{5}^{ - } + {\text{H}}_{ 2} {\text{A}}\mathop{\longrightarrow}\limits^{{k^{\prime } }}{\text{Products}} $$
(6)

Thus, the interaction between PtCl5 − and HA− in step (6) can be considered to be the rate limiting step, leading to the rate law (7)

$$ \frac{{ - {\text{d}}[{\text{H}}_{ 2} {\text{A}}]}}{{{\text{d}}t}} = \frac{{k^{\prime } K_{1} K_{2} [{\text{H}}_{ 2} {\text{A}}][{\text{Pt(IV)}}]}}{{(K_{2} + [{\text{Cl}}^{ - } ])(K_{1} + [{\text{H}}^{ + } ])}} $$
(7)

where [H2A] and [Pt(IV)] are the gross analytical concentrations of ascorbic acid and platinum(IV), respectively. Rearrangement of Eq. (7) gives Eq. (8);

$$ k = \frac{{k^{\prime } K_{1} K_{2} }}{{(K_{2} + [{\text{Cl}}^{ - } ])(K_{1} + [{\text{H}}^{ + } ])}} $$
(8)

where k is the observed second-order rate constant (Table 1).

Since the variation of chloride ion concentration was made at constant hydrogen ion concentration, Eq. (8) can be simplified to Eq. (9),

$$ k = \frac{{k^{\prime } K_{1} K_{2} ({\text{A}})}}{{\left( {K_{2} + [{\text{Cl}}^{ - } ]} \right)}} $$
(9)

where \( ( {\text{A)}} = \frac{1}{{\left( {K_{1} + [{\text{H}}^{ + } ]} \right)}} \) is a constant that co-relates to hydrogen ion concentration.

Taking double reciprocals of Eq. (9) and rearranging, Eq. (10) is obtained.

$$ 1 /k = \frac{{ [ {\text{Cl}}^{ - } ]}}{{kK_{1} K_{ 2} ( {\text{A)}}}} + \frac{ 1}{{kK_{1} ( {\text{A)}}}} $$
(10)

A plot of 1/k versus [Cl−] was made according to Eq. (10), giving a straight line with nonzero intercept. The gradient (G) and intercept (I) of such a plot are given by Eqs. (11) and (12), respectively.

$$ G = \frac{1}{{k^{\prime } K_{1} K_{2} ({\text{A}})}} $$
(11)

and

$$ I = \frac{1}{{k^{\prime } K_{1} ({\text{A}})}} $$
(12)

the value of K 2 was calculated from the ratio of intercept and gradient as (0.37 ± 0.10) mol dm−3.

Equations (11) and (12) in terms of hydrogen ion concentration can be rewritten and rearranged as Eqs. (13) and (14), respectively, as follows:

$$ G = \frac{1}{{k^{\prime } K_{2} }} + \frac{{[{\text{H}}^{ + } ]}}{{kK_{1} K_{2} }} $$
(13)

and

$$ I = \frac{1}{{k^{\prime } }} + \frac{{[{\text{H}}^{ + } ]}}{{k^{\prime } K_{1} }} $$
(14)

Further plots according to Eq. (13) between gradient (G) and [H+] (Fig. 2) and (I) versus [H+] from Eq. (14) (Fig. 3) also yielded straight lines with nonzero intercepts. The values of k′K 2 and k′K 1 K 2 were calculated from the intercept and slope of Figs. 2 and 3, and the ratio of these two yielded K 1 as (0.51 ± 0.10) mol dm−3. Similarly, k′ was calculated from the intercept as (0.38 ± 0.10) mol dm−3.

Fig. 3
figure 3

A plot of intercept versus [H+]. I = 1.0 mol dm−3; 30 °C

Full size image

Conclusion

The previous kinetic analysis of this system was deficient; in that, the effect of chloride ion concentration on the rate of the reaction was not treated properly. In fact, the ionic strength has a pronounced effect on the rate, whereas chloride ion inhibits the rate. The revised mechanism proposed in this paper accounts separately for the effects of both ionic strength and chloride ion. The reaction appears to occur via an outer-sphere mechanism.