Changes in the Mechanism of the Fenton Reaction

Abstract

The kinetics of modified versions of the model of the Fenton reaction have been investigated. In these versions, radicals are produced by splitting FeO²⁺ (dissociation product of Fe²⁺ ozonide Fe²⁺O₃) into Fe³⁺ and OH^. The analysis shows that the revised models have the same shortcomings as the corresponding models of Haber and Weiss and of Barb et al. A nonradical model, based on an intact FeO²⁺ as an intermediate, accounted satisfactorily for the kinetics of the reaction under the same conditions. The amphoteric nature of FeO²⁺ to form FeOH³⁺ and HOFeO⁺ in reactions with H⁺ and OH⁻, respectively, extends its activity to a wide range of pH values.

1. Introduction

The mechanism of the reaction between H₂O₂ and Fe²⁺ (the Fenton reaction) has been intriguing researchers for over a century [1,2,3,4,5,6,7,8]. Models suggested for the reaction are based on two types of intermediates: (1) the free radicals OH^. and HO₂^. and (2) the non-radical species FeO²⁺. A combination of the two types has also been suggested to operate in different pH ranges. Active intermediates are present in reaction mixtures at low concentrations. Their identification is therefore difficult and often subject to controversies. A new line of investigations of these systems has started with the study of ozonide complexes of metal ions.

2. Discussion

Ozone reacts with Fe²⁺ ions in water, forming the complex (FeO₃)²⁺. It decomposes spontaneously to form FeO²⁺ and O₂ [9].

(FeO₃)²⁺ → FeO²⁺ + O₂

This equation suggests that FeO²⁺ may also be the intermediate in the reaction between H₂O₂ and Fe²⁺—in contradiction to the free radical theory. In an effort to resolve this contradiction, it was suggested by Batanieh et al. that FeO²⁺ is split in acidic solutions (pH < 3) by H⁺ to yield the radical OH^. [10].

FeO²⁺ + H⁺ → Fe³⁺ + OH^.

(1)

The OH^. radical was assumed to initiate a set of free radical reactions forming a free radical scheme for the Fenton reaction. The purpose of the present investigation is to elucidate the pathways of evolution of O₂ in such a system. The possible reactions (excluding reactions between two radicals) are

OH^. + H₂O₂ → HO₂^. + H₂O

(2)

HO₂^. + H₂O₂ → O₂ + OH^. + H₂O

(3)

OH^. + Fe²⁺ → Fe³⁺ + OH⁻

(4)

HO₂^. + Fe²⁺ → Fe³⁺ + HO₂⁻

(5)

HO₂^. + Fe³⁺ → Fe²⁺ + O₂ + H⁺

(6)

Fe²⁺ + H₂O₂ → Fe³⁺ + OH⁻ + OH^.

(7)

With HO₂^. being the precursor of O₂, H₂O₂ must be present in the system: step 2 must be a part of it. At t = 0, [Fe²⁺] and [Fe³⁺] are zero. The course of the reaction is then 1–2–3. After initiation in step 1, the pair 2–3 forms a chain reaction converting H₂O₂ to O₂ and H₂O (2 H₂O₂ = 2 H₂O + O₂). This scheme, together with the chain-terminating step 4, called the mechanism of Haber and Weiss, was the first mechanism for the Fenton reaction based on free radical intermediates [11]. It was later criticized as it did not explain the existence of an upper limit to the amount of O₂ produced at high [H₂O₂] levels (an effect discovered later by Barb et al.) [12] The Haber–Weiss chain can be eliminated from the scheme by removing reaction 3 from it. Using the remaining set of free radical steps and by using the steady state approximation for the concentrations of radical intermediates, we obtain

(d[OH^.]/dt) = k₁ [H⁺] [FeO²⁺] + k₇ [Fe²⁺] [H₂O₂] — (k₂ [H₂O₂] + k₄ [Fe²⁺]) [OH^.] = 0

(8)

(d [HO₂^.]/dt) = k₂ [H₂O₂] [OH^.] – (k₅ [Fe²⁺] + k₆ [Fe³⁺]) [HO₂^.] = 0

(9)

[OH^.] = (k₁ [H⁺] [FeO²⁺] + k₇ [Fe²⁺] [H₂O₂])/(k₂ [H₂O₂] + k₄ [Fe²⁺])

(10)

[HO₂^.] = k₂ [H₂O₂] [OH^.]/(k₅ [Fe²⁺] + k₆ [Fe³⁺])/DENOM

(11)

DENOM = (k₂ [H₂O₂] + k₄[Fe²⁺]) (k₅ [Fe²⁺] + k₆ [Fe³⁺])

The rate of evolution of O₂ becomes

d[O₂]/dt = (k₁ [H⁺] [FeO²⁺] + k₇ [Fe²⁺] [H₂O₂]) × R

(12)

R = k₂ k₆ [H₂O₂] [Fe³⁺]/DENOM

Because of the presence of the term k₁ [H⁺] [FeO²⁺] (rate of generation of OH^. radicals) in Equation (12), increasing [H⁺] will cause an increase in the rate of O₂ evolution, contrary to the experiment. At the start of the reaction, [Fe³⁺] and the rate of O₂ evolution are zero. As the reaction progresses, Fe³⁺ will be produced in step 1 and converted to Fe²⁺ in step 6; O₂ evolution will start also via steps 7–2–6. As time progresses, FeO²⁺ will gradually fall to a low level so that the rate of step 1 will become negligible beside that of step 7. The rate of O₂ evolution will then become

d[O₂]/dt = k₇ [Fe²⁺] [H₂O₂] × R

(13)

No dependence on [H⁺] remains in the rate expression: a change in the course of the reaction has occurred. The remaining steps of the scheme are then (7)–(2)–(4)–(5)–(6).

This is the free radical mechanism of Barb et al. [12]. In this mechanism, the O₂-producing step of Haber and Weiss (3) has been replaced by step (6), with the purpose of accounting for the existence of the upper limit of O₂ at high [H₂O₂]. It was found, however, that the revision did not solve the problem. No upper limit to [O₂] was predicted by the revised scheme either [13]. Since the course of the reaction in Batanieh et al.’s scheme becomes identical with that of Barb et al., the failure of the latter also implies failure of the former. A different breakup of FeO²⁺ to yield free radicals, not involving H⁺, can also be assumed (Logager et al., Equation (11) in [9]).

FeO²⁺ + H₂O → Fe³⁺ + OH^. + OH⁻

(14)

Obviously, hydrolytic and H⁺-induced breakups of FeO²⁺ are mutually exclusive. Both are assumptions. With H₂O₂ in the system, the course of the reaction in the two cases is analogous, except that in the case of hydrolytic splitting, [H⁺] does not appear in the rate equations. The failure to explain the limit of [O₂] applies also to this case. It has been concluded, therefore, that the case of free radical schemes in acidic solutions is not well established.

By investigating the Fenton reaction at higher pH values (6–7), Batanieh et al. concluded, on the basis of Mössbauer spectroscopic measurements on frozen reaction mixtures, the existence of a species which could be identified best with FeO²⁺ [10]. It appeared, therefore, that the identity of the active intermediate in the Fenton reaction changes from free radicals in acidic solutions to the non-radical species FeO²⁼ in neutral ones. A similar research group, however, performing similar Mössbauer measurements with the same system at pH = 1 (0.1 M HCl), found also that the spectrum was characteristic of being FeO²⁺. Although frozen reaction mixtures are not identical with those in water at room temperature, the results provide support to the conclusion that FeO²⁺ is the intermediate formed both in acidic and neutral solutions [14].

In a different field, Baerends and coworkers showed, by performing density function calculations, that the species formed in the reaction between Fe² and H₂O₂ is FeO²⁺, preceded by a very short living radical, OH^. [15]. Thus, the case for a nonradical intermediate seems to be supported by different sources. FeO²⁺ was first suggested as an intermediate in this reaction in 1932. Due to the lack of methods at that time to follow concentrations of components during the reaction as functions of time, no detailed mechanism could be given [16].

Returning to the kinetics of the Fenton reaction in acidic solutions, one can modify step 1: instead of breaking up FeO²⁺, H⁺ may be attached to the O atom of FeO²⁺ in the equilibrium

FeO²⁺ + H⁺ ⇌ Fe(OH)³⁺

(15)

Fe(OH)³⁺ is the protonated form of FeO²⁺. Assuming that Fe(OH)³⁺ reacts with Fe²⁺ to produce two ions of Fe³⁺ and the form FeO²⁺ with H₂O₂ to produce one molecule of O₂, an increase in [H⁺] (at constant [Fe²⁺] and [H₂O₂]) will shift the balance towards more Fe(OH)³⁺ and less FeO²⁺, causing a decrease in the rate of O₂ evolution, as actually observed [17,18]. There is thus a variation in products as a function of pH in the range of pH = 0–3. The variation is not connected to any change in the number of equivalents of oxidation transferred in the reaction: two equivalents are transferred either from FeO²⁺ to H₂O₂ to form O₂ or from Fe(OH)³⁺ to Fe²⁺ to form two Fe³⁺ ions. Since FeO²⁺ is considered to be a two-equivalent oxidizer also at higher pH values (6–7), there is no difference, in this respect, between intermediates in the reaction in pH ranges 1–3 and 6–7 [10]. It has been reported that by studying the Fenton reaction in non-coordinating buffers above pH = 5, the rate constant of the reaction becomes proportional to [OH⁻] [19]. The phenomenon can be accounted for by assuming the equilibrium

HO⁻ + FeO²⁺ ⇌ HOFeO⁺

(16)

FeO²⁺ is thus a versatile intermediate. It can account, due to its amphoteric character, for catalysis in the Fenton reaction in a wide range of pH values.

3. Conclusions

3.1. Mechanisms and Assumptions

Assumptions are indispensable in constructing mechanisms of chemical reactions. Steps in a proposed mechanism are based on assumptions. There are, however, restrictions and pitfalls. The proposed steps must follow a logical path from reactants to the products. The same holds for assumptions. Correctness of a proposed mechanism is judged by its ability to account for experimental results. Measurements are, however, performed in a limited section of the reaction for a limited number of reactants and/or products (occasionally of intermediates). Agreement between calculated and measured data supports a mechanism but cannot prove its correctness beyond doubt. It is generally accepted, however, that we interpret agreement as proof for a mechanism. On the other hand, if calculated and measured data disagree, then the mechanism is definitely wrong.

3.2. Thermodynamic Limitation of Free Radical Reactions

At present, the postulation that free radicals play a central part in oxidation–reduction reactions in aqueous media is widely accepted. “The importance and ubiquity of OH^. radicals have been increasingly apparent” (Walling) [1]. A similar opinion has been expressed by Koppenol and Stanbury [19,20]. However, basing mechanisms on free radicals with compulsory steady states of their concentrations puts severe restrictions on the rate equations. This was the reason for the failure of free-radical-based schemes to account for the existence of an upper limit of O₂ evolution (a non-radical scheme later achieved this [13]). Thermodynamic considerations also limit the formation of free radicals. Considering the two basic radical forming reactions, reaction (7) mentioned above and

Fe³⁺ + H₂O₂ → Fe²⁺ + H⁺ + HO₂^.

(17)

of the two, only step (7) is possible. Step (17) is connected with an increase of the free energy, and is therefore forbidden. Kinetic models including step (17) are not valid [20,21,22].

Conclusion: ideas accepted in consensus are not always correct.

This article is dedicated to the memory of Professor Gabriel Stein of the Hebrew University.