Kinetics and Mechanism of the Autocatalytic Oxidation of Bis

Article pubs.acs.org/IC

Kinetics and Mechanism of the Autocatalytic Oxidation of Bis(terpyridine)iron(II) by Peroxomonosulfate Ion (Oxone) in Acidic Medium Gábor Bellér,* Gábor Lente, and István Fábián Department of Inorganic and Analytical Chemistry, University of Debrecen, H-4032 Debrecen, Egyetem tér 1, Hungary S Supporting Information *

ABSTRACT: The autocatalytic oxidation of the bis(terpyridine)iron(II) complex, Fe(tpy)22+ by peroxomonosulfate ion (PMS) proceeds via the formation of the corresponding iron(III) complex (Fe(tpy)23+) as the primary oxidation product. The proton-assisted dissociation of Fe(tpy)22+ and subsequent oxidation of Fe2+ are side reactions in this system. In the initial stage of the reaction, a 1:1 adduct is formed between PMS and bis(terpyridine)iron(II), which decomposes in an intramolecular electron transfer reaction step. The autocatalytic role of Fe(tpy)23+ was also confirmed in the overall process. This effect is interpreted by considering the formation of an additional adduct between PMS and Fe(tpy)23+. The decomposition of the adduct yields two strong oxidizing intermediates, an Fe(IV) species and SO4−•, which consume the iron(II) complex in rapid reaction steps. A detailed kinetic model was postulated for the overall oxidation of Fe(tpy)22+ by PMS. The equilibrium constants for the formation of the adducts between PMS and complexes Fe(tpy)22+ and Fe(tpy)23+ were estimated as 129 ± 18 M−1 and 87 ± 10 M−1, respectively. In contrast to the closely related Fe(phen)32+-PMS reaction, the N-oxide derivative of the ligand (tpyO) does not have any kinetic role in the overall process because of the very slow formation of the N-oxide in the reaction.

■

bipyridine complexes, and the half-life of the first order decomposition is 6.4 min in 0.5 M H2SO4.13 There are several reports in the literature on the kinetics of the oxidation of Fe(tpy)22+ under strongly acidic conditions. The oxidation is fast with O3,14 Mn(IV),15 Mn(III),16 and Co(III).17 In these systems, the stopped-flow technique was used for monitoring the absorbance change, and the reactions follow overall second-order kinetics (the order with respect to both Fe(tpy)22+ and the oxidants is one), with the exception of O3. The reaction of the complex with excess ozone does not obey a simple first-order rate equation. The deviation from the first-order pattern in the early stage of the reaction was interpreted by assuming a process between Fe(tpy)22+ and hydroxyl radical (OH·), an intermediate of the oxidation: this additional step results in a second-order contribution of Fe(tpy)22+ in the rate law. Interestingly, the order with respect to the complex becomes one in the presence of chloride ion, which was explained by the scavenging effect of Cl− for OH·. Peroxomonosulfate ion (PMS) is an environmentally benign, nonchlorine oxidizing agent, used in a wide variety of industrial and consumer applications (e.g., decolorizing agent in denture cleansers, shock-oxidizer for swimming pools, repulping agent in papermaking, etc.). Lately, the use of PMS has increased rapidly in organic syntheses.18,19 The main reasons behind its

INTRODUCTION The complexes of terpyridine (2,2′:6′,2″-terpyridine, tpy) have been receiving growing attention in various fields of chemistry. Recent reviews discuss the role of tpy complexes in supramolecular chemistry and their applications in photochemistry, catalysis, or biomimetic studies.1,2 The bis(terpyridine) complex of iron(II) (Fe(tpy)22+) is characterized by an intense purple color and high thermodynamic stability (lgß = 20.4).3 Its use as an analytical reagent dates back to the 1930s when a method was described for the determination of iron in seawater and in marine plankton.4 More recently, the iron(II) complex and its derivatives have been used for catalytic transformations, e.g., the epoxidation5 and hydrosilylation of alkenes,6 and amide formation from aldehydes.7 Extensive computational studies have also been made on the coordination environment, which controls the ligand field strength8 and the photophysical properties of these systems.9 Lately, it has been shown that Fe2+ quenches the fluorescence of the tpy ligand and stopped-flow fluorescence spectroscopy can be used to follow the kinetics of the coordination reaction.10 In acidic solutions, the reverse reaction occurs and the bis complexes of both Fe2+ and Fe3+ dissociate to give the protonated ligand and noncomplexed metal ions. The pH dependence of the rate of acid dissociation of Fe(tpy)22+ has been studied in detail.11,12 Bis(terpyridine)iron(III) dissociates significantly faster than the corresponding phenanthroline and © 2017 American Chemical Society

Received: April 17, 2017 Published: June 26, 2017 8270

DOI: 10.1021/acs.inorgchem.7b00981 Inorg. Chem. 2017, 56, 8270−8277

Article

Inorganic Chemistry

the molar absorption coefficient of Fe(tpy)22+: λmax = 552 nm and ε = 1.222 × 104 M−1 cm−1.30 The Preparation of the Fe(III) Complex. The bis(tpy)iron(III) complex cannot be prepared by mixing Fe(III) and the ligand. Thus, it was synthesized by oxidizing Fe(tpy)22+.31 The typical chemical oxidants used for this purpose are lead(IV) oxide, aqueous cerium(IV) salts, or chlorine. The disadvantages of the use of chlorine were discussed earlier in the literature.25 In this study, two oxidants, Pb(IV) oxide and Ce(IV), were used to produce Fe(tpy)23+, and each procedure served different purposes. Similarly to Fe(phen)32+,32 Ce(IV) oxidizes Fe(tpy)22+ in a fast reaction. This provides a rapid and quantitative conversion of the iron(II) complex into the iron(III) complex without the interference of the dissociations of these complexes. By using Ce(IV), a reliable molar spectrum of the product was easily obtained, and the protonassisted dissociation of Fe(tpy)23+ was investigated in further experiments (see Supporting Information). However, this is not the most advantageous way to produce bis(terpyridine)iron(III) for our kinetic studies because the byproduct of the oxidation, Ce(III), is also oxidized by PMS.23 This problem can be circumvented if lead dioxide is used instead of Ce(IV). PbO2 was employed earlier in several cases in the literature to obtain the corresponding Fe(III) complexes when the direct mixing of the metal ion and the ligand is not feasible.25,30,31,33 In the oxidation by PbO2, the excess of oxidant and the byproduct of the reaction (PbSO4) can be removed by filtration. In this procedure, a large excess of the oxidant was used in order to achieve the oxidation in a shorter time because the rate of the dissociation of Fe(tpy)23+ is relatively fast. The optimal conditions were established as follows: ∼0.3 M H2SO4, 25 °C, ∼0.4 mM Fe(tpy)22+, ∼40 mg PbO2 (in 10 mL overall volume) and a stirring time of 1 min. The reagent solution was obtained by filtration with a glass filter. It should be mentioned that the yield of Fe(tpy)23+ was about 80− 85%, and a small amount (1−3%) of the iron(II) complex remained unconverted (SI Figure S1A,B). From the instant of mixing the iron(II) complex with the acidic slurry of PbO2 until starting to record the spectrum of the iron(III) complex, the procedure takes at least 2− 2.5 min. In this time, both complexes dissociate to some extent. Thus, the smaller than 100% yield was most probably due to the length of the procedure and the relatively fast dissociations of the complexes. Yet, this way of preparing the iron(III) complex was suitable to investigate the effect of Fe(tpy)23+ on the reaction of Fe(tpy)22+ with PMS. Instrumentation and Computation. UV−vis spectra and kinetic curves were recorded on PerkinElmer Lambda 25, Shimadzu UV 1800 scanning, and HP-8543 diode-array spectrophotometers at constant temperature maintained by the use of different thermostats with the various instruments. All measurements were performed at 25.0 ± 0.1 °C. Standard 1.000 cm quartz cuvettes were used. Fast kinetic experiments were carried out in an Applied Photophysics SX 18-MV stopped-flow instrument. Absorbance traces were collected in an optical cell of 1.000 cm path length. The kinetic runs were repeated 3−5 times, and the recorded signals were averaged to increase the signal-to-noise ratio. The oxidation of Fe(tpy)22+ by Ce(IV) was studied by this method, but this reaction is not discussed in detail in the present paper. It follows straightforward second order kinetics, and the corresponding rate constant is shown in the Supporting Information. The iodometric measurements were performed with a Metrohm 785 DMP Titrino automatic titrator equipped with a 6.0451.100 combined electrode. Electrospray ionization mass spectrometric (ESI-MS) analysis was carried out by a MicrOTOF-Q instrument (Bruker Daltonik, Bremen, Germany) in the positive ion mode. The mass spectra were calibrated using the exact masses of the clusters generated from the electrosprayed solution of sodium trifluoroacetate (NaTFA). The spectra were analyzed with the DataAnalysis 3.4 software from Bruker. The aim of the MS experiments was to identify the intermediates and products of the reactions in a quick and sensitive, but strictly qualitative, way.

popularity are favorable features such as stability, simple handling, nontoxic nature, good solubility in water, versatility of the reagent, and low costs. The oxidation reactions of PMS with noncomplementary reducing agents such as iron(II) and its complexes are interesting because the formation of reactive sulfur intermediates is highly probable in these systems. Studies of such oxidations could provide valuable information about sulfur intermediates and may help to gain a better insight into advanced oxidation processes performed by PMS, where sulfur radicals are often considered to be the active oxidizing species.20−22 During the oxidation of aqueous Fe2+, unexpectedly complex kinetic phenomena were found partly due to the noncomplementary features of the redox process.23 In order to get a better understanding of the intimate nature of such reactions, the oxidations of further iron(II) complexes (tris(phenanthroline)iron(II), Fe(phen)32+, and tris(bipyridine)iron(II), Fe(bipy)32+) were investigated. Highly unusual and diverse kinetic behavior was observed.24,25 In the initial stages of the reactions, the formation of an adduct between the reactants was confirmed in both systems. However, the studied redox reactions feature significant differences involving autoinhibition and ligand oxidation at longer reaction times. This paper reports the kinetics and mechanisms of the redox reaction of PMS with the bis(terpyridine)iron(II) complex. Our goal is to demonstrate that, in spite of rather common and plausible expectations, the redox reactions of very closely related complexes with the same oxidant exhibit very distinct kinetic and mechanistic features.

■

EXPERIMENTAL SECTION

Materials. All chemicals utilized in this study were of analytical reagent grade, purchased from commercial sources and used as received without further purification. Doubly deionized and ultrafiltered water from a Millipore Q system was used to prepare the stock solutions and samples. 2,2′:6′,2″Terpyridine was acquired from Sigma-Aldrich. Potassium peroxomonosulfate stock solutions were freshly prepared every day from Oxone (2KHSO5·KHSO4·K2SO4, Aldrich) and standardized by iodometric titration. These acidic stock solutions were found to be stable for at least 24 h. The constant pH of the reaction mixtures was maintained with a sufficient amount of sulfuric acid (H2SO4). The pH is defined as −lg[H+] throughout this paper and calculated from the analytical concentrations of the reagents by taking their acid dissociation into account.26 According to earlier literature reports, PMS decomposes to give H2O2 and H2SO4 under strongly acidic conditions (pH < 1).27,28 However, our experiments clearly show that within the time range of our kinetic studies, H2O2 does not form even when the H2SO4 concentration is as high as 2.75 M. The presence of H2O2 was tested by using titanium(IV) oxysulfate reagent (TiOSO4) to produce pertitanic acid. This method is based on the procedure reported by Eisenberg in 1943.29 Our results are reported in the Supporting Information. The iron(II) complex was prepared by directly reacting appropriate amounts of iron(II) sulfate (FeSO4·7H2O, Reanal) and the ligand. The complex was precipitated with a large excess of perchlorate ion from the concentrated solution as the perchlorate salt. The mixture was cooled down in a refrigerator, and then the precipitate was filtered, washed with cold water, and dried to constant mass at 45 °C. The product was soluble enough in pure water to prepare samples with the concentration required in the kinetic study. The stock solution of the iron(II) complex was stable for several months. The concentration of the samples were standardized by UV−vis spectrophotometry using 8271


Article

Inorganic Chemistry The primary data sets of the measurements were processed with the instrument controlling software. Further data evaluation was carried out with Microsoft Excel. It was also used for the initial rate method analysis of the kinetic data in complex systems. The initial rate was calculated from the initial 15−20% absorbance loss at 552 nm. The points were fitted to a second-order polynomial function, and the first degree term represents the initial rate. The software Scientist was used for nonlinear least-squares fittings of the kinetic traces at a given wavelength and for the evaluation of the concentration dependence of the rate constants.34 The matrix rank analysis of the time-resolved spectral changes was carried out by Matlab.35 The concentrations of various absorbing species were calculated by the direct linear algebraic method reported earlier,25,36 which is based on an overdetermined system of simultaneous linear equations. It was convenient to use matrix formalism with the software Matlab, i.e., eq 1:

⎛ε A ⎜ 1 ⎜ε A ⎜ 2 ⎜ ... ⎜ A ⎝ εn

ε1B ε1i ⎞ ⎛ ⎞ ⎟⎛ cA ⎞ ⎜ A1 ⎟ B Ai ⎟⎜ ⎟ ε2 ε2 ⎜ A 2 ⎟ or ε c = A ⎟ cB = ... ... ⎟⎜⎜ ⎟⎟ ⎜ ... ⎟ c ⎟⎝ i ⎠ ⎜⎝ A n ⎟⎠ εnB εni ⎠

Typical kinetic curves in the presence and absence of the oxidant are shown in Figure 1. The most notable feature of the

Figure 1. Representative kinetic traces recorded in the reaction between Fe(tpy)22+ and PMS. Circles: experimental data. Only 4% of the recorded points are shown for clarity. Lines: results of the fit of the curves to eqs 14 and 15. [Fe(tpy)22+]0 = 52.3 μM; [PMS]0 = 24.1 mM (a); 9.39 mM (b); 2.68 mM (c); no PMS is added (d); [H2SO4] = 0.323 M; T = 25.0 °C; λ = 552 nm; optical path length = 1.000 cm.

(1)

The concentration vector can be calculated using the pseudoinverse, pinv, of the molar absorbance matrix ε (eq 2).

c = pinv( ε)A

traces is the linear absorbance decay at higher initial PMS concentrations. Thus, the curves cannot be interpreted by a straightforward pseudo-first order approximation. The almost perfectly linear concentration change of Fe(tpy)22+ implies that the order with respect to the Fe(II) complex becomes zero under certain conditions. This is consistent with a complex mechanism. The initial rate method was used to study the reaction, which is free from interference from later processes. The initial rate was derived from the initial rate of absorbance change at 552 nm, which is the absorption maximum of the iron(II) complex, and was defined as follows (eq 4).

(2)

■

RESULTS AND DISCUSSION Initial Stage of the Oxidation of Fe(tpy)22+ by PMS. The complex is known to undergo proton-assisted dissociation (eq 3) in acidic medium.11,12 The observations taken under oxidant-free conditions were consistent with earlier reports: the dissociation follows first order kinetics with respect to Fe(tpy)22+ (eq 3), and the pseudo-first order rate constant, kobs, increases with decreasing pH. It was also found that Fe(tpy)22+ is stable above pH = 2.5. However, below pH ≈ 0.7, the dissociation of the complex goes to completion. The experiments were done under highly acidic conditions (at pH = 0.42, in 0.323 M H2SO4).

v0 = −

(3)

where k3 is a pH dependent apparent rate constant which is reported in Table 1. Table 1. Thermodynamic and Kinetic Parameters for the Reaction between Fe(tpy)22+ and PMS parameter k3 K5

k6 k9 k10 k11 K16 k17

value (4.0 ± 0.1) 129 ± 18 27 63 (4.8 ± 0.2) (1.6 ± 0.1) 1.8 × 10−3 3.7 × 104 (5.4 ± 0.4) 87 ± 10 (4.9 ± 0.2)

× 10−4

× 10−4 × 10−3

× 10−3 × 10−3

unit

remark

s−1 M−1 M−1 M−1 s−1 s−1 s−1 M−1 s−1 M−1 s−1 M−1 s−1

a a b c a a 23 °C, 0.1 M H2SO413 25 °C, 0.1 M H2SO423 60 °C, 1.00 M H2SO4 a a

(4)

This method can be used either when the contribution of the products to the absorbance is negligible or the stoichiometry is uniform over the course of the reaction. In this case, the first condition was satisfied, and the initial rate was used to explore the basic kinetic features of the reaction. In a large excess of the oxidant, the rate varies linearly with the Fe(tpy)22+ concentration (Figure 2A). It was also confirmed that the kinetic traces are not affected by the addition of the free ligand (tpy) in excess. The initial rate shows saturation with an intercept as a function of PMS concentration (Figure 2B). The intercept gives the rate of the proton-assisted dissociation of the complex at the given acidity (eq 3). Iron(II) produced in this reaction is rapidly oxidized by PMS.23 In principle, the noted saturation character of the concentration dependency can be consistent with a kinetic model that includes first the initial dissociation of a ligand from the complex and then the oxidation of Fe(tpy)2+ by PMS in a subsequent reaction step. In this case, the rate-determining step would shift from the oxidation to the initial conversion of the complex as the initial PMS concentration increases. This model implies that the order for Fe(tpy)22+ should deviate from one. Furthermore, the reaction rate should be dependent on the free ligand concentration but such an effect was not observed when tpy was added to the reaction mixture. Thus, this interpretation was ruled out. In analogy with the oxidation of Fe(phen)32+ and Fe(bipy)32+ by PMS, the formation of an adduct between the

Fe(tpy)2 2 + + 4H+ → Fe 2 + + 2H 2tpy 2 + v = k 3[Fe(tpy)2 2 + ]

dA(552 nm) dt

This work: 25.0 °C, 0.323 M H2SO4. bAnalogous formation equilibrium constant for [Fe(bipy)32+·HSO5−].24 cAnalogous formation equilibrium constant for [Fe(phen)32+·HSO5−].24

a

8272


Article

Inorganic Chemistry

the Fe(phen)32+-PMS and Fe(bipy)32+-PMS are listed in the table as well. The concentration dependencies of the initial rate strongly suggest the formation of the adduct (eq 5) and ESI-MS results in the related Fe(phen)32+-PMS reaction24 provided further evidence for the existence of such species. In this system, we could not detect the ion pair by the ESI-MS method. The possible reasons for this problem are discussed in the end of the next section. The UV−vis spectra of the adduct and Fe(tpy)22+ are indistinguishable implying that the coordination environment around the metal center must be rather similar in the adduct and the parent complex. Kinetic Studies of the Further Stage of the Reaction. Unique kinetic curves were recorded at various Fe(tpy)22+ concentrations in a large excess of the oxidant (Figure 3). A

Figure 2. Dependence of the initial rate on the reactant concentrations. [H2SO4] = 0.323 M; T = 25.0 °C; λ = 552 nm; path length = 1.000 cm. (A) dependence on Fe(tpy)22+. [PMS]0 = 14.8 mM. (B) dependence on PMS. [Fe(tpy)22+]0 = 52.3 μM. The two sets of data were evaluated together using proportional weighting. Lines: results of best fit to eq 8.

Figure 3. Kinetic traces detected in the reaction between Fe(tpy)22+ and PMS. Circles: experimental data. Only 4% of the recorded points are shown for clarity. Lines: results of the fit of the curves to eqs 14 and 15. [PMS] = 14.8 mM; [Fe(tpy)22+]0 = 9.34 μM, 18.7 μM, 37.4 μM, 52.3 μM and 74.7 μM in the order of increasing absorbance; [H2SO4] = 0.323 M; T = 25.0 °C; λ = 552 nm; path length = 1.000 cm. Along the horizontal red line, the Fe(tpy)22+ concentrations are the same in each kinetic trace.

reactants seems to be the simplest explanation for the experimental observations.24 Thus, the appropriate kinetic model includes the acid catalyzed dissociation of the complex (eq 3), an adduct formation of the reactants (eq 5), the rate determining electron transfer step (eq 6) and a fast oxidation by SO4−• (eq 7): Fe(tpy)2 2 + ( +4H+) → Fe 2 + + 2H 2tpy 2 +

(3)

closer inspection of these curves reveals a small but well-defined acceleration of the absorbance change before the traces become linear. At 552 nm, the only absorbing species is Fe(tpy)22+ in this system, thus, the horizontal red line marks the reaction times at which the concentrations of the reactants can be considered identical for all practical purposes. (Since PMS is used in large excess, its concentration is practically constant and the same in each kinetic run.) Apparently, the higher the initial Fe(tpy)22+ concentration, i.e., the more the reaction proceeds to reach the same concentration of Fe(tpy)22+ (cf. the horizontal line in Figure 3), the higher the rate is. The simplest explanation for this observation is autocatalysis. This assumption was tested in additional experiments. The reaction was allowed to reach completion (cycle 1), then the consumed amounts of Fe(tpy)22+, PMS, and H+ was added to the aged solution in order to have identical initial conditions except for the oxidation products (cycle 2). This procedure was repeated as many times as the volume of the cuvette enabled us to do so (up to five cycles). As Figure 4 shows, the reaction is getting faster in every successive cycle due to the increasing amount of oxidation products. This clearly proves again the autocatalytic nature of the reaction. The kinetic model for the overall reaction is summarized in Scheme 1, which postulates two alternative reaction paths. The identification of the autocatalyst in this system is essential to describe the overall redox process in sufficient detail. In order

v = k 3[Fe(tpy)2 2 + ]

Fe(tpy)2 2 + + HSO5− ⇌ [Fe(tpy)2 · HSO5]+ K5 =

(5)

[Fe(tpy)2 · HSO+5 ] [Fe(tpy)22 + ][HSO−5 ]

[Fe(tpy)2 · HSO5]+ (+ H+) → Fe(tpy)2 3 + + SO4 −• + H 2O (6) +

v = k6[Fe(tpy)2 · HSO5 ] Fe(tpy)2 2 + + SO4 −• → Fe(tpy)2 3 + + SO4 2 −

(7)

v = fast On the basis of the kinetic model, the following expression was derived for the initial rate (eq 8): ⎛ 2k6K5[PMS]0 ⎞ v0 = [Fe(tpy)22 + ]0 ⎜k 3 + ⎟ 1 + K5[PMS]0 ⎠ ⎝

(8)

The rate constant of the proton-assisted dissociation of the complex was determined under oxidant-free conditions and held fixed during the fitting procedure. Parameters K5 and k6 were estimated by nonlinear least-squares fitting and are listed in Table 1. Analogous equilibrium constants for the ion pairs in 8273


Article

Inorganic Chemistry

oxide(s) was taken into consideration in this system as well (eq 11). However, it was found that the oxidation of tpy by PMS practically does not occur in metal-free solutions and strongly acidic medium at 25 °C (SI Figure S2). Still, the effect of the oxidized ligand (tpyO) on the oxidation of Fe(tpy)22+ was tested, but no acceleration was observed. Both the dissociation of Fe(tpy)23+ and the oxidation of Fe2+ (eqs 9 and 10) result in the formation of Fe3+. As a potential product of the reaction, it was added to the initial reaction mixture, but it did not enhance the rate of the reaction, either. To sum up, the protonated ligand (H2tpy2+), the oxidized ligand (tpyO), Fe2+ and Fe3+ were excluded as autocatalysts. Although the bis(terpyridine)Fe(III) complex (Fe(tpy)23+) is the primary product of the oxidation (eq 6), it is, in fact, an intermediate on a longer time scale, and as such, it should also be considered as the catalyst of the main reaction. At 702 nm, it has a significantly lower molar absorption (ε = 7.05 × 102 M−1 cm−1) than that of Fe(tpy)22+ at 552 nm (ε = 1.222 × 104 M−1 cm−1). By properly choosing the initial concentration of Fe(tpy)22+, both peaks can be monitored simultaneously (SI Figure S3). The time-dependent concentration profiles of Fe(tpy)22+ and Fe(tpy)23+ were calculated from the corresponding spectra.25 Figure 5 shows that the initial formation of the

Figure 4. Proof of the autocatalysis in the reaction between Fe(tpy)22+ and PMS. Effect of increasing the amount of the aged reaction mixture on the rate of oxidation. After the completion of a cycle, the reaction was restarted by replenishing the reactants to their initial concentrations. [PMS]0 = 45.0 mM; [Fe(tpy)22+]0 = 47.2 μM; [H2SO4] = 0.323 M; T = 25.0 °C; λ = 552 nm; path length = 1.000 cm. a: cycle 1, b: cycle 2; c: cycle 3; d: cycle 4; e: cycle 5. See details of the procedure in the text.

Scheme 1. Summary of the Processes in the PMS−Fe(tpy)22+ Reaction System under Acidic Conditions

to elaborate this issue, the considerations based on the following known processes need to be taken into account. Fe(tpy)2 3 + ( +4H+) → Fe3 + + 2H 2tpy 2 +

Figure 5. Concentration profiles of the tpy complexes in the reaction between Fe(tpy)22+ and PMS. Points: experimental data; only 10% of the recorded points are shown for clarity. Lines: results of the simultaneous fit of the two curves to eqs 14 and 15 (only the points before the break point were fitted). [Fe(tpy)22+]0 = 234 μM; [PMS]0 = 26.8 mM; [H2SO4] = 0.323 M; T = 25.0 °C.

(9)

v = k 9[Fe(tpy)2 3 + ] 2Fe2 + + HSO5− + H+ → 2Fe3 + + SO4 2 − + H 2O

v = k10[Fe2 +][PMS]

(10)

k10 = 3.7 × 104 M−1 s−1

tpy + HSO5− → tpyO + HSO4 −

iron(III) complex is followed by a decrease due its protonassisted dissociation (eq 9). The increase of [Fe(tpy)23+] ceases when the iron(II) complex is completely consumed, as indicated by the break points in the kinetic traces at the same instant. In order to test the possible role of Fe(tpy)23+ as an autocatalyst of the oxidation process, the complex was prepared by the oxidation of the corresponding iron(II) complex using solid PbO2. The experimental results clearly confirm that the addition of Fe(tpy)23+ to the reaction mixture accelerates the disappearance of the absorbance peak of the bis(terpyridine)iron(II) complex (SI Figure S4). The initial rate is linearly dependent on [Fe(tpy)23+] (Figure 6); thus, the reaction is first order with respect to the autocatalyst. The intercept corresponds to the rate of the noncatalytic reactions (the sum of the noncatalytic oxidation, eq 6, and the proton-assisted dissociation of the complex, eq 3). The autocatalytic effect of Fe(tpy)23+ can be interpreted in terms of a simplified kinetic model (eqs 12 and 13).

(11)

v = k11[tpy][PMS] The proton-assisted dissociation of Fe(tpy)22+ (eq 3) was briefly discussed earlier. It has a half-life of approximately 30 min under the acidity applied. The Fe(tpy)23+ complex undergoes dissociation (eq 9), too, and its half-life is about 7 min. The rate constants of these reactions are listed in Table 1. Both reactions produce the protonated ligand (mainly H2tpy2+ under such acidic conditions).37 When the free ligand was added to the initial reaction mixture, it did not influence the rate at all. This observation rules out the ligand as the autocatalyst. The oxidation of Fe2+ by PMS (eq 10) was confirmed to be fast on the time scale of the present study,23 and this reaction ensures that Fe2+ cannot accumulate as long as excess PMS is present. This excludes the presence of iron(II) in the aged solution; thus it cannot be the autocatalyst, either. Since the oxidation of the ligand plays a significant kinetic role in the Fe(phen)32+-PMS reaction,25 the formation of N-

Fe(tpy)22 + → Fe(tpy)32 + 8274

v = kobs1[Fe(tpy)22 + ]

(12)


Article

Inorganic Chemistry

Figure 6. Effect of Fe(tpy)23+ on the reaction between Fe(tpy)22+ and PMS. Dependence of the initial rate on the Fe(tpy)23+ concentration. [Fe(tpy)22+]0 = 117 μM; [PMS]0 = 321 mM; [H2SO4] = 0.323 M; T = 25.0 °C; λ = 552 nm; path length = 1.000 cm.

Figure 7. Dependence of the observed rate constant (kobs2) of the autocatalytic pathway on the PMS concentration. [Fe(tpy)22+]0 = 52.3 μM; [H2SO4] = 0.323 M; T = 25.0 °C; path length = 1.000 cm. Line: result of best fit to eq 19.

Fe(tpy)22 + + Fe(tpy)32 + → 2Fe(tpy)32 + v=

kobs2[Fe(tpy)32 + ]

discussed before, the noncatalytic path is also consistent with the formation of an adduct.

(13)

Fe(tpy)2 3 + + HSO5− ⇌ [Fe(tpy)2 · HSO5]2 +

where kobs1 and kobs2 are the pseudo-first order rate constants of the noncatalytic and autocatalytic pathways of the oxidation, respectively. Since PMS was used in large excess and in constant concentration, this species is not written explicitly in the rate expressions; thus, kobs1 and kobs2 are dependent on the PMS concentration. The inclusion of the acid catalyzed dissociations of the complexes leads to the following differential equations. d[Fe(tpy)22 + ] dt

dt

(17)

v = k17[Fe(tpy)2 · HSO5 ]

Fe(tpy)2 2 + + SO4 −• → Fe(tpy)2 3 + + SO4 2 −

(7)

v = fast

(14)

Fe(tpy)2 2 + + Fe(O)(tpy)2 2 + + 2H+ → 2Fe(tpy)2 3 + + H 2O

(18)

v = fast

(15)

The catalytic route of the oxidation is zeroth order with respect to the initial complex. This is only possible if the Fe(tpy)22+-consuming process occurs after the rate-determining step. Furthermore, after the total amount of iron(II) is oxidized to iron(III) (after the breakpoint shown in Figure 5), the rate of the consumption of the Fe(tpy)23+ does not depend on the PMS concentration. In agreement with this finding, when PMS was added to the freshly prepared iron(III) complex, the initial rate of the loss of Fe(tpy)23+ was independent of [PMS]0 (SI Figure S5). These observations strongly suggest that there is a direct reaction between Fe(tpy)23+ and PMS, which requires the presence of Fe(tpy)22+. This unique feature can be interpreted by assuming that the adduct between Fe(tpy)23+ and PMS decomposes in an equilibrium electron transfer process (eq 17). This reaction is almost completely shifted to the left when Fe(tpy)22+ is absent. However, the sulfate ion radical is quickly consumed in the presence of the iron(II) complex (eq 7). This reaction is much faster than the backward step of reaction 17. Consequently, reaction 17 practically becomes irreversible. Thus, the autocatalytic role of the iron(III) complex is attributed to the generation of two powerful oxidizing intermediates (sulfate ion radical and an iron(IV) species) in the rate-determining step (eq 17) which, in turn, quickly oxidize the iron(II) complex (eqs 7 and 18). The involvement of Fe(IV)tpy species (FeIV=O and FeIV=NR) has earlier been postulated as active reaction intermediates in

The kinetic traces were fitted to eqs 14 and 15 using a nonlinear least-squares algorithm. During the fitting procedure, k3, k9, and kobs1 were held fixed and only kobs2 was allowed to float. The value of kobs1 was calculated in each run by using [PMS]0, K5 and k6 according to the rearranged eq 8. Eqs 14 and 15 provide excellent fit of the experimental data up to ∼90− 95% consumption of the initial iron(II) complex. ⎛ 2k6K5[PMS]0 ⎞ v0 = [Fe(tpy)22 + ]0 ⎜k 3 + ⎟ 1 + K5[PMS]0 ⎠ ⎝ = k 3[Fe(tpy)22 + ]0 + kobs1[Fe(tpy)22 + ]0

[Fe(tpy)32 + ][HSO−5 ]

+

= kobs1[Fe(tpy)22 + ] + kobs2[Fe(tpy)32 + ] − k 9[Fe(tpy)32 + ]

[Fe(tpy)2 · HSO52 +]

[Fe(tpy)2 · HSO5]2 + ⇌ Fe(O)(tpy)2 2 + + SO4 −• + H+

= −k 3[Fe(tpy)22 + ] − kobs1[Fe(tpy)22 + ] − kobs2[Fe(tpy)32 + ]

d[Fe(tpy)32 + ]

K16 =

(16)

(8)

The individual evaluation of the kinetic traces in Figure 3 confirmed that the fitted rate constants are independent of the initial complex concentration and identical within the margin of error. The continuous lines in Figure 3 are the result of the simultaneous fit of the traces. As shown in Figure 7, the rate constant for the catalytic route also exhibits saturation as a function of PMS concentration. We propose a kinetic model, eqs 16−18, which postulates the formation of an adduct between Fe(tpy)23+ and HSO5− (eq 16). This is not unreasonable because the formation of such an intermediate seems to be a common feature in the oxidation of iron complexes of N-heteroaromatic ligands,24,25 and, as 8275


Article

Inorganic Chemistry catalytic cycles.5,7 A high-valent FeIV-oxo complex was detected by high-resolution electrospray ionization mass spectrometry (ESI-MS) analysis of the reaction mixture of the [Fe(Cl3tpy)2]2+-catalyzed epoxidation of cis-stilbene when PMS was used as the terminal oxidant.5 On the basis of these considerations, the experimental data were fitted to eq 19. The results are listed in Table 1, and the quality of the fit is demonstrated in Figure 7. kobs2 =

k17K16[PMS]0 1 + K16[PMS]0

It was confirmed that the reaction is autocatalytic due to the direct reaction between Fe(tpy)23+ and PMS. The experimental data are consistent with the formation of an adduct between the autocatalyst and PMS. Subsequent reactions include the formation of Fe(O)(tpy)22+ and SO4•− in a rate-determining step and the reactions of these intermediates with Fe(tpy)22+ in the final stage of the oxidation. Kinetic data suggest the formation of an iron(IV) species similar to the ones detected as active oxidizing species in epoxidation reactions by PMS where [Fe(Cl3tpy)2]2+-was used as a catalyst.5 Our results may contribute to the better understanding of catalytic reactions in which iron(II)/(III) complexes are used as catalysts and PMS is applied as the oxidant. It is worth comparing the oxidation reactions of three similar iron(II) complexes (Fe(tpy)22+, Fe(phen)32+, and Fe(bipy)32+) by PMS.24,25 In all systems, the primary oxidation product is the corresponding iron(III) complex, and the initial stage of the oxidation can be interpreted with a common kinetic model which includes an adduct formation between the reactants. However, these systems feature significant differences in the later stage of the reaction: the oxidation of Fe(bipy)32+ does not feature exotic kinetic behavior, that of Fe(tpy) 2 2+ is autocatalytic, while in the Fe(phen)32+-PMS system, the oxidation of the ligand triggers unexpected phenomena. In our earlier study, we showed that the ligand (phen) and the metal center of the Fe(phen)32+ complex are oxidized on a similar time scale,25 and the oxidation of phen yields the corresponding N-oxide (phenO).19 This product inhibits the oxidation of Fe(phen)32+ by inducing temporary reformation of the iron(II) complex from Fe(phen)33+. In contrast, the oxidations of the ligands bipy and tpy are significantly slower than those of the corresponding complexes, and the oxidized ligands (bipyO and tpyO) apparently play no role in the redox processes between PMS and these complexes.

(19)

ESI-MS measurements were carried out in order to identify the intermediates and products of the reaction. The peaks of the oxidation product, Fe(tpy)23+ were found in the spectrum. However, the proposed adducts could not be detected in contrast to the results obtained in the Fe(phen)32+-PMS reaction where the corresponding ion pair gave identifiable peaks.24 The association constant of the adduct (K5) is relatively high; thus, this species formed in substantial amount during the reaction [PMS]0 is large. The failure of confirming the existence of the adduct is due to experimental limitations. Most importantly, the ESI-MS experiments cannot be performed at the high acidity used in the kinetic studies. For the mass spectrometric studies, reaction mixtures of pH ≈ 3.5− 4.0 were prepared. Under these conditions, the oxidation of the complex is considerably faster than in strongly acidic medium (SI Figure S6). In principle, the large excess of the oxidant favors the formation of the adduct, but this transient species is undetectable because the reaction goes to completion by the time the sample is injected into the MS instrument. Under such mildly acidic conditions (pH = 3.5−4.0), the oxidation of the ligand is also faster than in strongly acidic medium. Traces of the N-oxide derivative of the ligand, tpyO, were among the oxidation products, but no experimental observations indicate the formation of this species under strongly acidic conditions. The equilibrium constants for the formation of the adducts are very similar (Table 1). These species can be considered as regular ion pairs. However, the stability constants are considerably higher than expected on the basis of the Fuoss equation.38 This and the fact that the stability constants do not differ for the +2 and +3 species strongly suggest that the formation of the adducts between PMS and the iron complexes is not governed by electrostatic forces. Most likely, some kind of interaction between the aromatic electron cloud and PMS controls the adduct formation.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.7b00981. Figures of the spectral changes and kinetic curves in the decomposition reaction of Fe(tpy)23+, kinetic traces in the reaction between tpy and PMS, spectral changes in the reaction between Fe(tpy)22+ and PMS, effect of Fe(tpy)23+ on the reaction between Fe(tpy)22+ and PMS, dependence of the initial rate of the loss of Fe(tpy)23+ on [PMS]0, kinetic curve in the reaction between Fe(tpy)22+ and [PMS] under mildly acidic conditions, table of the results of the tests of H2O2 (PDF)

■

CONCLUSION During the oxidation of Fe(tpy)22+ by peroxomonosulfate ion (PMS), the formation of Fe(tpy)23+, as the primary oxidation product was identified and unusual kinetic behavior was observed. Essentially, the results confirm the following reaction sequence for the primary oxidation reaction: (i) the acid assisted dissociation of the Fe(II) complex; (ii) the formation of an intermediate, an 1:1 adduct between the reactants, which is considered a fast pre-equilibrium; (iii) the rate determining intramolecular electron transfer step followed by the dissociation of the adduct into the corresponding Fe(III) complex and sulfate ion radical; and (iv) the fast oxidation of another Fe(II) complex by SO4•−. On the basis of the observations a kinetic model is proposed which elaborates the intimate details of these reactions.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel: + 36 52 512900/22327. Fax: + 36 52 518-660. ORCID

Gábor Bellér: 0000-0003-4005-1534 Notes

The authors declare no competing financial interest. 8276


Article

Inorganic Chemistry

■

(18) Hussain, H.; Green, I. R.; Ahmed, I. Journey Describing Applications of Oxone in Synthetic Chemistry. Chem. Rev. 2013, 113, 3329−3371. (19) Bellér, G.; Szabó, M.; Lente, G.; Fábián, I. Formation of 1,10Phenanthroline-N,N′-dioxide under Mild Conditions: The Kinetics and Mechanism of the Oxidation of 1,10-Phenanthroline by Peroxomonosulfate Ion (Oxone). J. Org. Chem. 2016, 81, 5345−5353. (20) Anipsitakis, G. P.; Dionysiou, D. D. Degradation of Organic Contaminants in Water with Sulfate Radicals Generated by the Conjunction of Peroxymonosulfate with Cobalt. Environ. Sci. Technol. 2003, 37, 4790−4797. (21) Zhang, T.; Zhu, H.; Croué, J.-P. Production of Sulfate Radical from Peroxymonosulfate Induced by a Magnetically Separable CuFe2O4 Spinel in Water: Efficiency, Stability, and Mechanism. Environ. Sci. Technol. 2013, 47, 2784−2791. (22) Guan, Y.-H.; Ma, J.; Li, X.-C.; Fang, J.-Y.; Chen, L.-W. Influence of pH on the Formation of Sulfate and Hydroxyl Radicals in the UV/ Peroxymonosulfate System. Environ. Sci. Technol. 2011, 45, 9308− 9314. (23) Lente, G.; Kalmár, J.; Baranyai, Z.; Kun, A.; Kék, I.; Bajusz, D.; Takács, M.; Veres, L.; Fábián, I. One- Versus Two-Electron Oxidation with Peroxomonosulfate Ion: Reactions with Iron(II), Vanadium(IV), Halide Ions, and Photoreaction with Cerium(III). Inorg. Chem. 2009, 48, 1763−1773. (24) Beller, G.; Bátki, G.; Lente, G.; Fábián, I. Unexpected Adduct Formation in the Reaction of Peroxomonosulfate Ion with the Tris-(2, 2′-bipyridine)iron(II) and Tris-(1, 10-phenanthroline)iron(II) Complexes. J. Coord. Chem. 2010, 63, 2586−2597. (25) Bellér, G.; Lente, G.; Fábián, I. Central Role of Phenanthroline Mono-N-oxide in the Decomposition Reactions of Tris(1,10phenanthroline)iron(II) and -iron(III) complexes. Inorg. Chem. 2010, 49, 3968−3970. (26) Lente, G.; Fábián, I. Ligand Substitution Kinetics of the Iron(III) Hydroxo Dimer with Simple Inorganic Ligands. Inorg. Chem. 2002, 41, 1306−1314. (27) Robert, A.; Meunier, B. Oxygenation of Hydrocarbons by KHSO5 Catalyzed by Manganese Porphyrin Complexes. New J. Chem. 1988, 12, 885−896. (28) Meunier, B. Potassium Monopersulfate - Just Another Primary Oxidant or a Highly Versatile Oxygen Atom Donor in Metalloporphyrin-mediated Oxygenation and Oxidation Reactions. New J. Chem. 1992, 16, 203−211. (29) Eisenberg, G. Colorimetric Determination of Hydrogen Peroxide. Ind. Eng. Chem., Anal. Ed. 1943, 15, 327−328. (30) Ford-Smith, M. H.; Sutin, N. The Kinetics of the Reactions of Substituted 1,10-Phenanthroline, 2,2′-Dipyridine and 2,2′,2″-Tripyridine Complexes of Iron(III) with Iron(II) Ions 1. J. Am. Chem. Soc. 1961, 83, 1830−1834. (31) Reiff, W. M.; Baker, W. A.; Erickson, N. E. Binuclear, OxygenBridged Complexes of Iron(III). New Iron(III)-2,2′,2″-Terpyridine Complexes. J. Am. Chem. Soc. 1968, 90, 4794−4800. (32) Dulz, G.; Sutin, N. The Kinetics of the Oxidation of Iron(II) and its Substituted tris(1,10-Phenanthroline) Complexes by Cerium(IV). Inorg. Chem. 1963, 2, 917−921. (33) Eichler, E.; Wahl, A. C. Electron-exchange Reactions between Large Complex Cations 1. J. Am. Chem. Soc. 1958, 80, 4145−4149. (34) Scientist, Version 2.0; Salt Lake City, UT, USA Patent. (35) MatLab for Windows, Version 4.2c1; Natick, MA, USA Patent. (36) Kerezsi, I.; Lente, G.; Fabian, I. Kinetics of the Light-Driven Aqueous Autoxidation of Sulfur(iv) in the Absence and Presence of Iron(ii). Dalton Trans. 2006, 955−960. (37) Offenhartz, P. O. D.; George, P.; Haight, G. P. The Ionization Constants for the Ligand 2,2′,2″-Tripyridine. J. Phys. Chem. 1963, 67, 116−118. (38) Fuoss, R. M. Ionic Association. III. The Equilibrium between Ion Pairs and Free Ions. J. Am. Chem. Soc. 1958, 80, 5059−5061.

ACKNOWLEDGMENTS This research was supported by the Hungarian Science Foundation (OTKA: NK-105156), as well as by the EU and cofinanced by the European Regional Development Fund under the project GINOP-2.3.2-15-2016-00008. This research was also supported by the European Union and the State of Hungary, cofinanced by the European Social Fund in the framework of TÁ MOP 4.2.4. A/2-11-1-2012-0001 ‘National Excellence Program’.

■

REFERENCES

(1) Hofmeier, H.; Schubert, U. S. Recent Developments in the Supramolecular Chemistry of Terpyridine-Metal Complexes. Chem. Soc. Rev. 2004, 33, 373−399. (2) Constable, E. C. 2,2′,6′,2″-Terpyridines: From Chemical Obscurity to Common Supramolecular Motifs. Chem. Soc. Rev. 2007, 36, 246−253. (3) Martin, R. B.; Lissfelt, J. A. The 2,2′,2″-Tripyridine System. J. Am. Chem. Soc. 1956, 78, 938−940. (4) Cooper, L. H. N. Iron in the Sea and in Marine Plankton. Proc. R. Soc. London, Ser. B 1935, 118, 419−438. (5) Liu, P.; Wong, E. L.-M.; Yuen, A. W.-H.; Che, C.-M. Highly Efficient Alkene Epoxidation and Aziridination Catalyzed by Iron(II) Salt + 4,4′,4″-Trichloro-2,2′:6′,2″-terpyridine/4,4″-Dichloro-4′-OPEG-OCH3−2,2′:6′,2″-terpyridine. Org. Lett. 2008, 10, 3275−3278. (6) Kamata, K.; Suzuki, A.; Nakai, Y.; Nakazawa, H. Catalytic Hydrosilylation of Alkenes by Iron Complexes Containing Terpyridine Derivatives as Ancillary Ligands. Organometallics 2012, 31, 3825− 3828. (7) Chow, T. W. S.; Chen, G. Q.; Liu, Y.; Zhou, C. Y.; Che, C. M. Practical Iron-catalyzed Atom/group Transfer and Insertion Reactions. Pure Appl. Chem. 2012, 84, 1685. (8) Bowman, D. N.; Bondarev, A.; Mukherjee, S.; Jakubikova, E. Tuning the Electronic Structure of Fe(II) Polypyridines via Donor Atom and Ligand Scaffold Modifications: A Computational Study. Inorg. Chem. 2015, 54, 8786−8793. (9) Nance, J.; Bowman, D. N.; Mukherjee, S.; Kelley, C. T.; Jakubikova, E. Insights into the Spin-State Transitions in [Fe(tpy)2]2+: Importance of the Terpyridine Rocking Motion. Inorg. Chem. 2015, 54, 11259−11268. (10) Munzert, S. M.; Schwarz, G.; Kurth, D. G. Kinetic Studies of the Coordination of Mono- and Ditopic Ligands with First Row Transition Metal Ions. Inorg. Chem. 2016, 55, 2565−2573. (11) Burgess, J.; Twigg, M. V. Kinetics of Aquation of Bis(2,2′;6′,2″Terpyridyl)iron(II) in Acidic Media and its Reaction with Cyanide Ion. J. Chem. Soc., Dalton Trans. 1974, 2032−2036. (12) Farina, R. D.; Hogg, R.; Wilkins, R. G. Rate-pH Profile for the Dissociation of Iron(II)- and Cobalt(II)-2,2′,2″-Terpyridine Complexes. Inorg. Chem. 1968, 7, 170−172. (13) Chen, Y. W. D.; Santhanam, K. S. V.; Bard, A. J. Solution Redox Couples for Electrochemical Energy Storage: I. Iron (III)-Iron (II) Complexes with O-Phenanthroline and Related Ligands. J. Electrochem. Soc. 1981, 128, 1460−1467. (14) Tomiyasu, H.; Gordon, G. Kinetics and Mechanism of the Oxidation of Bis(terpyridine)-iron (II) by Ozone in Acidic Aqueous Solutions. Ozone: Sci. Eng. 1989, 11, 59−67. (15) Satyanarayana, T.; Anipindi, N. R. Kinetics and Mechanism of Oxidation of Bis(2,2′,6′,2″-Terpyridine) Iron(II) by Manganese(IV). React. Kinet. Catal. Lett. 1992, 47, 111−118. (16) Satyanarayana, T.; Anipindi, N. R. Kinetics and Mechanism of Electron-Transfer between Bis(2,2′,6′,2″-Terpyridine)iron(II) and Manganese(III) in Sulphuric Acid Medium. Transition Met. Chem. 1992, 17, 533−534. (17) Satyanarayana, T.; Anipindi, N. R. Kinetics and Mechanism of Cobalt(III) Oxidation of Bis(2,2′,6′,2″-Terpyridine) Iron(II) in Sulfuric Acid Solutions. React. Kinet. Catal. Lett. 1992, 47, 333−341. 8277


Kinetics and Mechanism of the Autocatalytic Oxidation of Bis

Recommend Documents