chromium(IV) by hexacyanoferrate(III) in strongly alkaline media ... Kinetics of oxidation of the macrocyclic Cr(III) complex in alkaline solution has been studied ...
Transition Metal Chemistry 29: 634–643, 2004. 2004 Kluwer Academic Publishers. Printed in the Netherlands.
634
Kinetics and mechanism of a macrocyclic chromium(III) complex oxidation to chromium(IV) by hexacyanoferrate(III) in strongly alkaline media Janusz Chatłas, Olga Impert, Anna Katafias*, Przemysław Kita and Grzegorz Wrzeszcz Department of Chemistry, N. Copernicus University, 87-100 Torun´, Poland Jette Eriksen and Ole Mønsted Department of Chemistry, University of Copenhagen, Denmark Andrew Mills Department of Pure and Applied Chemistry, University of Strathclyde, Glasgow, UK Received 16 March 2004; accepted 20 March 2004
Abstract Oxidation of the macrocyclic Cr(III) complex cis-[Cr(cycb)(OH)2]+, where cycb ¼ rac-5,5,7,12,12,14-hexamethyl1,4,8,11-tetraazacyclotetradecane, by an excess of the hexacyanoferrate(III) in basic solution, slowly produces Cr(V) species. These species, detected using e.p.r. spectroscopy, are stable under ambient conditions for many hours, and the hyperfine structure of the e.p.r. spectrum is consistent with the interaction of the d-electron with four equivalent nitrogen nuclei. Electro-spray ionization mass spectrometry suggests a concomitant oxidation of the macrocyclic ligand, in which double bonds and double bonded oxygen atoms have been introduced. By comparison basic chromate(III) solutions are oxidized rapidly to chromate(VI) by hexacyanoferrate(III) without any detectable generation of stable Cr(V) intermediates. Kinetics of oxidation of the macrocyclic Cr(III) complex in alkaline solution has been studied under excess of the reductant. Rate determining formation of Cr(IV) by a second order process involving the Cr(III) and the Fe(III) reactants is seen. This reaction also involves a characteristic higher order than linear dependence on the hydroxide concentration. Reaction mechanisms for the processes, including oxidation of the coordinated macrocyclic ligand – under excess of the oxidant- are proposed.
Introduction The thermodynamic stability of chromium species in different oxidation states in aqueous solutions depends strongly on the ligand environment and the acidity of their solutions [1–2]. Thus, chromium(III), is stable as the hexaaquachromium(III) complex in acidic media, where chromium(II) and chromium(VI) species, such as hexaaquachromium(II) and dichromate, behave as reducing and oxidizing agents, respectively. In weakly basic solution, chromium(III) initially forms a dark green gel of Cr(OH)3, which is soluble in excess base, forming chromate(III), frequently written as [Cr(OH)6]3), but essentially a poorly defined complex system of monomers and polymeric species. Such solutions are easily oxidized and chromium(VI) is the stable oxidation product in the form of CrO42). Solutions of chromium species at the intermediate oxidation states, IV and V, have been studied intensively in recent years especially because of their suspected biological importance including attempts to
* Author for correspondence
understand the mutagenic and carcinogenic effects of chromium(VI) [3–9]. Chromium(IV) and chromium(V) species coordinated with water and deprotonated water ligands are unstable in aqueous solution irrespective of the acidity. In an acidic solution, chromium(V) is a 0 very strong two-electron oxidant, ECr V =CrIII ¼ 1:7 V, or a moderately strong one-electron oxidant, 0 ECr V =CrIV ¼ 1:34 V. Chromium(IV) is an even stronger 0 oxidant with ECr IV =CrIII ¼ 2:1 V in 1 M acid [1]. Given these high oxidation potentials it is no surprise that chromium(IV) and chromium(V) are very reactive, unstable species. Several preparative methods have been developed to produce stable chromium(V) complexes including: (i) reduction of chromium(VI) in the presence of a suitable chelating ligand, e.g. [8, 10–13], (ii) intermolecular oxidation of some chromium(III) species, e.g. [14–16], (iii) intramolecular photoredox chromium(III) transformation into chromium(V), e.g. [17–18], and (iv) oxidation of chromium(0) species in the presence of chromium(V) stabilizing ligands, e.g. [19]. Although chromium(IV) complexes are much more difficult to observe in solution, some notable studies have been carried out, e.g. [7, 20–22].
635 Mechanistic studies on successive steps of oxidation of chromium(III) in aqueous alkaline media are almost terra incognita, with the exception of a few studies on the oxidation of chromium(III) by hydrogen peroxide and hypochlorates [15, 23–24] in alkaline solutions. The main goal of the present study was to examine the kinetics of oxidation of chromium(III) to chromium(IV) in alkaline solutions using the much simpler oxidant [Fe(CN)6]3), which is an established mild one-electron oxidant [25]. Additionally, a stoichiometric, kinetic and e.p.r. investigation was undertaken in order to elucidate the metal (and ligand) oxidation in the further reaction stages. The standard electrode potential of the [Fe(CN)6]3)/4) couple is +0.355 V, a value which is constant from weakly acidic to strongly basic solution. This potential is sufficiently high as to enable oxidation of chromium(III) to chromium(VI) in alkaline media, as 0 0 ECrO ¼ 0:11 V and ECrO ¼ 2 2 4 =CrðOHÞ3 ðsÞ 4 =chromateðIIIÞ 3) 0:72 V [1]. A characteristic feature of the [Fe(CN)6] oxidant is its simple electron transfer action, involving generation of the [Fe(CN)6]4) complex without any major change in the coordination sphere of either hexacyanoferrate complex. An important practical limitation, however, of [Fe(CN)6]3) as an oxidant in spectroscopic studies carried out in situ, is the strong absorption at wavelengths below 500 nm, covering possible chromium(IV) and chromium(V) absorption bands. The tetraazamacrocyclic chromium(III) complex, cis-[Cr(cycb)(OH)2]+, was selected as chromium(III) reactant, since stabilization of unusual oxidation states is well established for this type of macrocyclic ligands, cycb ¼ rac-5,5,7,12,12,14-hexamethyl-1,4,8,11tetraazacyclotetradecane. Structural details of the cis[Cr(cycb)(OH)2]+ complex, which is a well characterized [26], hydrolytically stable monomeric species [27] are given in Figure 1.
Experimental Materials cis-[Cr(cycb)(OH2)2]Br3 Æ 2H2O was prepared and characterized as reported previously [28]. Other chemicals were the best commercially available analytical grades, which were used without further purification. All the cis-[Cr(cycb)(OH)2] +
N N2 N1
3
Cr
O1 O2
cycb-ligand
Cr — O1
192 pm
Cr — N1
214 pm
N3 — Cr — N4 165.30° N1 — Cr — N2 94.85°
N
N
N
N
N
N
N
N
N4
Fig. 1. X-ray data for cis-[Cr(cycb)(OH)2]+ ion and structure of the cycb ligand.
solutions used in the preparations and kinetic measurements were made using water redistilled from alkaline permanganate and purged with argon or dinitrogen to avoid contamination with CO2. Freshly prepared solutions of chromate(III) were used for all experiments and were made via the addition of a slight excess of 4 M KOH to a 0.2 M solution of Cr(NO3)3 Æ 9H2O. Epr measurements In situ e.p.r. studies of the reaction of the macrocyclic chromium(III) complex were performed on solutions containing 1–2 mM cis-[Cr(cycb)(OH)2]+, 2–200 mM [Fe(CN)6]3) and 0.04–0.2 M KOH at 295 K. The e.p.r. spectra were recorded using a Radiopan EPR SE/X 2547 spectrometer in the X-band, ca. 9.25 GHz, with a 100 kHz modulation. The microwave frequency was monitored with a frequency divider and the magnetic field was measured with an automatic n.m.r. type magnetometer. A flat quartz cell was used for this work and the e.p.r. spectra were recorded every 5 min after the initial preparation of the reaction mixture. In the case of the oxidation of chromate(III), the solution concentrations were [CrIII]: 5–20 mM , [FeIII]: 0.2 M and [OH)]: 0.3–0.7 M . E.p.r. spectra were recorded at 298 K using a continuous flow technique and measurements were recorded 0.8–3 s after the initial mixing of reagents. Stoichiometry of the reaction of the macrocyclic chromium(III) complex The stoichiometry of the oxidation process for the macrocyclic complex was examined using an excess of iron(III) oxidant over the chromium(III) reactant. Concentrations were varied within the ranges: [CrIII]: 1.3–5.4 mM , [FeIII]: 20–45 mM , [FeII]: 7.5 mM and [OH)]: 0.11–0.71 M . The FeIII/CrIII ratio was varied between 7 and 29. The experiments were performed at 298 K and the ionic strength was maintained at 1.0 M by the addition of NaClO4. Three types of analysis were performed: Potentiometric measurements using a platinum indicator electrode for the determination of the [FeIII]/ [FeII] ratio and thereby following the amount of hexacyanoferrate(II) produced during the reaction. Sampling, followed by acidification and immediate permanganate or cerium(IV) titration, allowing the determination of the amount of hexacyanoferrate(II) produced. Sampling, followed by acidification and immediate ion exchange separation of cationic complexes, allowing the determination of unreacted chromium(III) complex in addition to the detection of cationic complexes produced during the reaction. Results of the first two methods are in good agreement at lower degrees of conversion. At the end of the reactions discrepancies are seen, which can be ascribed
636 2.5
2.5
[OH−] = 0.7 M
[OH— ] = 0.1M
2
∆c Cr(III) (mM)
∆c Cr(III) (mM)
2
1.5
1.5
1
1
0.5
0.5
0
0 0
5
10
15
20
0
25
5
∆c Fe(III)(mM)
10
15
20
25
∆c Fe(III)(mM)
Fig. 2. Stoichiometry of the oxidation of cis-[Cr(cycb)(OH)2]+ complex by [Fe(CN)6]3) ion in 1.0 M NaClO4 at 298 K. Boxed prints are these for which chromatograms are given in Figure 3.
to significant changes in the ionic medium affecting concentration determinations by potentiometry through changes in ionic activity coefficients and diffusion potentials. It should specifically be pointed out that all analyses of acidified solutions have to be performed immediately after acidification, otherwise complexation of the hexacyanoferrate complexes to the chromium complexes will occur. Two typical experiments are given in Figure 2. Chromatographic separations Analyses of acidified solutions were performed chromatographically using a Pharmacia FPLC apparatus equipped with a Mono S HR 5/5 column. Gradient elution with NaBr from 0.0 to 1.0 M was used to separate the chromium(III) species. The acidity was maintained at 0.001 M HBr. Chromium concentrations were determined by integration of the relevant chromatographic peaks, cf. Figure 3. Integrated peak areas for the parent macrocyclic complex were standardized against values obtained for solutions of the pure component at known concentrations. Solutions for the mass spectrometric measurements were prepared by reacting 50 mg of the macrocyclic chromium(III) complex dissolved in 1 M NaOH(aq) with 300 mg K3[Fe(CN)6] dissolved in the minimum amount of 1 M NaOH(aq) at room temperature for 60 s. The reaction was stopped by acidification with 100 cm3 0.02 M HBr(aq) and the resulting solution was
applied to a 2 · 10 cm Sephadex SP-C25 filled column, which had been pretreated with 0.001 M HBr(aq). Washing with 0.001 M HBr(aq) removed remaining hexacyanoferrate ions, and elution with 0.001 M HBr(aq) + 0.5 M NaBr(aq) partly separated the cationic components. Fractions containing orange-yellow product species, eluted prior to the red parent cis-[Cr(cycb)(OH2)2]3+ complex, were collected, and evaporated to dryness at room temperature in vacuum. The resulting semi-solid material was extracted with ethanol. The alcoholic solution was evaporated to dryness, dissolved in water, reevaporated to dryness to remove traces of ethanol, and redissolved in water. Solutions prepared this way were dominated by the two species marked by the two arrows in Figure 3, but in varying relative amounts. Such solutions were used directly for the mass spectrometric measurements after approriate dilution with water. Mass spectrometric measurements Electro-spray ionization mass spectrometric measurements were conducted by a Micromass Q-Tof instrument [29], on solutions prepared as described above. The measurements demonstrated species with m/z values for the dominant isotopic combination of 360, 364, 378, 402 and 406, with the first two by far dominating. These dominating mass peaks were accompanied by satellite peaks with an isotopic pattern expected for the natural isotopic distribution of chromium and carbon. The
Absorbance
[OH−] = 0.7 M
Absorbance
[OH−] = 0.1 M
Elution time
Elution time
Fig. 3. Chromatographic separation of cationic species produced by the oxidation of cis-[Cr(cycb)(OH)2]+ complex by [Fe(CN)6]3) ion, cf. Figure 2. Dominant peaks to the right in the chromatograms are the parent cis-[Cr(cycb)(H2O)2]3+ complex. Arrows mark chromatographic peaks corresponding to the dominant reaction products.
637 parent cis-[Cr(cycb)(OH)2]+ complex has m/z ¼ 370 for the dominant isotopic combination, and control experiments on solutions of this complex showed no fragmentation at the experimental conditions used. The dominant m/z values of 360 and 364 for reaction products correspond to the two chromatographic peaks marked by arrows in Figure 3. Kinetic measurements The kinetics of the oxidation of the macrocyclic chromium(III) complex by hexacyanoferrate(III) were followed spectrophotometrically using a Hewlett Packard 8453 diode-array spectrophotometer, equipped with Peltier Hewlett Packard 89090A Temperature Controller. The reaction was studied under pseudo-first order limiting conditions by using either an excess of the chromium(III) reactant over the oxidant, or an excess of hexacyanoferrate(III) over the chromium(III) complex. In the first case, reactant concentrations were varied within the limits: [FeIII]: 0.2–0.5 mM and [CrIII]: 5– 20 mM , and spectral scans were recorded over the spectral range 275–475 nm, cf. Figure 4a. In this wavelength range the absorbance decreases due to the reduction of [Fe(CN)6]3) to [Fe(CN)6]4). In the second case, the iron(III) concentration was varied between 0.13 and 0.27 M , and the chromium(III) concentration was Absorbance
0.6
0.9
420 nm
0.5 0.4 0.3
0
50
100
150
200
t (s)
Absorbance
0.7
0.5
0.3
0.1 275
325
375
425
475
(a) 0.5
500 nm Absorbance
0.6
Absorbance
0.4
0.5 0.4 0.3 0.2 0.1 0
0.3
300
600
900
1200
1500
t (s ) 0.2
0.1 480
530
580
630
680
(b) Fig. 4. Spectral changes during oxidation of cis-[Cr(cycb)(OH)2]+ complex by [Fe(CN)6]3) ion; (a) [CrIII] ¼ 10 mM , [FeIII] ¼ 0.4 mM , [OH)] ¼ 0.5 M , I ¼ 2.0 M , T ¼ 322 K, scans every 3 s; (b) [CrIII] ¼ 2 mM , [FeIII] ¼ 0.2 M , [OH)] ¼ 0.2 M , I ¼ 2.0 M , T ¼ 295 K, scans every 50 s.
kept at 1.2 mM . The increase of absorbance around 500 nm, cf. Figure 4b, is in qualitative agreement with formation of chromium(V). Hydroxide concentrations were varied in the range 0.1–1.2 M , and the ionic strength was kept constant by addition of KBr at 2.0 M . Each kinetic run was repeated 2–3 times; the time scale of rate measurements varied between 35 and 1200 s. Absorbance–time data fits very well to a pseudo-first order rate expression for experiments with an excess of chromium(III). For experiments with iron(III) in excess the fit is poorer, but still acceptable. This is explainable as discussed later. The relative standard errors of a single pseudo-first order rate constant were usually smaller than 1.3% and between 1% and 3% for the average value of the rate constant. Oxidation of chromate(III) Addition of an excess of hydroxide to fresh precipitates of chromium(III) hydroxide gives a complex mixture of mono- and polynuclear species, collectively referred to as chromate(III). Such solutions have not been characterized in details but contain oligomers of chromium(III), as recently demonstrated by an EXAFS investigation [23]. Such chromate(III) solutions are rapidly oxidized to CrO42) by [Fe(CN)6]3). Preliminary kinetic experiments on oxidation of chromate(III) were conducted with the concentrations [FeIII]: 0.3–0.7 mM and [CrIII]: 0.01–0.025 M , for experiments with chromium(III) in excess over iron(III) and [FeIII]: 0.15–0.30 M and [CrIII]: 5 mM , for experiments with iron(III) in excess over chromium(III). The experiments were performed using an HP spectrophotometer combined with a home-made stopped-flow apparatus thermostated by an external Julabo F 25 cryostat. Hydroxide concentrations were varied within 0.2–0.8 M , and the ionic strength was kept constant by addition of KBr at 1.0 M . The spectral changes observed during the oxidation of chromate(III) are presented in Figure 5. Under a large excess of chromium(III) the reaction obeys a first order rate expression, cf. Figure 5a. The rates of the reaction were found to be independent of the age of the chromate(III) solution over a period of 1000 s, and the first order rate constant exhibits a linear dependence on the chromium(III) concentration, consistent with an overall second order process (Table 1). The hydroxide dependence has the form of a linear increase of the second order rate constant with respect to the hydroxide concentration, with a significant value for the intercept, suggesting two parallel reaction paths. The rate of the reaction is strongly accelerated by an increase of the ionic strength, with the rate constant increasing by nearly an order of magnitude with an increase in the ionic strength from 0.2 to 1.5 M , consistent with a reaction between like charged ionic species. The oxidation of chromate(III) under the large excess of hexacyanoferrate(III) behaves differently (Table 1), cf. Figure 5b. Rapid spectral changes are seen less than 0.1 s after mixing of the reactants, but chromium(V) could not
Absorbance
638 1.2
1
0.9 0.7
420 nm 0.5 0
20
40
60
80
t(s)
Absorbance
0.8
0.6
0.4
0.2
0 300
350
400
450
500
λ (nm)
(a) 0.25
Absorbance
0. 2
(2)
0.15 0. 1
(1) 0.05 0 500
550
(b)
600
650
700
750
λ (nm)
Fig. 5. Spectral changes during oxidation of chromate(III) by [Fe(CN)6]3) ion; (a) [CrIII] ¼ 10 mM , [FeIII] ¼ 0.5 mM , [OH)] ¼ 0.5 M , I ¼ 1.0 M , T ¼ 278 K, scans every 4 s; (b) [CrIII] ¼ 5 mM , [FeIII] ¼ 0.2 M , [OH)] ¼ 0.5 M , I ¼ 2.0 M , T ¼ 298 K, scans every 0.2 s, (1) – spectrum of the reactants before mixing (at t ¼ 0 s), (2) – spectrum taken 0.1 s after mixing of the reactants.
be detected by continuous flow e.p.r. in the time range 0.8–3 s during the reaction in spite of the changes in the visible spectra that occurs within that time. The initial rapid reaction is followed by a slower pseudo-first order reaction with an absorbance decrease which is consistent with formation of the practically non-absorbing in this wavelength range CrO42) anion. The pseudo-first order rate constants for this reaction vary linearly with the hydroxide concentration with an intercept equal to zero, and the rates are independent of the iron(III) concentration in the range 0.15–0.30 M . Both these observations are different from the kinetic behaviour in solutions with chromium(III) in excess over iron(III). Obviously, the mechanism of chromate(III) oxidation, is complicated, and cannot be decided at present. Rapid oxidation of chromium(V), once formed, seems likely, however.
Results and discussion Oxidation of the macrocyclic chromium(III) complex The cis-[Cr(cycb)(OH2)2]3+ ion undergoes a reversible two step deprotonation, without any observable signs of
decomposition or isomerization. The deprotonation is manifested by distinct spectral changes within the visible region: the cis-diaqua complex is red and the fully deprotonated product, the dihydroxo complex, is blue [28]. Oxidation of the chromium(III) complex by [Fe(CN)6]3) was carried out in strongly alkaline media. Chromium-species formed during the oxidation were initially identified by an e.p.r. signal characteristic for chromium(V) and a high absorbance near 500 nm, cf. Figure 4b. The value for the apparent molar absorption coefficient of the chromium containing reaction product, about 300 M )1 cm)1, is qualitatively characteristic for chromium(V) [30]. In the oxidation of the macrocyclic chromium(III) complex under a large excess of the oxidant, the spectral changes given in Figure 4b show a regular increase of absorbance at 500 nm that correlates with an increase in the chromium(V) concentration, observed as an increase of the e.p.r. signal intensity. Figure 6a and b show two e.p.r. spectra. The spectrum in Figure 6a is recorded at a time when the absorbance at 500 nm is rapidly increasing. The spectrum in Figure 6b is recorded at the time when the absorbance becomes almost constant, and this e.p.r. signal is stable for several hours. The initial e.p.r. spectrum, in Figure 6a, is complicated and because the e.p.r. signal of chromium(III) consist of a very broad singlet and chromium(IV) produces an e.p.r. signal only at very low temperature [31], the observed signals are assigned to the multicomponent mixture of chromium(V) species. In the course of the reaction the e.p.r. spectrum is simplified, cf. Figure 6b and long-lived chromium(V) complexes are apparently generated that exhibits a strong isotropic e.p.r. signal due to 90.5% of the chromium isotopes with nuclear spin of 0 at giso ¼ 1.987. The signal is resolved into nine lines, because of electron-nuclei hyperfine interactions with four equivalent 14N-nuclei, each with a nuclear spin of 1, and the coupling constant a(14N) ¼ 2.36 · 10)4 cm)1, with the expected intensity ratio being 1:5:15:30:45:30:15:5:1. This hyperfine structure reveals that the Fermi-contact interaction of four nitrogen nuclei occurs to the same extent and corresponds to a spin density of 0.0046 on each nitrogen [32]. A small isotropic satellite signal, split into a quartet due to hyperfine coupling by the less abundant 53Cr isotope (9.5%) with a nuclear spin of 3/2, is also observed with a coupling constant a(53Cr) ¼ 17.1 · 10)4 cm)1. To conclude: the e.p.r. data illustrated in Figure 6b are consistent with the reaction product being long-lived chromium(V) species coordinated to four equivalent or almost equivalent nitrogens. The stoichiometry of the oxidation process is given by Equation 1: CrIII ðcycbÞ þ nFeIII ! CrV ðcycbox Þ þ nFeII
ð1Þ
with values for n which are significantly larger than the 2. This shows that the formation of chromium(V) is
639 Table 1. Pseudo-first order rate constants for oxidation of chromate(III) by hexacyanoferrate(III) in K(Br/OH) as function of temperature, hydroxide concentration, ionic strength and initial reactant concentrations T (K)
I (M)
[FeIII] (mM)
[CrIII] (M)
[OH)] (M)
102kobs (S )1)
278 278 278 278 278 278 278 278 278 278 278 278
1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0
0.50 0.50 0.50 0.30 0.50 0.70 0.50 0.50 0.50 0.50 0.50 0.50
0.010 0.010 0.010 0.010 0.010 0.010 0.015 0.020 0.025 0.010 0.010 0.010
0.2 0.3 0.4 0.5 0.5 0.5 0.5 0.5 0.5 0.6 0.7 0.8
4.00 4.67 5.42 5.99 5.84 5.68 8.44 11.8 13.8 6.25 6.47 7.22
278 278 278 278 278
0.2 0.5 0.75 1.00 1.50
0.50 0.50 0.50 0.50 0.50
0.010 0.010 0.010 0.010 0.010
0.2 0.2 0.2 0.2 0.2
0.77 1.94 3.16 4.00 6.65
278a 283a 288a 293a 298a
1.0 1.0 1.0 1.0 1.0
0.50 0.50 0.50 0.50 0.50
0.010 0.010 0.010 0.010 0.010
0.2 0.2 0.2 0.2 0.2
0.77 1.02 1.36 1.91 2.63
[FeIII] (M)
[CrIII] (mM)
0.2 0.2 0.15 0.30 0.2 0.2 0.2 0.2
5.0 5.0 5.0 5.0 5.0 5.0 5.0 5.0
298 298 298 298 298 298 298 298 a
2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0
kobs (s)1) 0.2 0.3 0.3 0.3 0.4 0.5 0.6 0.7
0.55 0.81 0.77 0.77 1.06 1.35 1.59 1.83
These data define an apparent energy of 42.5ð15Þ kJ mol)1.
accompanied by transformation of the coordinated macrocyclic ligand to an oxidized form, denoted here as cycbox. Details of two stoichiometric experiments are given in Figure 2. These two experiments give initial values for n around 9 at the lower hydroxide concentration and of ca. 7 at the higher hydroxide concentration. It is further noted that there is an increase in n as the reaction proceeds at the higher hydroxide concentration, whereas the stoichiometric factor seems to be more constant at the lower concentration of hydroxide. Acidification of basic reaction mixtures reduces the generated chromium(V) to chromium(III), as indicated by the disappearance of the chromium(V) e.p.r.-signal. This reduction is not accompanied by reoxidation of the iron complex, as indicated by the stoichiometric studies. Consequently, further oxidation of the macrocyclic ligand must take place. This oxidation of the ligand has further been substantiated by chromatographic analyses of acidified reaction mixtures as shown by the examples given in Figure 3. In addition to the peak from the parent chromium(III) complex at least five welldefined chromatographic peaks are seen, as is an illdefined broad peak. These two chromatograms further illustrate a significant difference in distribution of
oxidation products at the two hydroxide concentrations, for a very similar overall degree of conversion of the initial reactants, cf. Figure 2. This presence of a multitude of reaction products is further demonstrated by mass-spectrometric analyses of solutions of cationic reaction products isolated by macroscopic ion exchange chromatography in acidic solution. Two major components in oxidized solutions were identified with m/z values for the dominant peaks of 360 and 364, respectively, corresponding to the two arrows in Figure 3. This corresponds to singly charged dihydroxochromium(III) complexes in which the macrocyclic ligand has been oxidized to loose 10 or 6 hydrogen atoms, corresponding to incorporation of five or three double bonds in the ligand. Reduction of chromium(V) to chromium(III) on acidification with concomitant ligand oxidation demonstrate the presence of four or two double bonds in the chromium(V) complex prior to the acidification. It is realized that these two products will correspond to stoichiometric factors of n ¼ 10 and n ¼ 6, respectively. Additional minor peaks with m/z values of 378, 402 and 406 dominating, accompanied by expected satellite peaks, are identified as species having been further oxidized to
640 corresponding to formation of an even greater number of isomers of the chromium complex. Incorporation of two imine functions in the ligand increases the number of possible ligand isomers to 8, again with the possibility of an even greater number of complexes. Further ligand oxidation is not expected to simplify this pattern, and it is consequently understandable that only complex mixtures of chromium reaction products have been obtained. Further support for the production of such a types of oxidized complexes with a significant number of coordinated imine-type nitrogens is provided by the similarity of the e.p.r. parameters for the generated species to those of perchloratooxochromium(V) porphyrins [33], but with a somewhat higher giso-value for the present chromium(V) complexes.
600
400
Intensity[a.u.]
200
0
-200
-400
-600
-800 331.25
331.75
332.25
332.75
(a)
333.25
333.75
334.25
334.75
B[ mT]
2000
1500
Intensity[a.u.]
1000
Kinetics of oxidation of the macrocyclic complex
500
Kinetic studies on the oxidation of the macrocyclic chromium(III) by hexacyanoferrate(III) were performed at a series of constant hydroxide concentrations and at pseudo-first order conditions with respect to either the chromium(III)- or the iron(III)-reactant. In the first case, the reaction between cis-[Cr(cycb)(OH)2]+ and [Fe(CN)6]3) ions was carried out with a 20-fold excess of the chromium(III) reactant. A factor analysis of the spectral changes presented in Figure 4a demonstrate the disappearance of the iron(III)-oxidant in a well-defined manner consistent with the production of hexacyanoferrate(II) ions, without spectral interference from chromium complexes. The absorbance–time data at two wavelengths, i.e. 303 and 420 nm, fit very well a first order rate expression giving practically the same values of the pseudo-first
0
-500
-1000
-1500
-2000 331.25
331.75
332.25
332.75
(b)
333.25
333.75
334.25
334.75
B[ mT]
Fig. 6. E.p.r. spectra of chromium(V) species generated via oxidation of cis-[Cr(cycb)(OH)2]+ complex by [Fe(CN)6]3) ion. Conditions: time of reaction, (a) 5 min; (b) 140 min; microwave frequency, (a) 9.2511 GHz, (b) 9.2521 GHz; time constant, 0.1 s; modulation width, 0.05 mT.
incorporate double bonded oxygen atoms in the macrocyclic ligand, cf. Scheme 1.
CH3 CrIII
N
-2 e-
H H
CH3 CrV
N
H H
-2 H+
III
Cr
NH
H CH3
-2 e
-
V
Cr
H CH3
NH
H H
III
Cr
NH
H H
O
-2 H+
III
Cr
N
H CH3 H
H CH3
+H 2O
-2 e-
CrV
N
-2e-
V
Cr
NH
H CH3 O
-2 H+
H CH3 H
Scheme 1.
This scheme demonstrate the formation of a single C@N double bond. It is realized, that the structure of the ligand allows four such bonds to be formed
order rate constant. Dependence of this rate constant on the hydroxide concentration was studied at three temperatures. Additionally, the rate constant dependence
641 Table 2. Pseudo-first order rate constants for oxidation of cis-[Cr(cycb)(OH)2]+ by hexacyanoferrate(III) in K(Br/OH) as function of temperature, hydroxide concentration, ionic strength and initial reactant concentrations [OH)] (M)
I (M)
[FeIII] (mM)
[CrIII] (mM)
102kobs(s)1) 302 K
312 K
322 K
333 K
0.10 0.10 0.10
2.0 1.0 0.12
0.5 0.5 0.5
10 10 10
– – –
0.202 – –
0.38 0.434 1.29
0.77 – –
0.30 0.30 0.30 0.30
2.0 2.0 2.0 2.0
0.5 0.5 0.5 0.5
5 10 15 20
– – – –
– 0.61 – –
0.55 0.97 1.50 1.95
– 1.95 – –
0.30 0.30 0.30 0.10 0.30 0.50 0.70 0.80 0.90 1.0 1.2
2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0
0.2 0.4 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5
10 10 10 10 10 10 10 10 10 10 10
– – – – – – – – – – –
– – – 0.202a 0.61a 1.00a 1.61 – 2.42 2.93 4.40
0.96 0.97 0.97 0.38a 0.97a 1.95a 3.06 – 4.69 5.85 9.10
– – – 0.77a 1.95a 3.95a 5.72 7.20 9.01 – –
[FeIII] (M ) 0.20 0.30 0.40 0.50 0.60 0.70
2.0 2.0 2.0 2.0 2.0 2.0
0.2 0.2 0.2 0.2 0.2 0.2
1.2 1.2 1.2 1.2 1.2 1.2
0.434b 0.564b 0.68b 0.87b 1.10 1.32
0.920b 1.27b 1.60b 1.95b 2.37 2.91
1.74b 2.78b 3.50b 4.25b 5.13 6.35
– – – – – –
0.40 0.40 0.40
2.0 2.0 2.0
0.13 0.2 0.26
1.2 1.2 1.2
– 0.68 –
– 1.60 –
1.95 3.50 5.11
– – –
These values defines second order rate constants, at 323 K, and activation energies: a 4.55(13) M)1 s)1, 57(3) kJ mol)1 and b 0.461(19) M)1 s)1, 67(3) kJ mol)1. a,b
on the ionic strength, and reactant concentrations was examined at a single temperature. The results are summarized in Table 2 and illustrated in Figure 7. The pseudo-first order rate constant which characterizes the rate of the redox reaction at a constant hydroxide concentration, is linearly dependent on the chromium(III) concentration. Thus, at a constant hydroxide concentration, the disappearance of the oxidant obeys a second order rate law, first order in [FeIII] and in [CrIII]: d½FeIII =dt ¼ kFe ½CrIII ½FeIII ¼ kobs;Fe ½FeIII
ð2Þ
The significant increase of kobs with a decrease of the ionic strength, as demonstrated by the data in Table 2, accords with a secondary kinetic salt effect and oppositely charged reactants. The rate constant dependence on the hydroxide concentration, as shown in Figure 7a, illustrates an initially linear dependence of the rate on the hydroxide concentration which turns into a higher order than linear dependence at the highest hydroxide concentrations. An analogous dependence was found before [27] for simple base accelerated hydrolysis of a number of macrocyclic chromium(III) complexes. These
complexes are potential Brønsted acids in strongly alkaline media and the coordinated hydroxyl ligand or/and the amine nitrogen atoms can be deprotonated. In the case of the studied redox process the model implies the decisive role of chromium(III) species deprotonation and the higher reactivity of the reactant in the form of the conjugate base. Deprotonation inforces p-antibonding character of the chromium(III) t2g electrons and increases electron donor capability of the chromium(III) centre. The dependence of the second order rate constants on the OH) ion concentration presented in Figure 7a is very well described by the rate expression: k ðIIÞ ¼
kobs a þ b½OH ¼ ½Cr 1 þ c½OH
ð3Þ
consistent with the assumption that the proton transfer from the cis-[Cr(cycb)(OH)2]+ to OH) ion is retarded by the ion-pair formation between the chromium(III) complex and Br) ion. The mechanistic interpretation of the empirical a, b and c parameters (Equation 3) leads to the three physico-chemical quantities: two rate constants and an equilibrium constant [27]. However,
642 12
charged anions, dominating ion-pair formation with the cationic chromium(III) reactant in experiments with the hexacyanoferrate(III) reactant in excess. Otherwise the functional dependence of these rate constants on reactant and hydroxide concentrations is similar to that described above for [CrIII][FeIII], but with rate constants significantly smaller. This is an independent verification of the reaction stoichiometry since the stoichiometric reaction path:
333 K 322 K
10
-1
10 k obs (s )
8
312 K
2
6
4
2
CrIII ðcycbÞ þ nFeIII ! CrV ðcycbox Þ þ nFeII
0 0
0.2
0.4
0.6
0.8
1
1.2
1.4
−
(a)
[OH ]( M)
7
-1 2
10 k obs (s )
5
3
ð5Þ
followed rapidly by further oxidation, in accordance with the stoichiometry, gives rise to:
6
4
with a rate determining step governed by the second order rate constant, k, for the process: CrIII ðcycbÞ þ FeIII ! CrIV ðcycbÞ þ FeII
322 K
ð4Þ
d½FeIII =dt ¼ n k½CrIII ½FeIII ¼ kobs;Fe ½FeIII for ½CrIII ½FeIII
312 K
ð6Þ 2
302 K 1
d½CrIII =dt ¼ k½CrIII ½FeIII ¼ kobs;Cr ½CrIII for ½FeIII ½CrIII
0 0
(b)
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
ð7Þ
[OH − ]( M)
Fig. 7. Dependence of the pseudo-first order rate constant for the oxidation of cis-[Cr(cycb)(OH)2]+ complex on [OH)]; (a) [CrIII] ¼ 10 mM , [FeIII] ¼ 0.5 mM , I ¼ 2.0 M (KBr); (b) [CrIII] ¼ 1.2 mM , [FeIII] ¼ 0.2 M , I ¼ 2.0 M (KBr).
close to the )0.5 value of the c parameter for the present system disenables calculation of two among three of the mentioned quantities. The reaction at hexacyanoferrate(III) in excess over chromium(III) appear to be more complicated than the experiments just described in which chromium(III) species were in excess over hexacyanoferrate(III). Spectral changes indicate the production of more than one chromium complex, as verified directly by chromatographic analysis and mass spectrometric investigations of acidified reaction mixtures. Kinetic traces at wavelengths around 500 nm indicate an approximate pseudofirst order behaviour, but the fitting to simple first order kinetics is not quite as well defined as that seen with the experiments in which chromium(III) were in excess over hexacyanoferrate(III), as expected due to the spectral influence of a succession reaction products. The observed rate constants are given in Table 2. The rate constant dependence on the hydroxide concentration, as shown in Figure 7b, appears to be somewhat more linear than for the data with chromium(III) in excess over iron(III), although an upwards curvature is noticeable at the highest hydroxide concentrations, cf. Figure 7b. A possible reason for this apparent difference may be sought in the high concentration of highly
and consequently kobs,Fe » nÆkobs,Cr for rate constants from the two types of experiment at the same hydroxide concentration. A quantitative comparison of the linear portion of the data in Table 2 and Figure 7 at hydroxide concentrations below 0.5 M gives a value for n » 9.7, in good agreement with results from the stoichiometric determination in Figure 2, despite the reservations referred to concerning the maintaining of a constant ionic medium with the highly charged reactant and product anions. The rate determining formation of chromium(IV), involving deprotonation of the chromium(III) reactant, is obviously followed by further oxidation. The further fast reaction steps were not identified. However, absorbance–time data within 480– 700 nm were processed using the Specfit software based on the factor analysis and it has been found that disproportionation of CrIV cannot play any important role in the CrV formation. Conclusions Chromium(III) is oxidized by hexacyanoferrate(III) in basic solution to give ultimately chromate(VI). Comparing the results obtained for the two chromium(III) reactants employed in this study indicates clearly that the macrocyclic ligand, although concomitantly oxidized, stabilizes the chromium(V) oxidation state in basic solution, by making the chromium(V) fi CrO42) process very slow. In the absence of the macrocyclic
643 ligand, this latter step is rapid. Not surprisingly, the slowest step of the overall redox process correlates with the largest structure changes within the inner coordination sphere of the chromium complexes. Thus, in the oxidation of the macrocyclic chromium(III) species, the chromium(V) fi CrO42) transformation does not take place until after further oxidative degradation of the macrocyclic ligand, whereas, in the oxidation of chromate(III) it is probably the configurational change from hexacoordinated chromium(IV) to tetrahedral chromium(V) which is the rate determining step. The rate of oxidation for the both studied reductants increases with an increase of [OH)] as well for the CrIII–CrIV as for CrIV–CrV stages. The opposite effect of [OH)] on the rate of the chromates(III) oxidation by H2O2 found by other group [23] correlates, in our opinion, with the deprotonation of the oxidant and not with reactions of CrIII species with OH) ion. One can expect much lower reactivity (as an electron acceptor) of HO2) anion than its protonated form– H2O2 molecule.
Acknowledgments The authors wish to thank Solveig Kallesøe, Department of Chemistry, University of Copenhagen, for the electro-spray mass spectrometric measurements and N. Copernicus University students Joanna Blajchert and Małgorzata Szabłowicz for valuable assistance with the preliminary experimental work.
References 1. D.F. Shriver and P.W. Atkins, Inorganic Chemistry, 3rd edit., Oxford University Press, 1999. 2. S.J. Slattery, J.K. Blaho, J. Lehnes and K.A. Goldsby, Coord. Chem. Rev., 174, 391 (1998). 3. A. Levina, A.M. Bailey, G. Champion and P.A. Lay, J. Am. Chem. Soc., 122, 6208 (2000). 4. P. Belanzoni, M. Rosi, A. Sgamelloti, L. Bonomo and C. Floriani, J. Chem. Soc., Dalton Trans., 1492 (2001). 5. A. Nemes and A. Bakac, Inorg. Chem., 40, 2720 (2001).
6. O.A. Babich and E.S. Gould, ibid., 40, 5708 (2001). 7. A.E. Meier-Callahan, A.J. Di Bilio, L. Simkhovich, A. Mahammed, I. Goldberg, H. Gray and Z. Gross, ibid., 40, 6788 (2001). 8. A. Levina, L. Zhang and P.A. Lay, ibid., 42, 767 (2003). 9. A. Levina, R. Codd, C.T. Dillon and P.A. Lay, Progr. Inorg. Chem., 51, 145 (2003). 10. E.S. Gould, Coord. Chem. Rev., 135/136, 651 (1994). 11. D.I. Pattison, P.A. Lay and M.J. Davies, Inorg. Chem., 39, 2729 (2000). 12. H.A. Headlam and P.A. Lay, ibid., 40, 78 (2001). 13. C.M. Cawich, A. Ibrahim, K.L. Link, A. Bumgartner, M.D. Patro, S.N. Mahapatro, P.A. Lay, A. Levina, S.S. Eaton and G.R. Eaton, ibid., 42, 6458 (2003). 14. D.K. Geiger, Coord. Chem. Rev., 152, 359 (1996). 15. R. Rajan, N.B. Unni and T. Ramasami, Inorg. React. Mech., 1, 247 (2000). 16. T. Birk and J. Bendix, Inorg. Chem., 42, 7608 (2003). 17. A. Malcolm and P.C. Ford, Coord. Chem. Rev., 208, 47 (2000). 18. A. Hori, T. Ozawa, H. Yoshida, Y. Imori, Y. Kuribayashi, E. Nakano and N. Azuma, Inorg. Chim. Acta., 281, 207 (1998). 19. A.E. Meier-Callahan, H.B. Gray and Z. Gross, Inorg. Chem., 39, 3605 (2000). 20. R. Codd, C.T. Dillon, A. Levina and P.A. Lay, Coord. Chem. Rev., 216/217, 537 (2001). 21. J.F. Perez-Benito, C. Arias and R.M. Rodriguez, J. Phys. Chem. A., 105, 1150 (2001). 22. A.K. Rappe and M. Jaworska, J. Am. Chem. Soc., 125, 13956 (2003). 23. L. Rao, Z. Zhang, J.I. Friese, B. Ritherdon, S.B. Clark, N.J. Hess and D. Rai, J. Chem. Soc., Dalton Trans., 267 (2002). 24. M. Knoblowitz and J.I. Morrow, Inorg. Chem., 15, 1674 (1976). 25. J. Leal, B. Garcia and P. Domingo, Coord. Chem. Rev., 173, 79 (1998). 26. E. Bang and O. Mønsted, Acta Chem. Scand., A38, 281 (1984). 27. E. Madej, O. Mønsted and P. Kita, J. Chem. Soc., Dalton Trans., 2361 (2002). 28. J. Eriksen and O. Mønsted, Acta Chem. Scand., A37, 579 (1993). 29. U.N. Andersen, C.J. McKenzie and G. Bojesen, Inorg. Chem., 34, 1435 (1995). 30. A. Levina, P.A. Lay and N.E. Dixon, ibid., 39, 385 (2000). 31. M. Krumpolc, B.G. Deboer and J. Rocˇek, J. Am. Chem. Soc., 100, 145 (1978) and refs therein. 32. B.A. Goodman and J.B. Raynor, Adv. Inorg. Chem. Radiochem., 13, 135 (1970) and refs therein. 33. H. Fujii, T. Yoshimura and H. Kamada, Inorg. Chem., 36, 1122 (1997).
TMCH 5932
