Mechanism of reaction of alkyl radicals with (NiIIL)2+

0 downloads 0 Views 208KB Size Report
The reactions of methyl radicals, •CH3, with the macrocyclic complexes NiIIL1–5 (L1–5 = cyclam derivatives .... and the dose per pulse was calculated assuming G(SCN)2. •− or .... (L(H2O)NiIII–CH3)2+ + •CH3 → (NiIIL)2+ + C2H6 + H2O. (13) ..... intermediates in enzymes like CO dehydrogenases29–33 and acetyl. Co–M ...
View Online

www.rsc.org/dalton | Dalton Transactions

PAPER

Mechanism of reaction of alkyl radicals with (NiII L)2+ complexes in aqueous solutions Tamar Kurzion-Zilbermann,a Alexandra Masarwa,a Eric Maimon,*a,b Haim Cohena,c and Dan Meyerstein*a,c

Downloaded on 09 February 2011 Published on 27 July 2007 on http://pubs.rsc.org | doi:10.1039/B706491H

Received 30th April 2007, Accepted 19th June 2007 First published as an Advance Article on the web 27th July 2007 DOI: 10.1039/b706491h The reactions of methyl radicals, • CH3 , with the macrocyclic complexes NiII L1–5 (L1–5 = cyclam derivatives, vide infra) and NiII edta in aqueous solutions were studied. Methyl radicals react with all these nickel complexes, forming intermediates with NiIII –C r-bonds. The Lm NiIII –CH3 complexes are formed in equilibria processes with relatively fast forward rate constants of kf > 1 × 108 M−1 s−1 (except in the case of NiL2 -trans I cyclam, where the reaction is slower). In all cases the decomposition of the transient complexes occurs via the homolytic cleavage of the metal–carbon r-bond. When the homolysis is relatively slow, an isomerisation process of the transient is also observed with the exception of NiL2 , where no isomerisation was observed. The results suggest that the strength of the NiIII –CH3 r-bond is mainly affected by steric hindrance.

Introduction In the past, several r-bonded alkyl intermediates derived from NiII cyclam (two isomers, a (cis) and b (trans-III)) were studied by pulse radiolysis1,2 and flash photolysis.3 It was shown that alkyl radicals react with the NiII complexes to yield short lived intermediates with metal carbon r-bonds of the type Lm NiIII –R in an equilibrium process.1,3 Lm NiII + • R + H2 O  Lm (H2 O)NiIII –R

(1)

Rates of formation and decomposition are rather sensitive to the structure of the Ni(II) complex (2 orders of magnitude change in the equilibrium constant of the transient formation between the two isomers of the cyclam complex).3 In principle, the stability of the NiIII –C r-bonds is expected to depend on two major factors, a) the redox potential of the NiIII/II Li couple, and b) steric hindrance introduced by the ligands Li . It was therefore of interest to study the effect of ligand structure in a series of related ligands. In this study the mechanisms of reaction of several macrocyclic amine complexes Ni(II)L1–5 (L1–5 = cyclam derivatives) and Ni(II)edta with the methyl radical, • CH3 , were studied. The transients in those systems are of biological importance being analogues to the transients reported in two types of bacterial enzymes: methyl coenzyme M reductase4–7 and acetyl coenzyme A synthase.4,8

Materials and methods Materials All materials used were of AR grade and were purchased from Merck, Aldrich, or Fluka. The solutions were prepared using distilled water that was further purified using a Millipore Milli-Q a Department of Chemistry, Ben-Gurion University of the Negev, Beer-Sheva, Israel b Nuclear Research Centre Negev, Beer-Sheva, Israel c Department of Biological Chemistry, College of Judea and Samaria, Ariel, Israel

This journal is © The Royal Society of Chemistry 2007

system. The final resistance was better than 10 MX cm−1 . All pH measurements were performed using a Corning 220 or a HANNA HI 9017 pH meter, and the pH was adjusted by HClO4 and/or NaOH. The macrocyclic complexes were prepared according to literature procedures.9–14 The complexes were characterised using CHN elemental analysis and UV–Vis and IR spectroscopy. The results for all the complexes were in excellent agreement with the expected formula and the literature data. N2 O gas was purchased from Maxima. Dioxygen traces were removed by passing the gas (He, Ar or N2 O) through a washing bottle containing aqueous V2+ (0.1 M) in dilute H2 SO4 over Zinc amalgam (the V2+ aq was prepared by reduction of NaVO3 by the Zinc amalgam), and a washing bottle containing distilled water. All solutions were saturated with the desired gases by bubbling the gas through the solution in a glass syringe for 15 minutes.15,16 Dalton Trans., 2007, 3959–3965 | 3959

View Online

Downloaded on 09 February 2011 Published on 27 July 2007 on http://pubs.rsc.org | doi:10.1039/B706491H

Irradiations The pulse radiolysis experiments were carried out using the Varian 7715 linear electron accelerator at the Hebrew University of Jerusalem. The pulse duration was 0.1–1.5 ls with a 200 mA current of 5 MeV. The dose per pulse was dependent on its length. A 4 cm Suprasil cell was used, with the analyzing light passing through 3 times to yield an optical path length of 12.3 cm. A 150 W xenon arc lamp produced the analyzing light. The entire experimental setup has been previously described.17,18 The exact dose was determined using a 1 × 10−3 M KSCN (or an Fe(CN)6 4− ) solution saturated with N2 O. The yield of (SCN)2 • − (or Fe(CN)6 3− ) was measured using a molar extinction coefficient of e475 = 7600 M−1 cm−1 (or e420 = 1000 M−1 cm−1 respectively), and the dose per pulse was calculated assuming G(SCN)2 • − or G(Fe(CN)6 3− ) = 6.0.19 The dose per pulse was set so that the initial radical concentration was 2–20 lM. The dose delivered to vials, identical to those irradiated for final product analysis, was measured using the Fricke dosimeter for high dose rates using G(Fe(III)) = 1.68 and e302 = 2205 M−1 cm−1 .20 The values of the molar extinction coefficients calculated from the dosimetry measurements have an error limit of 15% due to the scatter in the pulse intensity and due to uncertainties in G values. The equilibrium constant of a reaction can be obtained spectroscopically, if the OD (optical density) of the resulting species is measured at a point where only this species absorbs. Thus for a reaction A + B ↔ C, K= [C]/[A][B], which transforms with [B] = [Bt ]−[C] (one of the reacting species in excess) and [C] = OD/el into 1/[A] = (K[Bt ]el/OD) − K. Thus a plot of 1/[A] vs. 1/OD results in −K as the intercept. When second-order decompositions of transients of the type L M–R are observed two plausible mechanisms have to be considered: (ML)n+ + • R + H2 O  (L(H2 O)Mn+1 –R)n+

(2)

followed by either (L(H2 O)Mn+1 –R)n+ + • R → (ML)n+ + R–R + H2 O

(3)

R + • R → R–R

(4)

or •

Reaction (2) followed by (3) leads to the following rate law (assuming steady-state conditions for the • R radical): −d[(L(H2 O)Mn+1 −R)n+ ] = dt 2k3 k−2 [(L(H2 O)Mn+1 − R)n+ ]2 k2 [(ML)n+ ] + k3 [(L(H2 O)Mn+1 −R)n+ ]

(6)

Therefore, plotting the observed rate of decomposition, kobs vs. 1/[(ML)n+ ] yields a straight line that passes through the origin, if this assumption is correct. From the slope of the line the quotient 2k3 /K 2 can be calculated. 3960 | Dalton Trans., 2007, 3959–3965

−d[(L(H2 O)Mn+1 −R)n+ ] −d[R] = = k4 [R]2 = dt dt k4 [(L(H2 O)Mn+1 −R)n+ ]2 K22 [(ML)n+ ]2

(7)

which will result in the observed rate of decomposition being inversely proportional to the square of the initial complex concentration. A Noratom and a Nordion 60 Co c-source with dose rates of 3.5–8.0 Gy min−1 and 38–82 Gy min−1 (over the time range of the research) were used for low dose-rate experiments and product analysis. GC Analysis All GC analyses were done on a Varian 3700 gas chromatograph with FID and TCD detectors, and on a Poropak Q column. Formation of radicals with ionizing radiation. In this study the formation and disappearance of short-lived intermediates that form in reactions of alkyl radicals with nickel complexes in aqueous solutions were followed. The radicals were formed by irradiating the solutions with ionizing radiation (c radiation or fast electrons). When ionizing radiation is absorbed by a dilute aqueous solution the following initial products are formed:21 c ,e−

H2 O −−−→ • H(0.60), • OH(2.65), e−aq (2.65), H2 O2 (0.75), H2 (0.45) (8) where the G values are given in parentheses (G values are defined as the number of molecules of each product per 100 eV of radiation absorbed by the solution). The distribution of these products in the solution after 1 × 10−7 s is homogeneous.21 The measurements were performed in times longer than 1 × 10−6 s, so that the measurements were performed on homogenous solutions. Reactions of hydroxyl radicals were studied by saturating the solutions with N2 O (0.022 M) to reduce interference from other initial radicals. This is due to the reaction of N2 O with the solvated electron (e− aq ):22 e− aq + N2 O → N2 + • OH + OH−

k = 8.7 × 109 M−1 s−1

(9)

It is important that [H+ ] be held below 1 × 10−3 M, as e− aq reaction is diffusion controlled with H+ (reaction (10)).22 Under these conditions the solvated electrons will all react with N2 O, yielding • OH as the major product. e− aq + H3 O+ → • H + H2 O k = 2.2 × 1010 M−1 s−1

(5)

If k2 [(ML)n+ ]  k3 [(L(H2 O)Mn+1 –R)n+ ], then this rate law is reduced to: 2k3 [(L(H2 O)Mn+1 −R)n+ ]2 −d[(L(H2 O)Mn+1 −R)n+ ] = dt k2 [(ML)n+ ]

If on the other hand reaction (2) is followed by reaction (4), the following rate law applies:

(10)

Dioxygen reacts quickly with the alkyl radicals, whose reactions are the subjects of this study, and thus interferes with the desired study. Therefore dioxygen must be removed from the system. Removal of dioxygen was accomplished by bubbling N2 O through the solutions for 15 min. Preparation of • CH3 radicals. The methyl radicals were formed in DMSO-containing aqueous solutions by the following reactions:23 •

OH + (CH3 )2 SO → (CH3 )2 S• (O)OH k = 7.0 × 109 M−1 s−1 (11) This journal is © The Royal Society of Chemistry 2007

View Online

(CH3 )2 S• (O)OH → CH3 S(O)OH + • CH3 k = 1.5 × 107 M−1 s−1

(12)

Thus over 90% of the initial radicals are converted to the methyl radical in under 10−6 s in solutions saturated with N2 O containing ≥0.1 M (CH3 )2 SO at pH ≥3.

The equilibrium constant of reaction (1 ), K 1 = 3300 ± 800 M−1 , can be obtained as the negative value of the intercept in a plot of [(NiII L1 )2+ ]−1 vs. (DOD)−1 (Fig. 2).

Results and discussion Reaction of NiII L1 (L1 = cyclam-trans III) with • CH3 radicals

Downloaded on 09 February 2011 Published on 27 July 2007 on http://pubs.rsc.org | doi:10.1039/B706491H

In the literature, the mechanism for the reaction of methyl radicals with NiII L, which in neutral aqueous solutions is a mixture of isomers of 85% L1 and 15% L2 9,10 was reported as:1–3 (NiII L)2+ + • CH3 + H2 O  (L(H2 O)NiIII –CH3 )2+

(1 )

(L(H2 O)NiIII –CH3 )2+ + • CH3 → (NiII L)2+ + C2 H6 + H2 O

(13)

At that time two time-separated reactions were observed (fast formation and subsequent decomposition in the 100 s time range). k−1 and consequently an apparent equilibrium constant K 1 was derived by competition kinetics with dioxygen.1,2 Re-examining the reaction with updated equipment (with mainly better digital analysis of the data) provided the following results. When N2 O saturated aqueous solutions containing Nicomplex ((1.0–6.0) × 10−4 M), (CH3 )2 SO (0.30 M), at pH 3–5 are irradiated by a short pulse, three distinct time-resolved reactions are observed. At the observation wavelength of 310 nm the first two reactions appear as formations and the 3rd as decay. In the first reaction, an unstable intermediate is formed within 20 ls under the experimental conditions with kmax = 310 ± 10 nm, (350 ± 10 nm sh) and emax = 1200 ± 200 M−1 cm−1 (950 ± 200 M−1 cm−1 ) (Fig. 1). Its kinetics of formation obeys a first-order rate law with the observed rate constants proportional to the concentration of (NiII L1 )2+ , and it is independent of the concentrations of the other components of the solution, pH or wavelength of observation. Thus, the second-order rate constant is derived as k1 = (5 ± 1) × 108 M−1 s−1 , from the slope of the graph of the observed first-order rate constants vs. [(NiII L1 )2+ ], in accord with the literature value1 and the known mechanism.1,3

Fig. 1 UV–Vis spectra of the intermediates formed in the reaction of the methyl radicals with (NiII L1 )2+ (a) intermediate formed 24 ls after the pulse. (b) intermediate formed 0.4 ms after the pulse. (Solution composition: 0.25 mM (NiII L1 )2+ ; 0.3 M (CH3 )2 SO; 0.022 M N2 O; pH 3.2; irradiated: 2.5 Gy).

This journal is © The Royal Society of Chemistry 2007

Fig. 2 The linear dependence of [(NiII L1 )2+ ]−1 on 1/OD310 nm . Solution composition: (0.1–0.6 mM) (NiII L1 )2+ ; 0.3 M (CH3 )2 SO; 0.022 M N2 O; pH 3.2; irradiated: 2.5 Gy; measured at k = 310 nm.

In the second reaction, the intermediate formed in the first reaction undergoes a transformation to a second intermediate in the ms time-range, with a similar spectrum (Fig. 1b, kmax = 310 ± 10 nm, (350 ± 10 nm sh) and emax = 1350 ± 200 M−1 cm−1 (1050 ± 200 M−1 cm−1 ). The transformation obeys first-order kinetics, its rate is independent of the initial complex concentration, with values of k14 = (1.8 ± 0.3) × 104 s−1 . The error in this measurement is large as the change in optical density is rather small. It is therefore reasonable to assume that the second intermediate is the isomerisation product of the first intermediate and can be represented in the following way: (L1 (H2 O)NiIII –CH3 )2+  (L1 (H2 O)NiIII –CH3 )2+ R k14 = (1.8 ± 0.3) × 104 s−1

(14)

This suggestion is based on the following observations: 1. The two intermediates have very similar spectra. 2. The disappearance of the first intermediate to yield the second intermediate obeys a first-order rate law, as expected for an isomerisation reaction. 3. The spectrum of the second intermediate is not identical to that of the first intermediate, ruling out the possibility that it is the same compound at higher concentration. There is no way to determine the exact structure of these intermediates, as they are very short-lived and can only be characterised by UV–Vis spectra, thus the second intermediate is represented as (L1 (H2 O)NiIII –CH3 )2+ R . Consequently, reaction (14) has to be added to amend the mechanism describing the reaction of methyl radicals with (NiII L1 )2+ complexes (reactions (1 ), (14), (13)). In the final reaction, a decay, obeying a second-order rate law, is observed with a half-life of several seconds. The rate of this process is inversely dependent on the initial Ni-complex concentration in accordance with the proposed mechanism (see Experimental).1 Thus it is evident that C2 H6 is indeed formed via reaction (13) and not via the possible recombination of two • CH3 radicals Dalton Trans., 2007, 3959–3965 | 3961

View Online

Table 1 Kinetic data of the formation and decomposition of Ni(III)Li –CH3 complexes Complex NiLi

K 1 /M−1 3300 ± 800

1 2 3 4 5 Edta

— 2200 ± 400 3900 ± 800 12 000 ± 2400 450 ± 100

K 1 K 14 /M−1

k1 /M−1 s−1

1.1 × 107 1 , a 9.1 × 106 3 , a — — — — —

(5 ± 1) × 108 (6.5 ± 0.7) × 108 1.513 ∼+0.938

Data estimated from comparative studies with scavengers for a mixture of L1 and L2 . b Based on ethane production and the length of the decomposition reaction (slow compared to all other complexes), and see text. c Too fast to be measured; it should be noted that the rate of the back reaction for complex 1 is 1.5 × 105 s−1 i.e. causes a half life of 4.6 × 10−6 s. The results indicate that in these systems the rate of the forward reaction is at least 5 times faster.

Downloaded on 09 February 2011 Published on 27 July 2007 on http://pubs.rsc.org | doi:10.1039/B706491H

a

(reaction (16)). From a plot of kobs vs. 1/[(NiII L1 )2+ ] the quotient 2k13 /K 1 K 14 is derived (Fig. 3), in analogy to the earlier study,1,2 and substituting K 1 K 14 obtained earlier, k13 is calculated as 2.3 × 108 M−1 s−1 . The derived kinetic data are tabulated in Table 1 and compared to literature values.

the direct homolysis of (L1 (H2 O)NiIII –CH3 )2+ R . The equilibrium constant obtained in this study thus relates to K 1 , while reported in earlier studies equals K 1 K 14 , therefore K 14 = 3300. Analysis of the final products of samples irradiated in the low dose rate 60 Co source proved that C2 H6 (85%) and CH4 (15%) are formed. The result is in accord with previous reports.1 Reaction of (NiII L2 )2+ (L2 = cyclam-trans-I) with • CH3 radicals The trans-I and trans-III isomers of [Ni(cyclam)]2+ are energetically the most stable of its possible conformations, and the energy difference between them is insignificant.9,10,24 Solution 1 H NMR studies indicate that both [Ni(cyclam)]2+ trans-I and trans-III forms coexist in aqueous solutions, the trans-I isomer comprising ∼15% of the total complex at room temperature.9,10 The rate of equilibration: trans-I-(Ni(cyclam))2+  trans-III-(Ni(cyclam))2+

Fig. 3 The linear dependence of the rate constant of the decomposition of (L1 (H2 O)NiIII –CH3 )2+ on [(NiII L1 )2+ ]−1 . Solution composition: 0.1–0.6 mM (NiII L1 )2+ ; 0.3 M (CH3 )2 SO; 0.022 M N2 O; pH 3.2; irradiated: 2.5 Gy; measured at k = 310 nm.

Comparing the values of the equilibrium constant of methyl complex formation derived in this study to those of the earlier studies, it is clear that two different reactions are observed. In the present study the initial formation of the (L1 (H2 O)NiIII –CH3 )2+ complex and its isomerisation could be monitored, as faster time scales could be utilised. Consequently, the complete mechanism has to be updated and can be described in the following:

1

(NiII L1 )2+ + • CH3 + H2 O  (L1 (H2 O)NiIII –CH3 )2+

(1 )

(L1 (H2 O)NiIII –CH3 )2+  (L1 (H2 O)NiIII –CH3 )2+ R

(14)

III

(L (H2 O)Ni –CH3 )

2+

+ CH3 → (Ni L ) + C2 H6 + H2 O (13 ) •

R

II

1 2+



The results can not distinguish between the formation of the methyl radicals via the reverse of reactions (14) and (1 ), or 3962 | Dalton Trans., 2007, 3959–3965

(15)

is rapid at high pHs, but extremely slow at acidic pHs. Thus, studies of trans-I-(NiII L2 )2+ were performed only at a relatively low pH. When N2 O saturated aqueous solutions containing (NiII L2 )2+ ((1.0–10.0) × 10−4 M), (CH3 )2 SO (0.30 M), at pH 3.2 are irradiated by a short pulse, two distinct time-resolved reactions are observed (Fig. 4). At the observation wavelength of 300 nm the first reaction appears as a first-order formation reaction in the millisecond time range. The rate of the reaction depends only very slightly on the initial (NiII L2 )2+ concentration with kobs = 1.4 × 104 s−1 . The second reaction obeys a second-order rate law and finishes within 40 seconds under the experimental conditions. No reaction at shorter time scales could be observed. These results can be explained in three ways: 1. On the assumption that the observed reaction is reaction (1), thus kobs = k1 [(NiII L2 )2+ ] + k−1 . If k−1  k1 [(NiII L2 )2+ ] then kobs = k−1 and therefore the rate of formation is independent of [(NiII L2 )2+ ]. Moreover, if k−1 is in the order of 104 s−1 , k1  1 × 106 M−1 s−1 is therefore much lower than the rate constant for the isomeric complex (NiII L1 )2+ . It should be mentioned here that if the formation reaction is that slow, then radical–radical reactions have to also be considered: 2 • CH3 → C2 H6

k = 1.6 × 109 M−1 s−1 2

25

(16) III

2. The second possibility is that the (L (H2 O)Ni –CH3 )2+ complex is formed very quickly (reaction (1)) and that the observed reaction is the isomerisation of that first intermediate (reaction This journal is © The Royal Society of Chemistry 2007

Downloaded on 09 February 2011 Published on 27 July 2007 on http://pubs.rsc.org | doi:10.1039/B706491H

View Online

Fig. 4 The formation and decomposition processes observed for the reaction between • CH3 and (NiII L2 )2+ . Solution composition: 0.5 mM (NiII L2 )2+ ; 0.3 M (CH3 )2 SO; 0.022 M N2 O; pH 3.2; irradiated: 2.5 Gy; measured at k = 300 nm.

(14)). This option is highly unlikely, as no other intermediate was observed. 3. The third alternative is that reaction (1) is very slow and that the reaction determining the observed rate is reaction (16) and reaction (1) is practically a side reaction. In that case the contribution of reaction (1) to the overall observed rate is very small. The timescale (ms) of the observed reaction is in accord with that of the radical–radical dimerisation reaction under our experimental conditions. Also, this assumption leads to the conclusion that k1  1 × 106 M−1 s−1 . Considering all the above options, it is clear that reaction (1) is rather slow for this isomer of the Ni(II)L2+ complex. The decomposition of the intermediate (L2 (H2 O)NiIII –CH3 )2+ takes place over a time range of 40 seconds. A plot of the observed rate constants vs. 1/(NiII L2 )2+ yields a straight line and thus confirms that the main pathway of decomposition is via reaction (13) and not (16).

Fig. 5 (a) The processes observed for the reaction between • CH3 and (NiII L3 )2+ measured at k = 370 nm. Solution composition: 1 mM (NiII L3 )2+ ; 0.3 M (CH3 )2 SO; 0.022 M N2 O; pH 7.4; irradiated: 2.5 Gy. (b) The processes observed for the reaction between • CH3 and (NiII L3 )2+ measured at k = 300 nm. Solution composition: 1 mM (NiII L3 )2+ ; 0.3 M (CH3 )2 SO; 0.022 M N2 O; pH 7.4; irradiated: 2.5 Gy.

Reaction of (NiII L3 )2+ (L3 = hexamethylcyclam) with • CH3 radicals When N2 O saturated aqueous solutions containing (NiII L3 )2+ ((0.5–5.0) × 10−4 M), (CH3 )2 SO (0.30 M), at pH 7.4 are irradiated by a short pulse, three distinct time-resolved reactions are observed at 300 and 370 nm (Fig. 5). The first reaction, observed best at 370 nm, is too fast to be measured even at low complex concentrations, as the homolysis reaction k−1 is very fast. Its equilibrium constant can be derived as K 1 = 2200 ± 300 M−1 as the negative value of the intercept in a plot of [(NiII L3 )2+ ]−1 vs. (DOD)−1 . In the second reaction, a relatively fast reorganisation of the initially formed intermediate is observed in the 100 ls timescale (Fig. 5 and Fig. 6) accompanied by a significant change in spectra. The transformation obeys a first-order rate law, its rate is independent of the initial complex concentration, with a value of k14 = (1.0 ± 0.2) × 105 s−1 . The spectral changes suggest, that in the case of the (NiII L3 )2+ complex a rearrangement of the formed complex from an octahedral to a five-coordinate environment is observed, i.e. probably (L3 (H2 O)NiIII –CH3 )2+ → (L3 NiIII –CH3 )2+ . The spectrum of the rearranged product, featuring a weak absorbance peak around 500 nm, is analogous to the five-coordinated (NiIII L3 )aq 3+ complex.26,27 This journal is © The Royal Society of Chemistry 2007

Fig. 6 Absorption spectra of the intermediates obtained after irradiation of solutions containing (NiII L3 )2+ . (a) intermediate formed 4 ls after the pulse. (b) intermediate formed 85 ls after the pulse. Solution composition: 1 mM (NiII L3 )2+ ; 0.3 M (CH3 )2 SO; 0.022 M N2 O; pH 7.4; irradiated: 2.5 Gy.

The final second-order decomposition reaction of this intermediate is inversely proportional to the initial (NiII L3 )2+ -complex concentration, indicating that this complex decomposes via an analogous mechanism to those of (L1 (H2 O)NiIII –CH3 )2+ and of (L2 (H2 O)NiIII –CH3 )2+ . Gas chromatography of the final products formed after low dose rate irradiation in the 60 Co c-source yields Dalton Trans., 2007, 3959–3965 | 3963

View Online

60 ± 5% ethane and 40 ± 5% methane. The large yield of methane in this system can be explained in two ways: (a) In the reaction of this complex more methane is produced due to H-abstraction from the methyl groups of the ligand. (b) The complex undergoes heterolysis, producing (NiIII (H2 O)2 L3 )3+ and methane. The absence of any absorbance after the decomposition reaction indicates possibility (a).

Downloaded on 09 February 2011 Published on 27 July 2007 on http://pubs.rsc.org | doi:10.1039/B706491H

Reaction of (NiII L4 )2+ (L4 = hexamethyl-1,7-diene cyclam) with • CH3 radicals When N2 O saturated aqueous solutions containing (NiII L4 )2+ ((1.0–10.0) × 10−4 M), (CH3 )2 SO (0.30 M), at pH 7.4 are irradiated by a short pulse, two distinct time-resolved reactions are observed at 310 nm. The first reaction is too fast to be measured even at low complex concentrations, as the homolysis reaction k−1 is very fast. Its equilibrium constant can be derived as K 1 = 3900 ± 800 M−1 as the negative value of the intercept in a plot of [(NiII L4 )2+ ]−1 vs. (DOD)−1 . The final decomposition reaction of this intermediate obeys a second-order rate law (within hundreds of microseconds) with a rate which is inversely proportional to the initial (NiII L4 )2+ concentration, indicating that this complex decomposes via an analogous mechanism to that of the other complexes studied. Gas chromatography of the final products yields 60 ± 5% ethane and 40 ± 5% methane in analogy to the (NiII L3 )2+ system. Reaction of (NiII L5 )2+ (L5 = TIM) with • CH3 radicals When N2 O saturated aqueous solutions containing (NiII L5 )2+ ((0.5–5.0) × 10−4 M), (CH3 )2 SO (0.30 M), at pH 7.4 are irradiated by a short pulse, three distinct time-resolved reactions are observed at 320 nm. The derived kinetic data are tabulated in Table 1. In the first reaction, an unstable intermediate is formed within 180 ls under the experimental conditions (Fig. 7). Its kinetics of formation obeys a first-order rate law with the observed rate constants proportional to the concentration of (NiII L5 )2+ , and independent of the concentrations of other components of the solution, pH or wavelength of observation. Thus the second-order rate constant is derived as k1 = (1.6 ± 0.3) × 108 M−1 s−1 , from the slope of the graph of the observed first-order rate constants vs. [(NiII L5 )2+ ]. The equilibrium constant of reaction (1), K 1 = 12000 ± 2000 M−1 , was obtained as the negative value of the intercept in a plot of [(NiII L5 )2+ ]−1 vs. (DOD)−1 . In the second reaction, the isomerisation of the initially formed intermediate is observed on a 400 ms timescale (Fig. 7). The transformation obeys first-order kinetics, its rate is independent of the initial complex concentration, with values of k = 25 ± 5 s−1 . The final decomposition reaction is very slow (minutes). Gas chromatography of the final products yields 85 ± 5% ethane and 15 ± 5% methane in analogy with the (NiII L1 )2+ system. The details of the decomposition reaction were not studied further, as the reaction proceeds over several minutes. Based on GC results, an analogous mechanism to the (NiII L1 )2+ system is corroborated for the (NiII L5 )2+ system, though some contribution of reaction 16 in the long low dose irradiation can not be ruled out. It should be 3964 | Dalton Trans., 2007, 3959–3965

Fig. 7 The UV–vis spectra of intermediates observed in the reaction between • CH3 and (NiII L5 )2+ . (a) Intermediate formed 180 ls after the pulse. (b) Intermediate formed 400 ms after the pulse. Solution composition: 0.25 mM (NiII L5 )2+ ; 0.3 M (CH3 )2 SO; 0.022 M N2 O; pH 7.4; irradiated: 2.5 Gy.

noted that the z-axis of (NiII L5 )2+ is the least sterically hindered of all the complexes studied.28 Reaction of NiII edta with • CH3 radicals When N2 O saturated aqueous solutions containing NiII edta ((1.0– 10.0) × 10−3 M), (CH3 )2 SO (0.30 M), at pH 7.0, NaH2 PO4 (0.02 M) and Na2 HPO4 (0.03 M) are irradiated by a short pulse, two distinct time-resolved reactions are observed at 310 nm. Although for this complex high concentrations were employed, only a negligible amount of the complex will react with e− aq to form the Ni(I)edta complex (k(e−aq + NiII edta) = 1 × 108 M−1 s−1 ,23 compared to reaction (3)). The first reaction is too fast to be measured even at low complex concentrations, as the homolysis reaction k−1 is very fast. Its equilibrium constant can be derived as K 1 = 430 ± 90 M−1 as the negative value of the intercept in a plot of [NiII edta]−1 vs. (DOD)−1 . The spectrum of the intermediate is shown in Fig. 8.

Fig. 8 The UV–Vis spectrum of the intermediate observed in the reaction between • CH3 and NiII edta. Solution composition: 5 mM NiII edta; 0.02 M NaH2 PO4 ; 0.03 M NaH2 PO4 ; 0.3 M (CH3 )2 SO; 0.022 M N2 O; pH 7.4; irradiated: 2.5 Gy.

This journal is © The Royal Society of Chemistry 2007

View Online

The final decomposition reaction of this intermediate obeys a second-order rate law (within several milliseconds) and is inversely proportional to the initial NiII edta concentration, indicating that this complex decomposes via a mechanism analogous to that of all other (L(H2 O)NiIII –CH3 )2+ complexes. Gas chromatography of the final products yields 80 ± 5% ethane and 20 ± 5% methane.

Downloaded on 09 February 2011 Published on 27 July 2007 on http://pubs.rsc.org | doi:10.1039/B706491H

Concluding remarks The kinetic and thermodynamic properties of all nickel complexes studied are summarised in Table 1. The results demonstrate that methyl radicals formally oxidise these nickel complexes, forming intermediates with NiIII –C r-bonded of the type [Li NiIII –CH3 ]2+ in an equilibrium process with a relatively fast forward rate of k1 > 1 × 108 M−1 s−1 (except the case of NiL2 -trans-I cyclam, where the reaction is considerably slower). In all cases the decomposition of the transient occurs via the homolytic cleavage of the metal–carbon r-bond. Two major factors are expected to affect the stability of NiIII –C r-bonds: (a) The redox potential of the NiIII/II Li couple. (b) Steric hindrance, which is expected to weaken the Ni–C bond. The results seem to suggest that the latter is the major factor affecting the strength of the NiIII –C r-bond. Thus though the redox potential of the NiIII/II L5 couple is the highest the kinetics of decomposition of (L5 (H2 O)NiIII –CH3 )2+ are the slowest. This study shows that intermediates with NiIII –CH3 r-bonds exist and can be studied in aqueous solutions. These results point out that the proposal of NiIII –CH3 species as crucial intermediates in enzymes like CO dehydrogenases29–33 and acetyl Co–M synthase34–36 is more than relevant.

Acknowledgements This study was supported in part by a grant from the Budgeting and Planning Committee of the Council of Higher Education and the Israel Atomic Energy Commission. DM wishes to thank the Alexander von Humboldt Foundation for support and Mrs Irene Evens for her ongoing interest and support.

References 1 A. Sauer, H. Cohen and D. Meyerstein, Inorg. Chem., 1988, 27, 4578– 4581. 2 R. van Eldik, H. Cohen, A. Meshulam and D. Meyerstein, Inorg. Chem., 1990, 29, 4156–4158. 3 D. G. Kelley, A. Marchaj, A. Bakac and J. H. Espenson, J. Am. Chem. Soc., 1991, 113, 7583.

This journal is © The Royal Society of Chemistry 2007

4 R. Cammack and P. Vliet, in Bioinorganic Catalysis (2nd edn), ed. J. Reedijk, Marcel Dekker, New York, 1999, ch. 4, pp. 231–233. 5 J. Telser, R. Davydov, Y. C. Horng, S. W. Ragsdale and B. M. Hoffman, J. Am. Chem. Soc., 2001, 123, 5853–5860. 6 T. Wondimagegn and A. Ghosh, J. Am. Chem. Soc., 2001, 123, 1543– 1544. 7 J. Telser, Y. C. Horng, D. F. Becker, B. M. Hoffman and S. W. Ragsdale, J. Am. Chem. Soc., 2000, 122, 182–183. 8 W. P. Lu, S. R. Harder and S. W. Ragsdale, J. Biol. Chem., 1990, 265, 3124–3133. 9 D. T. Pierce, T. L. Hatfield, E. J. Billo and Y. Ping, Inorg. Chem., 1997, 36, 2950–2955. 10 E. J. Billo, P. J. Connolly, D. J. Sardella, J. P. Jasinski and R. J. Butcher, Inorg. Chim. Acta, 1995, 230, 19–28. 11 M. Jaacobi, D. Meyerstein and J. Lilie, Inorg. Chem., 1979, 18, 429–433. 12 A. M. Tait and D. H. Busch, Inorg. Nucl. Chem. Lett., 1972, 8, 491. 13 F. V. Lovecchio, E. S. Gore and D. H. Busch, J. Am. Chem. Soc., 1974, 96, 3109–3118. 14 A. M. Tait and D. H. Busch, Inorg. Synth., 1978, 18, 2–26. 15 E. J. Hart, S. Gordon and J. K. Thomas, J. Phys. Chem., 1964, 68, 1271–1274. 16 J. K. Thomas, S. Gordon and E. J. Hart, J. Phys. Chem., 1964, 68, 1524–1527. 17 S. Goldstein, G. Czapski, H. Cohen and D. Meyerstein, Inorg. Chem., 1992, 31, 2439–2444. 18 M. Freiberg and D. Meyerstein, J. Chem. Soc., Faraday Trans. 1, 1980, 76, 1825–1837. 19 J. W. T. Spinks and R. J. Woods, An Introduction to Radiation Chemistry (3rd edn), John Wiley & Sons, New York, 1990, p. 305. 20 J. W. T. Spinks and R. J. Woods, An Introduction to Radiation Chemistry (3rd edn), John Wiley & Sons, New York, 1990, p. 105. 21 M. S. Matheson and L. Dorfman, Pulse Radiolysis, M.I.T. Press, Cambridge, MA, 1969. 22 G. V. Buxton, C. V. Greenstock, W. P. Helman and A. B. Ross, J. Phys. Chem. Ref. Data, 1988, 17, 513–886. 23 D. Veltwisch, E. Janata and K. D. Asmus, J. Chem. Soc., Perkin Trans. 2, 1980, 146–153. 24 E. Maimon, I. Zilbermann, D. Kost, H. Cohen, R. van Eldik and D. Meyerstein, Eur. J. Inorg. Chem., 2005, 4997–5004. 25 P. Neta, J. Grodkowski and A. B. Ross, J. Phys. Chem. Ref. Data, 1996, 25, 709–1050. 26 A. G. Lappin and A. McAuley, Adv. Inorg. Chem., 1988, 32, 241. 27 E. Zeigerson, G. Ginzburg, N. Schwartz, Z. Luz and D. J. Meyerstein, J. Chem. Soc., Chem. Commun., 1979, 241–243. 28 D. J. Szalda, E. Fujita, R. Sanzenbacher, H. Paulus and H. Elias, Inorg. Chem., 1994, 33, 5855–5863. 29 S. W. Ragsdale, Crit. Rev. Biochem. Mol. Biol., 1991, 26, 261–300. 30 S. W. Ragsdale and M. Kumar, Chem. Rev., 1996, 96, 2515–2539. 31 A. Volbeda and J. C Fontecilla-Camps, J. Biol. Inorg. Chem., 2004, 9, 525. 32 G. G. Riordan, J. Biol. Inorg. Chem., 2004, 9, 509. 33 C. L. Draman, T. I. Dovkov and S. W. Ragsdale, J. Biol. Inorg. Chem., 2004, 9, 511. 34 G. G. Riordan, J. Biol. Inorg. Chem., 2004, 9, 542. 35 P. A. Lindahl, J. Biol. Inorg. Chem., 2004, 9, 516. 36 T. C. Brunold, J. Biol. Inorg. Chem., 2004, 9, 533. 37 E. Maimon, PhD thesis, Ben Gurion University, Beer Sheva, Israel, 2001. 38 E. Fuchs, G. Ginsburg, J. Lati and D. Meyerstein, J. Electroanal. Chem., 1976, 73, 83–90.

Dalton Trans., 2007, 3959–3965 | 3965