Grantee Research Project Results
Final Report: Biomimetic Oxidation of Hydrocarbons Related to Bioremediation Processes
EPA Grant Number: R823377Title: Biomimetic Oxidation of Hydrocarbons Related to Bioremediation Processes
Investigators: Stavropoulos, Pericles
Institution: Boston University
EPA Project Officer: Aja, Hayley
Project Period: October 1, 1995 through September 30, 1998
Project Amount: $321,711
RFA: Exploratory Research - Engineering (1995) RFA Text | Recipients Lists
Research Category: Safer Chemicals , Land and Waste Management
Objective:
This project explored metal-mediated processes for the catalytic oxidation of alkanes and alkenes to biodegradable oxo products. This research is inspired by the action of iron-containing biological monooxygenases, especially methane monooxygenase and P-450. The type of catalysts employed in the present study mediate Gif-like chemistry, which demonstrates an unusual selectivity for the oxygenation of secondary C-H bonds over tertiary or primary sites. Despite structural similarities between precursor Fe(II) species observed in Gif chemistry and ferrous sites present in methane monooxygenase, the functional characteristics of the two systems differ substantially. The target of this work was to contribute toward unraveling the active oxidant(s) involved in Gif chemistry and developing mechanistic insight into the mode of action vis-a-vis unfunctionalized hydrocarbons.
Summary/Accomplishments (Outputs/Outcomes):
This section discloses key results that place this research into context. Two distinct classes of iron-based hydrocarbon-oxidizing reagents have been recognized, those that necessitate dioxygen and a reducing agent (but would dismutate H2O2), and those that require H2O2 (but will afford no oxo products with O2/Red). The distinction is made on the choice of carboxylic acid, which would influence the stability of FeII sites versus O2. These two classes of systems are treated here separately.
1. Systems Employing O2/Reducing Agent
[Fe3O(O2CCH3)6(py)3] py (or related species)/Zn/O2. Structural and functional aspects of
the behavior of this basic iron acetate system have been extensively explored. Details of this work have been presented in previous reports, can be found in a preliminary report (Journal of the American Chemical Society 1996;118:5824-5825) and an extensive account (Journal of the American Chemical Society 1997;119:7030-7047), and can be downloaded from the Internet (http://pubs.acs.org/cgi-bin/jtext?jacsat/118/i24/pdf/ja960529p and http://pubs.acs.org/cgi-bin/jtext?jacsat/119/i30/pdf/ja970562r ). Here, we concentrate on isolation of intermediates and kinetics.
The ferrous sites [FeII(O2CCH3)2(py)4] (referred to as 1*) and [FeII2(O2CCH3)4)(py)3]n (referred to as 2*) can be isolated under reducing conditions (Zn, Fe, H2/Pd) as the only relevant, equilibrium-dependent species in Gif solutions (py/AcOH 10:1). The intermediates and kinetics of their interaction with dioxygen have been investigated. Quantitative UV-vis data have indicated that, under anaerobic conditions, compound 1* ( λmax 388 nm ( M =2084)) is the only detectable species up to a concentration of 0.88 x 10-3 M. Beyond this threshold value, new species develop in solution in a concentration-dependent manner directly associated with the appearance of 2* as indicated by near-IR data. Admittedly, the nature of 2* in solution is not precisely known, but EI-MS data have yielded fragments assigned to [Fe(OAc)(py)]+, [Fe(OAc)(py)2]+, [Fe2(OAc)3(py)]+, [Fe2(OAc)3(py)2]+, [Fe3(OAc)5(py)]+, and [Fe3(OAc)5(py)2]+.
UV-vis studies on the oxidation of 1* with dioxygen in py/AcOH at concentrations below the 0.88 x 10-3 M threshold suggest that an intermediate ( λmax =338 nm) is formed that only slowly converts to the trinuclear cluster [Fe3O(OAc)6(py)3] (referred to as 3*) ( λ = 358, 416, 488, 580 nm). A broad but weak absorption centered at 658 nm accompanies the 338 nm feature. In addition to absorbances assigned to 3*, a feature at 474 nm grows in after prolonged periods, most likely due to a decay product. A tentative assignment for this EPR-silent intermediate is [Fe2( µ-O2)(OAc)4-(py)6] (referred to as 4*), based on EI-MS and NMR data. Interestingly, similar features at 474 and 670 nm (broad transient) recently have been detected in the oxidation of [Fe2(OH) (Me3-TACN)2(OAc)2]+ and attributed to diferric µ-oxo and µ-peroxo species, respectively.
The reaction of 1* with dioxygen in py/AcOH (10:1) also has been examined by variable-temperature stopped-flow kinetics using the unique Hitech cryostopped-flow instrument at the Institute of Inorganic Chemistry, University of Basel, in collaboration with Professor A. D. Zuberbuhler. Eight variable-concentration experiments were conducted ([1*] = (0.2 - 8) x 10-4 M) by mixing a solution of 1* in py/AcOH (10:1) with a dioxygen-saturated py/AcOH solution. The temperature was varied in the range -35?C to 25?C. Full mechanistic analysis is contingent upon a better understanding of the nature of the intermediates involved, but the following points can be made: (i) no fast preequilibria were discernible within the present concentration and temperature range; (ii) two primary relaxations were observed (the first at the limit of the SF range and the second outside the SF range), most likely due to consecutive reactions; (iii) the first relaxation is associated with consumption of 1* in an apparent third-order reaction (-d[1*]/dt =kobs[Fe]2[O2]); rate constants and activation parameters obtained are as follows: kobs(267 K) = 2 x 102 M-2 s-1 , kobs(297 K) = 3 x 103 M-2 s-1, H? = 55 + 3 kJ mol-1, S? 0 J K-1 mol-1; and (iv) the second relaxation is associated with slow conversion to 3*, although the order of this decay process is not currently known. The optical features, kinetics, and preliminary characterization indicate the intermediacy of ferric peroxo species 4, but this suggestion needs to be further substantiated by rigorous characterization.
[Fe3O(O2CCMe3)6(L)3](Cl) (or related species)/Zn/O2 (L = py, H2O). The basic iron pivalate system is of interest to us for several reasons: (i) the bulky carboxylate may facilitate O2 activation by influencing the lability of pyridine moieties, thus providing free coordination sites; (ii) the electron-donor capacity of pivalate stabilizes not only the mixed-valent [Fe3O(O2CCMe3)6(py)3] (referred to as 5*), but also the all-ferric [Fe3O(O2CCMe3)6(py)3]+ (referred to as 6*) in aerobic py/PivH solutions (pyridine and methanol adducts of the all-ferric species are known to epoxidize olefins by O2); and (iii) the weak acidity of pivalic acid reduces the amount of protonated pyridine and, therefore, minimizes the extent of coupling of substrate-derived alkyl radicals to pyridine to yield alkylpyridine byproducts.
Compounds 5* and 6* can be prepared easily in pyridine from the corresponding water adducts. Reduction of either 5* or 6* with Zn affords [Fe(O2CCMe3)2(py)4] (referred to as 7*) and [Fe2(O2CCMe3)4(py)2] (referred to as 8*). Compounds 7* and 8* (and possibly trinuclear [Fe3(O2CCMe3)6(py)2], referred to as 9*), detected by EI-MS, are in equilibrium in py/PivH (10:1) solutions. The solutions of these ferrous species are even more sensitive to dioxygen by comparison to the acetate analogs, eventually yielding a mixture of 5* and 6*. UV-vis investigation (-40?C to RT) of dilute solutions of 7* affords an intermediate similar to the one observed with the acetate analog (4*). At higher concentrations (>0.4 x 10-4 M), dioxygen activation becomes progressively more rapid than anticipated for a second-order reaction (vs. metal) and increasingly more complex with respect to the observed intermediates.
Apparently, the five-coordinate species 8*, prominently emerging at higher concentrations, can bind dioxygen instantaneously, also as indicated in non-pyridine solutions (CH3CN or CH2C12) in which 8* is the only detectable species. Kinetic analysis of this system currently is underway.
2. Systems Employing Hydrogen Peroxide
[Fe(Pic)2(py)2] (or related species)/H2O2. Picolinic acid has been shown to increase the rate of "oxygenated Fenton-type" and Gif-like oxidations by fiftyfold, but the relevant picolinate species and intermediates involved in these systems have remained speculative.
Polymeric [Fe(Pic)2]n (referred to as 10*) is precipitated as a red powder upon stirring metallic Fe and PicH in CH2Cl2 under dinitrogen atmosphere. [Fe(Pic)2]n dissolves in pyridine to afford orange-red crystals of analytically pure [FeII(Pic)2(py)2] (referred to as 11*). Solutions of 10* afford orange crystals of mononuclear [FeII(Pic)2(MeOH)2] (referred to as 12*) from methanol and [FeII(Pic)2(H2O)2] (referred to as 13*) from water, whereas red crystals of dinuclear [FeII2(Pic)4(DMF)2] (referred to as 14*) are obtained from dimethylformamide (DMF). Compounds 12*-14* are air-sensitive, yielding [FeIII2( -OMe)2(Pic)4] (referred to as 15*), [FeIII2 -OH)2(Pic)4] (referred to as 16*), and [FeIII2O(Pic)4(DMF)2] (referred to as 17*), upon exposure to dioxygen in the corresponding solvents. Importantly, 11* is not sensitive to dioxygen in pyridine, which may explain why Gif-like picolinate-based systems do not operate under O2/Red. However, ferric picolinate species can be obtained indirectly, by dissolving 17* in pyridine to afford brown-red crystals of [FeIII2O(Pic)4(py)2] (referred to as 18*). The same compound can be obtained from [Fe2OCl6]2- and NaPic. The bridging oxo group in 18* is sufficiently basic to regenerate 16* or 15* in pyridine, upon protonation by stoichiometric amounts of H2O or CH3OH, respectively. Compound 16 is unstable in pyridine, affording initially mononuclear [FeIII(Pic)2(py)(OH)] (referred to as 19*) and eventually intractable polymeric products. Excess water or CH3COOH transforms 18* to [Fe(Pic)3]?py (referred to as 20*).
Compounds 11* and 18*-20* react with H2O2 in pyridine to yield the same thermally unstable intermediate. We concentrate on the reaction of yellow 20* with H2O2, which requires smaller excess of H2O2. The reaction is initiated at -40?C with slow drop-wise addition of H2O2 (5 equivalents). UV-vis followup of the reaction indicates that a purple species is generated ( = 340, 420, 530 nm), starting at ~ -30?C, which decays at higher temperatures (~ -5?C) to afford 11*. Analysis of the reaction profile by EPR shows that the initial rhombic signal of 20* disappears, and a new signal (g = 4.34, 7.48) grows in within the temperature range of the stability of the purple species. At higher temperatures (>0?C), an extremely complex EPR spectrum is obtained (11* is EPR silent). The EPR spectrum of the purple species is characteristic of a mononuclear high-spin ( µeff = 5.7 µB) ferric compound. Based on EI-MS data and the similarity of this EPR signal to the one observed for the yellow hydroxy species 19, we tentatively assign this compound as [FeIII(Pic)2(py)(OOH)] (referred to as 21*). However, we cannot at this point exclude a side-on peroxo species, such as [PyNH][FeIII(Pic)2(O2)].
[Fe3O(O2CCF3)6(H2O3)] (or related species)/H2O2. The basic iron trifluoroacetate system exhibits more than one difference by comparison to the parent acetate analog. Entry to this system is achieved by replacing all acetate groups of [Fe3O(O2CCH3)6(H2O)3] upon reaction with CF3COOH. The resulting compound crystallizes as red [Fe3O(O2CCF3)6(H2O)3]?2.5H2O?CF3COOH (referred to as 22*) from TFA/H2O (4:l v/v). When dissolved in dimethyl sulfoxide (DMSO), 22* affords the DMSO adduct [Fe3O(O2CCF3(DMSO)3] (referred to as 23*). Surprisingly, solutions of 22* or 23* in pyridine, which would be expected to furnish the analogous pyridine adduct, afford quantitatively green [FeII(O2CCF3)2(py)4] (referred to as 24*) and red [FeIII2O(O2CCF3)4(py)6]?2py (referred to as 25*). Apparently, the trinuclear core structure dissociates completely according to valence requirements. Compounds 24* and 25* are separated by fractional crystallization, but also can be synthesized independently. Compound 24* is obtained from a pyridine solution of [FeII(O2CCF3)2]n (in turn prepared by dissolving Fe powder in TFA), whereas 25* is made from the reaction of [Fe2OCl6]2- and CF3CO2Na in pyridine. Compound 25* is reduced easily to 24*.
Compound 24* is completely stable versus dioxygen, but reacts with H2O2 in py/TFA to show unique characteristics. The initial yellow-green color of 24* disappears upon addition of H2O2 (3 equivalents) at -40?C to afford a green-brown intermediate. At temperatures >-10?C, this intermediate gives way to a progressively intensifying red-pink solution that is stable at room temperature. The following species can be isolated from this solution: red-pink [FeII(2,2'-bipy)3](O2CCF3)2 (referred to as 26*), brown [Fe2IIIO(O2CCF3)4(2,2'-bipy)2] (referred to as 27*), and trace amounts of starting material 24*. This is the only system in which 2,2'-bipy (formed via hydrohylation/dehydration of pyridine) is found to interfere with the outcome of stoichiometric reactions (although it is produced in small amounts, along with other bipyridines, in all catalytic Gif-type systems). Obviously, the high yield of 2,2'-bipy is a testimony of the oxidizing power of the present system, which otherwise resembles a typical Fenton reagent in its need for a not easily autooxidized FeII precursor and requirement of acidic conditions. In neat pyridine, the reaction of 24* with H2O2 affords 25* as the only detectable product. The present system allows us to study FeII/H2O2 and FeIII/H2O2 interactions within a context of well recognizable species. Detailed followup of these observations by a variety of spectroscopic probes is in progress.
FeCl3/H2O2 in the presence/absence of PicH. As already mentioned, the system FeCl3/PicH/H2O2 (1:4:4) has been the most exploited Gif reagent. Barton and coworkers have shown that FeCl3 and PicH in pyridine afford green [PyNH???Py][FeIII(Pic)2Cl2] (referred to as 28*). They also have indicated that a purple solid is precipitated from the reaction of 28* and H2O2 in acetone at low temperature. In our hands, the same reaction in the presence of Et3N in pyridine/acetone (1:1) at -20?C affords a purple microcrystalline material upon cooling (-60?C). Barton's assignment of this complex as the peroxo [PyNH???Py][FeIII(Pic)2O2] (referred to as 29*) has not as yet been confirmed. Reportedly, this purple compound does not react directly with cyclohexane, unless H2O2 is added to the reaction mixture. We also have isolated the reduced species [FeII(Pic)(py)3Cl] (referred to as 30*), which is not sensitive to dioxygen but reacts rapidly with H2O2. These transformations currently are under investigation.
The system FeCl3/H2O2 (95 percent) in CH3CN or pyridine (in the absence of any carboxylic acid) is unique inasmuch as oxygenation of alkanes affords predominantly alcohols rather than ketones, whereas olefins afford epoxides with good stereospecificity. The system in anhydrous CH3CN was first recognized by Sawyer. In anhydrous pyridine, FeCl3 exists as orange-red [mer-FeCl3(py)3] (referred to as 31*). Its reaction with only one equivalent of H2O2 at -78?C in py/CH2Cl2 (CH2Cl2 is added to allow work at low temperature) generates a deep green solution that persists up to -50?C. A substantial amount of dioxygen is released at higher temperatures and the green product decays to yellow-orange [Fe2III( -OH)2(py)4(Cl)4]?2py (referred to as 32*; solvate pyridine is hydrogen-bonded to bridging hydroxy groups). Most interestingly, 32* regenerates the deep green solution upon addition of H2O2 (1 equivalent) at -78?C. Spectroscopic analysis of the green product currently is underway.
3. Catalytic Oxidation of Hydrocarbons
A number of substrates (alkanes, alkenes) have been subjected to catalytic oxygenations mediated by the aforementioned systems. Here, we concentrate on adamantane, which has been traditionally used by Barton as a good mechanistic probe. These results indicate that, under N2, sec-adamantylpyridines are present in comparable yields to tert-adamantylpyridines in oxygenation mediated by [FeII(O2CCF3)2(py)4] (referred to as 24*), but the amounts of sec-adamantylpyridines are minimized when [FeIII2O(O2CCF3)4(py)6]?2py (referred to as 25*) is used. The same oxygenations under O2 reduce sec-adamantylpyridines to trace amounts (in favor of the oxo products), although substantial amounts of tert-adamantylpyridines remain (50-60 percent), while the tert/sec ratio under N2 or O2 is virtually the same. These results are qualitatively in agreement with the presence of tert- and sec-adamantyl radicals (reflecting the preferred trapping of tert-adamantyl radicals by pyridine), but there exists a quantitative disagreement by comparison to the behavior of authentic tert- and sec-adamantyl radicals in control experiments. Similar results are obtained for adamantane oxidation mediated by [FeII(Pic)2(py)2] (11*) or [FeIII2O(Pic)4(py)2] (18*), but reactions under O2 also are characterized by disappearance of tert-adamantylpyridines and a marked increase in the tert/sec ratio (~ 4.5), suggesting interference of a more selective oxidant. [Fe(Pic)3]?py (20*) behaves in an analogous fashion in pyridine, although no 2-ol or adamantylpyridines are obtained; the latter is most likely due to the absence of excess acid. Some very recent results (not shown) have indicated that the amount of sec-adamantylpyridines can be increased if a stream of argon is constantly passed through the FeIII-based catalytic systems. This also has been noted by Barton in regard to cyclohexane oxidation by FeIII/H2O2 systems, but has been interpreted as loss of bound (rather than dissolved) dioxygen from a putative FeIII-O-O-FeIII unit to yield Fe(II), thus entering the radical FeII/FeIV =O manifold. Naturally, the dioxygen-dependent systems 3*/Zn/O2 or 7*/Zn/O2 afford traces of sec-adamantylpyridines, and in the latter case only small amounts of tert- adamantylpyridines, due to the low acidity of pivalic acid.
4. Conclusions
The following are the principal findings of this investigation:
- The hydrocarbon-oxidizing systems (i) [Fe3O(O2CCH3)6(py)3]?py (or related species)/Zn/O2; (ii) [Fe3O(O2CCMe3)6(L)3](Cl) (or related species)/Zn/O2 (L = py, H2O); (iii) [Fe(Pic)2(py)2] (or related species)/H2O2; (iv) [Fe3O(O2CCF3)6(H2O)3] (or related species)/H2O2; and (v) FeCl3/H2O2 in the presence/absence of PicH have been investigated in pyridine or pyridine/AcOH solutions and in the absence of any hydrocarbon substrate to reveal a host of mononuclear, dinuclear, and polymeric ferrous or ferric species that are precursor or end-products of putative catalytic cycles.
- Upon exposure to dioxygen or air, the reduced Fe(II) sites generate mixed-valent Fe(II)/Fe(III) or purely Fe(III) centers, which, in turn, can be reduced to the starting Fe(II) sites, thus generating consistent chemical redox cycles that may be responsible for sustaining the catalytic conversions of Gif chemistry. These turnovers are brought about by a variety of reducing agents (Zn, Fe, electrochemical cathode) coupled to dioxygen, or by shunt reagents such as hydrogen peroxide or superoxide. However, the aforementioned hydrocarbon-oxidizing systems are specific in the type of oxygen-transfer apparatus employed.
- Analogies with structural elements of the diiron site in the hydroxylase component of soluble methane monooxygenase are apparent, especially between [FeII2(O2CCH3)4(py)3]n and the reduced diiron center of the biological hydroxylase. However, the nuclearity of the present systems upon chemical redox cycles is not conserved. In addition, product profiles and deuterium kinetic isotope effect values obtained from catalytic oxidation of a number of substrates reveal features that are not consistent with the functional characteristics assigned to methane monooxygenase activity.
- The catalytic systems based on hydrogen peroxide are mechanistically valuable, inasmuch as substrate oxygenations can be performed under dioxygen or inert atmosphere. In the presence of protonated pyridine (and more recently, nitroxyl radical traps) under N2, both tertiary and secondary alkyl radicals have been trapped as tert- or sec-alkylpyridine products. Although tert-alkyl radicals are observed in systems employing Fe(II) or Fe(III) "catalysts," sec-alkyl radicals are only detected in oxygenations mediated by Fe(II) precursor species. In contrast, diffusively free alkyl radicals are not obtained with methane monooxygenase, as they are instantaneously trapped by the FeIV-O(H) unit.
- The accumulated results suggest that Fe(II)/H2O2 systems resemble Fenton-type radical reagents, although the intermediacy of hydroxyl radicals has not yet been confirmed. In contrast, Fe(III)/H2O2 reagents remain open to further investigation, as the amount of generated sec-alkyl radicals is minimized in favor of sec-alkyl oxo products. Ferric (hydro)peroxo intermediates are under scrutiny.
Journal Articles on this Report : 4 Displayed | Download in RIS Format
Other project views: | All 4 publications | 4 publications in selected types | All 4 journal articles |
---|
Type | Citation | ||
---|---|---|---|
|
Singh B, Long JR, Papaefthymiou GC, Stavropoulos P. On the reduction of basic iron acetate: Isolation of ferrous species mediating gif-type oxidation of hydrocarbons. Journal of the American Chemical Society 1996;118(24):5824-5825. |
R823377 (Final) |
not available |
|
Singh B, Long JR, Fabrizi de Biani F, Gatteschi D, Stavropoulos P. Synthesis, reactivity, and catalytic behavior of iron/zinc-containing species involved in oxidation of hydrocarbons under Gif-type conditions. Journal of the American Chemical Society 1997;119(30):7030-7047. |
R823377 (Final) |
not available |
|
Singh K, Long JR, Stavropoulos P. Ligand-unsupported metal-metal (M=Cu, Ag) interactions between closed-shell d10 trinuclear systems. Journal of the American Chemical Society 1997;119(12):2942-2943. |
R823377 (Final) |
not available |
|
Singh K, Long JR, Stavropoulos J. Polynuclear complexes of copper(I) and the 2-(3(5)-pyrazolyl), 6-methylpyridine ligand: structures and reactivity toward small molecules. Inorganic Chemistry 1998;37(5):1073-1079. |
R823377 (Final) |
not available |
Supplemental Keywords:
hydrocarbon oxidation, remediation, Gif chemistry, Fenton chemistry, peroxo intermediates., Scientific Discipline, Toxics, Waste, Environmental Chemistry, HAPS, Bioremediation, Environmental Engineering, iron based oxidizing system, air pollutants, hydrocarbon, copper based oxidizing, hydrocarbon oxidation, ecological consequences, redox transformations, hydrocarbon degrading, hydroxylation of secondary carbonsProgress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.