Non-Heme-Iron-Mediated Selective Halogenation of Unactivated Carbon-Hydrogen Bonds.

Oxidation of the iron(II) precursor [(L1 )FeII Cl2 ], where L1 is a tetradentate bispidine, with soluble iodosylbenzene (s PhIO) leads to the extremely reactive ferryl oxidant [(L1 )(Cl)FeIV =O]+ with a cis disposition of the chlorido and oxido coligands, as observed in non-heme halogenase enzymes. Experimental data indicate that, with cyclohexane as substrate, there is selective formation of chlorocyclohexane, the halogenation being initiated by C-H abstraction and the result of a rebound of the ensuing radical to an iron-bound Cl- . The time-resolved formation of the halogenation product indicates that this primarily results from s PhIO oxidation of an initially formed oxido-bridged diiron(III) resting state. The high yield of up to >70 % (stoichiometric reaction) as well as the differing reactivities of free Fe2+ and Fe3+ in comparison with [(L1 )FeII Cl2 ] indicate a high complex stability of the bispidine-iron complexes. DFT analysis shows that, due to a large driving force and small triplet-quintet gap, [(L1 )(Cl)FeIV =O]+ is the most reactive small-molecule halogenase model, that the FeIII /radical rebound intermediate has a relatively long lifetime (as supported by experimentally observed cage escape), and that this intermediate has, as observed experimentally, a lower energy barrier to the halogenation than the hydroxylation product; this is shown to primarily be due to steric effects.

First-Generation Bispidine Chelators for 213 BiIII Radiopharmaceutical Applications.

Hepta- and octadentate bispidines (3,7-diazabicyclo[3.3.1]nonane, diaza-adamantane) with acetate, methyl-pyridine, and methyl-picolinate pendant groups at the amine donors of the bispidine platform have been prepared and used to investigate BiIII coordination chemistry. Crystal structure and solution spectroscopic data (NMR spectroscopy and mass spectrometry) confirm that the rigid and relatively large bispidine cavity with an axially distorted geometry is well suited for BiIII and in all cases forms nine-coordinate complexes; this is supported by an established hole size and shape analysis. It follows that nonadentate bispidines probably will be more suited as bifunctional chelators for 213 BiIII -based radiopharmaceuticals. However, two isomeric picolinate-/acetate-based heptadentate ligands already show very efficient complexation kinetics with 213 BiIII at ambient temperature and kinetic stability that is comparable with the standard ligands used in this field. The experimentally determined hydrophilicities (log D7.4 values) show that the BiIII complexes reported are relatively hydrophilic and well suited for medicinal applications. We also present a very efficient and relatively accurate method to compute charge distributions and hydrophilicities, and this will help to further optimize the systems reported here.

Spin state and reactivity of iron(iv)oxido complexes with tetradentate bispidine ligands.

The iron(iv)oxido complex [(bispidine)FeIV[double bond, length as m-dash]O(Cl)]+ is shown by experiment and high-level DLPNO-CCSD(T) quantum-chemical calculations to be an extremely short-lived and very reactive intermediate-spin (S = 1) species. At temperatures as low as -90 °C, it decays with a half-life of approx. two minutes, and this is the reason why, so far, it remained undetected and why it is extremely difficult to trap and fully characterize this interesting and extremely efficient oxidant. The large difference in reactivity between [(bispidine)FeIV[double bond, length as m-dash]O(Cl)]+ and [(bispidine)FeIV[double bond, length as m-dash]O(MeCN)]2+ (at least two orders of magnitude), while both oxido-iron(iv) complexes have very similar structures and an S = 1 electronic ground state, is presumably due to the large difference in the energy gap between the triplet and quintet electronic states. In presence of cyclohexane as substrate, [(bispidine)FeIV[double bond, length as m-dash]O(Cl)]+ oxidizes cyclohexane with a rate that is approx. 25 times faster than the self-decay of the oxidant, and selectively leads to chlorocyclohexane in moderate yield. The S = 1 electronic ground state of [(bispidine)FeIV[double bond, length as m-dash]O(Cl)]+ and a relatively low gap to the S = 2 state (approx. 6 kJ mol-1vs. approx. 75 kJ mol-1 for [(bispidine)FeIV[double bond, length as m-dash]O(MeCN)]2+) is also predicted by DLPNO-CCSD(T) quantum-chemical calculations. The method used was benchmarked with a set of six ferryl complexes with experimentally known electronic ground states.

Nonheme Iron-Oxo-Catalyzed Methane Formation from Methyl Thioethers: Scope, Mechanism, and Relevance for Natural Systems.

A range of nonheme oxo-iron(IV) model systems with tetra- or pentadentate ligands is shown to produce methane from methionine and other thioethers. This model reaction for the natural aerobic production of methane is shown to proceed via two sulfoxidation steps involving the oxo-iron(IV) complexes, with a bifurcation in the second step that either produces the sulfone or leads to demethylation with similar probabilities. In the presence of O2 , the resulting methyl radicals produce methanol and formate or, in an O2 -depleted environment, lead to formation of methane.

A Bispidine Iron(IV)-Oxo Complex in the Entatic State.

For a series of Fe(IV) =O complexes with tetra- and pentadentate bispidine ligands, the correlation of their redox potentials with reactivity, involving a variety of substrates for alkane hydroxylation (HAT), alkene epoxidation, and phosphine and thioether oxidation (OAT) are reported. The redox potentials span approximately 350 mV and the reaction rates over 8 orders of magnitude. From the experimental data and in comparison with published studies it emerges that electron transfer and the driving force are of major importance, and this is also supported by the DFT-based computational analysis. The striking difference of reactivity of two isomeric systems with pentadentate bispidines is found to be due to a destabilization of the S=1 ground state of one of the ferryl isomers, and this is supported by the experimentally determined redox potentials and published stability constants with a series of first-row transition metal ions with these two isomeric ligands.

Influence of ligand architecture on oxidation reactions by high-valent nonheme manganese oxo complexes using water as a source of oxygen.

Mononuclear nonheme Mn(IV)=O complexes with two isomers of a bispidine ligand have been synthesized and characterized by various spectroscopies and density functional theory (DFT). The Mn(IV)=O complexes show reactivity in oxidation reactions (hydrogen-atom abstraction and sulfoxidation). Interestingly, one of the isomers (L(1) ) is significantly more reactive than the other (L(2) ), while in the corresponding Fe(IV)=O based oxidation reactions the L(2) -based system was previously found to be more reactive than the L(1) -based catalyst. This inversion of reactivities is discussed on the basis of DFT and molecular mechanics (MM) model calculations, which indicate that the order of reactivities are primarily due to a switch of reaction channels (σ versus π) and concomitant steric effects.

Spin-state ordering in hydroxo-bridged diiron(III)bisporphyrin complexes.

We report the synthesis, structure, and spectroscopic characterization of 1,2-bis[µ-hydroxo iron(III) 5-(2,3,7,8,12,13,17,18-octaethylporphyrinyl)]ethane with PF6(­) and SbF6(­) counteranions. The two iron centers are nonequivalent with admixed intermediate spin state (S = 3/2 with a minor contribution of S = 5/2) on each metal both in the solid and in solution. The molecules are compared with previously known µ-hydroxo complexes with other counterions, such as I3(­), BF4(­), and ClO4(­), which demonstrates that the nature of the counterion can affect the spin-state ordering dramatically. To understand how the spin-state ordering is affected by external perturbations, we also have done a comprehensive computational study. The calculations show that subtle environmental perturbations affect the spin-state ordering and relative energies and are likely to be the root cause of the variation in spin-state ordering observed experimentally.

Computational modelling of oxygenation processes in enzymes and biomimetic model complexes.

With computational resources becoming more efficient and more powerful and at the same time cheaper, computational methods have become more and more popular for studies on biochemical and biomimetic systems. Although large efforts from the scientific community have gone into exploring the possibilities of computational methods for studies on large biochemical systems, such studies are not without pitfalls and often cannot be routinely done but require expert execution. In this review we summarize and highlight advances in computational methodology and its application to enzymatic and biomimetic model complexes. In particular, we emphasize on topical and state-of-the-art methodologies that are able to either reproduce experimental findings, e.g., spectroscopic parameters and rate constants, accurately or make predictions of short-lived intermediates and fast reaction processes in nature. Moreover, we give examples of processes where certain computational methods dramatically fail.

The computation of lipophilicities of 64Cu PET systems based on a novel approach for fluctuating charges.

A QSPR scheme for the computation of lipophilicities of 64Cu complexes was developed with a training set of 24 tetraazamacrocylic and bispidine-based Cu(II) compounds and their experimentally available 1-octanol-water distribution coefficients. A minimum number of physically meaningful parameters were used in the scheme, and these are primarily based on data available from molecular mechanics calculations, using an established force field for Cu(II) complexes and a recently developed scheme for the calculation of fluctuating atomic charges. The developed model was also applied to an independent validation set and was found to accurately predict distribution coefficients of potential 64Cu PET (positron emission tomography) systems. A possible next step would be the development of a QSAR-based biodistribution model to track the uptake of imaging agents in different organs and tissues of the body. It is expected that such simple, empirical models of lipophilicity and biodistribution will be very useful in the design and virtual screening of positron emission tomography (PET) imaging agents.

An efficient fluctuating charge model for transition metal complexes.

A fluctuating charge model for transition metal complexes, based on the Hirshfeld partitioning scheme, spectroscopic energy data from the NIST Atomic Spectroscopy Database and the electronegativity equalization approach, has been developed and parameterized for organic ligands and their high- and low-spin Fe(II) and Fe(III), low-spin Co(III) and Cu(II) complexes, using atom types defined in the Momec force field. Based on large training sets comprising a variety of transition metal complexes, a general parameter set has been developed and independently validated which allows the efficient computation of geometry-dependent charge distributions in the field of transition metal coordination compounds.

Ensemble and single-molecule studies on fluorescence quenching in transition metal bipyridine-complexes.

Beyond their use in analytical chemistry fluorescent probes continuously gain importance because of recent applications of single-molecule fluorescence spectroscopy to monitor elementary reaction steps. In this context, we characterized quenching of a fluorescent probe by different metal ions with fluorescence spectroscopy in the bulk and at the single-molecule level. We apply a quantitative model to explain deviations from existing standard models for fluorescence quenching. The model is based on a reversible transition from a bright to a dim state upon binding of the metal ion. We use the model to estimate the stability constants of complexes with different metal ions and the change of the relative quantum yield of different reporter dye labels. We found ensemble data to agree widely with results from single-molecule experiments. Our data indicates a mechanism involving close molecular contact of dye and quenching moiety which we also found in molecular dynamics simulations. We close the manuscript with a discussion of possible mechanisms based on Förster distances and electrochemical potentials which renders photo-induced electron transfer to be more likely than Förster resonance energy transfer.

Structure, bonding, and catecholase mechanism of copper bispidine complexes.

Oxygen activation by copper(I) complexes with tetra- or pentadentate mono- or dinucleating bispidine ligands is known to lead to unusually stable end-on-[{(bispidine)Cu}(2)(O(2))](2+) complexes (bispidines are methyl-2,4-bis(2-pyridin-yl)-3,7-diazabicyclo-[3.3.1]-nonane-9-diol-1,5-dicarboxylates); catecholase activity of these dinuclear Cu(II/I) systems has been demonstrated experimentally, and the mechanism has been thoroughly analyzed. The present density functional theory (DFT) based study provides an analysis of the electronic structure and catalytic activity of [{(bispidine)Cu}(2)(O(2))](2+). As a result of the unique square pyramidal coordination geometry, the d(x(2)-y(2)) ground state leads to an unusual σ/π bonding pattern, responsible for the stability of the peroxo complex and the observed catecholase activity with a unique mechanistic pathway. The oxidation of catechol to ortho-quinone (one molecule per catalytic cycle and concomitant formation of one equivalent of H(2)O(2)) is shown to occur via an associative, stepwise pathway. The unusual stability of the end-on-peroxo-dicopper(II) complex and isomerization to copper(II) complexes with chelating catecholate ligands, which inhibit the catalytic cycle, are shown to be responsible for an only moderate catalytic activity.

Calculation of exchange coupling constants of transition metal complexes with DFT.

A broken-symmetry method for the calculation of exchange coupling constants from DFT calculations, using the Heisenberg-Dirac-van Vleck spin Hamiltonian, has been validated for a dinuclear copper(II) complex. Hybrid functionals in combination with a large basis set on the metal centers and their first coordination sphere, and a smaller basis set on the ligand backbone are shown to be efficient and acceptable with respect to the computational cost and precision in comparison with experimental data. This method was thoroughly tested with a series of oligonuclear transition metal complexes with Cr(III), Cu(II), Fe(III), Mn(II), Mn(III), Mn(IV), Ni(II), and V(IV) as magnetic centers. The computed values of J are within approximately 50 cm(-1) of the experimental values for most of the examples; with combined basis sets, there generally is a similar accuracy to that obtained with a large basis set for the entire spin cluster but with significantly reduced computational expense. When the experimentally observed structural data are refined prior to the calculation of the exchange coupling constants, the computed values of J are in most cases in slightly better agreement with the experimental data than those obtained from single point calculations based on the X-ray data.

Trinuclear {M1}CN{M2}2 complexes (M1 = Cr(III), Fe(III), Co(III); M2 = Cu(II), Ni(II), Mn(II)). Are single molecule magnets predictable?

The reaction of the hexacyanometalates K3[M(1)(CN)6] (M(1) = Cr(III), Fe(III), Co(III)) with the bispidine complexes [M(2)(L(1))(X)](n+) and [M(2)(L(2))(X)](n+) (M(2) = Mn(II), Ni(II), Cu(II); L(1) = 3-methyl-9-oxo-2,4-di-(2-pyridyl)-7-(2-pyridylmethyl)-3,7-diazabicyclo[3.3.1]nonane-1,5-dicarboxylic acid dimethyl ester; L(2) = 3-methyl-9-oxo-7-(2-pyridylmethyl)-2,4-di-(2-quinolyl)-3,7-diazabicyclo[3.3.1]nonane-1,5-dicarboxylic acid dimethyl ester; X = anion or solvent) in water-methanol mixtures affords trinuclear complexes with cis- or trans-arrangement of the bispidine-capped divalent metal centers around the hexacyanometalate. X-ray structural analyses of five members of this family of complexes (cis-Fe[CuL(2)]2, trans-Fe[CuL(1)]2, cis-Co[CuL(2)]2, trans-Cr[MnL(1)]2, trans-Fe[MnL(1)]2) and the magnetic data of the entire series are reported. The magnetic data of the cyanide bridged, ferromagnetically coupled cis- and trans-Fe[ML]2 compounds (M = Ni(II), Cu(II)) with S = 3/2 (Cu(II)) and S = 5/2 (Ni(II)) ground states are analyzed with an extended Heisenberg Hamiltonian which accounts for anisotropy and zero-field splitting, and the data of the Cu(II) systems, for which structures are available, are thoroughly analyzed in terms of an orbital-dependent Heisenberg Hamiltonian, in which both spin-orbit coupling and low-symmetry ligand fields are taken into account. It is shown that the absence of single-molecule magnetic behavior in all spin clusters reported here is due to a large angular distortion of the [Fe(CN)6](3-) center and the concomitant quenching of orbital angular momentum of the Fe(III) ((2)T2g) ground state.

CopperII-mediated aromatic ortho-hydroxylation: a hybrid DFT and AB initio exploration.

Mechanistic pathways for the aromatic hydroxylation by [CuII(L1)(TMAO)(O)](-) (L1=hippuric acid, TMAO=trimethylamine N-oxide), derived from the O--N bond homolysis of its [CuII(L1)(TMAO)2] precursor, were explored by using hybrid density functional theory (B3LYP) and highly correlated ab initio methods (QCISD and CCSD). Published experimental studies suggest that the catalytic reaction is triggered by a terminal copper-oxo species, and a detailed study of electronic structures, bonding, and energetics of the corresponding electromers is presented. Two pathways, a stepwise and a concerted reaction, were considered for the hydroxylation process. The results reveal a clear preference for the concerted pathway, in which the terminal oxygen atom directly attacks the carbon atom of the benzene ring, leading to the ortho-selectively hydroxylated product. Solvent effects were probed by using the PCM and CPCM solvation models, and the PCM model was found to perform better in the present case. Excellent agreement between the experimental and computational results was found, in particular also for changes in reactivity with derivatives of L1.

DFT models for copper(II) bispidine complexes: structures, stabilities, isomerism, spin distribution, and spectroscopy.

Various DFT and ab initio methods, including B3LYP, HF, SORCI, and LF-density functional theory (DFT), are used to compute the structures, relative stabilities, spin density distributions, and spectroscopic properties (electronic and EPR) of the two possible isomers of the copper(II) complexes with derivatives of a rigid tetradentate bispidine ligand with two pyridine and two tertiary amine donors, and a chloride ion. The description of the bonding (covalency of the copper-ligand interactions) and the distribution of the unpaired electron strongly depend on the DFT functional used, specifically on the nonlocal DF correlation and the HF exchange. Various methods may be used to optimize the DFT method. Unfortunately, it appears that there is no general method for the accurate computation of copper(II) complexes, and the choice of method depends on the type of ligands and the structural type of the chromophore. Also, it appears that the choice of method strongly depends on the problem to be solved. LF-DFT and spectroscopically oriented CI methods (SORCI), provided a large enough reference space is chosen, yield accurate spectroscopic parameters; EDA may lead to a good understanding of relative stabilities; accurate spin density distributions are obtained by modification of the nuclear charge on copper; solvation models are needed for the accurate prediction of isomer distributions.

Coordination chemistry of a new rigid, hexadentate bispidine-based bis(amine)tetrakis(pyridine) ligand.

The hexadentate bispidine-based ligand 2,4-bis(2-pyridyl)-3,7-bis(2-methylenepyridine)-3,7-diazabicyclo[3.3.1]nonane-9-on-1,5-bis(carbonic acid methyl ester), L(6m), with four pyridine and two tertiary amine donors, based on a very rigid diazaadamantane-derived backbone, is coordinated to a range of metal ions. On the basis of experimental and computed structural data, the ligand is predicted to form very stable complexes. Force field calculations indicate that short metal-donor distances lead to a buildup of strain in the ligand; that is, the coordination of large metal ions is preferred. This is confirmed by experimentally determined stability constants, which indicate that, in general, stabilities comparable to those with macrocyclic ligands are obtained with the relative order Cu(2+) > Zn(2+) >> Ni(2+) < Co(2+), which is not the typical Irving-Williams behavior. The preference for large M-N distances also emerges from relatively high redox potentials (the higher oxidation states, that is, the smaller metal ions, are destabilized) and from relatively weak ligand fields (dd-transition, high-spin electronic ground states). The potentiometric titrations confirm the efficient encapsulation of the metal ions since only 1:1 complexes are observed, and, over a large pH range, ML is generally the only species present in solution.

Local molecular properties and their use in predicting reactivity.

Expressions for the local electron affinity, electronegativity and hardness are derived in analogy to the local ionization energy introduced by Sjoberg, Murray and Politzer. The local polarizability is also defined based on an additive atomic orbital polarizability model that uses Rivail's variational technique. The characteristics of these local properties at molecular surfaces and their relevance to electrophilic aromatic substitution, to S(N)2 reactivity and to the nucleophilicity of enolate ions are discussed.

AM1* parameters for phosphorus, sulfur and chlorine.

An extension of the AM1 semiempirical molecular orbital technique, AM1*, is introduced. AM1* uses AM1 parameters and theory unchanged for the elements H, C, N, O and F. The elements P, S and Cl have been reparameterized using an additional set of d orbitals in the basis set and with two-center core-core parameters, rather than the Gaussian functions used to modify the core-core potential in AM1. Voityuk and Rösch's AM1(d) parameters have been adopted unchanged for AM1* with the exception that new core-core parameters are defined for Mo-P, Mo-S and Mo-Cl interactions. Thus, AM1* gives identical results to AM1 for compounds with only H, C, N, O, and F, AM1(d) for compounds containing Mo, H, C, N, O and F only, but differs for molybdenum compounds containing P, S or Cl. The performance and typical errors of AM1* are discussed.