Bonding, structure, and energetics of gaseous E8(2+) and of solid E8(AsF6)2 (E = S, Se)

The attempt to prepare hitherto unknown homopolyatomic cations of sulfur by the reaction of elemental sulfur with blue S8(AsF6)2 in liquid SO2/SO2ClF, led to red (in transmitted light) crystals identified crystallographically as S8(AsF6)2. The X-ray structure of this salt was redetermined with improved resolution and corrected for librational motion: monoclinic, space group P2(1)/c (No. 14), Z = 8, a = 14.986(2) A, b = 13.396(2) A, c = 16.351(2) A, beta = 108.12(1) degrees. The gas phase structures of E8(2+) and neutral E8 (E = S, Se) were examined by ab initio methods (B3PW91, MPW1PW91) leading to delta fH theta[S8(2+), g] = 2151 kJ/mol and delta fH theta[Se8(2+), g] = 2071 kJ/mol. The observed solid state structures of S8(2+) and Se8(2+) with the unusually long transannular bonds of 2.8-2.9 A were reproduced computationally for the first time, and the E8(2+) dications were shown to be unstable toward all stoichiometrically possible dissociation products En+ and/or E4(2+) [n = 2-7, exothermic by 21-207 kJ/mol (E = S), 6-151 kJ/mol (E = Se)]. Lattice potential energies of the hexafluoroarsenate salts of the latter cations were estimated showing that S8(AsF6)2 [Se8(AsF6)2] is lattice stabilized in the solid state relative to the corresponding AsF6- salts of the stoichiometrically possible dissociation products by at least 116 [204] kJ/mol. The fluoride ion affinity of AsF5(g) was calculated to be 430.5 +/- 5.5 kJ/mol [average B3PW91 and MPW1PW91 with the 6-311 + G(3df) basis set]. The experimental and calculated FT-Raman spectra of E8(AsF6)2 are in good agreement and show the presence of a cross ring vibration with an experimental (calculated, scaled) stretching frequency of 282 (292) cm-1 for S8(2+) and 130 (133) cm-1 for Se8(2+). An atoms in molecules analysis (AIM) of E8(2+) (E = S, Se) gave eight bond critical points between ring atoms and a ninth transannular (E3-E7) bond critical point, as well as three ring and one cage critical points. The cage bonding was supported by a natural bond orbital (NBO) analysis which showed, in addition to the E8 sigma-bonded framework, weak pi bonding around the ring as well as numerous other weak interactions, the strongest of which is the weak transannular E3-E7 [2.86 A (S8(2+), 2.91 A (Se8(2+)] bond. The positive charge is delocalized over all atoms, decreasing the Coulombic repulsion between positively charged atoms relative to that in the less stable S8-like exo-exo E8(2+) isomer. The overall geometry was accounted for by the Wade-Mingos rules, further supporting the case for cage bonding. The bonding in Te8(2+) is similar, but with a stronger transannular E3-E7 (E = Te) bonding. The bonding in E8(2+) (E = S, Se, Te) can also be understood in terms of a sigma-bonded E8 framework with additional bonding and charge delocalization occurring by a combination of transannular n pi *-n pi * (n = 3, 4, 5), and np2-->n sigma * bonding. The classically bonded S8(2+) (Se8(2+) dication containing a short transannular S(+)-S+ (Se(+)-Se+) bond of 2.20 (2.57) A is 29 (6) kJ/mol higher in energy than the observed structure in which the positive charge is delocalized over all eight chalcogen atoms.     

Aminooxazolinate; a chiral amidinate analogue

High levels of diastereoselection with respect to chirality-at-metal are achieved at equilibrium for complexes containing a new and available range of diazaallyl ligands.     

Osmium(II) and ruthenium(II) arene maltolato complexes: rapid hydrolysis and nucleobase binding

Density functional calculations show that aquation of [Os(eta6-arene)(XY)Cl]n+ complexes is more facile for complexes in which XY=an anionic O,O-chelated ligand compared to a neutral N,N-chelated ligand, and the mechanism more dissociative in character. The O,O-chelated XY=maltolato (mal) [M(eta6-p-cym)(mal)Cl] complexes, in which p-cym=p-cymene, M=OsII (1) and RuII (2), were synthesised and the X-ray crystal structures of 1 and 22 H2O determined. Their hydrolysis rates were rapid (too fast to follow by NMR spectroscopy). The aqua adduct of the OsII complex 1 was 1.6 pKa units more acidic than that of the RuII complex 2. Dynamic NMR studies suggested that O,O-chelate ring opening occurs on a millisecond timescale in coordinating proton-donor solvents, and loss of chelated mal in aqueous solution led to the formation of the hydroxo-bridged dimers [(eta6-p-cym)M(mu-OH)3M(eta6-p-cym)]+. The proportion of this dimer in solutions of the OsII complex 1 increased with dilution and it predominated at micromolar concentrations, even in the presence of 0.1 M NaCl (conditions close to those used for cytotoxicity testing). Although 9-ethylguanine (9-EtG) binds rapidly to Os(II) in 1 and more strongly (log K=4.4) than to RuII in 2 (log K=3.9), the OsII adduct [Os(eta6-p-cym)(mal)(9EtG)]+ was unstable with respect to formation of the hydroxo-bridged dimer at micromolar concentrations. Such insights into the aqueous solution chemistry of metal-arene complexes under biologically relevant conditions will aid the rational design of organometallic anticancer agents.     

Extending ligand field molecular mechanics to modelling organometallic π-bonded systems: applications to ruthenium-arenes

The ligand field molecular mechanics method has been extended to treat η(6)-arene ligands coordinated to a ruthenium(II) centre by employing a dummy atom located at the centroid of the arene ring and distributing the forces on the dummy to the arene carbon atoms. Angular overlap model parameters based on orbital energies derived from Kohn-Sham density functional theory (KS-DFT) calculations show that, relative to the Ru-dummy vector, the arene behaves as a very strong π donor and weak σ donor. Based on KS-DFT geometries, partial atomic charges and potential energy scans for a series of homoleptic and half sandwich complexes spanning arene, am(m)ine, imine, pyridyl, hydride and chloride ligands, a new LFMM force field has been developed which accurately reproduces the KS-DFT data. This FF was validated against 47 half-sandwich complexes obtained from the Cambridge Structural Database which, after minor corrections to account for the systematic errors between our chosen functional (BP86) and the experimental structures, yields a 'structurally tuned' FF where 93% of the Ru-L contacts are reproduced to 0.05 ? or better and all bar two bond lengths are within 0.1 ? of experiment. Over half the systems have non-hydrogen-atom rmsds of less than 0.5 ?. Larger differences are usually due to rotation of the arene moiety which is shown by ligand field molecular dynamics (LFMD) simulations to be an inherently low-energy process. Comparisons between LFMD and Car-Parrinello MD for [Ru(p-cymene)(ethylenediamine)Cl](+)show that LFMD is equally accurate but much faster enabling modelling of dynamic properties which occur on a timescale beyond the scope of CPMD.     

DommiMOE: an implementation of ligand field molecular mechanics in the molecular operating environment

The ligand field molecular mechanics (LFMM) model, which incorporates the ligand field stabilization energy (LFSE) directly into the potential energy expression of molecular mechanics (MM), has been implemented in the "chemically aware" molecular operating environment (MOE) software package. The new program, christened DommiMOE, is derived from our original in-house code that has been linked to MOE via its applications programming interface and a number of other routines written in MOE's native scientific vector language (SVL). DommiMOE automates the assignment of atom types and their associated parameters and popular force fields available in MOE such as MMFF94, AMBER, and CHARMM can be easily extended to provide a transition metal simulation capability. Some of the unique features of the LFMM are illustrated using MMFF94 and some simple [MCl)]2- and [Ni(NH3)n]2+ species. These studies also demonstrate how density functional theory calculations, especially on experimentally inaccessible systems, provide important data for designing improved LFMM parameters. DommiMOE treats Jahn-Teller distortions automatically, and can compute the relative energies of different spin states for Ni(II) complexes using a single set of LFMM parameters.     

Modeling of the various minima on the potential energy surface of bispidine copper(II) complexes: a further test for ligand field molecular mechanics

Copper(II) complexes of bispidines (bispidine = tetra-, penta-, or hexadentate ligand, based on the 3,7-diazabicyclo[3.3.1]nonane backbone) display several isomeric forms. Depending on the substitution pattern of the bispidine and the type of coligands used, the structure elongates along one of the three potential Jahn-Teller axes. In an effort to develop a computational tool which can predict which isomer is observed, 23 bispidine-copper(II) complexes with 19 different ligands are analyzed theoretically by ligand field molecular mechanics (LFMM). With two exceptions, the lowest-energy LFMM structure and the experimental solid-state structure agree concerning the Jahn-Teller axis. However, in most cases and especially for six-coordinate complexes, LFMM predicts a second local minimum within a few kilojoules per mole. Although detailed analysis reveals that the current force field is too "stiff", reasonable quantitative reproduction of the structural data is achieved with Cu-L bond length root mean square (rms) deviations for nine complexes of 0.05 A or less and with 20 reproduced to a rms deviation of 0.1 A or less. Across all of the complexes, the Cu-amine and Cu-pyridyl bond length rms deviations are 0.07 and 0.12 A, respectively.     

General molecular mechanics method for transition metal carboxylates and its application to the multiple coordination modes in mono- and dinuclear Mn(II) complexes

A general molecular mechanics method is presented for modeling the symmetric bidentate, asymmetric bidentate, and bridging modes of metal-carboxylates with a single parameter set by using a double-minimum M-O-C angle-bending potential. The method is implemented within the Molecular Operating Environment (MOE) with parameters based on the Merck molecular force field although, with suitable modifications, other MM packages and force fields could easily be used. Parameters for high-spin d (5) manganese(II) bound to carboxylate and water plus amine, pyridyl, imidazolyl, and pyrazolyl donors are developed based on 26 mononuclear and 29 dinuclear crystallographically characterized complexes. The average rmsd for Mn-L distances is 0.08 A, which is comparable to the experimental uncertainty required to cover multiple binding modes, and the average rmsd in heavy atom positions is around 0.5 A. In all cases, whatever binding mode is reported is also computed to be a stable local minimum. In addition, the structure-based parametrization implicitly captures the energetics and gives the same relative energies of symmetric and asymmetric coordination modes as density functional theory calculations in model and "real" complexes. Molecular dynamics simulations show that carboxylate rotation is favored over "flipping" while a stochastic search algorithm is described for randomly searching conformational space. The model reproduces Mn-Mn distances in dinuclear systems especially accurately, and this feature is employed to illustrate how MM calculations on models for the dimanganese active site of methionine aminopeptidase can help determine some of the details which may be missing from the experimental structure.     

Comprehensive molecular mechanics model for oxidized type I copper proteins: active site structures, strain energies, and entatic bulging

The ligand field molecular mechanics (LFMM) model has been applied to the oxidized Type 1 copper center. In conjunction with the AMBER94 force field implemented in DommiMOE, the ligand field extension of the molecular operating environment (MOE), LFMM parameters for Cu-N(imidazole), Cu-S(thiolate), Cu-S(thioether), and Cu-O(carbonyl) interactions were developed on the basis of experimental and theoretical data for homoleptic model systems. Subsequent LFMM optimizations of the active site model complex [Cu(imidazole)2(SMe)(SMe2]+ agree with high level quantum results both structurally and energetically. Stable trigonal and tetragonal structures are located with the latter about 1.5 kcal mol-1 lower in energy. Fully optimized unconstrained structures were computed for 24 complete proteins containing T1 centers spanning four-coordinate, plastocyanin-like CuN2SS' and stellacyanin-like CuN2SO sites, plus the five-coordinate CuN2SS'O sites of the azurins. The initial structures were based on PDB coordinates augmented by a 10 A layer of water molecules. Agreement between theory and experiment is well within the experimental uncertainties. Moreover, the LFMM results for plastocyanin (Pc), cucumber basic protein (CBP) and azurin (Az) are at least as good as previously reported QM/MM structures and are achieved several orders of magnitude faster. The LFMM calculations suggest the protein provides an entatic strain of about 10 kcal mol-1. However, when combined with the intrinsic 'plasticity' of d9 Cu(II), different starting protein/solvent configurations can have a significant effect on the final optimized structure. This 'entatic bulging' results in relatively large fluctuations in the calculated metal-ligand bond lengths. For example, simply on the basis of 25 different starting configurations of the solvent molecules, the optimized Cu-S(thiolate) bond lengths in Pc vary by 0.04 A while the Cu-S(thioether) distance spans over 0.3 A. These variations are the same order of magnitude as the differences often quoted to correlate the spectroscopic properties from a set of proteins. Isolated optimizations starting from PDB coordinates (or indeed, the PDB structures themselves) may only accidentally correlate with spectroscopic measurements. The present calculations support the work of Warshel who contends that adequate configurational averaging is necessary to make proper contact with experimental properties measured in solution. The LFMM is both sufficiently accurate and fast to make this practical.