Conformation of tripod Metal Templates in CH₃C(CH₂PPh₂)₃ML_n (n = 2, 3): Neural Networks in Conformational Analysis

Abstract

The conformational space spanned by tripod metal templates CH3C(CH2Ph&M is analysed on the basis of the solid-state structures of 82 tripodCo templates in compounds tripodCoL, and tripodCoL3. Systematic analysis, including the techniques of conformational space group scatter graphs, principalcomponent analysis, and partial least squares, reveals a series of basic regularities: The torsion of the phenyl groups is strongly linked to the torsional skew of the bicyclooctanetype framework of the chelate cage. For one sense of this skew there are two classes of low-energy conformation that differ by the helicity of the phenyl arrangement and by the degree of torsional skew in the chelate backbone. From the scatter graphs it is evident that a change in helicity may ocworks / Factor analysis cur by one-or by two-ring flip mechanisms. The basic regularities found by the above methods are also evident from the analysis of the same data by a neural network approach. The fact that these regularities are found for tripodCoL, and tripodCoL3, irrespective of the widely different coligands L and crystal environments, means that the conformation of the tripod metal templates is governed by the forces acting within them and not so much by the forces imposed on them by their individual chemical or crystal environment. It is shown that the classifications, although derived from a data basis only containing Co compounds, are characteristic for tripod metal templates irrespective of the specific metal involved.In most cases specificity in coordination chemistry is dominated by a specific choice of the ligands surrounding the metal. Homogenous catalysisr'*'] and. even more so, the subset of enantioselective catalysis [7] are clear illustrations of this statement.Part of this specificity will arise from steric interactions. In order to be able to predict specificity a thorough knowledge of these steric interactions is therefore needed. One basis for acquiring such a knowledge is currently available in many cases: A considerable amount of information exists on solid-state structures (determined by X-ray analysis), most of it deposited in an orderly and easily retrievable way in internationally available databasesr41. The wealth of information contained in these data fles is, as yet, far from having been systematically exploited.The current paper deals with the systematic analysis of all such structural data describing the solid-state conformation of tripodCo templates [tripod = CH7C(CH'PPh2)7] for compounds tripodCoL, and tripodCoI+ The interest in this analysis stems from a research programme aiming at an understanding of the chemical reactivity of a subsel of ligand metal templates in which the ligand is a neopentanebased tripodal entity. The programme involves the three following steps: 1) synthesis of a library of neopentane-based tripod ligands RCH2C(CH2X)(CH2Y)(CH2Z) (X, Y, Z = donor groups), 2) understanding and modelling the shape of tripod metal templates tril,odM, 3) correlating the results Catalysis mediated by tripod metal t...

Structure Correlation and Chemistry

“…As far as structure correlation is concerned, there seems to be only one such application. It is part of a research program in stereospeci®c catalysis which involves: (i) the synthesis of a library of closely related metal complexes; (ii) measuring, understanding and modeling their shapes; and (iii) correlating shape to results of a catalytic reaction (Beyreuther et al, 1996). A set of 82 tripod Co II L 2 and tripod Co II L 3 complexes was retrieved from the Cambridge Structural Database [tripod CH 3 C(CH 2 P{C 6 H 5 } 2 ) 3 , see Fig.…”

Optimization and analysis of force field parameters by combination of genetic algorithms and neural networks

“…metal templates tripodM in the solid state are not largely influenced by the forces acting upon them within the crystal, and thus represent innermolecular forces. 9 This means that the conformations observed correspond to minima on the molecular energy hypersurface. A necessary condition for a force field description of these templates is therefore that the force field has to reproduce these conformations as at least local minima on the hypersurface generated by the force field.…”

How to Predict Activation Barriers – Conformational Transformations of Compounds CH3C(CH2PPh2)3–n[CH2P(oTol)2]nMo(CO)3 (n = 1–3): Force Field Calculations versus NMR Data

“…To generate the starting conformations, all three torsion angles of one conformer were set at the same value and two complete scans of φ values were made, one at τ ϭ 10°and the other at τ ϭ 31°. These starting values were selected because analyses of similar compounds [8] have shown the τ values to cumulate around these values. The search was only performed with positive starting values of τ owing to the fact that all conformations must exist in isoenergetic enantiomeric pairs such that for every conformation with a given set of τ and φ values, an inverted conformation exists where the τ and φ values each have opposite signs.…”

“…[10] The technique used to set these rotations at the values defined by the grid points has also been described previously. [10] The torsion of the chelate cage as described by τ, [8] defined as indicated in Figure 11, was set at a uniform starting value of 20°for all three "legs" of the ligand and it was found that the strain imposed on it by the forced rotation of the aryl groups is sufficiently strong to produce negative as well as positive torsions after refinement. The energies of the local minima obtained after refinement are distributed as shown in Figure 12a.…”