M. Waroquier

Partial optimization is a useful technique to reduce the computational load in simulations of extended systems. In such nonequilibrium structures, the accurate calculation of localized vibrational modes can be troublesome, since the standard normal mode analysis becomes inappropriate. In a previous paper [ A. Ghysels et al., J. Chem. Phys. 126, 224102 (2007) ], the mobile block Hessian (MBH) approach was presented to deal with the vibrational analysis in partially optimized systems. In the MBH model, the nonoptimized regions of the system are represented by one or several blocks, which can move as rigid bodies with respect to the atoms of the optimized region. In this way unphysical imaginary frequencies are avoided and the translational/rotational invariance of the potential energy surface is fully respected. In this paper we focus on issues concerning the practical numerical implementation of the MBH model. The MBH normal mode equations are worked out for several coordinate choices. The introduction of a consistent group-theoretical notation facilitates the treatment of both the case of a single block and the case of multiple blocks. Special attention is paid to the formulation in terms of Cartesian variables, in order to provide a link with the standard output of common molecular modeling programs.

Theoretical calculations are presented on elementary reactions which are important during coke formation in a thermal cracking unit. This process is known to proceed through a free radical chain mechanism. The elementary reaction steps that lead to the growth of the coke surface can be divided into five classes of reversible reactions: hydrogen abstraction, substitution, gas phase olefin addition to radical surface species, gas phase radical addition to olefinic bonds and cyclization. To identify the elementary reaction classes that determine the coking rate, all microscopic routes that start from benzene and lead to naphthalene have been investigated. It is found that initial creation of surface radicals, either by hydrogen abstraction or substitution and subsequent hydrogen abstractions, determines the global coking rate. The influence of the local polyaromatic structure on the kinetics of the hydrogen abstraction reactions is determined by performing calculations on a large set of polyaromatic hydrocarbons (PAHs). On basis of the BDE values six types of possible reactive sites at the coke surface can be distinguished. For the initial hydrogen abstraction the local polyaromatic structure strongly influences the reaction kinetics and abstraction is preferred from less congested sites of the polyaromatic.

A novel force-field development strategy is proposed that tackles the well-known difficulty of parameter correlations arising in a conventional least-squares optimization. In the first step of the new gradient curves method (GCM), continuity criteria are imposed to transform the raw multidimensional ab initio training data to distinct sets of one-dimensional data, each associated with an individual energy term. In the second step, the transformed data suggest suitable analytical expressions, and the parameters in these expressions are fitted to the transformed data; that is, one does not have to postulate a priori analytical expressions for the force-field energy terms. This approach facilitates the derivation of valence terms. Benchmarks have been performed on a set of small molecules. The results show that the new method yields physically acceptable energy terms exactly when a conventional parametrization would suffer from parameter correlations, that is, when an increasing number of redundant internal coordinates is used in the force-field model. The generic treatment of parameter correlations in the proposed method facilitates an intuitive physical interpretation of the individual terms in the force-field expression, which is a prerequisite for the transferability of force-field models.

Hydrogen abstraction reactions at polyaromatic hydrocarbons by a methyl radical are investigated from the viewpoint of DFT-based reactivity descriptors. The BMK functional succeeds in accurately reproducing experimental data for the global indicators. All species are found to be soft. The local HSAB principle shows an overall good qualitative agreement with kinetic barriers, and the local softness is successful for describing the general reactivity trends. However, the indicators do not succeed in predicting the particularly high barriers encountered in some abstraction reactions, as these barriers are mainly caused by steric hindrance effects in the transition structures.

1-Arylmethyl-2-(cyanomethyl)aziridines were transformed into 4-(N,N-bis(arylmethyl)amino)-3-(pyrrolidin-1-yl)butanenitriles and 4-(N,N-bis(arylmethyl)amino)-2-butenenitriles via 4-(N,N-bis(arylmethyl)amino)-3-bromobutanenitriles in high yields and purity. The key steps involve the unprecedented regiospecific ring opening of intermediate 2-(cyanomethyl)aziridinium salts by bromide and pyrrolidine in acetonitrile, exclusively at the substituted aziridine carbon atom. The results were rationalized on the basis of ab initio calculations.