Calculation theory, molecular dynamics simulation

The theory was very similar to that described earlier, but was simplified in view of the complexity of the problem. A number of reaction intermediates were considered explicitly, and the corresponding signals were calculated by molecular dynamics simulation. Kinetic equations governing the reaction sequence were established and were solved numerically. The main simplification of the theory is that, when calculating A5[r, r], the lower limit of the Fourier integral was shifted from 0 to a small value q. The authors wrote [59]... 

The theory was very similar to that described earlier but was simplified in view of the complexity of the problem. A number of reaction intermediates were considered explicitly, and the corresponding signals were calculated by molecular dynamics simulation. Kinetic equations governing the reaction... 

A promising method, developed in recent years, is the use of first principles molecular dynamics as exemplified by the Car-Parrinello technique (8]. In these calculations the interatomic potentials are explicitly derived from the electronic ground-state within the density functional theory in local or non-local approximation. It combines quantum mechanical calculations with molecular dynamics simulations and, therefore, overcomes the limitations of both methods. Actual computers allow only simulations of aqueous solutions of about 60 water molecules for several ps (10 s). This limit is still at least one order of magnitude shorter than the fastest directly measured water exchange rate, k = 3.5 x 10 s for [Eu(H20)8], i.e. one exchange event every (8 x 3.5 x lO s ) = 36 ps [9]. Nevertheless, several publications appeared in the late 1990s on solvated Be [10], K+ [11] and Cu + [12] presenting mainly structural results. 

Characterization of the structure of the decomposition products of LiPFs-based electrolytes (Figure 17.7, bottom insert) along with the rates of formation have eventually led to the elucidation of the thermal decomposition mechanism [30] Campion et al. [30] reported that during the thermal decomposition of LiPFe-based electrolytes PF5 reacted rapidly with trace protic impurities in the electrolyte, such as water and alcohol, to form OPF3, which then initiated an auto-catalytic decomposition of the electrolyte (Figure 17.7). Density functional theory calculations and molecular dynamics simulations also supported the reactivity of PF5 toward carbonate solvents [31]. The auto-catalytic cycle relied upon the formation of the reactive Lewis acids PF5 and POF3. Several studies indicated that the addition of a Lewis base or a... 

Specific solute-solvent interactions involving the first solvation shell only can be treated in detail by discrete solvent models. The various approaches like point charge models, siipennoleciilar calculations, quantum theories of reactions in solution, and their implementations in Monte Carlo methods and molecular dynamics simulations like the Car-Parrinello method are discussed elsewhere in this encyclopedia. Here only some points will be briefly mentioned that seem of relevance for later sections. 

Parchment et al. [271] have provided more recent calculations on the 3-hydroxypyrazole equilibrium at the ab initio level. They noted that tautomer 9, which was not considered by Karelson et al. [268], is the lowest-energy tautomer in the gas phase at levels of theory (including AMI) up to MP4/6-31G //HF/3-21G [271], Although 8 is the dominant tautomer observed experimentally in aqueous solution, in the gas phase 8 is predicted to be nearly 9 kcal/mol less stable than 9 at the MP4 level [271], Using a DO model with an unphysically small cavity radius of 2.5 A, Parchment et al. [271] were able to reproduce at the ab initio level the AMI-DO prediction of Karelson et al. [268], namely that 8 is the most stable tautomer in aqueous solution. With this cavity, though, 8 is predicted to be better solvated than 9 by -22.2 kcal/mol [271], This result is inconsistent with molecular dynamics simulations with explicit aqueous solvation [271], and with PCM and SCME calculations with more reasonable cavities [271] these predict that 8 is only about 3 kcal/mol better solvated than 9. In summary, the most complete models used by Parchment et al. do not lead to agreement with experiment... 

Various theoretical methods and approaches have been used to model properties and reactivities of metalloporphyrins. They range from the early use of qualitative molecular orbital diagrams (24,25), linear combination of atomic orbitals to yield molecular orbitals (LCAO-MO) calculations (26-30), molecular mechanics (31,32) and semi-empirical methods (33-35), and self-consistent field method (SCF) calculations (36-43) to the methods commonly used nowadays (molecular dynamic simulations (31,44,45), density functional theory (DFT) (35,46-49), Moller-Plesset perturbation theory ( ) (50-53), configuration interaction (Cl) (35,42,54-56), coupled cluster (CC) (57,58), and CASSCF/CASPT2 (59-63)). 

Fig. 4. Decaniobate Nbi0O286 ion. Symmetrically equivalent oxygen atoms are shown in different colors. The niobium ions are labeled 1,2,3, as shown in the Figure. The oxygen atoms are then labeled with respect to the numbers on the niobium atoms, and whether they are bridging (p) or terminal (r ). Bond lengths are compared for density functional theory in continuum solvent versus model predictions as averaged over a molecular dynamics simulation in water. DFT-calculated bond lengths are above, and model predictions are below, given in angstroms.

By contrast, few such calculations have as yet been made for diffusional problems. Much more significantly, the experimental observables of rate coefficient or survival (recombination) probability can be measured very much less accurately than can energy levels. A detailed comparison of experimental observations and theoretical predictions must be restricted by the experimental accuracy attainable. This very limitation probably explains why no unambiguous experimental assignment of a many-body effect has yet been made in the field of reaction kinetics in solution, even over picosecond timescale. Necessarily, there are good reasons to anticipate their occurrence. At this stage, all that can be done is to estimate the importance of such effects and include them in an analysis of experimental results. Perhaps a comparison of theoretical calculations and Monte Carlo or molecular dynamics simulations would be the best that could be hoped for at this moment (rather like, though less satisfactory than, the current position in the development of statistical mechanical theories of liquids). Nevertheless, there remains a clear need for careful experiments, which may reveal such effects as discussed in the remainder of much of this volume. 

---