This thesis

This thesis is about the study of chemical reactions in water without getting our hands wet. Solvent effects can have a very large influence on the thermodynamics and the kinetics of chemical reactions in water, but are usually poorly described or even neglected completely in theoretical models. Moreover, the understanding of solvation shell dynamics and solvent effects on chemistry on a microscopic level is in general still rather poor. Having said that, we will continue in the next chapter with the introduction of the techniques applied for our study. In particular, A) the Car-Parrinello technique which efficiently solves the problem of following the motions of the nuclei while simultaneously updating the electronic structure allowing for the simulation of chemical reactions in the condensed phase and B) the statistical thermodynamics techniques to solve the ``rare event'' problem of observing the crossing of a chemical reaction barrier whose height is several order of kT with the Car-Parrinello molecular dynamics technique. In chapter 3, we start with the investigation of the solvent effects on a textbook S$_\mathrm{N}$2 reaction. Chemically, this type of reaction is very well understood: the solvent effects are very large, leading to a decrease of the reaction rate of 13 orders of magnitude (!) in aqueous solution with respect to this reaction in the gas phase. We show that we can accurately calculate the solvent-solute interactions and compute the solvent effects on the reaction free energy barrier in aqueous solution. Moreover, we rationalize the strong solvent effects from the structural changes in the solvent during the reaction. After this prototype, yet very instructive, S$_\mathrm{N}$2 reaction, we move on to the study of transition metal catalyzed oxidation reactions in aqueous solution, a short introduction of which is given in chapter 4. In particular, we investigate the solvent effects on the catalyzed activation of hydrogen peroxide by iron(II) in chapter 5 and by iron(III) in chapter 6, also known as Fenton and Fenton-like chemistry, respectively. We show that in these cases the solvent molecules are actively involved in the reaction mechanisms, catalyzing the chemical transformations on the metal complexes via hydrogen and proton abstraction reactions. Spontaneous hydrolysis of the acidic metal complexes is seen to play an important role, again a process which cannot be observed without an accurate incorporation of the aqueous environment. We elucidate the surprisingly different reaction mechanisms for Fenton and Fenton-like chemistry in water and show that in the Fe$^\mathrm{II}$/H$_2$O$_2$ case the much contested [Fe$^\mathrm{IV}$O]$^{2+}$ ion is the predominant oxidative intermediate and not the hydroxyl radical as is generally believed. Instead, in the Fe$^\mathrm{III}$/H$_2$O$_2$ case, an H$^+$ is donated to the solvent to produce the iron(III)hydroperoxo species as the primary intermediate. In these chapters (chapter 5 and 6), we present a number of illuminating and illustrative simulations of the dynamics of the Fenton and the Fenton-like reactions in aqueous solution at room temperature. However, concern is in place with the representativeness of such reactive trajectories, in view of the unphysical geometric constraints applied in the construction of the systems in order to overcome the earlier mentioned rare event problem. Therefore, chapter 7 is devoted to the generation of representative reactive trajectories with no memory of the artificial construction of the system by way of the new transition path sampling technique. In chapter 8, we study the oxidation of methane to methanol by the ferryl ion ([Fe$^\mathrm{IV}$O]$^{2+}$), i.e. the active intermediate proposed for the Fenton reaction in chapter 5 and find that again the water environment has a large effect on the chemistry via the hydrophobic interactions with the hydrocarbon species and the solvated metal oxo and hydroxo complexes. Also, we confirm the mechanistic similarities between the organic oxidations by Fenton's reagent and those by such enzymes as MMO (methane mono-oxygenase) and cytochrome P450, as was recently suggested in literature. This thesis ends with a summary (also in Dutch) and acknowledgments to the people who have contributed to this work.