Abstract
The potential energy surfaces in gas phase and in aqueous solution for the nitration of benzene, chlorobenzene, and phenol have been elucidated with density functional theory at the M06-2X/6-311G(d,p) level combined with the polarizable continuum solvent model (PCM). Three reaction intermediates have been identified along both surfaces: the unoriented π-complex (I), the oriented reaction complex (II), and the σ-complex (III). In order to obtain quantitatively reliable results for positional selectivity and for modeling the expulsion of the proton, it is crucial to take solvent effects into consideration. The results are in agreement with Olah's conclusion from over 40 years ago that the transition state leading to (II) is the rate-determining step in activated cases, while it is the one leading to (III) for deactivated cases. The simplified reactivity approach of using the free energy for the formation of (III) as a model of the rate-determining transition state has previously been shown to be very successful for halogenations, but problematic for nitrations. These observations are rationalized with the geometric and energetic resemblance, and lack of resemblance respectively, between (III) and the corresponding rate determining transition state. At this level of theory, neither the σ-complex (III) nor the reaction complex (II) can be used to accurately model the rate-determining transition state for nitrations.