We have carried out ab initio electronic structure calculations on a portion of DNA, the results of which provide support for a mechanism that produces single-strand breaks (SSBs) with low-energy electrons. This mechanism involves attaching a low-energy electron (ca. 1 eV) to a π* orbital of a DNA base to form a shape-resonance state. This π* anion then undergoes a sugar-phosphate C-O bond rupture over a small barrier to produce SSBs. In addition to supporting the efficacy of such a mechanism, our results suggest that solvation plays a crucial role in the rate of SSB formation when such very short-lived shape resonances are involved. In particular, they suggest that either the π* anion must be rendered electronically stable by solvation or its detachment lifetime must be several orders of magnitude longer in the solvated species than in the nonsolvated species.

We demonstrate that a simple Coulomb-energy model can be used to predict the vertical electron detachment energy of an anion of charge (−n) given the detachment energy of the corresponding anion having one less charge (−n + 1). This model was applied earlier by other workers to dianions in which the two charged sites are quite distant. In this paper we show that it can also be applied to more spatially compact species as long as the two orbitals from which the electrons are removed are sufficiently noninteracting. We first demonstrate how to use this model by applying it to a series of electronically stable dianions (MgF4 2-, BeF4 2-, TeF8 2-, SeF8 2-, and TeCl8 2-) for which the (−2) to (−1) and (−1) to (neutral) electron detachment energies have been evaluated using conventional ab initio methods. These test calculations allow us to assess the predictive accuracy of the Coulomb model. We then extend the model's use to predict the energies of dianions and trianions that are not electronically stable (SO4 2-, CO3 2-, PO4 3-, and PO4 2-) and for which application of conventional quantum chemistry methods will not yield reliable predictions. That is, we predict at what energies metastable resonance states of these species will occur. Finally, we use the Coulomb nature of the long-range part of the electron−anion potential to estimate the lifetimes of these resonance states with respect to electron loss.

No abstract

It is known that SO 4 2Ϫ is not electronically stable as an isolated species but can be rendered stable by solvation ͑e.g., by adding a few H 2 O molecules͒. Recently, our group introduced a Coulomb repulsion model that offers an approximation to the energy instability and lifetimes of such species. In order to achieve an independent and likely more reliable estimate of the instability of SO 4 2Ϫ , we have undertaken a follow-up study of this dianion. Specifically, we apply a stabilization method to determine the vertical electronic energy difference between the metastable SO 4 2Ϫ dianion and its SO 4Ϫ1 daughter at several levels of theory. The particular variant of the stabilization method used here involves adding a partial positive charge to the central sulfur nucleus in order to confine the escaping electron. Our coupled-cluster data, which represent our highest level of theory, suggest that SO 4 2Ϫ is unstable by 1.1 eV and has a lifetime with respect to electron loss of 1.6ϫ10 Ϫ10 s ͑our earlier estimates were 0.75 eV and 2.7ϫ10 Ϫ8 s͒.

The possibility of electron binding to the complex (H3BNH3) was studied at the coupled cluster level of theory with single, double, and noniterative triple excitations. The staggered conformation (minimum), with a dipole moment of 5.3 Debye, binds an electron by 984 cm−1, whereas the eclipsed conformer (saddle point) possesses a larger dipole moment (5.5 Debye) and binds an electron by 1014 cm−1. The neutral parent of the (H3BNH3)− anion involves a dative bond that is responsible for a significant polarization of the neutral species and generates a significant dipole moment.