Many-body theory of positron binding and annihilation in polyatomic molecules

Whilst positron binding energies have been measured via resonant annihilation spectra for over ~80 molecules in the past two decades [1], an accurate ab initio theoretical approach has remained elusive. Measured and calculated binding energies exist together for only 6 molecules. For these, standard quantum chemistry approaches have proved deficient, agreeing with experiment to at best 25% accuracy. The theoretical difficulty lies in the need to accurately describe strong positron-molecule many-body correlations including polarisation, screening of the positron-molecule Coulomb interaction, and the unique process of virtual-positronium formation (where a molecular electron temporarily tunnels to the positron). Their specific role in positron-molecule binding has not yet been elucidated and is not well understood. Accurately accounting for the correlations is a formidable task even for positron-atom interactions, but molecules bring additional formidable computational challenges.

In this talk I will discuss an ab initio many-body theory approach to positron-molecule interactions developed in my group capable of calculating binding energies and (direct) annihilation in polyatomic molecules [2]. Specifically, we solve the Dyson equation for the positron quasiparticle wavefunction with self-energy calculated at the GW@BSE + virtual-Positronium + positron-hole T-matrix level, going much beyond current state-of-the-art. Delineating the effects of the distinct correlations, we show that the non-perturbative process of virtual-positronium formation is essential to support binding in non-polar molecules including CS$_2$ and CSe$_2$, and that it significantly enhances binding in organic polar molecules. Overall, we find a delicate interplay of the correlations, which results in binding energies in the best agreement with experiment to date, and in some cases within a few percent accuracy [2]. For non-binding molecules, we calculate the annihilation rate Zeff and find excellent agreement with experiment. 

[1] G. F. Gribakin, J. A. Young and C. M. Surko, Rev. Mod. Phys. 82, 2557 (2010).

[2] B. J. Cunningham, J. Horfierka, C. M. Rawlins, C. H. Patterson and D. G. Green, in preparation (2020).