Amorphous Li-ion conductors are important solid-state electrolytes. However, Li transport in these systems is much less understood than for crystalline materials. We investigate amorphous LiPON electrolytes via ab initio molecular dynamics, providing atomistic-level insight into the mechanisms underlying the Li+ mobility. We find that the latter is strongly influenced by the chemistry and connectivity of phosphate polyanions near Li+. Amorphization generates edge-sharing polyhedral connections between Li­(O,N)4 and P­(O,N)4, and creates under- and overcoordinated Li sites, which destabilizes the Li+ and enhances their mobility. N substitution for O favors conductivity in two ways: (1) excess Li accompanying 1­(N):1­(O) substitutions introduces extra carriers; (2) energetically favored N-bridging substitutions condense phosphate units and densify the structure, which, counterintuitively, corresponds to higher Li+ mobility. Finally, bridging N is not only less electronegative than O but also engaged in strong covalent bonds with P. This weakens interactions with neighboring Li+ smoothing the way for their migration. When condensation of PO4 polyhedra leads to the formation of isolated O anions, the Li+ mobility is reduced, highlighting the importance of oxygen partial pressure control during synthesis. This detailed understanding of the structural mechanisms affecting Li+ mobility is the key for optimizing the conductivity of LiPON and other amorphous Li-ion conductors.

Lithium phosphorus oxynitride, also known as Lipon, solid-state electrolytes are at the center of the search for solid-state Li metal batteries. Key to the performance of Lipon is a combination of high Li content, amorphous character, and the incorporation of N into the structure. Despite the material's importance, our work presents the first study to fully resolve the structure of Lipon using a combination of  ab initio molecular dynamics, density functional theory, neutron scattering, and infrared spectroscopy. The modeled and experimental results have exceptional agreement in both neutron pair distribution function and infrared spectroscopy. Building on this synergy, the structural models show that N forms both bridges between two phosphate units and nonbridging apical N. We further show that as the Li content is increased the ratio of bridging to apical N shifts from being predominantly bridging at Li contents around 2.5:1 Li:P to only apical N at higher Li contents of 3.38:1 Li:P. This crossover from bridging to apical N appears to directly correlate with and explain both the increase in ionic conductivity with the incorporation of N and the ionic conductivity trends found in the literature.

CRYSTAL is a periodic ab initio code that uses a Gaussian-type basis set to express crystalline orbitals (i.e. Bloch functions). The use of atom-centred basis functions allows treating 3D (crystals), 2D (slabs), 1D (polymers) as well as 0D (molecules) systems on the same grounds. In turn, all-electron calculations are inherently permitted along with pseudopotential strategies. A variety of density functionals is implemented, including global and range-separated hybrids of various nature and, as an extreme case, Hartree-Fock (HF). The cost for HF or hybrids is only about 3-5 times larger than when using the local density approximation (LDA) or the generalized gradient approximation (GGA). Symmetry is fully exploited at all steps of the calculation. Many tools are available to modify the structure as given in input and simplify the construction of complicated objects, such as slabs, nanotubes, molecules, clusters. Many tensorial properties can be evaluated by using a single input keyword: elastic, piezoelectric, photoelastic, dielectric, as well as first and second hyperpolarizabilies, etc. The calculation of infrared and Raman spectra is available, and the intensities are computed analytically. Automated tools are available for the generation of the relevant configurations of solid solutions and/or disordered systems. Three versions of the code exist, serial, parallel and massive-parallel. In the second one the most relevant matrices are duplicated on each core, whereas in the third one the Fock matrix is distributed for diagonalization. All the relevant vectors are dynamically allocated and deallocated after use, making the code very agile. CRYSTAL can be used efficiently on high performance computing machines up to thousands of cores.