Creating Interfaces by Stretching the Solvent Is Key to Metallic Ammonia Solutions

Abstract

Although the phenomenon of electron solvation has been known for at least one and a half centuries, [1,2] explanations of the detailed behavior of solvated electrons in molecular liquids, in particular in metal/ammonia solutions, [1][2][3] have remained controversial. At low concentrations, metal/ammonia solutions behave like electrolytes whereas they undergo a transition to a metallic liquid-heralded by changing their color from blue to bronze-on increasing the metal content. [2,3] Also, adding metal lowers the density of the solution drastically compared to the neat liquid. This is thought to lead to the formation of voids hosting the unbound electrons, [4,5] which percolate at sufficiently high metal content. [6,7] Distinct from this viewpoint is a more chemical perspective [8,9] whereby the solvated electron in ammonia is believed to be some sort of open-shell molecular species and, in the limiting case, even an anionic radical as such. However, with increasing concentration close to saturation the influence of the metal ion cores, particularly their solvation and ion-pair formation, can no longer be neglected. Indeed, recent diffraction data of Li/NH 3 solutions up to 21 mol % metal (21 MPM) clearly demonstrate that they are highly structured over both the short and intermediate length scales.[10] In particular, the studies suggest that these solutions can be constructed from two major structural units: tightly bound [Li(NH 3 ) 4 ] + tetrahedra and empty voids. [10] In addition, inelastic X-ray scattering close to the saturation limit uncovered not only plasmon resonances and strong deviations from the Jellium model of simple metals, but also excitations that are associated with vibrations of these quasimolecular complexes.[11] Finally, the protonic self-diffusion coefficient is nonmonotonic in the liquid state; it first increases with metal concentration before decreasing again on reaching saturation, an observation that is explained in terms of competing influences of electron and ion solvation.[12] Together these novel findings raise the question of how the intricate realspace structure and the delocalized electronic structure are interrelated in the metallic liquid.Unfortunately, known density functionals do not provide the correct structure and energetics of the ammonia-ammonia interactions in that they predict an essentially linear hydrogen bond for the dimer, similar to classical force field models.[13] To cure this deficiency in the description of the solvent the HCTH/407 + functional [13] has been designed for the simulation of ammonia systems. By using HCTH/407 + , a 21-MPM lithium/ammonia solution at 230 K was simulated in terms of 6 Li and 23 NH 3 in a periodic box of 11.314 by diagonalizing the finite-temperature free-energy functional [14,15] iteratively as implemented in CPMD [15,16] to account for metallicity in terms of partially occupied KohnSham orbitals. The orbitals were expanded in plane waves at the G point up to 50 Ry, H and N were represented as usual by one and five valence-electron...

A Molecular Perspective on Lithium–Ammonia Solutions

“…[53,54] Recent computational work on saturated lithiumammonia solutions by Chandra and Marx has led to the suggestion that the electrons do not reside within the voids, but rather near to the hydrogen atoms at the void/solvent interface. [55] Our findings will concur with this study. A large amount of theoretical work has also focused on the nature of the composition-induced TMS.…”

“…P(r) reached a maximum near the cavity radius, 3-3.2 , and decayed to zero about 10 away from the center of the cavity. [114] Thus, this work, Jortners, [114] and the Wannier function study by Chandra and Marx, [55] all show that the electron is certainly not in the center of the cavity, but rather near the cavity surface, i.e., the region around the nitrogen and hydrogen atoms. The answer to the question posed in the title of this section is therefore "No".…”

Lithium‐Ammoniak‐Lösungen: eine molekulare Betrachtung

“…[55] In der Literatur finden sich zahlreiche Berechnungen von Wasser-und Ammoniak-solvatisierten Elektronen, die einen etwas anderen Ansatz als den unsrigen verwenden (siehe Lit. [111]).…”

“…10 Abstand zum Zentrum des Käfigs gegen null. [114] Demnach zeigen unsere und Jortners Arbeit [114] sowie auch die Wannier-Studie von Chandra und Marx, [55] dass das Elektron nicht im Zentrum lokalisiert ist, sondern nahe der Oberfläche des Käfigs, d. h. in der Umgebung der N-und H-Atome. Die Antwort auf die im Titel dieses Abschnitts gestellte Frage lautet deshalb "nein".…”

“…[53,54] Neuere Computerstudien an gesättigten Li-NH 3 -Lösungen durch Chandra und Marx führten zu dem Vorschlag, dass sich die Elektronen nicht innerhalb der Lücken aufhalten, sondern nahe den Wasserstoffatomen an der Grenzfläche zum Lösungsmittel. [55] Unsere Befunde werden mit dieser Studie übereinstimmen. Viele theoretische Arbeiten haben sich auch mit der Natur des zusammensetzungsinduzierten TMS befasst.…”