Structural effects in molecular metal halides

Dimeric Molecular Structure of Molten Gallium Trichloride and a Hidden Evolution toward a Possible Liquid-Liquid Transition

Group 13 trihalides MY₃ (M = Al, Ga, and In; Y = Cl, Br, and I) mostly having a dimeric M₂Y₆ molecular structure in the solid state and a mixture of M₂Y₆ dimers and MY₃ monomers in the vapor phase are potential candidates for entropy-driven liquid-liquid transition M₂Y₆ ⇄ 2MY₃ at elevated temperatures. Using pulsed neutron diffraction and high-energy X-ray scattering supported by structural modeling, we show a dimer molecular structure of liquid GaCl₃ above the melting point at 351 K and midway between the boiling point (474 K) and the critical temperature (694 K) with almost hidden characteristic evolution toward a possible liquid-liquid transition. In contrast to edge-sharing (ES) dimers in solid and vapor of D_2h symmetry, the ES Ga₂Cl₆ molecules in the melt have a puckered structure of the central four-membered ring with shorter Cl-Cl (2.90-3.09 Å) and longer Ga-Ga (3.20-3.26 Å) second-neighbor correlations. The elongation of Ga-Ga intramolecular distances with increasing temperature simultaneously with diminished Cl-Cl nearest neighbor contacts destabilizes the ES dimers, indicating the first step toward dimer dissociation.

Structures and stabilities of group 17 fluorides EF3 (E = I, At, and element 117) with spin-orbit coupling

In this work, a recently developed CCSD(T) approach with spin-orbit coupling (SOC) as well as density functional theory (DFT) using various exchange-correlation (XC) functionals are employed to investigate structures and stabilities of group 17 fluorides EF(3) (E = I, At, and element 117). These molecules are predicted to have bent T-shaped C(2v) structures according to the second-order Jahn-Teller (SOJT) effects or the valance shell electron pair repulsion (VSEPR) theory. For IF(3) and (117)F(3), our results are consistent with previous SOC-DFT calculations. However, different XC functionals provide different results for AtF(3) and our SOC-CCSD(T) calculations show that both the C(2v) and D(3h) structures are minima on the potential energy surface and the C(2v) structure is the global minimum for AtF(3). The performance of XC functionals on structures and stabilities of IF(3) and AtF(3) is found to depend on the fraction of the Hartree-Fock exchange (HFX) included in the XC functionals and the M06-2X functional with 54% of HFX providing results that agree best with CCSD(T) results. In addition, although both the C(2v) and D(3h) structures are minima for AtF(3), the energy barrier between them is only 8 kJ mol(-1) for the C(2v) structure and 0.05 kJ mol(-1) for the D(3h) structure. This indicates that the D(3h) structure could not possibly be observed experimentally and AtF(3) can convert easily between the three C(2v) structures. The SOJT term is shown to be reduced by electron correlation for IF(3) and AtF(3). On the other hand, although SOC decreases the energy difference between the C(2v) and D(3h) structures and reduces the deviation of the C(2v) structure from the D(3h) structure, it decreases the frequency of the bond bending mode, which may indicate that SOC actually increases the SOJT term. This could be related to mixing of spin-singlet E' states to low-energy spin-triplet states due to SOC.