New trick for an old dog: from prediction to properties of "hidden clathrates" Ba₂Zn₅As₆ and Ba₂Zn₅Sb₆

The zinc–antimony phase space has been heavily investigated due to the structural complexity and abundance of high-performing thermoelectric materials. Consequentially, the desire to use zinc and antimony as framework elements to encage rattling cations and achieve phonon-glass-electron-crystal-type properties has remained an enticing goal with only two alkali metal clathrates to date, Cs₈Zn₁₈Sb₂₈ and K₅₈Zn₁₂₂Sb₂₀₇. Guided by Zintl electron-counting predictions, we explored the Ba–Zn–Pn (Pn = As, Sb) phase space proximal to the expected composition of the type-I clathrate. In situ powder X-ray diffraction studies revealed two "hidden" compounds which can only be synthesized in a narrow temperature range. The ex situ synthesis and crystal growth unveiled that instead of type-I clathrates, compositionally close but structurally different new clathrate-like compounds formed, Ba₂Zn₅As₆ and Ba₂Zn₅Sb₆. These materials crystallize in a unique structure, in the orthorhombic space group Pmna with the Wyckoff sequence i²h⁶gfe. Single-phase synthesis enabled the exploration of their transport properties. Rattling of the Ba cations in oversized cages manifested low thermal conductivity, which, coupled with the high Seebeck coefficients observed, are prerequisites for a promising thermoelectric material. Potential for further optimization of the thermoelectric performance by aliovalent doping was computationally analyzed. Data is provided from both experiment and theory: - CIF files containing the experimentally determined crystal structure for each compound. - Quantum ESPRESSO input- and output-files for SCF, density of states, and bands calculations - Boltztrap2 configuration file and output, containing electronic transport calculations - LOBSTER configuration file and crystal orbital Hamilton population (COHP) outputs