18-electron complexes questioned

Last year chemists reported that bonding long considered exclusive to transition metals could occur with barium, calcium, and strontium. Mingfei Zhou of Fudan University, Gernot Frenking of Nanjing Tech University and the Philipp University of Marburg, and colleagues reported that these atoms formed 18-electron complexes with 8 carbon monoxide molecules at 4 K (Science 2018, DOI: 10.1126/science.aau0839). But the unexpected results generated skepticism among other chemists. Clark R. Landis of the University of Wisconsin–Madison and colleagues argue that, though they don’t doubt that octocarbonyl complexes were made, the group’s bonding analysis ignored the known chemistry of metals and CO (Science 2019, DOI: 10.1126/science.aay2355).

Zhou and Frenking’s bonding analysis assumes that the metal and octocarbonyl are neutral fragments. Landis’s team redid the analysis assuming those fragments were charged, which Landis says better reflects the complexes’ electron density distribution. They did not observe the kinds of bonding characteristic of transition-metal complexes, a conclusion they supported with other analyses of the bonding and orbitals.

In response, Zhou and Frenking explained that their assumption of neutral Ca and (CO)₈ in that analysis applied to those species only before bonding, and they reaffirmed their conclusion that these elements bond like transition metals (Science 2019, DOI: 10.1126/science.aay5021). Zhou said all analyses show bonding in d orbitals.

These exotic molecules may be an edge case, but Thomas R. Cundari at the University of North Texas says the debate speaks to broader questions about orbitals and bonding.