TECHNICAL COMMENTS Cite as: C. R. Landis et al., Science 10.1126/science.aay2355 (2019). Comment on “Observation of alkaline earth complexes M(CO)8 (M = Ca, Sr, or Ba) that mimic transition metals” Clark R. Landis1*, Russell P. Hughes2, Frank Weinhold1 1Department of Chemistry, University of Wisconsin, Madison, WI 53706, USA. 2Department of Chemistry, Dartmouth College, Hanover, NH 03755, USA. *Corresponding author. Email: [email protected] Wu et al. (Reports, 31 August 2018, p. 912) claim that recently characterized octacarbonyls of Ca, Sr, and Ba mimic the classical Dewar-Chatt-Duncanson bonding motif of transition metals. This claim, which contradicts known chemistry and computed electron density distributions, originates in the assumption of a flawed reference state for energy decomposition analyses. Downloaded from The report by Wu et al. (1) concerning the existence and Natural bond orbital (NBO) (10) calculations compute a characterization of the unexpected calcium carbonyl, range of calcium charges (+1.15 to +1.55) that depend on the http://science.sciencemag.org/ Ca(CO)8, at low temperatures demonstrates the power of basis set; basis sets lacking representation of the nd func- modern spectroscopic methods used in combination with tions give higher charges, whereas the presence of nd func- modern electronic structure calculation. There is no doubt tions gives the lower value. Taken at face value, the NBO that the compound exists, as formulated, under the experi- results may seem to indicate a bonding role for the Ca 3d mental conditions. However, the description of these com- orbitals, as suggested in the original report. It is easily plexes as mimicking classic characteristics of transition shown, however, that such a role is specious and that the Ca metal carbonyls, such as backbonding from calcium d orbit- nd functions serve primarily to augment the diffuse, delocal- als into the π* orbitals of the carbonyl groups and conform- ized charge of the (CO)8 dianion. Similar conclusions, based ance to the 18-electron rule, conflicts with simple analysis of on basis set extrapolations, have recently appeared (11). on December 28, 2020 the computed electron density and the longstanding chem- Figure 1 depicts the highest occupied molecular orbital istry of CO in the presence of reducing metals. (HOMO) (eg symmetry) of Ca(CO)8, the eg HOMO of the 2– Calcium metal is a good reducing agent and its chemis- [(CO)8] cluster at the geometry of Ca(CO)8, and the nomi- try is dominated by the formation of “saline” compounds nal “Ca d orbital” that shows ~0.4e occupation in the default with Ca2+ ions, even for weakly electronegative counterions NBO analysis. These orbital depictions include two promi- 2– (2–8). For more than a century, the chemistry of carbon nent features: (i) The two electrons of the [(CO)8] cluster monoxide in the presence of reducing metals has been un- are highly delocalized among the π*CO orbitals and (ii) the derstood as reductive in nature; examples include elements nominal d orbital, which extends far from the calcium nu- from groups 1, 2, and 3 as well as a variety of organometallic cleus with a curious stem-and-bulb shape, consists almost complexes. As we demonstrate below, calcium reacts with solely of the outermost d Gaussian primitive. This orbital is carbon monoxide by transfer of two valence electrons to not a valence “Ca 3d” orbital, because no Ca nucleus is pre- 2– form the [(CO)8] dianion that serves as a σ– and π–Lewis sent in this calculation. When d-type functions are absent base to the resultant Lewis acid, Ca2+. Electron density dis- from the calcium basis set, NBO charges closely agree with tributions, bonding patterns, and Ca-CO bonding thermo- AIM charges. The bonding in Ca(CO)8 evidently arises from dynamics in Ca(CO)8 bear little resemblance to the common a combination of ionic forces and conventional Lewis-base 2– Dewar-Chatt-Duncanson motif of transition metal carbon- donation of [(CO)8] density into the vacant 4s and 3d or- 2+ 2+ 2– yls. bitals of the Ca Lewis acid in the ionic Ca [(CO)8] formu- The computed electron density distributions of Ca(CO)8 lation. 2+ 2– indicate an ionic formulation, Ca [(CO)8] . For example, Bond lengths and bonding patterns for the interaction the QTAIM (quantum theory of atoms in molecules) method of CO with Ca differ substantially from those with transition (9) results in an effective charge of +1.56 at calcium. This metals. The RCa-C bond length (2.6 Å) and first Ca-CO bond value is consistent over a variety of basis sets, density func- dissociation energy (∆ECa-CO = 9 kcal/mol) of Ca(CO)8 are far tional theory (DFT) functionals, and non-DFT methods. longer and weaker than those of Cr(CO)6 (RCr-C = 1.91 Å, Publication date: 9 August 2019 www.sciencemag.org 1 ∆ECr-CO = 37 kcal/mol). The computed enthalpies of succes- is a powerful reductant. From this starting point, the final sive CO additions to give singlet and triplet calcium carbon- charge distribution can be accommodated only by strong yls are shown in Fig. 2. Note that the first three additions of transfer of d electrons into the overlapping π*CO orbitals. CO to Ca are approximately thermoneutral and each subse- Thus, the appearance of strong d-orbital “backbonding” is quent addition of CO is exothermic by ~12 kcal/mol. Appar- an artifact that wholly results from selection of a high- ently, CO binding is enthalpically favorable only when there energy reference state that poorly describes the molecular are enough CO ligands (about four) to stabilize the accumu- electron density distribution. As shown in Table 1, alterna- 2+ 2– lation of negative charge on the ligands. In contrast, transi- tive EDA analysis using Ca and [(CO)8] as the reference tion metals commonly have large bond enthalpies for only fragments is demonstrably superior to the original treat- the first few CO additions (12), with subsequent CO addi- ment; the EDA orbital interaction energy is much smaller tions becoming notably weaker as a result of the change and the charge distribution better approximates the molecu- from 2c/2e bonding to 3c/4e bonding. In keeping with the lar environment. In this state, the metal acts as a good σ– 2+ 2– Ca [(CO)8] formulation, the spin density of the triplet and π–Lewis acid, as expected for a group 2 metal. state is >90% delocalized over the CO ligands. We conclude that the electronic structure of Ca(CO)8 at Downloaded from The shapes of molecular Ca(CO)n (n = 2 to 8) are easily the equilibrium geometry does not conform to the Dewar- deformed with multiple local minima. Optimization of the Chatt-Duncanson model used for transition metal complex- geometry using basis sets that lack d orbitals on Ca or opti- es. Although provocative, the interpretation posed by Wu et mization after deletion of all Fock matrix elements involving al. (1) contradicts fundamental charge distributions and calcium d-type natural atomic orbitals lead to modest long-established reactivity of the alkaline earth elements. http://science.sciencemag.org/ changes in geometry (e.g., an increase of the Ca-CO distance The numerical evidence that is offered to support the transi- of just 0.4 Å). Such small effects are consistent with nominal tion metal–like bonding of Ca and CO arises solely from the calcium-centered d-type basis orbitals acting primarily to assumption of a reference state that bears little resemblance 2– augment the basis functions of the [(CO)8] dianion. to the true molecular electron density distribution. This is Why did the initial energy decomposition analysis not a novel example; problematic interpretations arising (EDA) attribute the bonding in Ca(CO)8 to strong backbond- from ill-chosen EDA reference states were previously criti- ing from the d2 electron configuration of Ca? Put simply, the cized for neutral N-heterocyclic carbene complexes of cop- d2 configuration of Ca is a bad reference point for analyzing per, silver, and gold (13). the bonding in Ca(CO)8 as judged by a number of criteria. Any EDA-type bonding analysis requires a starting ref- on December 28, 2020 erence state from which one can assess the changes in den- REFERENCES AND NOTES sity distributions. The choice of the reference state is critical 1. X. Wu, L. Zhao, J. Jin, S. Pan, W. Li, X. Jin, G. Wang, M. Zhou, G. Frenking, Observation of alkaline earth complexes M(CO)8 (M = Ca, Sr, or Ba) that mimic for meaningful analysis and can usually be chosen sensibly transition metals. Science 361, 912–916 (2018). doi:10.1126/science.aau0839 on chemical grounds. For example, the diatomic molecule Medline LiF is well described as ionic at the equilibrium distance; 2. See (3–8) for pertinent summaries of CO reductive coupling chemistry in the the computed electron density partitions as nearly +1 and –1 presence of reducing metals. at Li and F, respectively, and the shape of the quantum me- 3. J. Liebig, Über das Verhalten des Kohlenoxyds zu Kalium. Ann. Chem. Pharm. 11, chanical dissociation energy curve closely matches the pure 182–189 (1834). doi:10.1002/jlac.18340110206 4. B. Wayland, X. Fu, Building molecules with carbon monoxide reductive coupling. electrostatic curve from near-equilibrium RLi-F = 1.6 Å out to ~6 Å. Thus, at Li-F distances around the equilibrium value, Science 311, 790–791 (2006). doi:10.1126/science.1123884 Medline the appropriate reference is Li+ F–. However, at very long 5. R. West, in Oxocarbons (Academic Press, 1980), pp.