Trends in Strong Chemical Bonding in C2, CN, CN–, CO, N2, NO, NO+, and O2

The strong chemical bonds between C, N, and O play a central role in chemistry, and their formation and cleavage are critical steps in very many catalytic processes. The close-lying molecular orbital energies and large correlation effects pose a challenge to electronic structure calculations and have led to different bonding interpretations, most notably for C 2 . One way to approach this problem is by strict benchmark comparison of related systems. This work reports reference electronic structures and computed bond dissociation enthalpies D 0 for C 2 , CN, CN - , CO, N 2 , NO, NO + , O 2 and related systems C 2 + and C 2 - at chemical accuracy (∼1 kcal/mol or 4 kJ/mol) using CCSD(T)/aug-cc-pV5Z, with additional benchmarks of HF, MP2, CCSD, explicitly correlated F12 methods, and four density functionals. Very large correlation and basis set effects are responsible for up to 93% of total D 0 . The order of the molecular orbitals 1π u and 3σ g changes, as seen in textbooks, depending on total and effective nuclear charge. Linear trends are observed in 2σ u -2σ g orbital splitting (R 2 = 0.91) and in D 0 of C 2 , C 2 - , and C 2 + (R 2 = 0.99). The correlation component of D 0 of C 2 is by far the largest (∼93%) due to a poor HF description. Importantly, density functional theory fails massively in describing this series consistently in both limits of effective nuclear charge, and Hartree-Fock exchange or meta functionals do not remedy this 100 kJ/mol error, which should thus be addressed in future density functional developments as it affects very many studies involving cleavage or formation of these bonds.