On the geometry dependence of the nuclear magnetic resonance chemical shift of mercury in thiolate complexes: A relativistic density functional theory study

Thiolate containing mercury(II) complexes of the general formula [Hg(SR) n] 2−n have been of great interest since the toxicity of mercury was recognized. 199Hg nuclear magnetic resonance spectroscopy (NMR) is a powerful tool for characterization of mercury complexes. In this work, the Hg shielding constants in a series of [Hg(SR) n] 2−n complexes are therefore investigated computationally with particular emphasis on their geometry dependence. Geometry optimizations and NMR chemical shift calculations are performed at the density functional theory (DFT) level with both the zeroth‐order regular approximation (ZORA) and four‐component relativistic methods. The four exchange‐correlation (XC) functionals PBE0, PBE, B3LYP, and BLYP are used in combination with either Dyall's Gaussian‐type (GTO) or Slater‐type orbitals (STOs) basis sets. Comparing ZORA and four‐component calculations, one observes that the calculated shielding constants for a given molecular geometry have a constant difference of ∼1070 ppm. This confirms that ZORA is an acceptable relativistic method to compute NMR chemical shifts. The combinations of four‐component/PBE0/v3z and ZORA/PBE0/QZ4P are applied to explore the geometry dependence of the isotropic shielding. For a given coordination number, the distance between mercury and sulfur is the key factor affecting the shielding constant, while changes in bond and dihedral angles and even different side groups have relatively little impact. Relativistic DFT calculations show that the percentage change in the chemical shift of Hg in various thiolate complexes is proportional to the percentage change in the mercury–sulfur bond length independent of the side groups of the thiolates. Changing the bond length by up to 10% can lead to changes in the chemical shift of Hg by up to 15%, while changing the bond angle has little effect on the chemical shift.