Molecular recognition in molecular tweezers systems: quantum-chemical calculation of NMR chemical shifts

Abstract

Quantum-chemical calculations for molecular tweezers systems are presented, where the focus is not only on the recognition process in the host-guest systems, but on the self aggregation of the tweezers host as well. Such intermolecular interactions influence the corresponding NMR spectra strongly by up to 6 ppm for proton chemical shifts, since ring-current effects are particularly important. The quantum-chemical results allow one to reliably assign the spectra and to gain information both on the structure and on the importance of intra- and intermolecular interactions. In addition, we study the accuracy of a variety of density functionals for describing the present host-guest systems, where we observe a considerable underestimation of ring-current effects on (1)H NMR chemical shifts at the density functional theory (DFT) level using smaller basis sets such as 6-31G**, so that larger bases like TZP are required. This stands in contrast to the behavior of the Hartree-Fock scheme, where small basis sets, such as 6-31G**, provide reliable (1)H NMR shieldings for molecular tweezers systems.

Quantum-chemical simulation of solid-state NMR spectra: the example of a molecular tweezer host–guest complex

“…In this way, we study the convergence of calculated proton NMR chemical shieldings at DFT and HF levels, where we expect that the major effects of neighboring groups are sufficiently accounted for by the tridecamer. This corresponds to the second challenge mentioned above, whereas the first and third challenges for the present system have been discussed earlier [12,35] and will be briefly re-examined below.…”

“…While we were able to gain useful information by quantum-chemical simulation of NMR spectra for a variety of similar systems [6,10,12,[35][36][37][38][39][40][41][42], we focus in our present work on the study of a dicyanobenzene (DCNB) tweezer host-guest complex (see Fig. 1) and compare the experimentally observed MAS-NMR proton shieldings to quantum-chemically calculated shieldings for different fragment sizes of the solid-state structure.…”

“…1) and compare the experimentally observed MAS-NMR proton shieldings to quantum-chemically calculated shieldings for different fragment sizes of the solid-state structure. The DCNB system has already been subject to earlier studies by us [10,12,35], where we were constrained to rather small fragments such as a trimer at most. Due to the availability of our linear-scaling code for the calculation of NMR shifts we are now able to perform full DFT and HF calculations for fragment sizes of up to a tridecamer, i.e., a DCNB tweezer monomer with the full first sphere of neighboring complexes, without having to resort to an incremental scheme [12] or other additional approximations.…”

“…It should be mentioned that the crystallographic structure data were obtained at a temperature of 298 K whereas the NMR spectrum was recorded at 321 K. The reliable simulation of solid-state NMR spectra by quantum-chemical methods poses three major challenges: (1) Structure determination of building blocks (monomers) and their arrangement in the solid phase must be followed by (2) convergence studies of the NMR data with respect to fragment size, where it is also necessary (3) to estimate error bars caused by the chosen quantum-chemical method and basis set (see as well discussion in Refs. [12,35]). …”

“…[51,52] to offer promising perspectives. However, as these methodologies are mostly still topics of active developments, the structures used in this study were all obtained using the pragmatic approach described above, which also ensures comparability to earlier studies [10,12,35].…”