Statistical learning of NMR tensors from 2D isotropic/anisotropic correlation nuclear magnetic resonance spectra

Simulating multipulse NMR spectra of polycrystalline solids in the frequency domain

An approach is presented for simulating multipulse nuclear magnetic resonance (NMR) spectra of polycrystalline solids directly in the frequency domain. The approach integrates the symmetry pathway concept for multipulse NMR with efficient algorithms for calculating spinning sideband amplitudes and performing interpolated finite-element numerical integration over all crystallite orientations in a polycrystalline sample. The numerical efficiency is achieved through a set of assumptions used to approximate the evolution of a sparse density matrix through a pulse sequence as a set of individual transition pathway signals. The utility of this approach for simulating the spectra of complex materials, such as glasses and other structurally disordered materials, is demonstrated.

Torsion Angle Analysis of a Thermally Activated Delayed Fluorescence Emitter in an Amorphous State Using Dynamic Nuclear Polarization Enhanced Solid-State NMR

The torsion angle between donor and acceptor segments of a thermally activated delayed fluorescence (TADF) molecule is one of the most critical factors in determining the performance of TADF-based organic light-emitting diodes (OLEDs) because the torsion angle affects not only the energy gap between the singlet and triplet but also the oscillator strength and spin-orbit coupling. However, the torsion angle is difficult to analyze, because organic molecules are in an amorphous state in OLEDs. Here, we determined the torsion angle of a highly efficient TADF emitter, DACT-II, in an amorphous state by dynamic nuclear polarization enhanced solid-state NMR measurements. From the experimentally obtained chemical shift principal values of ¹⁵N on carbazole, we determined the average torsion angle to be 52°. Such quantification of the torsion angles in TADF molecules in amorphous solids will provide deep insight into the TADF mechanism in amorphous OLEDs.

Machine Learning Full NMR Chemical Shift Tensors of Silicon Oxides with Equivariant Graph Neural Networks

The nuclear magnetic resonance (NMR) chemical shift tensor is a highly sensitive probe of the electronic structure of an atom and furthermore its local structure. Recently, machine learning has been applied to NMR in the prediction of isotropic chemical shifts from a structure. Current machine learning models, however, often ignore the full chemical shift tensor for the easier-to-predict isotropic chemical shift, effectively ignoring a multitude of structural information available in the NMR chemical shift tensor. Here we use an equivariant graph neural network (GNN) to predict full ²⁹Si chemical shift tensors in silicate materials. The equivariant GNN model predicts full tensors to a mean absolute error of 1.05 ppm and is able to accurately determine the magnitude, anisotropy, and tensor orientation in a diverse set of silicon oxide local structures. When compared with other models, the equivariant GNN model outperforms the state-of-the-art machine learning models by 53%. The equivariant GNN model also outperforms historic analytical models by 57% for isotropic chemical shift and 91% for anisotropy. The software is available as a simple-to-use open-source repository, allowing similar models to be created and trained with ease.

3D Relaxation-Assisted Separation of Wideline Solid-State NMR Patterns for Achieving Site Resolution

There are currently no methods for the acquisition of ultra-wideline (UW) solid-state NMR spectra under static conditions that enable reliable separation and resolution of overlapping powder patterns arising from magnetically...

Silicon-29 echo train coherence lifetimes and geminal 2J-couplings in network modified silicate glasses

The natural abundance ²⁹Si echo-train coherence lifetimes in network-modified silicate glasses were examined under static and magic-angle spinning (MAS) conditions. The nuclear magnetic properties of modifier cations were found to play a major role in determining ²⁹Si coherence lifetimes, leading to differences as large as three orders of magnitude. In compositions with abundant NMR active nuclei, such as alkali silicates, the ²⁹Si coherence lifetimes are dominated by coherent dephasing due to residual heteronuclear dipolar couplings, whereas in compositions dilute in NMR active nuclei, such as alkaline earth silicates, the ²⁹Si coherence lifetimes are dominated by incoherent dephasing due to paramagnetic impurities. Expressing the inverse of the coherence lifetime as a residual full width at half maximum (FWHM), we found that increasing rates of both MAS and a π-pulse train are effective in removing the residual ²⁹Si heteronuclear broadenings, with a near-linear relationship between FWHM and MAS rotor period and π-pulse spacing. Based on these results, we conclude that accurate ²⁹Si J coupling measurements will be the most challenging in lithium silicate glasses due to strong homonuclear dipolar couplings among ⁷Li nuclei, requiring MAS speeds up to 100 kHz, and be the least challenging in the alkaline earth silicate glasses. At a modest MAS speed of 14kHz, distributions of geminal J couplings across Si-O-Si linkages were measured in alkali and alkaline earth silicate glasses giving mean values of 4.2Hz and 5.1Hz in 0.4 CaO·0.6 SiO₂ and 0.33 Ba₂O·0.67 SiO₂ glasses, respectively, and 5.2Hz and 5.3Hz in 0.33 Na₂O·0.67 SiO₂ and 0.33 K₂O·0.67 SiO₂ glasses, respectively. We also observe greater variance in the J distributions of alkaline earth silicate glasses consistent with greater structural disorder due to increased modifier cation potential, i.e., the charge-to-radius ratio, Z/r of the cation.