Session ThOD. There are 5 abstracts in this session.



Session: COMPUTATION / THEORY 2, time: 10:45 - 11:10 am

Verification of Molecular Crystal Structures via NMR Crystallography


Cory Widdifield1; Andrew Tatton2; Paul Hodgkinson3
1Oakland University, Rochester, MI; 2GSK Medical Research Centre, Stevenage, UK; 3Durham University, Durham, United Kingdom

Solid-state NMR is increasingly used to “verify” structures obtained by diffraction methods. To obtain an unbiased view of the field of applications, we have systematically surveyed repeat structures in the Cambridge Structural Database. Although many pairs of structures converge on DFT-based geometry optimisation, a significant fraction do not. NMR chemical shifts are shown to be a powerful complementary test of structure validity in these cases. Computation is also invaluable for structures obtained by powder XRD, where solutions are potentially more ambiguous. Experimental NMR is, however, vital in understanding the behaviour of many materials, especially when disorder is involved; examples will be shown where NMR reveals behaviour that would neither obvious in diffraction studies nor easy accessible to computation.


Session: COMPUTATION / THEORY 2, time: 11:10 - 11:25 am

Undamped and Damped Bloch Equations: Exact Solutions by Path-Sum


Christian Bonhomme1; Pierre Louis Giscard2
1Sorbonne Universite, Paris, France; 2ULCO, Calais, France

Path-Sum will be used to solve exactly the undamped and damped Bloch equations for any shape of the RF field and any initial condition. Each component of the magnetization at the end of the pulse is calculated exactly using a finite number of operations on a continuous fraction of finite breath. Conversely, with path-sum we also tackle the general problem of “inverse pulse design” by exhibiting a large family of closed form analytical solution as well as condition for the solutions to be hitherto unknown higher special functions. 


Session: COMPUTATION / THEORY 2, time: 11:25 - 11:40 am

Implementing a Physically Accurate Model for Experimental Prediction and Guidance in SABRE


Shannon Eriksson; Jacob Lindale; Warren S. Warren
Duke University, Durham, NC

Hyperpolarization by Signal Amplification by Reversible Exchange (SABRE) imparts large non-thermal polarizations on target nuclei from singlet order parahydrogen. This transfer requires reversible interactions with an iridium catalyst creating a large exchanging spin system which has previously been prohibitively expensive to compute accurately. Following up on our implementation of a master equation derived to the infinite order in the exchange interaction, we expand to include a treatment of independent free ligand evolution, relaxation, and nonlinear, concentration dependent interactions between the free ligand species and bound species during exchange. With this highly accurate and rigorous treatment of SABRE dynamics, we can perform directed experiments to improve bulk production and explore novel pulse sequences to optimize SABRE for potential clinical application.


Session: COMPUTATION / THEORY 2, time: 11:40 - 11:55 am

Towards 2D RASER-detected NMR


Stephan Ginthör1; Matthias Bechmann1; Maria Theresia Zich1; Eisuke Chikayama1, 2; Norbert Müller1, 3
1Johannes Kepler Universität, Linz, Austria; 2Dep. of Inform. Systems, Niigata University, Niigata City, Japan; 3Faculty of Science, University of South Bohemia, České Budějovice, Czech Republic

RASER (Radio-frequency amplification by stimulated emission of radiation) phenomena in NMR occur under conditions requiring negative spin temperature, a high quality factor and slow relaxation. We present first results of 2D NMR spectra obtained by the detection of RASER emissions after heteronuclear coherence transfer using a standard high-resolution spectrometer without involving hyperpolarization techniques. In a proof-of-concept HMQC-type experiment we have overcome the problem of random emission start by inserting an extremely short active trigger pulse. Gaps in the indirect time domain data exist where the RASER threshold is not reached due to modulation in the evolution time. This missing data problem can be solved by interpolation with processing methodology derived from non-uniform sampling.


Session: COMPUTATION / THEORY 2, time: 11:55 - 12:10

Full-Scale Ab initio Simulation of MAS-DNP


Frédéric Perras1; Scott Carnahan2; Aaron Rossini1, 2; Takeshi Kobayashi1; Marek Pruski1, 2
1Ames Laboratory, Ames, IA; 2Iowa State University, Ames, IA

Magic-angle-spinning (MAS) dynamic nuclear polarization (DNP) is revolutionizing materials and biological NMR spectroscopy by providing unprecedented sensitivity improvements. Efforts aimed at the development of new or improved DNP methodologies, polarizing agents, etc. would likely benefit from accurate ab initio treatments of the technique. This is however, complicated by the massive scale of the spin systems involved in MAS-DNP (1000’s of nuclei for each electron) as well as the prolonged timescale of DNP experiments. We will show that these complications can be addressed by combining state space reductions and a Monte Carlo optimization algorithm. This approach enables for highly accurate ab initio simulations of DNP that quantitatively reproduce experimental enhancement factors and yield previously inaccessible insights into DNP processes.