Improving Understanding, Utility and Generality of Hyperpolarized, Long-lived Spin States in Magnetic Resonance

Duke University, Durham NC

Investigators

Abstract

Nuclear magnetic resonance (NMR) spectroscopy is a powerful tool for chemists; it is useful for determining molecular structure and for monitoring the progress of chemical reactions. NMR's clinical cousin, magnetic resonance imaging (MRI) is also a useful tool for producing images of soft tissues in the body. Both methods usually suffer from relatively low sensitivity - meaning that it cannot detect small amounts of sample. "Hyperpolarization" methods, which can increase the signal by a factor of 1000 or more, have been extensively explored over the last decade, but fewer than 1% of all NMRs and MRIs have access to this technology. Hyperpolarization is technically challenging and extremely expensive. With support from the Chemical Measurement and Imaging Program, and partial co-funding from the Chemical Structure, Dynamics, and Mechanisms - A Program, Professors Warren Warren, Steven Malcolmson, and Thomas Theis and their groups at Duke University are working to expand the availability and utility of hyperpolarization (a readily prepared reagent). They are developing new catalysts to speed sample preparation and an instrument that can be built for about 1% of the cost of competitive approaches. This research is both expanding the range of possible molecular targets and increasing the attainable sensitivity. The project could ultimately enable low-cost, portable MRI instruments. The team is providing research and training opportunities in these critical technologies for members of underrepresented groups, and providing substantial K-12 science outreach. The low sensitivity of NMR and MRI is often a severe limitation. Hyperpolarization methods (mostly dynamic nuclear polarization, DNP) can increase the observable signal as much as 10,000 times in virtually any organic molecule, but hyperpolarization decays back to thermal equilibrium at a rate given by the nuclear spin lifetime T1, which in solution is commonly seconds. These short relaxation times limit the processes that can be studied by the method. In addition, the associated apparatus is complicated and expensive (ca. $2.5M for clinical systems). The team at Duke has shown that pulse sequences can load and unload long-lived states in many molecules, effectively lengthening homogeneous relaxation times to many minutes. Their new approaches to hyperpolarization (SABRE-SHEATH and LIGHT-SABRE) have increased 15Nitrogen or 13Carbon polarization by up to five orders of magnitude (relative to thermal populations) using an ultralow-field apparatus (about 1% of the Earth's field) that can be built by undergraduates. They are now working to improve polarization rates, total polarization, and the generality of these methods. The aim of this research project is to reduce the sensitivity to molecular parameters and increase reliability. The work includes a mixture of discovery, basic theoretical extensions of the density matrix understanding of magnetic resonance, detailed quantum mechanical and molecule dynamics simulations, targeted synthesis, and demonstration of useful applications. The research is addressing fundamental questions about spin dynamics. The systems reflect a new class of "reactive intermediates" which go back to their original state, yet create a useful product (spin polarized nuclei). Other extensions focus on the complex role of molecular symmetry in isolating states from their environment, and on developing tools to access ever-better protected states. Broader impacts arise from the potential to create a wide variety of Magnetic Resonance-trackable molecular targets, with applications from organometallic chemistry to clinical imaging.