SABRE variants are beginning to offer competitive polarization levels on clinically interesting molecular targets. In SABRE, an iridium-based catalyst reversibly binds ligand(s) with nuclear spin target(s), and binds parahydrogen to make a hydride with singlet order. Under the right circumstances (usually expressed as a level anti-crossing), couplings flow spin order out of the hydrides into ligand polarization. The most common approach uses low fields; in the heteronuclear case, the optimal experimental continuous field is . SABRE operates in an unusual regime for magnetic resonance, where Zeeman splittings, scalar couplings, and exchange rates are all comparable. As a result, most of the assumptions that make NMR tractable are invalid. A practical consequence is that, while low fields offer the unique opportunity to fully manipulate fields in three dimensions, very little exploration of parameter space has been done. This is important because advantages of SABRE (low cost, rapid polarization, broad ligand suitability) are counterbalanced by generally lower polarization levels than d-DNP.
We have developed a comprehensive theoretical picture of SABRE evolution dynamics, validated by the recent development of extremely accurate simulation tools[1]. This leads to unexpected predictions validated by experiments. For example, sequences of rectangular or shaped z-fields which never approach a matching condition instantaneously or on average can improve polarization. [2,3] Here we report multiple approaches with multiaxial pulse sequences that improve magnetization yields up to 7-fold and singlet yields up to 4-fold. The first approach, analogous to high-field experiments, uses decoupling to preserve hydride singlet order. The second approach gets away from assuming sequence structure, and uses an evolutionary strategy to optimize an arbitrary multiaxial field. The field generated by this Multi-Axis Computer-aided HEteronucular Transfer Enhancement approach (MACHETE SABRE) improves polarization more than 7-fold over continuous field SABRE SHEATH. This gain is highly nonintuitive, but is robust to a wide range of exchange rates and SABRE geometries.
[1]Lindale, Eriksson, Tanner, Warren, Science Advances 6, abb6874 (2020)
[2] Eriksson, Lindale, Li, Warren, Science Advances 8, abl3708 (2022)
[3] Li, Lindale, Eriksson, Warren, Phys. Chem. Chem. Phys. 24 16462-16470 (2022).
Acknowledgements: This work was supported by NSF CHE2003109