Wednesday, June 8, 2016

Static DNP-NMR Spectroscopy to Characterize Active Pharmaceutical Ingredients #DNPNMR

Dynamic Nuclear Polarization in general is no new method, but the focus of modern applications has initially been on bio-macromolecules under magic-angle-spinning (MAS) conditions.

One application that came out-of-the-blue was using DNP-NMR spectroscopy to study surface materials by DNP-NMR spectroscopy (for example Lafon et al., 2011) opening up a complete new research area within material science that traditionally struggled with very low signal-to-noise (S/N) ratios.

Even the application of DNP-NMR spectroscopy to study small molecules was not immediately evident, but as demonstrated in Rossini et al, 2012 DNP offers the possibility to record 13C correlation spectra of unlabeled molecules such as glucose in just 16 hours. Without DNP this experiment would require months of spectrometer time.

The majority of the DNP-NMR experiments that have been reported in recent years use gyrotron-based DNP-NMR systems and MAS-DNP probes operating at about 100 K. Alternatively, there is a small group of researchers that use DNP systems based on a solid-state microwave source. These systems have are typically limited by their output power, which ranges between >80 mW at 263 GHz (400 MHz 1H NMR) to < 200 mW at 197 GHz (300 MHz 1H NMR). At lower frequencies the output power increases and > 500 mW can be reached for systems operating at 95 GHz. A comprehensive overview of low-power DNP-NMR systems can be found in Siaw et al., 2016.

Because of the limited output power, DNP experiments are performed at temperatures < 20 K, which requires cooling with liquid helium (very common for example in EPR experiments) and can be cost-effective when using a cryostat (e.g. at 10 K the consumption is about 0.5 l/hr). Furthermore, with the increasing popularity of cryogen-free systems some cryostats don't require any liquid cryogens anymore for cooling. The main advantage is the reduced cost since a solid-state source based DNP-NMR system typically comes at a 10th of the cost of a gyrotron-based system.

At first sight it seems as if the applications of static DNP are very limited. However, when I was at ENC this year I listened to a talk by David A. Hirsh entitled "35Cl Dynamic Nuclear Polarization Solid-State NMR of Active Pharmaceutical Ingredients". David is a graduate student in the group of Rob Schurko, University of Windsor and gave a very nice talk on using DNP-NMR spectroscopy to characterize Active Pharmaceutical Ingredients (API) using 35Cl solid-state NMR spectroscopy. Since 35Cl is a quadrupole nucleus the corresponding NMR spectra are typically very broad. MAS does only have a small effect, mainly on the center transition, and traditionally wide-line spectra of static solids are recorded.

To overcome sensitivity issues, the group has developed pulse sequences such as WURST-CPMG or BRAIN-CP to rapidly record broad 35Cl patterns even at moderate magnetic field strengths (e.g. 9.4 T, 400 MHz 1H NMR). However, recording a single spectrum often requires several hours of signal averaging to achieve a sufficiently high signal-to-noise (S/N) ratio. With the aid of DNP these acquisition times can be dramatically reduced to just minutes. In his talk at ENC David described using a grytron-based DNP-NMR system, equipped with a MAS-DNP probe head in his experiments. Polarizing the sample is done while the rotor is spinning, but the rotor is stopped prior to recording the wide-line NMR spectrum. 

This experiment seems to be ideally suited for a low-power DNP-NMR system for static solids, using a cryostat for sample cooling. This would greatly simplify the experiment because starting and stopping the rotor is not required anymore. Because the experiment is performed at much lower temperatures, there will be an additional boost in sensitivity and multi-dimensional correlation experiments should be possible, experiments that are close to impossible to perform without the aid of DNP.

In recent years the NMR community has witnessed the transition of DNP-NMR spectroscopy from an exotic method with a limited number of applications to a method with more and more applications. High-field DNP-NMR spectroscopy either based on a gyrotron or using a low-power solid-state source is still a very young method with many possibilities and I'm very excited to see what other applications lie in the future. I am however convinced that DNP-NMR spectroscopy will find their way into many more labs in the future and that the method will become an integral part of the NMR toolbox.