A semilog plot showing the decay of hyperpolarized Xe-129 as a function of time; the slope is the longitudinal relaxation time T1. Our group recently established a new record for longest T1 for Xe-129 in the gas phase. We have measured T1 > 25 h for Xe-129 at room temperature in an applied field of 8.0 T. The maximum T1 reported previously in the literature was a few hours. These times were achieved by first understanding a critical intrinsic mechanism for Xe-129 relaxation resulting from the rapid formation and breakup of Xe-Xe van der Waals molecules. This mechanism was first elucidated by the group of Thad Walker at the University of Wisconsin (see B. Chann, et al, Phys. Rev. Lett. 88, 113201, 2002). We showed experimentally that this mechanism can be suppressed at high magnetic fields (see B.N. Berry-Pusey, et al., Phys. Rev. A 74, 063408, 2006), leading to the extraordinarily long relaxation times. This work also showed that there are two distinct but related interactions that contribute to the relaxation during a molecular lifetime: spin-rotation and paramagnetic antishielding (also known as chemical-shift anisotropy). This work may have important implications for the production of large quantities of highly polarized Xe-129, whereby long relaxation times are needed to accumulate and store the hyperpolarized gas.
Research in the Saam Lab centers on both the basic physics and applications of hyperpolarized (HP) gases. The nuclei
of non-zero-spin noble gas isotopes (most notably He-3 and Xe-129) can be polarized to 10% or more by collisional
spin exchange with a laser-optically pumped and polarized alkali-metal vapor (Walker and Happer, Rev. Mod. Phys.
69, 629, 1997). The nuclear polarization so produced is typically four to five orders of magnitude larger than
the thermal-equilibrium value, and the non-equilibrium state can be preserved for many hours. HP gases are at once
elegant systems for teaching and exploring basic spin physics and important tools in many subfields of physics,
chemistry, even bioengineering and medicine.
The most spectacular application of HP gases is magnetic resonance imaging (MRI) of inspired gas and gas flow in
the lung. (See review articles by Leawoods, et al., Concepts Magn. Reson. 13, 277, 2001; and Moeller, et al., Magn. Reson. Med., 47, 1029-1051, 2002.) Breathing may be imaged with unprecedented temporal
and spatial resolution, making the technique very promising for aiding diagnosis and treatment of diseases such
as emphysema. A non-invasive picture of lung ventilation has the potential for tremendous impact on our understanding
of both healthy lung physiology and pulmonary disease. Much of Professor Saam's post-doctoral work with Mark Conradi
at Washington University was devoted to developing various He-3 MRI techniques for the study of emphysema; click
here to learn about their more recent results. Our current work in this area is on flow and diffusion MRI with
HP He-3. With colleagues at the Pacific Northwest National Laboratory in Richland, WA, we are using He-3 MRI to
characterize airflow in the mammalian respiratory tract. The ultimate goal is to validate computational fluid-dynamics
models of this airflow.
A potentially revolutionary application to physical chemistry involves polarization transfer from HP Xe-129 to
other nuclear species in order to enhance their NMR sensitivity, potentially by orders of magnitude. (See, for
example, J.C. Leawoods, et al., Chem. Phys. Lett. 327, 359, 2000; and G. Navon, et al., Science 271,
1848, 1996.) NMR is notoriously insensitive, due to the weak nuclear magnetic moment. This is especially true for
surface studies, because of the need to fill the NMR coil and the small number of surface spins relative to the
bulk in most materials. Yet NMR remains the technique of choice for elucidating the atomic and molecular structure
of many important materials. For example, it is the only technique available to study structure and function of
proteins in close to their natural state. Because of the sensitivity problem, great effort is spent struggling
for small gains in signal-to-noise ratio, whereas polarization transfer from HP Xe-129 could provide orders of
magnitude gain in sensitivity, with a corresponding increase in information obtained from the sample and/or decrease
in the amount of sample which is needed for study.
Recently, in working on polarization transfer, we came up with a new way of producing hyperpolarized liquid xenon
by a combination of spin exchange and phase exchange in an appropriately designed convection cell (see cell schematic
above). So far, using this technique we have been able to generate steady-state liquid Xe-129 polarizations of
5-10% in about 15 min. If all of the xenon in these cells were in the gas phase, the high xenon density would preclude
efficient SEOP due to the strong Rb-Xe spin-rotation interaction that rapidly relaxes the Rb electron spin. We
envision at least two possible uses for liquid xenon polarized by our new technique: First, rapid volatilization
of the liquid could provide a substantial quantity of highly polarized Xe-129 gas for various applications. It
is not yet clear the degree to which our scheme could compete with flow-through xenon accumulator systems in producing
large amounts of polarized Xe-129. (See, for example, B. Driehuys, et al., Appl. Phys. Lett.
69, 1668, 1996; and A.L. Zook, et al., J. Magn. Reson. 159, 175, 2002 and the Center
for Xenon Imaging at the University of New Hampshire.) Our scheme has the advantage of avoiding the solid phase,
whereby polarization losses in the freeze-thaw cycles can be a problem. However, our production rate is ultimately
limited by the 20-min relaxation time of liquid Xe-129. Second, as a solvent, the liquid column would be a novel
platform for polarization-transfer experiments, particularly since the Xe-129 polarization could be maintained
at a steady-state value.
We also generally interested in the basic physics of spin-exchange optical pumping (SEOP), including the various
rates and relaxation processes involved. The atomic and molecular physics of SEOP has been studied for more than
40 years, and yet several interesting problems and questions remain. This part of our effort also feeds back to
the applications in the form of more highly polarized and efficiently produced noble gases. Recently, we discovered
that most, if not all, glass vessels (cells) used for polarization and storage of He-3 exhibit ferromagnetic properties
that are reflected in hysteretic He-3 wall-relaxation rates. (See R.E. Jacob, et al., Phys. Rev. Lett. 87,
143004, 2001; and R.E. Jacob, et al., Phys. Rev. A 69 021401(R), 2004.) We have also developed and
verified a predictive model for wall relaxation rates on bare glass (no Rb) surfaces, which is an important model
system for the study of wall relaxation. (See R.E. Jacob, et al., Chem. Phys. Lett. 370, 261, 2003.)