Text version of the page 
Text version of the page 
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | SUMMER 2002 | | | | | | | | | | | | | | | | | | | | 100000 | | | | | | | | | | I 1 1 I 1 II • With3He □ No 3He | | | | | 10000 | • | \ | | | | | | ■S 1000 | • | | | | | | | 100 10 | ........ | | ... 1 | | | | | 0.1 | 7(K) | 1 | | | | | | | | | | | | | | | | | Low temperature hyperpolarisation of xenon | | | | | | | | | | New solutions Dr Biskup's team overcame this difficulty | | | | | | Dr Neven Biskup's team at the Fhysics Department, University of Massachusetts, Amherst, USA, investigated brute force polarisation of xenon at high magnetic field and ultra low temperatures using an Oxford Instruments Superconductivity's dilution refrigerator, professional thermometer and temperature driver. Matter prepared with hyperpolarised nuclear spins such as 3He and 129Xe have important applications. This new method, significant in the field of physics, also has potential importance in medicine, for example, to polarise a broad range of contrast agents to improve resolution of MRI scanning techniques. | | | | | | | | | | by plating xenon onto
| | | | | | | | | Figure 1: Spin lattice relaxation time T, measured for 129Xe as a function of temperature and 3He coverage | | | | | | a silica gel substrate,
| | | | | | | | | | with a very high specific surface area. A cell containing powdered silica gel with a In the absence of 3He, T1 is strongly sintered-silver heat exchanger manufacturers, was cooled temperature dependent. The mechanism in a dilution refrigerator in an 8 Tesla of relaxation without 3He is not known, NMR magnet. The silca gel was contained although it may reflect interactions with | | | | | | | | | | in an epoxy lower portion of the cell, extending into a small birdcage NMR resonator at 92 MHz, the Larmor frequency at 8T. The magnetic field could also be reduced to measure the 3He NMR signal at the same frequency. 3He liquid levels were monitored using a vibrating wire viscosimeter at the top of the cell. Xenon was introduced using a heated fill line maintained at 90K. The volume condensed was typically 20% of the available pore space - corresponding to around three atomic layers on the silica surface. Small angle tipping pulses, after either a magnetisation-inverting n pulse or a magnetisation-destroying comb of large angle pulses, was used to measure the spin-lattice relaxation of 129Xe. Magnetisation was sampled from 0.1 - 40,000 s ensuring the inclusion of rapidly or slowly relaxing spin populations. Findings Addition of 3He to the cell shortens T1 and changes the recovery curve from a stretched exponential, typically due to a wide distribution of T1 values for individual 129Xe spins, to a simple exponential. | | the silica surface and/or adsorbed impurities. When 3He is added to the cell, the relaxation time reduces to a temperature-independent value t1 ~1000 s (Figure 1). | | | | | | | | | | Conclusion Solid xenon can be brute-force polarised at dilution refrigerator temperatures usng a 3He-porous substrate method. Before becoming a practical method for the production of hyperpolarised 129Xe gas, further obstacles need to be overcome. It may be necessary, for example, to switch off the relaxation process before removing the xenon sample to low B/T conditions. Addition of 4He to the cell could provide such a switch, as 4He preferentially occupies sites next to solid surfaces. Importantly, spreading of 4He over the xenon surface occurs by superfluid film flow. Like the tunnelling process used to induce relaxation, this is a quantum process that proceeds at arbitrarily low temperatures. | | | | | | | | | | Background An obstacle to brute-force polarisation of 129Xe is the long spin-lattice relaxation time (T1), preventing bulk solid xenon from quickly reaching equilibrium polarisation. Surface nuclear spins in substances immersed in liquid 3He, however, are rapidly relaxed by quantum tunnelling of 3He atoms in the localised (solid-like) layer that forms near solid surfaces - a process that perssts to arbitrarily low temperatures. The main constraints for this process is the need for a large surface area since quantum tunnelling polarisation transfer will appear only on the surface in direct contact with 3He. | | | | | | | | | | Contact: N. Biskup, Physics Department, University of Massachusetts, USA biskup@:pia.oxinst.com | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
|