Cryogenic accumulator for spin-polarized xenon-129

United States Patent [w]

Cates, Jr. et al.

US005809801A [ii] Patent Number: [45] Date of Patent:

5,809,801 Sep. 22, 1998

[54] CRYOGENIC ACCUMULATOR FOR SPINPOLARIZED XENON-129

[75] Inventors: Gordon D. Cates, Jr., Skillman, N.J.;

Bastiaan Driehuys, Bristol, Pa.;
William Happer, Princeton, N.J.; Eli
Miron, Princeton, N.J.; Brian Saam,
Princeton, N.J.

[73] Assignee: The Trustees of Princeton University, Princeton, N.J.

[21] Appl. No.: 622,865

[22] Filed: Mar. 29, 1996

[51] Int. CI. 125.1 1/00

[52] U.S. CI 62/637; 62/919; 62/925

[58] Field of Search 62/637, 919, 925,

62/55.5, 3.1

[56] References Cited

U.S. PATENT DOCUMENTS

3,748,864 7/1973 Lofredo et al 62/919

4,080,429 3/1978 Koeppe et al 62/925

4,369,048 1/1983 Pence 62/925

4,417,909 11/1983 Weltmer, Ir. 62/925

4,599,462 7/1986 Michl 62/637

4,755,201 7/1988 Eschwey 62/55.5

4,977,749 12/1990 Sercel 62/55.5

5,007,243 4/1991 Yamaguchi et al 62/55.5

5,039,500 8/1991 Shino et al 62/925

5,161,382 11/1992 Missimer 62/55.5

OTHER PUBLICATIONS

Wagshul, et al., "Optical Pumping 01 High-Density Rb With A Broadband Dye Laser And GaAlAs Siode Laser Arrays: Application to 3He Polarization", Physical Review A, vol. 40, No. 8, pp. 4447^1454 (1989).

Gatzke, et al., "Extaordinarily Slow Nuclear Spin Relaxation In Frozen Laser-Polarized 129Xe", Physical Review Letters, vol. 50, No. 5, pp. 690-693 (1993). Cates, et al., "Laser Production 01 Large Nuclear-Spin Polarization In Frozen Xenon", Physical Review Letters, vol. 65, No. 20, pp. 2591-2594 (1990). Becker, et al., "Study 01 Mechanical Compression 01 Spin-Polarized 3He Gas", Nuclear Instruments And Methods In Physics Research, vol. A 346, pp. 45-51 (1994). Middleton, et al., "MR Imaging With Hyperpolarized 3He Gas", Magnetic Resonance In Medicine, vol. 33, pp. 271-275 (1995).

Cummings, et al., "Optical Pumping ol Rb Vapor Using
High-Power Ga2-,ALAs Diode Laser Arrays", Physical
Review A, vol. 52, No. 6, pp. 4842-4851 (1995).
Middleton, "The Spin Structure ol The Neutron Determined
Using A Polarized 3He Target", Ph.D Dissertation, Princeton
University (1994).

Primary Examiner—Ronald C. Capossela
Attorney, Agent, or Firm—Hoffmann & Baron, LLP

A method and apparatus for accumulation ol hyperpolarized 129Xe is disclosed. The method and apparatus ol the invention enable the continuous or episodic accumulation ol flowing hyperpolarized 129Xe in frozen form. The method also permits the accumulation ol hyperpolarized 129Xe to the substantial exclusion ol other gases, thereby enabling the purification ol hyperpolarized 129Xe. The invention lurther includes 129Xe accumulation means which is integrated with 129Xe hyper polarization means in a continuous or pulsed flow arrangement. The method and apparatus enable large scale production, storage, and usage ol hyperpolarized 129Xe for numerous purposes, including imaging ol human and animal subjects through magnetic resonance imaging (MRI) techniques.

41 Claims, 1 Drawing Sheet

1

CRYOGENIC ACCUMULATOR FOR SPIN-
POLARIZED XENON-129

This invention was made with Government support under Grant Nos. DAAH04-94-G-0204, DAMD1794J4469, and F49620-94-0466. The Government may have rights in this invention.

BACKGROUND OF THE INVENTION

The invention relates to apparatus and methods for hyperpolarizing a noble gas. Specifically, the invention relates to methods and apparatus for manufacturing and accumulating significant quantities of a hyperpolarized noble gas in a continuous manner.

Nuclear magnetic resonance (NMR) is a phenomenon which can be induced through the application of energy against an atomic nucleus being held in a magnetic field. The nucleus, if it has a magnetic moment, can be aligned within an externally applied magnetic field. This alignment can then be transiently disturbed by application of a short burst of radio frequency energy to the system. The resulting disturbance of the nucleus manifests as a measurable resonance or wobble of the nucleus relative to the external field.

For any nucleus to interact with an external field, however, the nucleus must have a magnetic moment, i.e., non-zero spin. Experimental nuclear magnetic resonance techniques are, therefore, limited to study of those target samples which include a significant proportion of nuclei exhibiting non-zero spin. A highly preferred such nucleus is the proton (1H), which is typically studied by observing and manipulating the behavior of water protons (1H20) in magnetic fields. Other nuclei, including certain noble gas nuclei such as 3He and 129Xe, are in principle suited to study via NMR. However, the low relative natural abundance of these isotopes, their small magnetic moments, and other physical factors have made NMR study of these nuclei difficult if not impossible to accomplish.

One important consideration in studying noble gas nuclei via NMR is that they normally yield only a very low NMR signal intensity. It is known, however, that the spin polarization of such noble gases as 3He and 129Xe can be increased over natural levels, i.e., populations of these isotopes can be artificially "hyperpolarized", to provide a much larger NMR signal. One preferred hyper polarization technique is known as spin exchange hyper polarization. Without describing this technique in exhaustive detail, in this scenario a noble gas is hyperpolarized via interaction with an alkali-metal vapor, such as rubidium, which itself has been polarized by absorption of laser energy of an appropriate wavelength. The polarized rubidium transfers its polarization to the noble gas through a phenomenon known as spin exchange transfer. The end result is that the noble gas becomes "hyperpolarized", i.e., more polarized than it would otherwise be. Details of the theory underlying the spin exchange hyper polarization technique are available in the literature.

While well established as a theoretical phenomenon, the actual practice of spin exchange hyper polarization has proven to be something of an art. The production and handling of hyperpolarized noble gases is not only logistically difficult, it is expensive as well. Moreover, due to the experimental nature of spin exchange studies, the production of hyperpolarized noble gases has typically been undertaken only on a small scale. Exquisite craftsmanship is typically required, involving expertise in a variety of fields including lasers, electronics, glass-blowing, ultra-high vacuum pump

2

operation, high-purity gas handling, as well as nuclear magnetic resonance spectroscopy.

For example, the production of a single sample of hyperpolarized noble gas has typically involved the fabrication of

5 a single-use sealed glass cell with a volume capacity of only a few tens to a few hundred cubic centimeters. Such cells have required delicacy in manufacture, yet their quality, as measured by their tendency to depolarize the noble gas, has not always been predictable. Moreover, use of such cells for

10 spin exchange requires that they be sealed with the alkali metal present therein. This has meant that care must be taken to remove impurities which can cause oxidation of the metal and consequent ruination of the cell. Other problems arise in the glass itself which can depolarize the noble gas faster than

15 it can be polarized. For study of polarized noble gas by NMR techniques, the sealed cell must be cracked open or destroyed to release the hyperpolarized gas into the NMR spectrometer. Proceeding to the next sample has required repeating all of these steps, including fabricating and filling

20 a new glass cell, which might or might not have similar qualities, resulting in a tedious and often unpredictable procedure.

Middleton established for the first time the possibility of making sealed pumping cells capable of containing larger

25 quantities of a noble gas for hyper polarization by the spin exchange technique. Middleton H., The Spin Structure of the Neutron Determined Using a PolarizeifHe Target, Ph.D. Dissertation, Princeton University (1994). Even so, the reliability of the procedures described in this publication

30 have not proven to be suited to routine use, in that sampleto-sample variability has remained a problem. Moreover, there is no disclosure in this document of any method of making refillable cells or cells which could be used on a continuous or flowing basis without significant rehabilita

35 tion. Accordingly, while progress in cell manufacture has occurred, the art has not provided means for making refillable or continuous flow spin exchange pumping cells.

It has also been known that hyperpolarized 129Xe can be frozen but yet retain a significant proportion of its polariza

40 tion. Indeed, it is known that freezing 129Xe can actually prolong the polarization lifetime beyond that which can normally be achieved by keeping the 129Xe in a gaseous state. Accordingly, sealed glass cells containing small amounts of hyperpolarized 29Xe have been frozen, stored,

45 and later thawed (sublimed) for use. See, e.g., Cates et al., Phys. Rev. Lett. 65(20), 2591-2594 (1990). The Cates document projects that small amounts (up to about 1 g/hr) of 129Xe could be accumulated, but provides no practical indication of how such a result might be achieved. This

50 paper also fails to provide any indication of whether the accumulation of larger quantities of frozen 129Xe would be possible.

Alternatively, a publication by Becker et al., Nucl Inst. &Meth. in Phys. Res. A, 346:45-51 (1994) describes a

55 method for producing hyperpolarized 3He by a distinctly different polarization method known as metastability exchange. This approach requires the use of extremely low pressures of 3He, i.e., about 0.001 atm to about 0.01 atm, and does not involve the use of an alkali metal; the 3He is

60 polarized directly by the laser. Significant accumulation of hyperpolarized 3He by this method is limited by the necessity of using huge pumping cells (i.e., about 1 meter long) and then compressing the gas to a useful level. The Becker et al. publication discloses an ingenious but technically

65 difficult approach which employs large volume compressors made of titanium for compressing the gas to about atmospheric pressure. Unfortunately, manufacture and operation 3

of such a system requires great engineering skill, limiting the reproducibility and operability of the system on a routine basis. The apparatus described by Becker et al. also requires significant amounts of floor space and cannot be moved. The Becker et al. paper also avoids the use of alkali metals in the 5 pumping cells, and does not disclose any method of producing hyperpolarized noble gas by spin exchange. Hence, the Becker et al. paper does not resolve the complexity of manufacturing pumping cells in which an alkali metal is employed. As a result, this publication does not describe or 10 suggest any method or apparatus related to the production and delivery of arbitrarily large or small quantities of hyperpolarized noble gas by spin exchange.

It was recently demonstrated that hyperpolarized noble gases can be imaged by nuclear magnetic resonance imaging :5 (MRI) techniques. See U.S. Pat. No. 5,545,396. In addition, because the noble gases as a group are inert and non-toxic, it was found that hyperpolarized noble gases can be used for MRI of human and animal subjects. As a result, there exists a growing need for the generation of larger quantities of 20 hyperpolarized noble gases. Moreover, because of medical and veterinary concerns, controlled uniformity and reliability in the purity of the gases and the amount of hyper polarization have become necessary. Also, the need for convenient and reliable generation of these hyperpolarized 25 gases has become important for use in a clinical setting in which medical technicians, having little or no specific training in the laboratory techniques described above, are still able to provide discrete or continuous hyperpolarized noble gas samples to subjects undergoing MRI. 30

In view of the above considerations, it is clear that the apparatus and methods in use in the existing art are limited in a number of ways. For example, the existing art does not provide any practical means for refilling a spin exchange polarization chamber (cell) once it has been used. Most 35 current chambers are either permanently sealed after the first filling or have been refilled with at best unsatisfactory results. Thus, it would be of benefit to develop means for effectively refilling a pumping chamber, or even for optically pumping in a continuous flow mode in the same 40 chamber, so as to decrease costs of materials and personnel.

Moreover, even successful fills for the permanently sealed cells used previously were accomplished via a significantly different system. In the past, an expensive ultra-high vacuum 4J system, with either oil-free pumps or cryotrapped oilcontaining pumps, has been required in order to produce a sufficiently clean apparatus for filling high quality polarization chambers. Such a system is expensive (about $30,000), not very compact (3 ft by 6 ft footprint), and requires high 5Q maintenance by a trained vacuum technician. Anew system, requiring only minimal maintenance and capable of being operated without specialized knowledge of vacuum technology, would be desirable. Also, a system having a more convenient size would be useful in clinical settings. 5J

In addition, there has been no practical way to produce hyperpolarized gas in a continuous fashion. For each spin exchange hyper polarization procedure, a new sealed sample has had to be prepared and introduced into the hyper polarization apparatus. It would, therefore, be desirable to 60 develop a system which overcomes this limitation to provide means for continuous hyperpolarization of flowing noble gas.

Systems for producing hyperpolarized gases have also been quite bulky, typically requiring separate rooms for their 65 installation. Such systems are not transportable or installable as a single piece of apparatus in a room used for various

4

other purposes. Thus, small, convenient hyperpolarizers would be advantageous. Also transportable systems would be of benefit in situations where space is a critical consideration.

Also, there has previously been no convenient way to store substantial quantities of hyperpolarized noble gases, especially 129Xe, for later distribution in discrete quantities of arbitrary amount (up to tens of liters of gas at atmospheric pressure). It would be important to overcome this limitation as well, to provide apparatus for continuous accumulation of a hyperpolarized noble gas, as well as storage and controlled release of the hyperpolarized gas on an as-needed basis, while still retaining substantial quantities of hyperpolarization.

SUMMARY OF THE INVENTION

Accordingly, as a result of the invention, there is now provided an improved apparatus and method for accumulating a hyperpolarized noble gas. In particular, there is provided apparatus for accumulating large quantities of high purity hyperpolarized noble gas for use in magnetic resonance imaging. Also a method and apparatus for continuous generation and accumulation of hyperpolarized noble gas, as well as for storage of large quantities of hyperpolarized noble gas.

The apparatus of the invention includes an accumulator system which permits the accumulation of hyperpolarized 129Xe in a continuous or semi-continuous mode. The accumulator system enables hyperpolarized xenon to be flowed through a cryotrapping reservoir and trapped efficiently and selectively as xenon ice before escaping. The accumulator also permits other gases to be passed through the system, thereby serving to selectively concentrate the hyperpolarized xenon. The accumulator preferably employs a cold trap reservoir, in which the trap is cooled to a temperature at or below the freezing temperature of xenon. The xenon passing into the accumulator then deposits efficiently on the walls of the accumulator reservoir in frozen form. Moreover, the accumulator permits xenon flowing within the reservoir to deposit on top of previously deposited xenon, thereby permitting the continuous or semi-continuous accumulation of the frozen gas. Because the solid form of hyperpolarized 129Xe has a much longer polarization lifetime than the gaseous form, the accumulator can serve as a storage device, allowing the accumulation of significant quantities of hyperpolarized gas for use at a later time. The invention further provides a method of using the variously described apparatus.

Optionally, the accumulator reservoir may be removable from the cooling apparatus and disconnectable from inflow and outflow conduits. Thus, a removable accumulation reservoir cartridge may be removed and stored separately in another cooling or refrigerating apparatus. In this way, the operation of the accumulator can be continued by installing another of the removable reservoir cartridges.

For purposes of accumulation of hyperpolarized 129Xe according to the invention, the hyperpolarization of xenon is preferably performed using xenon supplied in a gas mixture. The gas mixture includes xenon including at least a natural isotopic abundance of 129Xe. In addition, for purposes of enhancing the efficiency of spin-exchange polarization, a quenching gas, such as nitrogen or hydrogen, is also included in the gas mixture to suppress fluorescence of the alkali metal during the optical pumping procedure. It has now been observed that the hyperpolarization of high partial pressures of xenon is not as efficient as desired, i.e., high