SCALABLE CONTINUOUS-WAVE ION LINAC PET RADIOISOTOPE SYSTEM

Abstract

A continuous wave ion linear accelerator PET radioisotope system is disclosed. The system includes a high brightness H.sup. ion source, a continuous wave RF quadrupole structure, and continuous wave RF interdigital structures to accelerate the ion beam to about 14 MeV. A high energy beam transport system is also described that includes a photo-detachment beam splitter and a magnet lattice for forming the proton beam into a beam having a Waterbag beam profile. The system also includes one or more targets upon which the proton beam is incident. The targets are either a high power metallic target oriented at about 10 degrees or a low thermal conductivity target oriented at about 35 degrees. The invention includes a method of producing PET isotopes by use of the systems described.

Claims

1. A continuous wave ion linear accelerator PET radioisotope system comprising: a high brightness H ion source that produces a multi-milliampere beam of H ions; a continuous wave radio frequency quadrupole structure that accelerates the multi-milliampere H ion beam to about 1 MeV; one or more continuous wave radio frequency interdigital structures that accelerate the multi-milliampere H ion beam to about 14 MeV; a high energy beam transport system that transforms the accelerated H ion beam to a uniform transverse profile proton beam; one or more high power target stations that include a target capsule upon which the uniform transverse profile proton beam is incident to produce PET radioisotopes; and a target capsule transfer system for recovery of the produced PET radioisotopes.

2. The system of claim 1, wherein the high energy beam transport system comprises one or more laser photo-detachment beam splitters that transform a portion of the accelerated H ion beam to a Gaussian profile proton beam.

3. The system of claim 2, wherein the high energy beam transport system further comprises one or more non-linear focusing magnet lattices that defocus the Gaussian profile proton beam into the uniform transverse proton beam profile.

4. The system of claim 1, wherein the uniform transverse profile proton beam is incident on the high power target station along an axis.

5. The system of claim 4, wherein the high power target station is windowless with axially symmetric cooling inlet and outlet connections, and having a cooling system working fluid of either water or eutectic gallium tin alloy, the target plate further comprising a face that includes a substrate upon which the uniform transverse profile proton beam axis is incident.

6. The system of claim 5, wherein the high power target station comprises a metallic target plate oriented at about a 10 degree glancing angle relative to the transverse uniform proton beam axis.

7. The system of claim 6, wherein the metallic target plate is made of copper and the substrate is yttrium.

8. The system of claim 6, wherein the metallic target plate is made of silver and the substrate is nickel.

9. The system of claim 5, wherein the high power target station comprises a low thermal conductivity target plate oriented at about a 35 degree glancing angle relative to the transverse uniform proton beam axis.

10. The system of claim 9, wherein the low thermal conductivity target plate is made of iridium and the substrate is tellurium oxide.

11. The system of claim 9, wherein the low thermal conductivity target plate is made of iridium and the substrate is copper selenide.

12. A method of producing PET radioisotopes comprising: introducing a multi-milliampere H ion beam into a continuous wave linear accelerator via a low energy beam transport to form a low emittance, multi-milliampere H ion beam; focusing the H ion beam via a continuous wave radio frequency quadrupole to accelerate the low-emittance, multi-milliampere H ion beam to about 1 MeV; accelerating the focused H ion beam to an energy of about 14 MeV; transporting the accelerated H ion beam to a high energy beam transport system via a beam pipe; transforming a portion of the accelerated H ion beam into a Gaussian profile proton beam; forming the Gaussian profile proton beam into a uniform transverse profile proton beam; transferring the uniform transverse profile beam to a target station that includes a target capsule, the target capsule contains a target plate that includes a substrate upon which the uniform transverse profile proton beam is incident, thereby producing one or more PET radioisotopes; cooling the target plate with a coolant working fluid of either eutectic gallium tin alloy or water; transferring one or more of the target capsules from the target station to one or more radiochemical processing hot cells; and opening the target capsule to remotely remove the target plate within the radiochemical processing hot cell to recover the produced PET radioisotopes.

13. The method of claim 12, wherein the PET radioisotope produced is Zirconium 89 (89Zr) and the target substrate is yttrium 89 (89Y).

14. The method of claim 12, wherein the PET radioisotope produced is Iodine 124 (124I) and the target substrate is tellurium 124 enriched oxide (124TeO2).

15. The method of claim 12, wherein the PET radioisotope produced is Bromine 76 (76Br) and the target substrate is selenium 76 enriched copper selenide (Cu276Se).

16. The method of claim 12, wherein the PET radioisotope produced is Copper 64 (64Cu) and the target substrate is nickel 64 (64Ni).

17. The system of claim 1, wherein the produced PET radioisotope is Zirconium 89 (89Zr).

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0021] FIG. 1 is a simplified schematic illustration (a side elevation) of the Continuous-Wave (CW) PET radioisotope production structure (system) of the present invention. In this figure, the components which make up this system are illustrated lying substantially along, and in alignment with, a horizontal line which defines the operational axis (the beam axis) of the system 1B.

[0022] FIG. 2 presents the simplified schematic from an overhead (top) view of the system components which are also shown in FIG. 1. In FIG. 2, the two-dimensional nature of the beam axis in the HEBT 36 portion of the system of the invention is made apparent at the position of each laser-wire beam splitter component.

[0023] FIG. 3 presents a reduced scale overhead (top) view of the system as installed in a facility, explicitly demonstrating the scalability in a configuration with six high power target stations portion of the system of the invention.

[0024] FIG. 4 presents, on a slightly larger scale than that which is employed in FIG. 1, a fragmentary, isolated, more detailed side-elevational view of the DC volume-cusp negative ion source portion of the system of the invention.

[0025] FIG. 4A presents a still further enlarged, fragmentary, isolated, three-dimensional perspective, top view of the ion injector portion of the DC volume-cusp H ion source.

[0026] FIG. 5 is a fragmentary, isolated, opened up side-elevational view of just the ion injector portion of the DC volume-cusp H ion source shown in FIG. 4A.

[0027] FIG. 6 is a slightly enlarged, three-dimensional perspective view of the linear accelerator picturing the DC volume-cusp H ion source, LEBT, box-like radio-frequency quadrupole (RFQ), RF coupler MEBT, and radio-frequency interdigital (RFI) accelerating structures portion of the system of the invention.

[0028] FIG. 6A is a TRACE 3D particle-tracking simulation showing a number of beam parameters of the 14 MeV CW ion linac.

[0029] FIG. 7 is a slightly enlarged, fragmentary, isolated overhead (top) view of a segment of the HEBT portion of the system of the invention, showing the laser-wire beam splitter and non-linear focusing magnet lattice components which make up this portion of the system of the invention.

[0030] FIG. 8 is an isolated block diagram depicting the functional characteristics of the laser-wire beam splitter components.

[0031] FIG. 9 is a multi-trace graph depicting the flat-topping operation of the non-linear focusing magnet portion of the system of the invention on the Gaussian proton beam profile. The flat-topped beam increases power acceptance of the high-power target plate portion of the system of the invention.

[0032] FIG. 10 is an enlarged, fragmentary, isolated, top view of a high-power eight-sector collimator 48 for receiving the flat-topped beam profile of FIG. 9.

[0033] FIG. 10A is an enlarged, fragmentary, isolated front view along the beam axis 1B of the high-power eight-sector collimator 48.

[0034] FIG. 10B is an enlarged, fragmentary, isolated cross-sectional view of the high-power eight-sector collimator 48.

[0035] FIG. 10C is a three-dimensional, top perspective view of the high-power eight-sector collimator 48

[0036] FIG. 11 is a fragmentary, isolated, upper perspective view of the high-power target station portion of the system of the invention.

[0037] FIG. 12 is an enlarged, fragmentary, isolated schematic illustration of the target capsule for the high-power metallic target capsule portion of the system of the invention.

[0038] FIG. 13 is an enlarged, fragmentary, isolated schematic illustration of the target capsule showing the axial symmetry of the coolant inlet and outlet channels.

[0039] FIG. 14 is an enlarged, fragmentary, isolated schematic illustration of the high-power metallic target plate showing the elongated, elliptical deposition area for the target material.

[0040] FIG. 15 is an enlarged, fragmentary, isolated schematic illustration of the high-power metallic target plate showing the increased surface area of the cooling fins. This area is plated by chromium for corrosion resistance.

[0041] FIG. 16 is an enlarged, fragmentary, isolated, schematic illustration of the target capsule for the low thermal conductivity oxide target capsule portion of the system of the invention.

[0042] FIG. 17 is an enlarged, fragmentary, isolated, schematic illustration of the target plate for the low thermal conductivity oxide target capsule portion of the system of the invention. The depressions in the top view of the target plate are filled with enriched target substrate.

[0043] FIG. 18 is an enlarged, fragmentary, isolated schematic illustration of the low thermal conductivity target plate portion of the system of the invention showing the increased surface area of the cooling fins.

[0044] FIG. 19 is an enlarged, fragmentary, isolated schematic illustration of the blower assembly for the target capsule transfer system portion of the system of the invention.

[0045] FIG. 20 is an enlarged, fragmentary, isolated schematic illustration of the divertor assembly portion of the system of the invention. It allows for the loading/unloading of up to 6 target stations to/from up to 6 hot cell destinations by the target capsule transfer portion of the system of the invention.

[0046] FIG. 21 is an enlarged, fragmentary, isolated schematic illustration of a typical hot cell for receiving the target capsule for processing.

[0047] FIG. 21A is an enlarged, fragmentary, isolated schematic illustration of the target capsule remote handling assembly portion of the system of the invention. It allows for opening and closing the target capsule using master/slave remote manipulators within a hot cell for radioisotope processing.

DETAILED DESCRIPTION

[0048] The following description is merely exemplary in nature and is not intended to limit the present disclosure, application, or uses. It should be understood that throughout the drawings, corresponding reference numerals indicate like or corresponding parts and features.

[0049] Before the present invention is described in further detail, it is to be understood that the invention is not limited to the particular embodiments described, and as such may, of course, vary. It is also to be understood that the terminology used herein is for the purpose of describing particular embodiments only, and is not intended to be limiting, since the scope of the present invention will be limited only by the appended claims.

[0050] A number of materials are identified as suitable for various aspects of the invention. These materials are to be treated as exemplary and are not intended to limit the scope of the claims. Although any methods and materials similar or equivalent to those described herein can also be used in the practice or testing of the present invention, a limited number of the exemplary methods and materials are described herein.

[0051] It must be noted that as used herein and in the appended claims, the singular forms a, an, and the include plural referents unless the context clearly dictates otherwise.

[0052] In general, the meaning of the various terms and abbreviations as used herein is as they are generally used and accepted in the art, unless otherwise specified. In order to aid in the understanding of the invention, specific meanings of several terms are provided.

[0053] As used herein, large scale production when applied to longer lived PET isotopes zirconium-89, iodine-124, and copper-64 means the provision of sufficient radioactivity to meet the demand of one million or more patient doses per annum.

[0054] The term high current as applied to beam current means a beam current of two milliamperes or greater. Similarly, the term multi-milliampere refers to currents of at least about two milliamperes. High power acceptance when applied to target stations requires managing the heat fluxes associated with high beam currents. The term high power is used to refer to power acceptance ranging from ten to many tens of kilowatts (kW).

[0055] Brightness applied to ion source output is a figure of merit defined as the number given by output beam current divided by the square of the beam emittance. Emittance means the two dimensional area in ion beam phase space that is enclosed by an equal intensity contour line enclosing 90% of the total beam current, normalized. The term high brightness as applied to ion source output means a brightness figure of three or greater.

[0056] As used herein, the term low emittance refers to a normalized total emittance of about 1.0 mm-mrad.

[0057] A continuous-wave (CW) linac refers to the continuous in time beam current pulse repetition rate for a 100% beam duty factor.

[0058] A beam having a Waterbag beam profile refers to a beam having a two-dimensional transverse beam profile with an rms uniformity of 5% or better.

[0059] Several types of beam transports are described and used herein. A low-energy beam transport (LEBT) permits high acceptance (95%) of the injected H.sup. beam at an ion energy of a few tens of keV. A medium-energy beam transport (MEBT) refers to beam having an ion energy of about 1 MeV. A modular high-energy beam transport (HEBT) refers to a transport at the final ion energy of about 14 MeV.

[0060] The term low thermal conductivity refers to a thermal conductivity of less than about 3 Wm.sup.1K.sup.1.

[0061] In general terms, the invention is a continuous wave ion linear accelerator PET radioisotope system that includes a high brightness H.sup. ion source that produces a multi-miliampere beam of H.sup. ions. A continuous wave radio frequency quadrupole structure accelerates the H.sup. ion beam to about 1 MeV. One or more continuous wave radio frequency interdigital structures further accelerate the beam to about 14 MeV. A high energy beam transport system transforms the accelerated beam to a uniform transverse profile proton beam. The uniform proton beam is incident on one or more high power target stations that include a target capsule where the PET radioisotopes are produced. The radioisotopes are recovered in a target capsule transfer system. The invention also includes a method of producing PET radioisotopes.

[0062] The high energy beam transport system includes one or more photo-detachment beam splitters that transform at least a portion of the accelerated H.sup. ion beam into a Gaussian profile proton beam. One or more non-linear focusing magnet lattices defocus the Gaussian profile proton beam.

[0063] The high power target station is windowless with axially symmetric cooling inlet and outlet connections. The cooling system working fluid is typically water or eutectic gallium tin alloy. The target station includes a face having a substrate upon which the proton beam is incident. When the target station includes a metallic target plate, the plate is oriented at a glancing angle of about 10 degrees relative to the proton beam. In several exemplary embodiments, when the target plate is copper, the substrate is yttrium; when the target plate is silver, the substrate is nickel.

[0064] As an alternative to the metallic target plate, the plate can be a low thermal conductivity plate, which would then be oriented at a glancing angle of about 35 degrees relative to the proton beam. In an exemplary embodiment, when the target plate is iridium, the substrate is tellurium oxide or copper selenide.

[0065] Examples of PET radioisotopes that can be produced by the system and method of the present invention include the following: Zirconium 89 (.sup.89Zr) when the target substrate is yttrium 89 (.sup.89Y); Iodine 124 (.sup.124I) when the target substrate is tellurium 124 enriched oxide (.sup.124TeO.sub.2); Bromine 76 (.sup.76Br) when the target substrate is selenium 76 enriched copper selenide (Cu.sub.276Se); and Copper 64 (.sup.64Cu) when the target substrate is nickel 64 (.sup.64Ni).

[0066] The present invention is a continuous-wave, fault-tolerant RF negative ion linac deployed in a scalable configuration with laser-photo-detachment beam splitter and non-linear focusing magnet arrangement with field strengths to provide for beam uniformization to the Waterbag beam profile. In this context, and as will be seen, in addition to utilitarian uniqueness which is expressed in this invention through the capability for economically advantageous scalability for large-scale production of PET radioisotopes previously envisioned as technically not achievable, by its high-energy CW ion linac and particle-beam-transport components which make up portions of the system of the invention, this special nature leads to a uniquely flexible, high-reliability configuration simultaneously accommodating multiple high-power acceptance target stations.

[0067] These characteristics provide the system with the ability to be: (a) rapidly and economically scalable for centralized supply of longer-lived PET radioisotopes for widespread availability to support translation of radio-immunoimaging protocols (immuno-PET) into the clinical standard of care; (b) highly reliable in CW ion linac operations due to modularity of design in RF power amplification which grants fault-tolerance, and allows faulty modules to be hot-swappable, further minimizing down-time; (c) capable of delivering high beam currents simultaneously to multiple target stations; (d) capable of uniformization of Gaussian beam to a Waterbag beam profile to maximize power acceptance; (e) capable of withstanding extreme thermal stress on target plates during irradiation; (f) able to minimize personnel exposure by highly reliable transfer of irradiated target plates to hot cells for radiochemical processing; and (g) adopted to the limitations of master/slave remote manipulators for isotope recovery in hot cells to meet As Low as Reasonably Achievable (ALARA) exposure guidelines for handling of radioactive materials. Some or all of these may be achieved through implementation of the present invention.

[0068] The radioisotope production components of the proposed system are arranged initially in a straight-linear fashion, progressing through the system from the low-energy end to the high-energy end. The components of the system of the invention include: (a) a DC volume-cusp ion injector source capable of high-brightness H ion beams; (b) a low-energy beam transport (LEBT) permitting high acceptance (95%) of the injected H beam; (c) a radio frequency quadrupole (RFQ) structure; (d) an RF coupler for medium-energy beam transport (MEBT); (e) a series of radio frequency interdigital (RFI) linac structures; (f) in an arc fashion, a modular high-energy beam transport (HEBT) incorporating a laser-photo-detachment (laser-wire) beam splitter and non-linear focusing magnet lattice; (g) a high-power acceptance, sandwich-type metallic target station; (h) a high-power acceptance, low thermal conductivity (e.g., oxides) target station; and (i) target capsule transfer components, terminating in the radioisotope processing hot cell with target capsule opening/closing device adapted for master/slave remote manipulation of the irradiated target plate.

[0069] To aid in appreciating certain technical background information, the contents of U.S. Pat. No. 6,777,893, which may be helpful in understanding the nature of the present invention, is hereby incorporated by reference into this disclosure.

[0070] The present invention utilizes an RFQ linac structure for high-current (10-200 mA) operations. RFQ structures have small ion beam diameters, and because the transverse focusing of the RFI linac structure is electric, similar to that of the RFQ structure, the RFI will have the same small diameter ion beam. Consequently, matching the ion beam from an RFQ into the RFI linac is straightforward. The RFI linac structure offers improved capabilities to capture and accelerate low-energy ion beams. With its improved beam quality and higher RF power efficiency for high-duty-cycle operations, such as continuous-wave (CW), the RFI linac can accelerate much higher currents than achievable with cyclotron technology.

[0071] The CW ion linac PET radioisotope system of the present invention provides acceleration to ion energies in the range of 10 MeV to 15 MeV of high-intensity CW negative ion beams, with beam currents on the order of 10 mA readily obtainable. By melding electric focusing-long recognized as the best method for focusing low-energy ion and proton beamsand acceleration of a high-current, CW, negative ion beam, this leads to an important advance in longer-lived PET radioisotope production. This performance advance enables the system of the present invention to provide the multi-milliampere proton beam currents for the irradiation of high-power acceptance targets that are essential for clinical translation of medical applications using longer-lived PET isotopes, such as immuno-PET with zirconium-89 and iodine-124.

[0072] The CW ion linac PET radioisotope system of the present invention combines a high-brightness negative ion (H.sup.) ion source with the strong RF focusing of the RFQ linac and the efficient acceleration of the RFI linac to provide compact, commercially-viable, linear acceleration of a multi-milliampere CW H beam at a relatively low cost.

[0073] The CW ion linac PET radioisotope system of the present invention includes modularity in the design of the CW ion linac and high-energy beam transport system (HEBT) for a highly reliable platform technology for medical isotope production. This design provides longer-lived PET radioisotopes requiring protons in the 0.1 to 0.2 times the velocity of light. With radioactive half-lives between 16 hours and 4.18 days, these longer-lived isotopes and their radiopharmaceuticals are suited to centralized production and distribution. The CW ion linac radioisotope system provides for high reliability in its operations by its modular, fault-tolerant RF power architecture, and its axially-symmetric target capsule, which eliminates the frequent failures associated with axial positioning systems in target capsule transfer.

[0074] The CW ion linac PET radioisotope system of the present invention may provide for the simultaneous irradiation of up to six high-power acceptance targets. This feature enables the scalability required for centralized supply of late-stage clinical trials involving large patient cohorts, as well as post-approval unit dose manufacturing for millions of PET scans annually. The modular HEBT system incorporates laser-photo-detachment beam splitters which preserve the parent H.sup. beam parameters, and dipole bending magnets to separate the parent H.sup. and proton beams for transport to the subsequent optical interaction cavity for the former, and a non-linear focusing magnet system for the latter.

[0075] The CW ion linac PET radioisotope system of the present invention may provide for the uniformization of the proton fluence rate incident on the target plate by incorporation of a non-linear magnet focusing system. The multiply-scattered, Gaussian beam profile undergoes flat-topping to the uniform Waterbag beam distribution for effective management of peak heat fluxes by the high-power acceptance target stations.

[0076] The CW ion linac PET radioisotope system of the present invention may also provide the high-power target station with the capability to manage extremely high heat fluxes (exceeding 1 kw/cm.sup.2). The use of eutectic GaSn alloy as the cooling system working fluid for the high-power metallic target plate results in linear heat removal all the way up to its boiling point of 1200 C., permitting multi-milliampere target currents and power acceptance in excess of 50 kW. This is achieved while operating far from the melting points of the target metals.

[0077] The CW ion linac PET radioisotope system of the present invention may provide high-power acceptance for the low thermal conductivity target station, with no beam window (windowless) to result in a multiply-scattered, Gaussian, beam profile requiring rastering of the beam by x-y steering magnets. Rather, high-power acceptance in excess of 10 kW is achieved through the non-linear focusing magnet system's flat-topped (Waterbag) beam profile which mitigates peak heat fluxes for the low thermal conductivity (3 Wm.sup.1K.sup.1) tellurium-124 enriched tellurium oxide. Tellurium oxide also possesses a relatively low melting point at 733 C., and the crystalline form used is brittle, cracking easily due to thermal stresses. Tellurium oxide is preferentially used for iodine-124 production due to the simplicity of thermo-chromatographic recovery.

[0078] The CW ion linac PET radioisotope system of the present invention may also provide axially-symmetric target capsules common to both the high-power metallic target and low thermal conductivity target. Axial symmetry (about beam axis 1B) of the coolant inlet and outlet channels obviates the need for an axial positioning system for such inlet and outlet. These axial positioning systems provide a frequent mode of failure in target capsule transfer systems.

[0079] The CW ion linac PET radioisotope system of the present invention includes targets that may be shielded in all directions. As is well known to those generally skilled in this art, it is important that an overall device like that which is disclosed herein be adequately shielded to prevent exposure to radiation with respect to people who work near and around such a system. In one implementation of the present invention, only the incoming beam pipe is a source of neutron leakage, so there is a substantial reduction in decommissioning costs and environmental impacts due to radio-activation of concrete and other nearby structural materials.

[0080] The CW ion linac PET radioisotope system is rapidly scalable, comprised of CW H.sup. linac, high-energy beam transport (HEBT)laser photo-detachment (aka laser wire) beam splitter, non-linear focusing magnetsfor up to six targets. For incident heat flux of up to 1 kW/cm.sup.2 on the target plate, we use eutectic GaSn alloy cooling with Ar gas overpressure (to prevent Ga oxidation). The extreme-power metallic target incorporates a unique high-power eight-sector collimator and elongated octagonal target plate into a titanium capsule (rabbit) for transfer to the target processing hot cell by pneumatic transfer system. The axially-symmetric rabbit needs no axial positioning, the most common failure mode of such systems. The target's large surface area, low critical angle, explosion welded yttrium cladding, and finned back offer the optimum combination for eutectic GaSn alloy coolant. It is believed that this is the first system to successfully use the Waterbag beam profile-since our HEBT uses octupole/dodecapole magnets for flat-topping the Gaussian beam.

[0081] As will be seen from the description of the invention set forth below, the system of the present invention directly and effectively addresses various performance, scalability, economic, and reliability issues.

[0082] As will be seen, the present invention offers a long-lived PET radioisotope production system which is rapidly scalable, highly reliable by its fault-tolerant modular design, and provides for economical, widespread availability of these isotopes.

[0083] The characteristics of the system of the present invention are directed to (a) the invention's proposed unique continuous-wave (CW) negative ion linear accelerator with optimal 14 MeV final energy; (b) the invention's fault-tolerant accelerator design, lending high reliability to its operation; (c) the invention's capability to accelerate multi-milliampere total beam currents to such optimal final energy; (d) such invention's unique HEBT design capability for simultaneous delivery of proton beam to up to six target stations, while it also preserves all desirable parent beam parameters using laser photo-detachment as the beam splitter mechanism; (e) such invention's unique uniformization of the transverse proton beam profile incident on the target plates; (f) such invention's two target capsule designs provide for uniquely high power acceptance for both iodine-124 and for zirconium-89 radioisotope production; and (g) such invention's target capsules incorporate axial symmetry of cooling channels, obviating the need for an axial positioning system, and affording high reliability in post-irradiation capsule transfer from target station to hot cell for radioisotope processing.

[0084] The two radioisotopes which are most commonly used in immuno-positron emission tomography, or immuno-PET, are iodine-124 and zirconium-89, with half-lives of 4.18 days and 3.27 days, respectively. A widely-recognized, longstanding technological barrier has precluded their reliable, economical, widespread availability. The present invention for a CW ion linac PET radioisotope system provides for their highly scalable, reliable, and economical supply.

[0085] Various objects, advantages and novel features, and further scope of applicability of the CW ion linac PET radioisotope system will be set forth in part in the detailed description to follow, taken in conjunction with the accompanying drawings, and in part will become apparent to those skilled in the art upon examination of the following, or may be learned by practice of the invention. Other objects and advantages of the invention may be realized and attained by means of the instrumentalities and combinations particularly pointed out in the appended claims.

[0086] The CW ion linac PET radioisotope system comprises a high-brightness DC volume-cusp H.sup. ion source as the injector for the LEBT, which focusing the H.sup. beam for acceleration by the RFQ. The RFQ structure has been proven capable of accelerating up to 30 mA of beam current at high-duty-factor. An RF coupler, or MEBT, between the RFQ and RFI accelerating structures performs matching of the beam ellipses. The RFQ and RFI linac tanks receive their RF power for acceleration of ions from modular RF amplifier (RFA) power blocks, which are fault-tolerant: each RFA power block can deliver its full rated power with up to 10% failed modules. The number of RFA power blocks may be increased to provide power for acceleration of still higher beam currents. The system, though initially configured to accelerate 6 mA of H.sup. beam, may be rapidly and economically scaled to accelerate 10 mA, or more, to the final energy of 14 MeV.

[0087] Modularity of design and fault-tolerance are highly-desired advantages in scaling production capacity to meet demand for production of medically urgent isotopes for cancer diagnosis and therapy. Scalability in isotope production capacity is further enabled by the modular nature of the HEBT portion of the system of the invention. Highly efficient beam splitting, while maintaining beam parameters that are key figures of merit for beam quality, i.e., size, divergence, energy, energy spread, and phase spread, is an important process which is incorporated in the HEBT design of the present invention. HEBT modules may be added to support up to six high-power target stations. Optimizing target yields for the long-lived PET radioisotopes identified as critical needs by the National Cancer Institute and IAEA is addressed in the present invention by the non-linear focusing magnet lattice and high-power target station portions of the system of the invention.

[0088] Turning attention now to the drawings, referencing first of all FIG. 1, shown is a side elevation view of the CW ion linac PET radioisotope system 22. The beam axis 1B shows the direction of charged particle beam transmission in the system 22. Components 23-33 comprise the CW ion linac and deliver 14 MeV H.sup. small diameter (low emittance) beam via beam pipe 37 to the high-energy beam transport (HEBT) 36 portion of the system 22. Each HEBT 36 system is comprised of beam splitter components 38-41 and non-linear focusing magnet lattice 43 for delivery of Waterbag profile beam via beam pipe 37 connected by bellows 47 and eight-sector high-power collimator 48 to high-power target station 49. Each high-power target station 49 has a target station cooling system 54, typically comprised of coolant working fluid reservoir, pump, and heat exchanger with connecting tubing, flow switches, gauges, and control system 55.

[0089] Completing a description of what is shown in FIG. 1, indicated by block forms at 28, 35, and 55, are appropriately programmed digital computers which are operatively connected to various electronically controllable components in system 22. These direct the overall operation of the system, while the computers, their operational software, and their specific connections to system 22, do not form any part of the present invention.

[0090] Shown in FIG. 2 is an overhead, top view, of the CW ion linac PET radioisotope system 22. The modular nature of the HEBT 36 system is depicted, with two such portions of the system of the invention displayed in FIG. 2. Each HEBT 36 beam pipe 37 delivers 14 MeV Gaussian proton beam to a non-linear focusing magnet lattice 43 for uniformization to the Waterbag beam profile as shown in FIG. 9. Simultaneous irradiations at multiple high-power target stations 49 may be performed as shown in FIG. 2 and FIG. 3.

[0091] The full CW ion linac PET radioisotope system, as shown in FIG. 3, demonstrates the highly scalable, modular nature. In FIG. 3, the CW ion linac portion of the system 22 is configured with a complement of five RF amplifier (RFA) and internal power blocks 31/31A. Controls at 35 insure that the phase of the RF power at each drive loop for RFA 31 be the same within a few degrees. With the multiple drive loop configuration shown in FIG. 3, the signals required for this control are the summation of the forward RF power on all drive lines and the summation of the reflected RF power on all drive lines. Resonance will be signaled by minimizing the voltage standing wave ratio (VSWR). Control of the phases will be signaled by maximizing the sum of the forward RF powers. The system 22 beam pipe 37 supplies a complement of six FIG. 2 HEBT 36 beamlines with six high-power target stations 49. FIG. 3 also depicts six target capsule transfer tubing lines 53, serviced by two target capsule divertors 57 short-connected by target capsule transfer tubing 53, with one target capsule blower/suction transfer station 56, for target capsule transfer to/from six radiochemical processing hot cells 58 by a manifold comprised of target capsule transfer tubing 53.

[0092] Shown in FIG. 4 is an enlarged, three-dimensional perspective illustration of the complete DC volume-cusp H.sup. ion source system portion of the invention, with ion injector assembly 23. Vacuum pumps 24 include mechanical roughing pumps for initial pump down and backing pressure for the turbomolecular pump mounted to the side of the ion source vacuum box. A low voltage rack 27 supplies the vacuum pumps, gauges, and controls. A high-voltage (HV) rack 26 is comprised of power supplies for ion source operation of the plasma lens 68, extraction lens 69, filament, and plasma heating arc in the ion source body 67, mounted in Faraday cage 25, which has an interlock system to isolate personnel from the high-voltage portion 26 of the system when energized.

[0093] Referring to FIG. 5, an enlarged, cross-sectional view of the DC volume-cusp H.sup. ion source injector 23 and its internal structure 23A; this provides H.sup. ions at up to 15 mA at 25-30 keV with a 4 normalized emittance of less than 1.00 mm-rad emittance. The DC volume production H.sup. ion source internal structure 23A with multi-cusp plasma confinement, shown in FIG. 5, produces a high-brightness, stable beam of ions in the 25-30 keV energy region. As is known to those skilled in the art, the ion source 23A maintains a confined plasma from which H.sup. ions are preferentially extracted. The plasma confinement region is the ion source body 67, supplied with H.sub.2 gas, where heating by a DC filament results in thermionic emission of electrons, ionizing the gas. The ion source body 67 is maintained at negative 25-30 kV. Volume-production processes exhibit lower emittance and are most useful for the generation of high-brightness H.sup. beams. In volume production, the dominant reactions in a confined hydrogen plasma are given by this two-step process, referred to as dissociative attachment,

e+H.sub.2(v=0).fwdarw.H.sub.2(v>0)+e

H.sub.2(v>0)+e.fwdarw.H.sub.2.sup.(v>0).fwdarw.H.sup.+H

where v is the vibrational quantum number. Whereas the first reaction peaks for energetic electrons (100 eV), the yield of H.sup. in the second reaction of the process is maximized for lower energy, cold, electrons (1 eV). A bias voltage applied to the plasma lens 68 increases production of secondary electrons. This necessitates a magnetic filter to eliminate the energetic electrons from the cold region. In the extraction region, a permanent magnet filter in the extraction lens 69 removes any electrons from the beam before being brought to its final energy of 25-30 keV by the ground lens 70. The permanent magnet filter obviates the need for an electron trap in the LEBT 29 to prevent these electrons from being accelerated along with the H.sup. ion beam by the RFQ 30. The beam waist 71 is created in the vicinity of the ground lens 70. The vacuum box 72 incorporates a small x-y steering magnet to ensure the beam ellipse is properly centered for acceptance by the LEBT 29.

[0094] Shown in FIG. 6 is an enlarged, three-dimensional perspective, top view of an embodiment of the CW ion linac PET radioisotope system 22. The DC volume-cusp H.sup. ion source 23 provides a highly collimated multi-milliampere beam of H.sup. ions. For the best injection into the RFQ 30, an azimuthally symmetric (round), strongly convergent beam is needed. The beam from the ion source 23 is round and diverging. The beam is allowed to expand at the entrance into low energy beam transport system (LEBT) at 29, which focuses the H.sup. beam into the ideal shape and steers the H.sup. ion beam into RFQ 30 which has a capture efficiency of 95%. The ion source has vacuum pumps 24 which provide the high vacuum for beam transmission between the ion source, through the LEBT 29, to the entrance of the RFQ 30. The RFQ linac 30 uses RF electric fields to focus and accelerate the CW H.sup. ion beam from around 25 keV to 1.0 MeV. This small diameter H.sup. ion beam from the RFQ 30 is injected by the MEBT (RF coupler) at 32, into the multiple-tank RFI linac at 33. A simple, resonance-control system, based on temperatures of the linac structures, keeps the relatively broad-band RFQ 30 in resonance with the RFI tanks 33. This simplicity obviates the need for an accurate frequency source, low-level RF power amplifier chain, and the associated power supplies. The RFI linac, with its RF electric focusing and acceleration fields maintains the small beam diameter while the CW H.sup. beam is accelerated to its final energy of 14 MeV. Also shown in FIG. 6, numerous water cooling channels are machined directly into the wall of the RFI linac tank 33. Controls are used in the multiple-tank RFI linac 33 for controlling the relative phase of the accelerating gradients in each tank such that incoming H.sup. ion bunches arrive at the center of each accelerating gap at the proper phase for acceleration. The H.sup. ion beam can then be injected into the high-energy beam transport (HEBT) 36 system portion of the invention where it is utilized for production of long-lived PET imaging isotopes for medical applications. The basic parameters for a preferred embodiment of the CW ion linac are presented in Table 1.

TABLE-US-00001 TABLE 1 Accelerated Particle H- Resonant Frequency 200 MHz Ion Source Output Energy 27 keV RFQ Output Energy 1.0 MeV RFI Output Energy 14.0 MeV Beam Current 10.0 mA Beam Duty Factor 100% RFI Tanks 4 RFI Linac Length 5 m RF Cavity Power (Peak) 590 kW RF Beam Power (Peak) 140 kW RF Total Power (Peak) 720 kW RFQ Transmission Efficiency 95 % RFI Transmission Efficiency 100 % Total Length, Including Ion Source 8.26 m

[0095] In FIG. 6A is a plot of a TRACE3D particle-tracking simulation of the 14 MeV CW ion linac portion of the system of the invention 22. The phase spaces and profiles are for the entire sample of ten thousand particles. The output of the 14 MeV RFQ/RFI linac demonstrates the small beam diameter achieved, with a normalized, rms emittance of 1.2 mm mrad. The FIG. 6A values of the CW ion linac's emittance, Twiss alpha parameters (ax and ay, both unitless), and Twiss beta (bx and by, cm/mrad) are used to determine the octupole and dodecapole magnet strengths in the non-linear focusing magnet lattice 43. The Twiss parameter values can be changed through the use of a single quadrupole 40 lens if needed.

[0096] Now referring to FIG. 7 is a slightly enlarged, top view of the modular HEBT 36 portion of the system of the invention 22. Beam pipe 37, a drift tube, transfers H.sup. beam from the RFI linac 33 to the optical box 38, in which the interaction cavity for the CW laser 39 beam crosses the H.sup. beam axis 1B. Photo-detachment of the first electron of the H.sup. ions for a fraction of the linac beam yields neutral H.sup.0 beam, which passes through a hybrid boron-carbon (HBC) foil in foil holder 41. High stripping efficiency results in near 100% of the H.sup.0 beam exiting foil holder 41 as protons. This is necessary to extract protons, while it Coulomb multiple scatters the beam, ensuring a Gaussian beam distribution. Residual H.sup.0 beam is collected in a water-cooled beam dump 42. Dipole bending magnets 40 provide horizontal deflecting forces to direct both the remaining H.sup. ions to the next optical box 38 via beam pipe 37, and the proton beam to the non-linear focusing magnet lattice 43. The proton beam is manipulated by two octupole magnets 44, to form a beam with a two-dimensionally uniform transverse intensity distribution. The octupole magnets 44 are separated at the x- and y-beam waist points by a doublet of quadrupole magnets 45 to suppress the betatron coupling which the octupoles introduce. The use of two octupole magnets 44 enables the tuning of octupole-focusing effects independently in the two transverse dimensions, minimizing this coupling effect. To obtain an rms uniformity of 5% or better, the flat-top Waterbag distribution, requires the higher order multipole of the dodecapole magnet 46. The flat-topped proton beam is delivered to the high-power target station 49 via the high-power eight-sector collimator 48, which provides for final shaping of the beam, feedback for beam steering, and essential safety interlock functionality for under-focused, over-focused, and slightly drifted beam conditions.

[0097] Referring to FIG. 8, a schematic representation of the beam splitter portion of the system of the invention reveals how the linac H.sup. beam undergoes charge exchange reactions to extract high current proton beam. The relativistic H.sup. beam from the linac 33 can be stripped by laser photo-detachment of the first electron since the threshold is only 0.75 eV. The H.sup. beam enters an interaction cavity in optical box 38 via beam pipe 37, as shown in FIG. 7, where the laser beam crosses the linac beam multiple times due to high reflectivity surfaces, shown in FIG. 8.

[0098] As an example it can be shown that this technique is effective for low energy (10-15 MeV) beams because the yield is inversely proportional to , the particle velocity in units of the velocity of light. Here, we have =0.17 for H.sup. ions at 14 Mev. The detachment cross-section is 3.510.sup.17 cm.sup.2 at 1.17 eV (1064 nm). In the H.sup. rest frame, the relativistic shifted, or Lorentz boosted, laser photon energy is E.sub.CM=E.sub.L [1 cos(.sub.L)], with and the Lorentz parameters of the H.sup. beam and .sub.L is the laboratory angle of the laser beam relative to the H.sup. beam. For a Gaussian laser beam with N.sub.L photons intercepting a Gaussian H.sup. beam of current I.sub.b at angle .sub.L, the yield Y.sub.1, the number of neutral hydrogen atoms produced per laser-H.sup. beam crossing, is given by the following approximation:

[00002]$Y_1=\frac{I_b.Math.N_L}{e.Math..Math..Math..Math.c}.Math.\frac{1-.Math..Math.\cos .Math..Math.{}_L}{\sin .Math..Math.{}_L}.Math.\frac{{}_N\left(E_{\mathrm{CM}}\right)}{2.Math..Math..Math..Math.{}_b.Math.{}_L}.Math.{}_{-}^{}.Math.e^{-\left(x^2/2.Math..Math.{}_b^2\right)}.Math.e^{-\left(x^2/2.Math..Math.{}_L^2\right)}.Math.\mathrm{dx}=\frac{I_b.Math.N_L}{\sqrt{2.Math..Math.}.Math.e.Math..Math..Math..Math.c}.Math.\frac{1-.Math..Math.\cos .Math..Math.{}_L}{\sin .Math..Math.{}_L}.Math.\frac{{}_N\left(E_{\mathrm{CM}}\right)}{\sqrt{{}_b^2+{}_L^2}}$

where .sub.b and .sub.L are the transverse rms sizes of the H.sup. and laser beams normal to the plane of incidence, and .sub.N(E.sub.CM) is the photo-detachment cross-section at photon energy ECM in the H.sup. rest frame. For illustration, in one embodiment using a 10 mA, 14 MeV H.sup. beam, N.sub.L=2.6810.sup.17, .sub.L=85, c=510.sup.9 cm/s, .sub.N (E.sub.CM)=3.510.sup.17 cm.sup.2, .sub.b=.sub.L=0.2 cm. Yield is enhanced above the fractional yield F.sub.1 of a single crossing by reflecting the laser beam through the H.sup. beam a number of times, N, to give the approximate fractional yield,

F.sub.N=1(1F.sub.1).sup.N

with Y.sub.1 at 1.510.sup.8 H.sup.0 atoms per 10 ns CW mode-locked laser 39 pulse, 0.5 mA, for N=8 mirror reflections, the total H.sup.0 current is 5 mA, effectively splitting the beam. The neutral H.sup.0 beam maintains nearly identical parameters as the parent H.sup. beam, including size, divergence, energy, energy spread, and phase spread.

[0099] The remaining electron is stripped by a long-lifetime hybrid boron-carbon (HBC) foil in foil holder 41 with an efficiency of nearly 100% given by the fractional yield as a function of H.sup.0 beam velocity in units of the speed of light, parametrized as follows:

[00003]$H^{+}=1-H^0$$H^0=\left(\frac{a}{a+b-c}\right)\left[\exp \left(-\mathrm{cdt}\right)-\exp \left(-\left(a+b\right).Math.\mathrm{dt}\right)\right]$

[0100] where

[0101] a=0.4791018 cm.sup.2/.sup.2

[0102] b=0.00851018 cm.sup.2/.sup.2

[0103] c=0.1871018 cm.sup.2/.sup.2

[0104] d=foil density in atoms/g

[0105] t=foil thickness in g/cm.sup.2

[0106] =relativistic factor (0.17 at 14 MeV)

[0107] The remainder of the neutral beam (<100 W) is directed to a beam dump 42. As shown in FIG. 2, FIG. 3, and FIG. 7, the parent H beam, viewed along beam direction 1B, is diverted by a dipole bending magnet 40 providing a 10 transverse kick to the left, while the proton beam receives a 5 transverse kick to the right from a second dipole bending magnet 40, directing the proton beam to the non-linear focusing magnet lattice 43 for flat-topping to the desirable Waterbag beam profile.

[0108] Referring to FIG. 9 is a multiple-trace plot of scaled proton beam intensity as a function of distance from the center of the beam axis 1B. FIG. 9 presents the effects of proton beam manipulation by the non-linear magnet lattice 43 on the CW ion linac 33 Gaussian beam profile, shown as the relatively peaked NO MULTIPOLE FOCUSING trace. Beam optics considerations by those skilled in the art for generating a 2D uniform distribution require octupoles for both transverse directions. Nonlinear multipole magnetic forces inevitably couple the vertical and horizontal motion. The nonlinear focusing magnet lattice 43 places the x-direction octupole 44 near the y-direction beam waist, with two focusing quadrupoles 45 next. The second octupole 44, spreading in the y-direction, is near the x-direction beam waist point. However, the uniform beam area is still surrounded by an overshoot wall of higher intensity, as shown by the OCTUPOLE FOCUSING trace in FIG. 9. The area of uniform intensity decreases with increased octupole field strength. This is because the tail of the Gaussian distribution is folded inside the octupole field when the octupole magnet is excited. Higher odd-order fields are required for perfect uniformization of the Gaussian beam. Addition of the dodecapole magnet 46 results in the flat-topped, or Waterbag, beam profile indicated by the OCTUPOLE+DODECAPOLE trace in FIG. 9, for delivering a constant proton fluence rate to each collimator 48 at each target station 49 of the CW ion linac PET radioisotope system 22 shown in FIG. 3.

[0109] In FIG. 10 is a side view of the high-power eight-sector collimator 48. It permits dividing the high incident heat load of the steep edge of the Waterbag beam profile among eight water-cooled sectors. The sectors are machined from low-alloyed aluminum to minimize radio-activation. The sectors are shown in FIG. 10A, a cross-sectional view of the high-power eight-sector collimator 48. With high-power metallic target currents as high as 4.0 mA, at 56 kW of power acceptance, each of the sectors should accept at least 250 A, or 3.5 kW of power. For practical ease of operation, the enhanced power acceptance is needed to provide for individual adjustment of octupole magnet 44 strengths in each direction. The power acceptance of the collimator 48 is also a necessary feature to provide safety signals for under-focused, over-focused, and slightly drifted beam conditions (off-center relative to beam axis 1B) of the proton beam to the accelerator control system (ACS) computer 35 to minimize the potential for equipment damage due to such high beam intensities.

[0110] Shown in FIG. 11 is a high-power acceptance target station 49 portion of the system of the invention. The target station 49 and bellows 47 are electrically isolated from the adjustable stand 50 and the incoming beam pipe 37 for target current measurement. The adjustable stand 50 provides fine x-, y-, and z-axis position adjustment for the target station relative to the incident proton beam delivered via beam pipe 37. Target capsule transfer tubing 53 delivers the remotely assembled target capsule comprised of high-power metallic target plate 60, target capsule male portion 61, and target capsule female portion 62 to the target capsule receiving station 52 of the target station 49. Pneumatic actuators of target capsule receiving station 52 position the assembled target capsule in the target capsule locking mechanism 51, which with pneumatic force closes the rubber-like elastomer seals of the axially symmetric cooling channels of target capsule male portion 61 and target capsule female portion 62 into the target station 49 for irradiation by proton beam via beam pipe 37. Following irradiation, pneumatic unlocking by target capsule locking mechanism 51, extraction of the assembly of high-power metallic target plate 60, target capsule male portion 61, and target capsule female portion 62 to target capsule receiving station 52 positions it for transfer via target capsule blower/suction station tubing 56A.

[0111] In FIG. 12 is a cross-sectional side view of the high-power metallic target capsule comprised of target capsule male portion 61 and target capsule female portion 62 remotely assembled in radiochemical processing hot cell 58 using target capsule remote manipulator opening/closing device 59 by means of twist-to-lock bayonet-type fittings. From the left, it accepts 35 mm diameter Waterbag beam onto target plate 60 at a 10 glancing angle relative to the beam direction 1B. Axial symmetry about beam axis 1B means that no axial positioning system is required. Such positioning systems represent one of the most common failure modes of target capsule transfer systems. The back side of target plate 60 is cooled with a eutectic GaSn alloy coolant in the turbulent flow regime. Indium wire seals are used for semi-permanent assembly of high-power metallic target capsule female portion 62, minimizing maintenance needs, and limiting the use of rubber-like elastomers. Such elastomers, used in O-ring sealing technology, have inherent thermal performance limitations and are degraded by ionizing radiation.

[0112] In FIG. 13 is a cross-sectional end view of the high-power metallic target capsule female portion 62 showing the axial symmetry of the coolant inlet and outlet channels relative to beam direction TB.

[0113] Next in FIG. 14 is a top view of the high-power metallic target plate 60, and the elongated, elliptical, cross-hatched area represents the target substrate location. The target plate 60 is inclined at a 100 glancing angle between the target and the beam axis 1B. Typical embodiments utilize either a natural yttrium layer, explosion welded to an oxygen-free, high conductivity (OFHC) copper target plate due to its very high electrochemical potential, or nickel-64, electroplated to a silver target plate.

[0114] In FIG. 15 is a bottom view of the high-power metallic target plate 60, showing the cooling fins used to increase the available surface area for heat transfer to the eutectic GaSn alloy coolant. The bottom of the target plate is coated with chromium to protect the copper from corrosion by gallium. Chromium possesses excellent heat conductivity at 94 Wm.sup.1K.sup.1. Due to the corrosion rate of chromium due to exposure to gallium at 390 C. capable of dissolving a 15 m thick layer in 24 hours, a 30 m thick protective chromium layer is used on the back side of target plate 60 in FIG. 15.

[0115] Multiphysics simulations can predict how target plate 60 performs under real-world conditions. Here, finite element analysis (FEA) utilized a 3D mesh for high-power metallic target plate 60 with 110.sup.6 nodes and 5.710.sup.6 elements. Eutectic GaSn coolant at 15 dm.sup.3 min.sup.1 across the back of the target plate 60 as shown in FIG. 15 yields a pressure drop of 7.0 bar through the target capsule. This figure is safe for the yield strength of the copper used in the high-power metallic target plate 60, so no deformation is anticipated. A 14 MeV proton beam at 4.5 mA, 63 kW beam power, on the high-power metallic target plate 60 yields a maximum yttrium temperature of 453 C. The maximum coolant temperature is 307 C. This is far from the melting point of any of the target metals, but is approaching the 327 C. operating limit for the high-temperature O-ring seals. A reasonable margin to allow for beam current fluctuations is 4.0 mA, 56 kW beam power, on the high-power metallic target plate 60. This is over 20-fold greater power acceptance than present 2 kW cyclotron targets.

[0116] Referring now to FIG. 16 is a cross-sectional side view of the low thermal conductivity (e.g., oxides) target capsule, consisting of low thermal conductivity target plate 63, low thermal conductivity target capsule male portion 64, and low thermal conductivity female portion 65. The target capsule is windowless, designed to accept 35 mm diameter Waterbag profile beam from the left side of the low thermal conductivity target capsule male portion 64. Axial symmetry about beam axis 1B means that no axial positioning system is required. Such positioning systems represent one of the most common failure modes of target capsule transfer systems. The low thermal conductivity target plate 63 is inclined at a 35 glancing angle relative to the beam axis 1B. This glancing angle reduces the proton beam energy to 4 MeV, below the nuclear reaction threshold, reducing the thermal load on the oxide target substrate in the rectangular depressions, as shown in FIG. 17, of the low thermal conductivity target plate 63. As a water-cooled target, the limiting factor is the Critical Heat Flux (CHF), at the boiling point of water. Beyond this limit, heat flow is unstable, vapor blankets the heated surface of the back side of the low thermal conductivity target plate 63, and temperatures jump to very large values. This regime is governed by film boiling and radiative heat transfer, called burnout.

[0117] Multiphysics simulations predict how target plate 63 performs under real-world conditions. Here, finite element analysis used a 3D mesh having 110.sup.6 nodes and 5.610.sup.6 elements for low thermal conductivity target plate 63. The back side of target plate 63 is cooled using 20 dm.sup.3 min.sup.1 of water at 20 C. inlet temperature. The pressure drop through the target capsule under these conditions is 3.5 bar. With inlet water pressure at 10 bar, a pressure drop of 3.5 bar, the pressure of the water behind the target plate 63 is 6.5 bar. The boiling point of water at this pressure is 162 C. Multiphysics simulations with 14 MeV proton beam at 1 mA beam current yield a maximum water temperature of 157 C. The maximum tellurium oxide temperature is 410 C., much less than the 550 C. limit, where the vapor pressure of the tellurium oxide is only 510.sup.4 mbar, ensuring enriched tellurium-124 losses are less than 0.1%. Cyclotron technology has not provided a low thermal conductivity target capable of more than 150 A with dual beam. The CW ion linac PET radioisotope system can deliver up to 6 mA total beam current to six low thermal conductivity target stations, a 20-fold increase in power acceptance.

[0118] Shown in FIG. 17 is a top view of the low thermal conductivity target plate 63, with rectangular depressions machined in the iridium target plate surface for deposition of the tellurium-124 enriched oxide target material. These depressions prohibit overflowing of the molten target material during thermo-chromatographic recovery of radioiodine.

[0119] Next, in FIG. 18 is a bottom view of the low thermal conductivity target plate 63, showing the cooling fins machined in the surface of the back of the target plate to increase the surface area in contact with coolant for improved heat transfer.

[0120] Referring now to FIG. 19 is a three-dimensional, top perspective view of the target capsule transfer blower/suction station 56. This provides the pneumatic forces for transfer of the target capsule assemblies through the target capsule transfer tubing 53 to/from the target stations 49 from/to the radiochemical processing hot cells 58.

[0121] In FIG. 20 is a three-dimensional, top perspective view of the target capsule divertor assembly 57. It provides the mechanism for selecting the sending/receiving stations of target station 49 and radiochemical processing hot cell 58. By connecting two divertor assemblies 57, one reversed as shown in FIG. 3, one can transfer target capsule assemblies between any of a maximum of six target stations 49 and six radiochemical processing hot cells 58. An appropriately configured manifold of target capsule transfer tubing 53 provides for flexibility to meet various isotope scalability requirements.

[0122] Next, FIG. 21 is a cross-sectional, side view of a radiochemical processing hot cell 58, illustrating the trenched entry and exit points for the target capsule transfer tubing 53, and the radiochemical processing hot cell internal structure 58A includes a target capsule opening/closing device 59 adapted to the limitations of master/slave remote manipulators.

[0123] Finally, FIG. 21A is an enlarged, three-dimensional, top perspective view of the target capsule remote manipulator opening/closing device 59. Shown in the open position is a target capsule assembly, with high-power metallic target plate 60, high-power metallic target capsule male portion 61, and high-power metallic target capsule female portion 62. Numeric values and ranges are provided for various aspects of the implementations described above. These values and ranges are to be treated as examples only and are not intended to limit the scope of the claims.

[0124] While the invention has been described in conjunction with specific exemplary implementations, it is evident to those skilled in the art that many alternatives, modifications, and variations will be apparent in light of the foregoing description. Accordingly, the invention is intended to embrace all such alternatives, modifications, and variations that fall within the scope and spirit of the appended claims.