A report written by a US nuclear consultancy firm calls the bluff on the Howard government’s claim that a new nuclear reactor is required in Australia to produce medical isotopes.

The report, titled Alternatives to a 20 Megawatt Nuclear Reactor for Australia, was written by nuclear physicist Dr. Robert Budnitz and energy and technology consultant Dr. Gregory Morris, both from the California-based consultancy firm Future Resources Associates. It was commissioned by the Sutherland Shire Council, which is fighting the federal government’s plan to build a new reactor in the southern Sydney suburb of Lucas Heights.

The production of isotopes for medical diagnosis and therapy has been the major selling point for the planned new reactor. This argument has necessitated and endless stream of misinformation since there is no doubt that Australia could do as most other countries do - rely on particle accelerators such as cyclotrons to produce some isotopes, and to import others.

The government’s case was not helped when the president of the Australian and New Zealand Association of Physicians in Nuclear Medicine, Dr. Barry Ellison, admitted that he did not know that the existing HIFAR reactor at Lucas Heights was shut down for maintenance for three months last year.

The Future Resources Associates report, released on August 8, argues that “importation of radioisotopes and more extensive use of accelerators for isotope production represent a viable alternative to the building of a new reactor in Australia.”

The report argues that this approach would have several benefits when compared with the plan for a new reactor, including reduced generation of radioactive waste, possible cost benefits, similar or better employment prospects, and better intellectual property opportunities (arising from the development of accelerator technology).

The report argues that Australia should develop and implement an accelerator method to produce technetium-99m, the isotope used in about 75% of all nuclear medicine procedures. Accelerators, such as the cyclotrons located in Sydney and Melbourne, already produce 20% of the isotopes used in nuclear medicine, and implementation of an accelerator method of technetium-99m production would dramatically reduce the reliance on reactors, whether located in Australia or overseas.

Future Resources Associates specifically identifies two accelerator methods of producing technetium-99m which could be pursued in Australia - an electron accelerator method developed at the Idaho National Engineering and Environmental Laboratory, and a proton accelerator method developed by researchers at the University of California and elsewhere.

The Future Resources Associates report argues that, “Development of accelerator based production of Tc-99m would probably require a one-to-two year effort involving several person-years of work, and a few million dollars of investment. The pay-off would be that Australia would develop and possess valuable expertise in a nearly radioactive waste and proliferation free route to the production of the world’s medically most important radioisotope.”

Some scientists, along with environmental and anti-nuclear groups, have long argued for research in Australia into accelerator methods of technetium-99m production. Public pressure forced the Australian Nuclear Science and Technology Organisation (ANSTO), operator of the Lucas Heights nuclear plant, to agree to a collaboration with the University of California in 1993.

However, ANSTO quietly shelved the proposed collaboration and its treatment of Dr. Manuel Lagunas-Solar, head of the University of California research team, led Lagunas-Solar to write to Prime Minister John Howard in September 1997 saying, "It is my understanding that my work has been reviewed by ANSTO, without the benefit of my direct participation, and clearly using outdated and incomplete information. ANSTO also provided statements to Parliament based on information (also out of date) available through our internet site. Based upon a general analysis of ANSTO's review, I strongly feel that it does not provide an objective and balanced review of the actual status or the conclusions of our work."

ANSTO operates the National Medical Cyclotron in Sydney but has never used it to explore innovations such as cyclotron production of technetium-99m.

In 1994, the federal Labor government rejected thorough, costed plans to develop a cyclotron method of technetium-99m production.

Tracie Sonda, mayor of the Sutherland Shire Council, said the Future Resources Associates report will be forwarded to the federal and state governments and opposition leaders, seeking their responses. Sonda said, “Building and running another nuclear reactor will not just cost Australians half a billion dollars, it will cost us the chance to lead the world in nuclear medical technology. If we are truly serious about ‘innovation’ and ‘knowledge nation’, then this report cannot be ignored.”

Introduction

ANSTO, the Australian Nuclear Science and Technology Organisation, has owned and operated the 10 MW HIFAR nuclear reactor for more than 30 years in the Lucas Heights region outside of Sydney.  This reactor, which is used for a variety of purposes, is nearing the end of its useful operating life, and will be retired from service unless a substantial investment is made in the next few years to upgrade it to meet modern safety standards.  ANSTO has been promoting the concept of building a new 20 MW reactor at the same location, which would be used for essentially the same purposes as the present reactor.

Local opposition to the commissioning of a new nuclear reactor in Australia has led to a national policy debate over its merits.  One of the important services that is provided by the HIFAR is the production of radioisotopes for a variety of important applications in Australia.  Radioisotopes are taking on increasing importance in medical practice and other areas, and their future availability is one of the important issues that must be addressed in the debate.

Some of the radioisotopes that are currently made in HIFAR could be made using particle accelerators in Australia.  Accelerators are smaller and cheaper than nuclear reactors, and produce significantly less radioactive waste.  The remainder of the reactor-produced isotopes would have to be imported into the country from foreign sources, most likely North America and Europe.  Australia already participates in the international isotope market, importing a variety of radioisotopes into the country.  Radioisotopes are hazardous materials, and require special shipping and handling considerations.  This becomes a particular problem for radioisotopes that have short half-lives.  Nevertheless, importation of radioisotopes and more extensive use of accelerators for isotope production represent a viable alternative to the building of a new reactor in Australia.

Most of the increased radioisotope importation that would be associated with a decision not to go forward with a new nuclear reactor could be mitigated if Australia decided to get involved with the development of an accelerator approach to the production of Technetium-99m (Tc-99m).  Tc-99m accounts for almost ninety percent of the amount of radioisotopes used today in nuclear medicine.  All Tc-99m today is produced in nuclear reactors.  Two different approaches to its production using accelerators have been extensively developed in the U.S.  Australia might consider making a national commitment to the finalization of the development and commercialization of one of these technologies, which would probably require a one-to-two year effort involving several person-years of work.  The payoff would be that Australia would not only minimize the amount of increased importation of radioisotopes in the absence of a reactor, it would develop and possess valuable expertise in a nearly radioactive waste-free route to the production of the world’s medically most important radioisotope.

Table 1 shows a matrix of the attributes of the three basic alternatives available to Australia with respect to the procurement of radioisotopes:
- build a new reactor
- no reactor, but develop an accelerator method for the production of Tc-99m
- no reactor, no accelerator production of Tc-99m

The entries in the table are discussed in detail in the text that follows.

Production of Radioisotopes

A wide variety of radioactive isotopes are used for medical, industrial, research, and commercial applications. Radioisotopes are classified as hazardous materials that require specialized storage and handling techniques.  Radioisotopes are manufactured using highly specialized equipment at a limited number of locations in the world.  Many radioisotopes have short half-lives, which means that they must be manufactured continuously, and used as quickly as possible after they have been prepared.

Some radioisotopes can be produced by any one of several different nuclear reaction pathways, but the number of options is limited due to physical properties of the production process.  Radioisotopes are produced by irradiating target materials with nuclear particles.  The target atoms either transmute directly into the desired isotope, or a parent radioisotope is produced that decays and subsequently generates the desired product.  The latter is the case, for example, in the production of the important medical imaging isotope Technetium-99m.  In current practice reactors are used to produce Molybdenum-99, which is the parent of Tc-99m.  The Mo-99 is packed into generators that are shipped to users of Tc-99m.  Technetium is eluted from the generators as needed for use.

Two different technologies are used to provide the source of radiation for radioisotope production: nuclear reactors, which produce a flux of neutrons, and particle accelerators, which produce a flux of charged particles, usually protons, but sometimes electrons or other particles.  For the most part these technologies are complementary, allowing the broadest range of nuclear reactions to be carried out.  Most radioisotopes in current and expected use can be manufactured either only in a reactor, or only in an accelerator.  Some radioisotopes can be produced in either a reactor or an accelerator.

Nuclear production reactors used for radioisotope production are major pieces of capital equipment, often costing in the hundreds of millions of dollars.  Most of the reactors used in the world for the production of radioisotopes are government owned, and were built for multi-use purposes.  These reactors typically are not optimized for the production of radioisotopes, and radioisotope production has to compete with a variety of other applications for reactor time.

Particle accelerators used for radioisotope production tend to be much smaller and cheaper than nuclear reactors, typically costing in the tens of millions of dollars, and sometimes less.  Several private international pharmaceutical manufacturers own their own accelerators, which are dedicated to the production of radioisotopes.  Accelerators tend to produce significantly less radioactive waste than reactors, although they are not waste-free.

Value of domestic source of supply vs. imports

Should Australia decide to forego the building of a new nuclear reactor, a few of the radioisotopes that would have been produced by the reactor could be produced using existing or new accelerators in Australia.  The rest of the radioisotopes that would have been produced by the reactor will need to be imported, mostly from North American or European suppliers.  Most observers consider the international radioisotope supply network to be secure and possessed of sufficient production capacity that importation does not represent a substantial risk to Australia’s access to supplies of radioisotopes, at least for the foreseeable future.

Importation does, however, entail certain constraints that need to be taken into account in making national policy decisions.  Importation adds one-to-three days to the time between the production of a radioisotope product, and the point at which a user in Australia has access to the material, as compared to a domestic source for the same material.  For isotopes with short half-lives, a good deal of the product can be lost in shipment.  Shipping delays can compound the problem.  Isotopes with half-lives of a few hours or less simply cannot be imported into Australia.  On the other hand, Australia has a long history of successfully importing radioisotopes from international suppliers, including Tc-99m, the most important medical radioisotope.  Moreover, HIFAR operations include periodic long-term outages for maintenance, and these extended outages have been managed without supply disruptions to the country’s use of radioisotopes.

The Market for Radioisotopes in Australia

A great variety of radioisotopes are currently used in Australia, for a variety of different applications.  In many cases radioisotopes make processes and products possible that cannot be provided in the absence of a radioisotope material.  Many of the great advances that have been made in medicine over the past couple of decades, for example, are in the area of, or dependent on, nuclear medicine.  These trends are likely to continue.

Technetium-99m has become the most important radioisotope in the world, due to its special value in medical imaging.  Technetium is used in 80-90 percent of all nuclear medical procedures.  Tc-99m is currently produced in nuclear reactors.  Most of the world’s supply comes from a single supplier, MDS Nordion of Canada, which has recently commissioned two new “MAPLE” reactors that will be used to produce the bulk of their Technetium generators.  Australia currently produces the majority of its Technetium requirements domestically using the HIFAR reactor, and imports the remainder of its needs from foreign sources.  Many other radioisotopes are imported into the country as well.

Applications for Radioisotopes in Australia

Radioisotopes are used for medical, industrial, research and commercial applications.  Several isotopes have already assumed important roles in the Australian economy.  For example, thallium-201 is used for medical cardiac imaging, calcium-44 is used in bone growth studies, iridium-192 is used for nondestructive testing of construction and other materials, and americium-241 is used in smoke detectors.  In addition, rapid advances in nuclear medicine and other fields are focusing attention on a number of isotopes that have not previously been considered commercially important.

Table 2 illustrates the breadth of applications used for radioisotopes (IOM 1995).  Radioisotopes are used in everyday products, in routine industrial practice, and in medical applications for both diagnostic and therapeutic purposes.  Nuclear medicine continues to grow in its importance to modern medical technology.

Radioisotopes are used in medicine for both diagnostic and therapeutic applications.  Radioisotopes are particularly useful for imaging applications, and are crucial for such non-invasive imaging techniques as positron emission tomography (PET).  Isotopes used for imaging, such as Tc-99m, are administered as radiopharmaceuticals, which are biologically active molecules labeled with a radioactive isotope that target particular organs or parts of the body.  Radioisotopes are also being used increasingly for therapeutic applications, mainly cancer treatment.  Isotopes are directed to or inserted in the diseased tissue, where they deliver a gauged and localized radiation dose to the cancerous growth.   Radioisotopes used for therapeutic purposes are administered as radiopharmaceuticals, or in sealed packages that are surgically inserted.

More than 10 million diagnostic procedures and nearly 100 million laboratory tests are performed annually in the United States using radioisotopes (IAEA 1990).  In addition, radioisotopes are used therapeutically for more than 200,000 patients each year.  Table 3 shows some of the radioisotopes that are in common use for medical applications.

Tc-99m, the most important radioisotope for medical use

Technetium-99m is the workhorse of the nuclear medical imaging world.  The physical properties of Tc-99m allow it to be combined with a wide variety of substrates.  In this way it can be delivered to very specific target tissues and organs.  Tc-99m produces high-energy gamma rays, which makes detection easy, and its short half-life of 6 hours minimizes the radiation dose delivered to the patient.  Nuclear medical imaging is unique in its ability to analyze organ structure and functioning.  It allows the gathering of diagnostic information that would otherwise require surgery, or not be available at all.  It also promotes early detection and treatment of cancers and other problems.

Tc-99m has been used widely since the late 1960s, and its use continues to grow.  Tc-99m is the decay daughter product of Mo-99, an isotope with a half-life of 66 hours that is produced in nuclear reactors as a fission product of the neutron bombardment of a uranium target material.  This is the source of all of the Tc-99m used in the world today.  After irradiation, Mo-99 is recovered from the target, and packed into a production device from which technetium can be eluted in quite pure form as needed for preparing radiopharmaceuticals for nuclear medical procedures.  ARI produces Tc-99m generators in Australia from Mo-99 produced in the HIFAR reactor.  Domestic production accounts for approximately 80 percent of the current use of Tc-99m in Australia.  The remainder of the supply is imported from North America or Europe.

Domestic supply and importation for the Australian radioisotope market

Australia produces a wide variety of radioisotopes for its domestic market, but it is not completely self-sufficient.  Domestic radioisotopes are produced mainly by Australian Radioisotopes (ARI), a government-owned private enterprise that manufactures and distributes radioisotope products for the domestic and export markets.  ARI is organized within the framework of the Australian Nuclear Science and Technology Organisation (ANSTO).  ARI produces radioisotopes in the HIFAR nuclear reactor, and in several particle accelerators located in Australia.  The HIFAR reactor is aging and scheduled for retirement within the next decade.  In addition, some radioisotopes are produced in non-ARI accelerators in Australia.

Several international pharmaceutical companies, including Nycomed Amersham, Tyco-Mallinckrodt, and Syncor, market radioisotope products in Australia that are imported into the country.  Most of the imported radioisotopes are produced in North America and Europe.  In addition, Isoflex USA, a company that is an international marketer of stable isotopes from Russian supplies, is moving into the radioisotope business, and has expressed an interest in selling Russian-sourced radioisotopes in Australia.  Due to their classification as special materials, radioisotope imports into Australia require special handling and shipping considerations.  Generally, access to dependable supplies is not a major consideration.

Shipping and handling considerations are particularly important for radioisotopes that have short half-lives.  These materials decay quickly, so it is important to use them as rapidly as possible after their manufacture.  Due to Australia’s remote location, shipments from North America or Europe take a significant amount of time.  Most radioisotope imports are transshipped, meaning that they come on at least two different flights from the source country. Transshipping, and the special handling requirements and paperwork associated with the importation of hazardous materials, mean that importation typically delays the delivery of radioisotopes by 1-3 days as compared with the supply of material from domestic production sources.  For an isotope with a half-life of 66 hours, such as Mo-99, the parent molecule for Tc-99m, this means that a generator that is imported to Australia will provide 25-50 percent less technetium than an identical generator manufactured in Australia.  For isotopes with half-lives of a day or less, importation becomes almost unfeasible.

As a general rule, the proton-rich radioisotopes that are produced in accelerators have relatively short half-lives, while the neutron-rich radioisotopes that are produced by reactors have relatively longer half-lives.  As a result, most of the radioisotope products that can only be made in a reactor can be supplied by foreign sources should Australia not build a new nuclear reactor.  There are, however, important exceptions.  Moreover, the presence of a domestic source of these materials does confer important market and supply-security benefits for the country.

Production of Radioisotopes

By their nature radioisotopes are unstable materials that must be manufactured in order to be available for the various applications they serve.  Radioisotopes are created via nuclear processes in which atomic nuclei targets are irradiated, resulting in a nuclear transformation reaction.  The two most common types of transformations that are used for radioisotope production are initiated by neutron and proton bombardment.  Neutron fluxes are usually provided by nuclear reactors, while proton fluxes are produced in particle accelerators.

Production Options for Radioisotopes

Radioisotopes can be produced in either nuclear reactors, or in particle accelerators.  Nuclear reactors are used to perform transformations involving neutron reactions, while accelerators are used for proton or electron promoted transformations.  Of the broad range of radioisotopes used in nuclear medicine and other applications, some of the radioisotopes can only be produced in particle accelerators, some can only be produced in nuclear reactors, and some can be produced in either an accelerator or a reactor.   Table 4a shows a list of radioisotopes that are only made in accelerators, Table 4b shows a list of radioisotopes that are only made in reactors, and Table 4c shows a list of radioisotopes that can be made using either technology (Schenter, R., 2001).

It is interesting to note that Tc-99m, which is used for as much as 90 percent of all nuclear medical procedures, is listed as a reactor-only isotope.  All of the Tc-99m used medically in the world today is produced in reactors.  Due to the overriding importance of this isotope, a great deal of research has been done on developing alternative sources of supply.  Two different approaches using particle accelerators (discussed below) have been identified that show promise of leading to a commercial alternative to reactor production of Mo-99.  Once one of these methods has been fully proven, Tc-99m can be switched from Table 4b to Table 4c.

Use of accelerators to produce radioisotopes

The use of particle accelerators to produce artificial radioisotopes dates back to the beginnings of the accelerator era under Lawrence in Berkeley in the early 1930s.  The use of proton accelerators in particular provides a rich potential, and hundreds of isotopes of various types are at least technically feasible.  Other beam types, such as deuteron or alpha beams, can also be used and indeed have been.  Also, producing neutrons from a proton accelerator through spallation processes allows the production of different isotopes (see the next section.)  Finally, a high-energy electron beam can produce still other isotopes through gamma-n and gamma-p reactions.

Many important isotopes are produced in commercial quantities today using particle accelerators.  For some of these isotopes there is no other feasible way to make them, meaning that accelerators occupy a vital niche in the overall isotope-production scheme.  For those isotopes whose half-lives are so short that they cannot feasibly be imported, the only approach is a production facility located in Australia.  For the others, with long enough half-lives, importation is fully feasible, and indeed is the accepted approach today in Australia for many products.

Unfortunately, and crucially, there is a list of radioisotopes (Table 4b) that simply cannot be made by accelerator techniques.  Physics limitations govern many of these.  In some cases the nuclear reactions simply do not exist, or the physical cross-sections are too small to be of commercial interest.  In other cases, the production of the desired isotope is practical but its production is accompanied by other isotopes of the same element, which would require difficult separation processes to make enough material, or pure enough material, for practical use.

Spallation accelerators as a neutron source

In recent years the technology of spallation-source neutron production has received increasing attention around the world.  A few such facilities have existed for some time and all of the basic problems with their operation have been overcome, at least insofar as the accelerator-technology aspects are concerned.  The basic idea is to use a proton accelerator to bombard a heavy-element target, producing several neutrons for each proton in the incident beam.  These so-called spallation neutrons can then be collimated into a beam useful for scientific studies, or otherwise used to initiate nuclear reactions.  A host of potential nuclear reactions is feasible, depending on whether one desires fast neutrons emerging directly from the target, or thermal neutrons produced by slowing down the emerging fast-neutron flux in a thermalizing element such as water.

The number of feasible neutron reactions for producing radioisotopes is large, and using the fluxes available from such a facility the potential exists for producing all sorts of isotopes for medical or other applications.

One might envision two different approaches to the development of such facilities in Australia.  In the first, a spallation facility would be built and used as a multiple-function facility, for use in basic and applied scientific studies as well as for routine isotope production.  In the second, the spallation facility would be dedicated entirely (or mostly) to isotope production.  The former might make sense for Australia, but would require a national decision to build the facility for the broader applications to which it is suited.  Such a facility would be expensive and could be justified only through its broad range of applications.

Whether the latter approach (building a dedicated spallation facility in Australia dedicated to isotope production) makes sense is difficult to assess.  We know of no studies that have explored how such a facility might work in terms of size, cost, throughputs, other applications, and potential technical problems.

Production of Mo-99 (Tc-99m parent) in accelerators

Although essentially all of the Tc-99m used in nuclear medicine today is produced in nuclear reactors by the fission process, it is technically feasible to produce it using an accelerator to effect the desired nuclear transformation.  Two such schemes have been reported in the literature; however, neither of them has been developed to the point where a full-scale production facility exists.  These accelerator schemes both share an advantage in that far less nuclear waste is produced than in the existing reactor-fission scheme.  The entire Australian market for Tc-99m could be served by one or two accelerators of either variety.  The two accelerator schemes are technically quite different and hence are discussed separately in the following sections.

In the first approach, which has never been executed but whose individual elements have been studied extensively, an electron accelerator in the 40 MeV range would be used to produce a gamma-ray beam through bremsstrahlung in a high-Z target.  The gamma rays would strike a molybdenum target that has been highly enriched in the isotope Mo-100 (which occurs naturally in molybdenum only at the 13% level.)  The gamma rays convert the Mo-100 to Mo-99 (the radioactive parent of Tc-99m) directly by knocking a neutron out of the target, in a so-called gamma-n reaction.  The Mo-99 decays to Tc-99m with a 66-hour half-life, and the Tc-99m must then be extracted from the molybdenum target.  The scheme investigated at the Idaho National Engineering and Environmental Laboratory (INEL) would perform the extraction using a thermal-separation method (Bennett, R., et. al., 1999, Christian, et. al., 2000).

A project carried out several years ago at INEL explored the technical potential for this approach in great detail, including identifying the key technical difficulties.  There are a number of complexities in this approach, but through some very detailed experimental work supported by analysis, many of them have been overcome.  In particular, the approach seems capable of being tailored nicely to the actual scheduling needs of hospitals and other medical facilities for Tc-99m.  Also, the electron accelerator needed for this scheme is well within existing technology today, and the size and cost of the proposed facility are modest enough (a few million US dollars for the facility, about 1 million US dollars annual operating cost) that distributing two or three of these accelerators around Australia would be feasible from both a logistical and engineering perspective, and could perhaps meet Australia's needs for Tc-99m very nicely.

The Idaho group developed the process quite far, to the point that only a few technical problems remain unsolved.  Several impressive technical advances were made, so that the scheme as proposed optimizes the beam energy and production efficiency, makes an end-product with acceptably small impurities so as to meet the medical requirements, and produces the short-half-life Tc-99m in a form not very different from that now available using the reactor-production scheme.  Perhaps the most vexing problem that remains to be solved is to demonstrate a reliable thermal-extraction process for removing the desired Tc-99m from the molybdenum target.  Although promising technical advances have been made, this aspect definitely requires more experimental development.  Otherwise, the next logical step would be a full-scale demonstration of the approach by actually building and operating such an accelerator.

The Idaho group reported (Bennett, private communication, 2001) that interest was expressed by commercial enterprises, to the extent that quite serious explorations of the technical details were pursued by at least one firm, but in the end no commercial development of the process has occurred.

The proposed scheme has several advantages.  One major advantage is that by design the approach is aimed at producing Tc-99m in several small accelerator facilities (that could be distributed regionally) rather than in one larger national facility.  The projected cost per unit Tc-99m dose also seems to be in a reasonable range, although this cannot be known in detail until the additional technical work is completed that provides the full scientific and engineering basis for the scheme.

Our evaluation is that a few technical person-years of effort would be required to establish a firm enough technical foundation to allow this approach to be pursued to maturity.  Although the scheme shows great promise, and has important advantages, there is no guarantee at this stage that all of the technical issues can be resolved.  On balance, even if adequate resources were devoted to the development of this scheme, its full-scale technical feasibility would probably require a few calendar years of work.  Then a full-scale facility would need to be designed, built, and operated before this scheme could be pronounced as established.  This is an area in which Australia could make a major contribution.

Direct production of Tc-99m in accelerators

In this approach a proton accelerator would be used to produce Tc-99m directly from a target material.  The process involves bombarding a target of highly-enriched Mo-100, which undergoes a so-called (p,2n) reaction in which two neutrons are knocked out of the target nucleus.  This method is explored in a paper that describes experimental investigations using a cyclotron that accelerates protons in the 12-to-22-MeV range (Lagunas-Solar, M., and Zeng, N., 1997).

The yield of Tc-99m from the reaction is acceptably high, and the purity of the end-product meets or exceeds medical criteria.  The method has the potential to produce rather large quantities of Tc-99m using one-hour bombardment times, which is crucial since the 6-hour half-life of the Tc-99m means that the turn-around times in the accelerator must be very short indeed.

The direct-production approach for technetium eliminates the need for the second step (elution of Tc-99m from Mo-99) that is a central feature of the standard reactor-production method.  However, the key liability is that Tc-99m has a short 6-hour half-life, which requires that the production accelerator facility be relatively close to the medical end-use points.

A facility along the lines of the one explored in the cited paper would likely produce enough Tc-99m for all of Australia's needs.  However, having only one such facility in Australia could be a problem given the 6-hour half-life of the isotope and the transportation logistics (Perth is about one half-life away by air from the eastern cities.)

Modern proton accelerators with large beam currents are now available for this application, and the remaining elements of the proposed scheme seem feasible in principle; indeed, many of them have been demonstrated already or are not likely to be problematical.  One key issue is that the practicality of the approach depends on achieving high specific activities of Tc-99m in the target, since many of the important medical applications require such.  This has not yet been demonstrated fully although it seems feasible in principle.

Several technical and engineering issues remain to be addressed, and to do so would probably require several person-years of scientific and engineering effort.  The cost of the end product also needs to be understood better, as do the logistical issues involved in taking the Tc-99m as produced and transforming it for medical use.  Only then would it be feasible to build a production facility, assuming that each unknown issue can be worked-out favorably.

Sources of import supply for reactor-derived material

ANSTO currently operates a nuclear reactor (HIFAR), and a particle accelerator (NMC).  Both of these units are used for the production of radioisotopes, as well as for a number of other purposes.  ANSTO-produced radioisotopes are used to supply the domestic market for these materials, as well as for export sales.  The domestic market for radioisotope products cannot be fully supplied from ANSTO’s current production capabilities.  Approximately twenty percent of the country’s Tc-99m use is supplied by import sources, and various other isotopes are also imported into Australia.

The proposed new Australian reactor has a power rating that is twice as large as the existing HIFAR reactor.  The increased size and capabilities designed into the new reactor are not being sought because of a deficiency with respect to domestic radioisotope production capabilities in Australia.  In fact, the decision to import a portion of the country’s radioisotope supply has to do with competing uses for the reactor, and with post-reactor processing requirements for various products.  It is likely that even with a new 20 MW reactor Australia will import some radioisotopes, and will export some radioisotopes.  Thus, the additional cost of the larger reactor design has to be judged on issues other than adequacy for radioisotope supply.

The major alternative to the building of a new nuclear reactor in Australia for the supply of reactor-produced radioisotopes is importation of these materials from international sources, most of which currently are in North America and Europe.  Since most of the important radioisotopes that are produced in reactors have reasonably long half-lives, timing is not a major hurdle in obtaining import supplies.  Technetium-99m generators, which have half-lives of 66 hours (Mo-99), lose 25-50 percent of their cumulative production capacity compared with domestically-produced sources, but technetium generators are currently successfully imported into Australia, and import supplies could be expanded if the country did not have a production reactor.  In this case, Australia might decide to support some of the research that is still needed in order to bring one of the accelerator methods of Tc-99m production into commercial development.  Having a domestic source of Tc-99m would minimize the overall quantity of radioisotopes that would have to be imported to Australia if a replacement reactor were not built.

Environmental and Economic Implications

Nuclear reactors and particle accelerators both produce radioactive waste, but the amount of waste generated by an accelerator is much less than that associated with a reactor.  Both technologies require very expensive equipment.  Reactors generally are much larger projects than accelerators.  Neither of the two types of systems employ a large number of staff, but the jobs required by both include a high percentage of positions requiring advanced technical skills.

Environmental consequences of reactor vs. accelerator operation

Except for the generation of radioactive wastes, neither the reactor option nor the accelerator option is associated with important environmental consequences.  The routine emissions of airborne or waterborne pollutants are minor for both options, as are the other routine and expected consequences (land use, occupational safety and health, risks associated with transportation, etc.)  Safety concerns are always important for any reactor.  Although the small size of the proposed new Australian replacement reactor for HIFAR means that the safety concerns are less than for a large electric-power reactor with a power level a hundred-fold greater, a there is still an open question as to whether the design will be adequate to protect against radioactive releases that might harm offsite populations and the environment.  The reactor’s design requires careful analysis to assure that it embodies all of the latest technical advances in reactor safety, and until the design is finalized such a safety analysis cannot be performed with the requisite detail.  The comparable risks from accelerator operation are unimportant.

Radioactive wastes are, however, clearly different:  Whereas an accelerator produces only minor amounts of such wastes from the activation of its materials, any nuclear reactor produces fission products proportional to the integrated power produced, and these include the full spectrum of long-lived fission products that must be isolated from the environment for very long times, some measured in millennia.  This issue is well known and well understood.  Since the new Australian reactor is being designed for twice the power level of the existing HIFAR, it will therefore produce annually about twice as much in the way of fission-product high-level radioactive waste.

It is important to note that most or all of the countries that supply highly enriched fuel for reactors like the one under consideration in Australia also remove spent fuel from the country as one of the conditions of sale.  This is done for reasons of global security.  The spent fuel from the HIFAR reactor, for example, is ultimately destined for return to Europe (originally the U.K., more recently France), although the timeliness of the cycle puts its effectiveness into question.  What the ultimate fate of the waste from the spent fuel will be is also in doubt, due to uncertainties about the fate of the material after it is returned to its point of origin.  The plan for the spent fuel from HIFAR, for example, is that it will be reprocessed in the Europe, with radioactive waste returned to Australia for ultimate disposal.  Uncertainties about the future of reprocessing, as well as global political uncertainties, put the ultimate fate of spent fuel from reactor operations in Australia in some degree of doubt, regardless of the contractual arrangements made for the procurement of fuel for the new Australian reactor.

The other environmental consequences seem minor.  Neither the reactor option nor the accelerator option would have much impact on global warming, although each would consume electricity in the course of operations, and electricity involves greenhouse-gas emissions with global-warming implications (the accelerator would consume more such electricity, but the absolute amount is not large for either option).  And either would be a “good neighbor” environmentally in terms of routine impacts, which would likely be dominated by the normal human activities associated with running any high-tech facility, such as traffic and its associated impacts from the staff commuting to work, trash from their normal day-to-day activities, etc.

Investment, job creation, multiplier effects

Nuclear reactors and particle accelerators are both highly capital-intensive equipment.  Reactors tend to be much larger than accelerators, with accelerators typically costing in the millions to tens of millions of dollars, and reactors in the hundreds of millions.  The new reactor proposed by ANSTO is estimated to cost approximately $300 million (Aus.), and would take about four years to build.  Accelerators, being smaller projects than reactors, and producing much less nuclear waste, can be designed, licensed, and built more quickly.

Employment is created during both project construction, and project operations.  The amount of construction employment that is associated with these types of projects is probably in the range of 100-500 on-site workers during the construction period.  Once the facilities are commissioned, the amount of employment associated with their operations is modest, probably in the range of 50-100 workers for each facility.  Many of the jobs are high-level technical jobs.  In addition, the operation of these types of facilities provides educational and job-upgrading opportunities for students and others in Australia.

The investment capital necessary to build a new nuclear reactor in Australia is substantial.  Due to the capital-intensive nature of the equipment required, and the fact that a foreign vendor will be used, the public-sector investment in a new reactor will have a low economic multiplier effect compared to other possible uses of the capital.  A network of particle accelerators in Australia would probably have a slightly higher multiplier effect, because most of the equipment could be produced domestically.  In addition, a network of accelerators would, in the aggregate, cost significantly less than a single nuclear reactor.

The alternative to a new reactor for the supply of many important radioisotopes in Australia is importation.  This is true in the short term for Tc-99m, the most widely used radioisotope in medicine, although in the longer term Australia could play a key role in the development of an accelerator method for technetium production, which would provide the world with an alternative that produces far less radioactive waste than the conventional reactor-production method.

A key consideration in comparing isotope importation with domestic production in a new reactor is the price of the radioisotopes for the user.  Price, however, is a very elusive quantity in the radioisotope world, due in large part to the fact that virtually all nuclear reactors in the world used for radioisotope production are either government owned or substantially subsidized by the public sector.  While many radioisotope production operations, including those using the HIFAR reactor in Australia, are charged “at-cost” for use of space and neutron flux in the reactor, determining the true cost of this service is impossible.  Various studies available on radioisotope production costs for HIFAR vs. the cost of imported isotopes show everything from domestic production at half the price of imports, to imports being able to undercut domestic sources.  The truth is that a variety of purely subjective criteria are tied to the pricing of both domestic and imported radioisotopes, and various suppliers may manipulate these prices for their own reasons.  As long as the international supply of radioisotopes remains in relatively plentiful supply, as it currently is, there is no way to determine a price advantage or disadvantage for radioisotopes produced from a new reactor in Australia.

 References

1. Institute of Medicine (IOM), Adelstein, SJ, and Manning, F., eds., Isotopes for Medicine and the Life Sciences, National Academy Press, Washington, D.C., 1995.

2. International Atomic Energy Agency (IAEA), Isotopes in Everyday Life, Report No. IAEA/PI/A6E, Vienna, Austria, 1990.

3. Schenter, R., Battelle Pacific Northwest National Laboratory (personal communication), March, 2001.

4. Bennett, R., et. al., “System of 99mTc Production Based on Distributed Electron Accelerators and Thermal Separation,” Nuclear Technology Vol. 126, page 102-121, April 1999.

5. Christian, et. al., “Advances in Sublimation Separation of Technetium from Low-Specific-Activity Molybdenum-99,” Ind. Eng. Chem. Res., Vol. 39, No. 9, page 3157-3168, 2000.

6. Lagunas-Solar, M., and Zeng, N., “Accelerator Production of Technetium-99m as an Alternative to Reactor Methods,” Submitted to Australia-New Zealand Journal of Nuclear Medicine, October 22, 1997.

Table 1 - Attributes for Alternative Scenarios for Radioisotope Procurement in Australia
Table 2 - Applications for Radioisotopes
(Email [email protected] if you want MSW versions of these two tables.)

Table 3 - Radioisotopes Used for Medical Applications

Carbon-11, Fluorine-18, Nitrogen-13, Oxygen-15:  Used in PET imaging.
Chromium-51:  Labeling red blood cells.
Cobalt-60:  External beam radiotherapy.
Gallium-67:  Detection and localization of tumors and lesions.
Iodine-125:  Used in laboratory analyses such as radioimmuno assays.
Iodine-131:  Detection and therapy of cancer, especially of the thyroid.
Indium-111:  Used for brain studies.
Iridium-192:  Tumor treatment.
Iron-59:  Used in studies of iron metabolism in the spleen.
Potassium-42:  Used for coronary blood flow studies.
Samarium-153:  Used in treatment of bone cancer.
Sodium-24:  Used for studies of electrolytes within the body.
Technetium-99m:  The most commonly used isotope, technetium is used for medical imaging applications in all parts of the body.
Thallium-201:  Diagnosis of artery and heart disease.
Xenon-133:  Used for pulmonary ventilation studies.
Ytterbium-169:  Used for cerebrospinal fluid studies in the brain.
Yttrium-90:  Used for cancer therapy and arthritis treatment.

A variety of other radioisotopes are also either in use or under study for use in medical applications.

Table 4a, Accelerator Produced Radioisotopes

1. As72
2. As73
3. At211 = Alpha emitter = RIT (Brain Cancer treatment)
4. Ba128
5. Be7
6. Bi205
7. Bi206
8. Bi207
9. Br75
10. Br76
11. C11 =  Very short lived PET
12. Co55
13. Co56
14. Cu61
15.  F18 = Very short lived PET
16. Fe52
17. Ga67
18. Ga68
19. Gd146
20. Gd148
21. Ge68
22. Hf172
23. Hg195m
24. I123
25. I124
26. In111 = Diagnostics
27. Kr81m
28. Lu172
29. Lu173
30. Mg28
31. Mn52
32. N13 = PET
33. O15 =  PET
34. Pb203
35. Pm145
36. Rb81
37. Rb82
38. Ru97
39. Se72
40. Sr82
41. Ta178
42. Te118
43. Ti44
44. Tl201- Blood flow studies
45. V48
46. W178
47. Y86
48. Y87
49. Zn62
50. Zr88

Table 4b, Reactor Produced Radioisotopes

1. Ac227 = Parent of Ra223 = RIT
2. Ag111
3. Au198
4. Au199
5. Br82
6. C14
7. Ce141
8. Cf252 = Brain cancer treatment
9. Co60 = “Gamma knife” Cancer treatment
10. Cs137 - FP
11. Dy165 = Arthritis treatment (2 hr.)
12. Er169
13. Fm255
14. Ga67 = SPECT
15. Gd153 = SPECT calibration = Osteoporosis detection
16. H3
17. Ho166 = Multiple Myeloma treatment = 1.1 day half-life
18. I129
19. I131 = Cell directed therapy (“Smart Bullets”) = “RIT” = several forms of cancer
20. Ir192
21. Lu177 = RIT
22. Mo99 = Diagnostic = 40,000 procedures a day in US = comes from Canada, etc. - (Parent for Tc99m)
23. Os191 = Sent to Children’s Hospital, Boston from FFTF 1992
24. Os194 = RIT
25. P32 Heart disease treatment
26. P33  Requires high energy neutrons
27. Pd109
28. Pt195m
29. Re186
30. Re188 = From W188 generator = for RIT cancer, heart disease treatment
31. Re188 = From Re187 = for RIT cancer, heart disease treatment
32. Rh105
33. Ru105
34. S35
35. Sb119
36. Sc47 = RIT
37. Se75 = Sent to NIH for research from FFTF
38. Sm153 = Bone Cancer Pain Relief = Has FDA approval = “QUADRAMET”
39. Sn113
40. Sr85
41. Sr89 = Bone Cancer Pain Relief = Has FDA approval = “METASTRON”
42. Sr89 = from Y89 = “Carrier free Sr89” = bone pain relief = requires high energy neutrons
43. Te123m
44. Th228 = Alpha emitter grand parent of Bi212 (AML, RIT, et al)
45. Th229 = Alpha emitter grand parent of Bi213 (AML, RIT, et al)
46. Ti44
47. Tm170
48. Xe133
49. Y90 - FP
50. Y90 = From Y89 = Liver cancer treatment = “microseeds”
51. Y91 = RIT
52. Yb169

Table 4c, Accelerator or Reactor Produced Radioisotopes

1. Ag105
2. Ag108m
3. Ag109m
4. Al26
5. As74
6. Br77
7. Br80m
8. Cd109
9. Ce139
10. Cr51
11. Co57
12. Cu62
13. Cu64 = RIT
14. Cu67 = RIT = Accelerators can’t keep up with demand (see Denardo’s comments)
15. F18
16. Fe55
17. Fe59
18. Hg197
19. Ho163
20. I125 = Brachytherapy  (“seeds”) = Prostate, Breast, et al treatments
21. Na22
22. Pd103 = Brachytherapy  (“seeds”) = Prostate, Breast, et al treatments
23. Pm149
24. Rb83
25. Rb86
26. Rh105
27. Ru103
28. Sc44
29. Sc46
30. Sc47
31. Si32
32. Sm145
33. Sn117m
34. Ta179
35. Tc95m
36. Tc96
37. Xe-22
38. Xe-27 = Diagnostics = “Xe-127 production continues to be a problem for the combined capabilities of BNL and LANL…”
39. Y88
40. Zn65
41. Zr89