A New Look at the Cyclotron for Making Short-Lived Isotopes (First Printed in 1966)

Michel M. Ter-Pogossian and Henry N. Wagner, Jr

This reprint of an article that first appeared in Nucleon- tially powerful to01 to biomedical centers. With this ics in 1966 provides a unique perspective of the accelerator One can produce a variety of short-lived introduction of the cyclotron into clinical medicine nuclides that are unavailable from other sources. and medical research. The cyclotron offers a poten-

tially. This meansthat these nuclides must be This prescient arricle first appeared in Nucleonics, Vol. 24, pp. 50-62, in 1966 and is being reproduced in full, with prepared locally at the laboratory or hospital that kindpennission bythe McGraw-Hill Company. Copyright @ plans to use them-a situation chat requires consid- byMcGruw-Uill Company, Inc, 1966. Reproduced by erably more efforton the part of the user. Neverthe- permission. Further reproduction forbidden wirhoutpermis- less, the increasingly important placeof short-lived sion. Please noIe thnt the Mus No. is shown to the right ofthe nuclides in biomedicine’J justifies the additional element symbol as It originully appeared. effortand has resulted in two approaches for preparing the radionuclidesclose to where they will OST RADIONUCLIDES used today in bio- be used. The first-and most widely used-is to M medical research and diagnostic medicine make the isotopes in nuclide generatorsor “cows.” decay with half-lives that are long enough so that The second-whichwill be discussed in this ar- they can be transported from the place where they ticle-is to use cyclotrons. areprepared to the laboratories wherethey are Nuclide generators consist of a long-lived “par- used.With fewexceptions thedecay of these ent” nuclide that produces a short-lived “daugh- radionuclides does not impose stringent limits on ter” nuclide as it decays. At intervals theuser the speed of delivery or the length of storage time separates the daughter, usually by chemical means, before use. and the parent is leftto generate a fresh But shorter-lived radionuclides that decay witha The first wide-spread application of a generator half-life less than - 10 hr cannot be shipped easily systemin nuclear medicine was the Te132-IL32 to distant laboratories without decaying substan- system developedby Brookhaven National Labora- toryin 1954. Other systems, such as thewidely used Tc~~~generator, have since been devised at MI^ M. ~~R-POGOSSIANreceived hisPhD in nuclear Brookhavenand elsewhere (two new cows for physics From Washington University in 1950. Be joined the In113m andare described on pages 57 and 60) faculty of Washingtoon University School of Medicine in 1950 where he holds a Professorship in Radiation Physics to supplyradionuclides with half-lives varying and in Physiology in addition to being Head of the Division from a few minutes to several days. Thesesystems of Radiation Physics and Nuclear Medicine. are simple, relativelyinexpensive and easily in- HENRY WAGNERreceived his Doctor of Medicine from Johns Hopkins University School of Medicine in 1952. He stalled in even modest laboratories. joined the faculty of Johns Hopkins in 1959 where he holds But unfortunately, few useful short-lived radionu- Associate Professorships in Medicine, Radiologyand Radio- clides can be preparedby this method. Therefore in logical Science in addition to being chief of Nuclear most cases the optimum useof short-lived radionu- Medicine. clides in a biomedical environment requires their preparation by a reactor or accelerator. If we examine a chart of all radionuclides, we see that From Washington Universiry School of Medicine, St. Louis, about half of them belong to the group of neutron- MO: ad the Johm Hopkim Medical Instittrtions, Balti- more, MD. excess nuclides while the other half are neutron Address reprint requests to Henry N. Wagner; JK, MD, deficient. In general, neutron-excess nuclides are Professor of Medicine, Radiology und Radiution Health Sci- producedin reactors while neutron-deficient nu- ences. TheJohns Hopkins Medical Institutions, Room 2001, clides =E produced in positive-ion accelerators School of Hygiene and Public Health, 6!5 North Wove Street. such as the cyclotron. And because most of the Baltimore, MD 21205. Copyright 0 1998 by WB. Saunders Company short-lived radionuclides that are particularly use- 0001-2998/98/2803-000388~00/0 ful in biology and medicine are of the neutron- A NEW LOOK AT THE CYCLOTRON 203 deficient type, the positive-ion accelerator holds resolution one can achieve with external counting out the potential of becoming a powerful tool for a is low-approximately 1/25th of that with conven- well-equipped biomedical center. tional radiographic procedures-primarily because At the present time there are only two cyclotrons of the statistical fluctuations in counting rate. These in medical centers-ane in London and the other in result from the limits imposed on the total amount St. Louis. In the very nearfuture a third one will be of activity that can beused without radiation functioning at Massachusetts General Hospital in damage to the organism. Boston. In addition, several medical centers in this This situation improves considerably with the country are contemplating installing cyclotrons in short-lived nuclides. Because of their short effec- their laboratories. As these instruments become tive half-lives, more information (in the form of available in larger numbers there should be a rapid higher counting rates) is available with short-half- increase in both the fundamental andmedical life nuclides for a given dose of radiation delivered studies that are possible with short-lived radionu- to the system than witha longer-lived nuclide of the clides. same element. In addition, a cyclotron located in a biomedical Of course the optimization of a given in vivo center can be used to prepare longer-lived radionu- measurement with a radionuclide requires one to clides such as Na24and as a source of both fast: and match the half-life of the nuclide to the phenom- slow neutrons for radiobiological research, activa- enon being studied. It is obvious thata radionuclide tion analysis and-perhaps in the future-radiation cannot be used tostudy a physiological process that therapy. is considerably longer than the half-life of the nuclide. WHY SHORT-LIVED TRACERS? From the standpoint of radiation safety,the Why are these short-lived nuclides so useful in optimal physical half-life of a nuclide used to study biology and medicine-perhaps even more so than a physiologcal phenomenon in a living organism is their longer-lived counter-parts? Four distinct-but 0.69 times the elapsed time between administration unrelated-advantages indicate the reason for their of the labeled substance and measurement of the importance: emitted radiation.xAlthough this time is less than a 1 When they are-administered to a living organ- few hours in most medical applications, only seven ism, short-lived radionuclides result in less of the nuclides nowin use (Ga68,F18, SrS7m, radiation dose. Tc99m, Ba137mand In113m)have half-lives (12 e -The radiation emitted by certain short-lived hours. Consequently, there is a strong need for a radionuclides makes them more desirable than wider choice of short-lived nuclides that can be longer-lived isotopes of the same nuclide. given in large quantities without exposing the 0 A certain number of nuclides that are useful in organism-and particularly the patient-to exces- biomedical research have only short-lived sive radiation. Short-lived nuclides also have the radionuclides. advantage of letting one make repeated measure- 0 And in certain cases a short-lived radionuclide mentsin the same system because the rapidly can be prepared with a higher specific activity disappearing activity does not interfere with subse- than its longer-lived isotope. quent measurements. Clearly the first of these advantages is shared by In addition to the rate of decay, the type of decay all short-lived nuclides while the others are charac- that a nuclide undergoes is an important factor in its teristic only of certain nuclides. usefulness in biology and medicine. Only x-rays, Any assessment of these advantages in biology gamma rays, and positron annihilation radiation and medicine requires a discussion of certain provide photons that can be detected outside a aspects that are specific to tracer methodology in living organism. Beta particles, conversion elec- living systems. The information elicited from a trons and low-energy photons are absorbed within living organism by a radioactive tracer is always the tissue wherethey contribute to the radiation coded in the form of the radiation emitted by the dose but provide no useful information. Therefore radionuclide: therefore the amount of information in medicine and biology we need nuclides that are extracted is limited by statistical fluctuations. characterized by both optimum half-life and decay For example, at present the degree of spatial characteristics. Many short-lived radionuclides have 204 TER-POGOSSIAN AND WAGNER these desirable characteristics: for example, C”, that can be prepared witha positive-ion accelerator Fe52,Cu61, Ce9and Hg199m. and compares them with reactor-produced isotopes Many short-lived radionuclides are potentially of thesame element. The list also includes a veryuseful in biologyand medicine not only number of accelerator-produced longer-lived nu- because of their short half-lives or mode of decay, clides thatare potentially useful in biomedical but also because there are no longer-lived radionu- research. clides of the element. Examples are OI5, N13, F18 Positive ions can be accelerated by several types and a number of others. of instruments; the most common are cyclotrons If we examine a comprehensivelist of all and Van de Graaff generators. While thelatter have radionuclides, we see that the most common half- been used in biomedical. work, at the present time life is - 1 hr; -24% of the nuclides (-290) have a theyappear to be vastly inferior to cyclotrons half-life of 1 hr- 1 day, 20% (240) have a half-life of because the sizes needed to produce radionuclides 1 day- 1 year and 83 have a half-life > 1 year. Of are cumbersome and difticult to house. Other types thesenuclides a surprisingly smallnumber have of acclerators such as Cockroft-Walton accelerators been used in biology and medicine. For example, a and “Dynamitrons” which are capable of the recentsurvey shows that 34 nuclides of 26 ele- energies needed for production purposes have not ments have been used at one time or another to been developed. studymore than 70 body function^.^ However, most of these investigations (>70% of the types of ENERGY CONSIDERATIONS study) were carried out with only 9 nuclides of 6 The cross section of nuclear reactions resulting elernent~-P~~,CP, TcWm,NaZ2, W5, from positive-ionbombardment depends on the HgI9’ and HgZo3.With the exception of Tcgqmthese energy of the ions. These reactions can be broadly nuclides are far from ideal for use in man because classifiedinto two categories: (1) exoergic (or they have neitherthe optimum decay characteristics exothermic) reactions and (2) endoergic or (endo- nor half-livessuited for many of their uses. Thus under thermic) reactions. In exoergic reactions energy is the circumstances it isquite apparent that short-lived released subsequent to the interaction of the posi- radionuclides offer great promise as tracersin tive ion with the nucleus.Consequently, these biology and medicine because of their half-lives, reactions can be initiated even by “zero-energy” decay characteristics and chemical properties. positive ions. Exoergicreactions with positively charged particles cantake place onlyif the bombard- POSITIVE-ION ACCELERATORS ing particle carries sacient kineticenergy to Inaddition to their advantages for producing overcome the electrostatic repulsion (Coulomb neutron-deficient nuclides, positive-ion accelera- barrier) of thepositively charged nucleus. This tors have two other features that make them ideal barrier increases withthe charge on the nucleus and nuclide sources for medicine and biology. In the the charge of the bombarding particle. The energy first place, the isotopes produced are carrier free-a required to overcome the Coulomb barrier, how- great asset in studyingtrace constituents of the ever, does not intervene in the nuclear reaction. body which one wants to study without perturbing Endoergic reactions, on the other hand, absorb the physiological system. In the second place, energy and can take place only if the impinging positron-emitting nuclides are produced easily in particle carries a suf6cient “thresholdenergy” positive-ioiaccslerators-giving one a hlgher de- below which the reaction wiU not take place. The gree of spatial resolution in measurements when amount of energy either absorbed or released in a they are used with a detection system designedfor nuclear reaction is expressed as the Q value for the this purpose. reaction. Q is negative for endoergic reactions and One should keep in mind that positive-ion accel- positive for exoergic reactions.’O Table 1 gives the erators can prepare any nuclide that cad be obtained Q values for the nuclear reactions listed. from a reactor if one uses the accelerators as In general, the cross section of a nuclearreaction neutron sources. In most instances, however, it is resulting from positive-ion bombardment increases simpler and cheaper to prepare a neutron-generated with the energy of the particle above the threshold radionuclide by irradiation in a reactor. value,reaches a maximum value andthen de- Table 1 lists a number of short-lived nuclides creases as Fig. l shows. The rise in cross section A NEW LOOK AT THE CYCLOTRON 205

Table 1. Short-Lived Isotopes From Accelerators and Reactors

Accelerator-Produced Radionuclides Reactor-Produced Radionuclides

Decay and Decay and Principal Principal Atomic Radiations Method of a Radiations Method of NumberElementNuclide Half-life Emitted(MeV)Generation (Mevl Nuclide Half-lifeEmitted (Me4 Generatlon

4 Be 53 d EC; 70.48 Li7(p, n)Be7-1.64 6 C 20.5 min p+0.96 Bio(d,6.47 n)C” C14 5,730 y p-0.156 Ni4(n, p)C14 7 N 9.96 min p+1.19 n)N’3B”’(u, 1.06 8 0 124 sec p’1.74 N14(d, n)O’5 5.06 9 F 110 min p+(97%)0.649; Ol6(a,pn)F18 -18.6 F18 110 min p’0.649 EC (3%)

17 CI 32 min p+2.5, 1.3; P3’(a, n)CPm-5.57 Cl36 3 x 105y p-(98%)0.71, y2.04. . . EC (2%) 17 CI CIS 37.3 min p-4.8,1.1,2.8; y2.2, 1.6 19 K 7.7 min p+2.7; 72.2 CP5(cr,n)K38 -5.87 19 K 12.4 hr p-3.53,Z.Ol; K41(d, p)K42 5.30 K42 12.4 hr p-3.53.2.01; K4’ ( n, y)K42 71.52, . . y1.52,. . 20 Ca 4.7 d p-1.98.0.67; CaTd, p)Ca4’ 5.08 Ca45 165 d p-0.25 y1.31,. . 26 Fe 8.3 hr p+(57%)0.80, CP(a,2n)Fe52 -15.65 Fe55 2.7 y EC (2.63); EC (43%) 70.17 26 Fe Fe54 45 d p-0.46.0.27, 1.56; yl.29, 1.10,0.19,. . 30 Zn 30 min 8+(93%)2.35; CuYd, 2n)i!nm -6.37 Zns5 245 d EC (97.5%); EC (7%); p+(2.5%) y0.67,0.97, 0.33;y1.12 0.81-2.9 53 I 13 hr EC; y0.159, . . TeT2’(p,n)I123 1120 25.0 min p-(94%)2.12, 1.67,. . ; EC (6%); yO.44, 0.53 53 I 4.2 d EC (70%); Sb’Z’(a, n)l’24 -7.91 l1S0 12.5 hr p-1.02, 0.60; !3+(30%)1.55, y0.66,0.53, 2.15,. . ; 0.74,1.15, y0.60.1.7, 0.41 0.65-2.9 53 I 13.2 d EC (58%); Sb123(a, n)1126 -6.92 1131 8.05 d @-O.Sl,0.25- Fission p-(40%)0.38; 0.81; y0.36. p+(2%)1.13,. . 0.0800-0.72 - 1.25; y0.39.0.67, 0.48-141

results mainly from the increased facility with positive ions impinging upon the target-can be which the charged particle penetrates the Coulomb predicted from the cross section of the reaction. If barrier, whle the decrease results chiefly from the target material is infinitely thin and if the competition with other nuclear reactions that may particles lose very little energy while they are arise at higher energies10 The excitation curve can passing through the target, the cross section re- i: also exhibit resonance peaks. mains constant, and the curve for yield a.~ a When one prepares a radioactive isotope by function of energy follows the same pattern as the bombarding a target material with accelerated excitation curve. If, on the other hand, the target is positive ions, the yield of the reaction+xpressed, thick-as is usually the case for radionuclide i for example, as activity produced per number of preparation-the particles lose their energy in the 206 TER-POGOSSIAN AND WAGNER

Fig 1. Cross-section for reactionresulting from positive ion bombardment-shown atI& for Nap (d,.p)NaZ4 reaction- increases with particle energy above thresholdvalue, reaches maximum and then decreases.la target where they are eventually stopped, and the curve shows an initial fast rise (Fig 2) followed by a much slower rise, (Fig 3)." Fig 3. Yield of productionof 015 by N14 (c/,n)015 reaction as The selection of the energy rating for a positive- function of deuteron kinetic energyis shown at right." ion accelerator is determined by the Q-value one needs to initiate the particular nuclear reactions. yield of a nuclear reaction for a thick target One mustkeep in mind that the energy of the increases with the energy of the accelerated par- accelerated positive ions must exceed the Q value ticles, it is usually desirable whenpreparing a of the reaction by the energy needed to overcome radioactive isotope to exceed the Q value of the the Coulomb barrier of the nucleus. Also, if an nuclear reaction by several MeV. However, for external beam of charged particles is used to certain radioisotopes it may be preferable not to prepare the isotope, the energyof the particles must exceed a given energy for the bombarding particles exceed the Q value of the reactions by the energy to prevent the formation of undesirable impurities lost by the particles as they traverse the window that may begenerated at higher energies. This is the separating the evacuated cyclotron chamber from case when one prepares 015 by deuteron irradiation the target material. Such an energy drop depends on of nitrogen. In this exoergic reaction one should the energy of the particles and is of the order of 1 maintain the energy of the deuterons close to -3 Mev per 0.05 mm of aluminum for 10-Mev deuter- MeV. Indeed, for hsreaction the yield does not ons. increase substantially with energies above 3 MeV, Because of these considerations and because the and the use of higher-energy deuterons increases the amount of undesirable CILcontamination." Any radionuclide is potentially useful for bio- medical applications. Therefore, it appears that specifications for a cyclotron designed to prepare biologically useful radioactive tracers should call for a high-beam intensity,high-energy machine capable of initiating any nuclear reaction resulting in a useful nuclide. A high-energy rnachme of this kind would also have the advantage over lower- energy accelerators of producing higher radiois+ tope yields. IT lower energy ions are needed from a high-energy machine, the energy of the ions can always be reduced byabsorption. Accelerators capable of generating positive ions with an energy Fig 2. Yield of Nax prepared by Nan (d,p) NaU reaction on thick target as function of deuteron kinetic energy is shown of - 10 Mev/nucleon can produce, for all practical above.'" purposes,any desired nuclide; therefore, much A NEW LOOK AT THE CYCLOTRON 207 higher-energy bombarding particles areunneces- Table 2. Cyclotron Parameters for Fixed Outer Radius sary for isotope production. Fixed Fixed Magneric Frequency Field COST CONSIDERATIONS E - Be E - e2/m The discussion of energy considerations does not Ion B E fu E take into account the cost limitations imposed on (Proton) H' 1 1 1 1 (1) the space needed, (2) the construction and (3) (Deuteron) HZ 2 2112 112 (Triton) HA 3 3 1/3 1/3 the operation (including personnel) of a large (Helium nucleus) He3 312 3 2/3 4/3 accelerator. Unfortunately, the size of a cyclotron, (Alpha particle) He4 2 4 112 1 its cost, the cost of the building that houses it and *In these calculations rn is replaced in the cyclotron equa- the shielding itneeds, the cost of operating and tions bv the mass numberA. maintaining themachine, the salaries of the person- nel to operate it and the general difficulties encoun- generate with the lower ez/m projectile. For ex- tered in its operation increase sharplywith the ample Fe5' can be prepared with a large cyclotron energy capabilities of the machine. by the a, 2n reaction on Cr50 with a Q value of A gross rule of thumb is that the cost of the -15.6 Mev or by the He3, 312 reaction on Crj? cyclotronis a power function with an exponent whichgives a goodyield with a lower-energy between 2 and 3 of theenergy of the particles machine. Of course, in choosing a particular nuclear accelerated; on the other hand, its housing, the reaction for preparing an isotope onemust also personnel required for its operation and the mainte- consider the cross section of the reaction, the nance and operating cost are linear functions of the isotopic abundance of the target material and the energy. possible generation of radioactive impurities. At times a compromise must be madebehveen a To summarize, larger cyclotrons are preferable cyclotron ideally designed for biomedical research to smaller ones for biomedical research because of (-20 Mev deuterons) and a smaller, cheaper and the broader spectrumnuclear reactions that are simpler machine, which-while unsuitable for ini- possiblewith them and because of the greater tiating certain nuclear reactions-is adequate for yieldsthat are achieved with higher energies. generatingmost useful isotopes. In fact, such However, the selection of a smaller unit does not compromise between the energy capabilities of a result in too stringent compromises because of the cyclotron and itscost is only moderately restrictive wide choice of nuclear reactions leading to produc- in biomedical research for the following reason. tion of a particular isotope and because the amounts In most instances the isotope one wants can be of activity required in macer work in general are prepared by several nuclear reactions on different modest. An examination of the Q values in Table I nuclides, each with a different Q value. The energy shows that a cyclotron capable of accelerating capabilities of a cyclotron for accelerating positive deuterons to an energy of -8 MeV, protons to an ions are given by the formula E = B2Rde2Emin energy of 16 MeV, alpha particles to an energy of I6 which E is the kinetic energy of the accelerated ion, MeV and He3 nuclei to an energy of -26 Mev is a B is the flux density of the magnetic field, Ro is the good source of most of the isotopes listed. A unit radius of the outer-most orbit, m is the mass of the with this capability is relatively inexpensive, and particleand e is the elementary charge. The re- its installation and operation in a hospital are quired oscillator frequency (f)is fo = Bebmn. uncomplicated. Thus the energy capabilities of the cyclotron are The usefulness of He3 as a projectile for various limited for a given ion by the flux density of the nuclear reactions deserves to be emphasized. The magnetic field and bythe maximum size of the pole eZ/mratio of 4/3 for this paaicle makes it particu- pieces of the magnet; moreover, the energy of the larly desirable because of the high energy that can positive ionaccelerated is proportional to ez/rn. be imparted to the particle even in a modest-size Table 2 shows the approximate energes of cyclotron; and a great number of nuclear reactions variousions accelerated in a cyclotron. With a that use He3-such as He3, 3n: He3, 2n; and H3. p lower-energy cyclotron one can prepare a radioiso- -lead to the generation of a number of radioactive tope by selecting a suitable bombarding particle isotopes that are useful in biomedical research. even though the nuclide maybe impossible to Despite its advantages, the acceleration of He' 208 TER-POGOSSIAN AND WAGNER ions does impose certain constrictions on the cyclotron. In the first place the high cost of He3 makes a recovery system in the cyclotron for this gas mandatory; however, such a recovery system is relatively inexpensive and functions well. In the second place, adequate focusing of He3 is easier in an azimuthally focused rather than in a uniform- field machine. The choice betweenthe more desirable, but costly and complex, high-energy machme, and the more limited, but cheap, lower-energy machine is illustrated bythe two cyclotrons thatare now Fig 5. Rear aspect of Washington University Medical School cyclotron (Reproduced by courtesy of Am J Roentgenol, Rad, installed in medical centers. These are the Medical Therapy & Nuc Med). Research Council Cyclotron located at the Hammer- smith Hospital (capable of accelerating deuterons around it must be shielded against these radiations. to 15 MeV) in LondonlzJ3 and the Washington The Washington University cyclotron is housed in University Medical Cyclotron (4-8-Mev deuterons) a room 14 ft wide and 24 ft long. The room, which located in St. Louis, Fig 4, 5, Table 3. The cost is 9 ft high, is sunk underground to a depth of 5 ft 4 items for the Washington University Medical Cyclo- in. Additional shielding for this installation is 3 ft of tron are: concrete. This arrangement was dictated by the Cost of unit -$200,000 space available and by the undesirability of disturb- Room and shielding -$ 60,000 ing the footings of the Barnard Hospital building in Yearly operation and which the unit is housed.A maze shields the access maintenance costs to the room. The adequacy of the shielding was (for heavy operation) -$ 15,000 determined for both gamma ray and neutron haz- Required personnel 2 operators ards created by bombarding copper and beryllium targetswith deuterons. The maximum measured Housing and Shielding radiation fields in normally occupied areas are Because a cyclotron produces neutrons and elec-

c11 AND 015 FROM ACCELERATOR We shall discuss only two cyclotron-produced nuclides, C1l and 0l5,which are of such importance Fig 4. Linediagrams of WashingtonUniversity Medical School cyclotron (Reproduced by courtesy of Am J Roent- in biology that their availability in a medical center genol Rad Therapy & Nuc Med). gives the most promise. A NEW LOOK AT THE CYCLOTRON 209

Table 3. Specifications for Two Medical Cyclotrons

Type: Fixed frequency-Uniform magnetic field cyclotron Type: Isochronous, fixed frequency, azimuthally varying Over-all Dimensions: 6 x 9 x 16 ft field-cyclotron. Housing for Acceleratorand Power Supplies: Over-all Dimensions: 7 x 7 x 7 ft 24 X 14 X 9-ft shielded room Housing for Accelerator and Power Supplies:15 x 15 ft Weight: Cyclotron and oscillator: 56,300 Ib shielded room Power supplies: 9,800 Ib Weight: Cyclotron: 30,000 Ib Control console: 1,500 Ib Power supplies:4,000 Ib Utilities: Power: 150 kva, 440 v Controls: 2,000 Ib 3phase60cps Utilities: Power: 150 kva, 480 v, 3 phase 50 or 60 cps Cooling water: 230 literslmin 25 kva, 115 v, 1 phase Magnet: Pole diameter:36 in. Water: 50 gallonshin Pole tip diameter: 33 in. Magnet: Pole tip diameter:30 in. Exit radius: 14in. Average field: 16.5 kilogauss Mean fieldat exit radius: 14 kilogauss d-c power required:40 kw Maximum current in conductor:650 amp Current regulationaccuracy: 1/104 d-c power required:40 kw Current regulation accuracy: 11105 Oscillator System: Operating Oscillator System: Operating frequencies: 21.3 Mdsec (protons) frequencies: 25 Mdsec for protons 10.65 Mdsec (deuterons) 12.5 Mclsec for deuterons 14.2 Mc/sec (He3) 16.7 Mc/sec forHeZ particles Number of dees: 1 Number of dees: 2 (120 deg) Dee to ground peak voltage: 50 kv Dee to groundpeak potential: 30 kv Available d-c power at rectifier:50 kw Dee voltage regulation: 1/102 Deflector maximum voltage:75 kv Dee frequency regulation:11105 Ion Source: Type: Hooded- hot cathode Ion Source: Penning ion gage(PIG) source. Maximum filamentcurrent: 200 amp Performance: F ast Thermal Fast Filament power supply maximum voltage:6 volts Particle:ProtonDeuteronnerrtron[ll He' neutroni21 Performance: Oeuteron energy: 6-8 Mev (nlcmzlsec)(nicmzlsec) (The cyclotron designallows acceleration of deuterons Energy: 15 Mev 7.5 Mev 20 Mev up to8 Mev, however, the machine is presently Resolution: 75 kev 38 kev 100 kev used at6 MeV) Extracted MeasuredDeuteron internal beam: -200 pa Beam Values: Deuteronexternal beam stable opera- Current: 50pa50pa50pa 2X 10l2 5X IO4 tion: 40 Fa maximum ("Be9(d,n) Blothicktargeti2)With paraffin moderator measured: 60 pa [ Beam size at exit port: 0.5 in. diameter Proton energy: 12-16 Mev 0.05 radian divergence PotentialHelium-3 ionenergy: 16-21 Mev PerformanceHelium-4 ion energy: 12-16 Mev Beam size at exit port: 2 rnm x 25 mm (approx.)

rising C" Although the most important radioactive tracer in biological research is probably reactor-produced CI4,this nuclide suffersfrom three deficiencies: (1) its long half-life of over 5,000 years precludes its use in human in vivo tracer work for compounds with a long biological half-life, (2)its long half-life prevents one frommaklng repeated experiments on the same system because of rising background and (3) it decays with the emission of soft beta rays which for all practical purposes cannot be detected in in vivo experiments. Because of thesedraw- backs, CI4 never been widely used in nuclear Fig 6. Azimuthally varying field (AVF) designed cyclotron has C yc lotron Cop. by Cyclotron medicine. 210

C", on the other hand,does not have these gas as is done at Harnrners~thin England or by limitations. Its half-life is only 20.4 min,15 and it the gas from the irradiatedchamber after decays by emitting - 100% 0.972-Mev positrons irradiation as is donein the Wasfington University with emission of annihilation radiation which can CYclotron (Fig 7). If one wants carbon monoxide, be detected easily at great distances from the site of thecarbon dioxide can be removed from he emission either by a single counter or by two mixture by passing the radioactive gasesover soda counters in coincidence allowing precise spatial lime. On the other hand, if carbon dioxide is more localization. Electron capture is -0.19%.l6 The useful, the carbon monoxide can be oxidized inthe main disadvantage of C" as a tracer is its short mixture of he twogases by oxidation in the half-life which limits its use to compounds that can presence of a catalyzer. The cross section of the be rapidly labeled and to experiments in which the nuclear reaction leading to C" generation lets one label is followed for only a short period of time. In preparevery high activities of C" evenwith certain cases the first drawback can be circum- modest cyclotron beam currents. vented by using fast labeling techniques such as fast synthesis, exchange reactions or recoil label- Using 015 ing."However, the second defect presents an 015 isthe longest-lived radioactive isotope of insurmountable barrier to the use of CL1 in a oxygen, decaying with a half-life of 122 sec.I5 It number of studies. Nevertheless, the importance of decays by emitting 1.726-Mev positrons.'j It might carbon in biology is so great that there seems to be appear that its short-half-life would preclude its use little doubt Cll can be used in many studies that in biological research; however, 0'5 is a very have been impossible before. valuable tracer in biology because of the combina- Cll was used as a radioactive Racer in biological tion of two factors: (1) oxidation is the fundamental :- experiments even before the discovery of C14. A phenamenon in the life of higher organisms, and large number of organic compounds, including probably the most reliable index of the metabolism acetic acid, lactic acid, succinic and fumaric acids of a tissue is its rate of oxidation. (2) Metabolic were studied after labeling with C11.18In spite of its oxidation is a process which inmost of its phases is short half-life, C" has also been used as a label in comparable in time scale to the half-life of oxygen; biosynthesis.19 More recently Stenstrom has de- therefore OL5can indeed be used to study many scribed a method for labeling inulin with Cl1 by phases of metabolism. Thus the combination of the recoil.20 need for a radioactive tracer of oxygen and the fact Cll hasbeen used extensively in the form of carbonmonoxide and carbon dioxide inhuman experimentation. The first application of he nu- clide was in a medical study in which it was used for labeling carbon monoxide.21J2 Carbon-labeled carbon monoxide provides an excellent and simple way to label red blood cells for localizing blood pools by external scanning (for the diagnosis of pericardial effusionsor for localization of placenta) or for repeated blood-volume determinations which are facilitatd by the short half-life of the radioac- tive label. P-labeled carbon monoxide and carbon dioxide have also been used to study pulmonary function.23 C" is prepared in the El1 (dp) C" reactiunm by bombardingboron oxide (B203)with cyclotron- accelerated deuterons." The C" escapes the solid target in he form of carbon monoxide and carbon Fig 7. Irradiation chamberfor preparing C"-labeled carbon monoxide and carbon dioxide by deuteron bombardment of dioxide and can be extracted and used either by boron oxide. Radioactivegases are withdrawn by one of two sweeping these gases by a continuous flow of inert syringas shown. A NEW LOOK AT THE CYCLOTRON 21 1

Acrlvored Monqonese Pume

. . .. gram of system used in conjunc- tion with Washington University Oeuleron MedicalSchool cyclotron for preparation, purification and uti- lization of 015 used in medical studies. Input that most phenomena studied that way are short in mum energy for OI5 production is 3 MeV; above durationhas made OI5 a usefuland particularly this energy the yield doesnot increase appreciably, attractive tracer in biological and medical studies. while at 2 Mevthis yield is appreciably lower OI5 has been used in biology and medicine as a (Fig 3). tracer for o~ygen,?~-~~carbon mon~xide,~~.~~ carbon The target material used isair at NTP circulating dio~ide~~.~~.~~and water.35In general terms, oxygen continuously in the cyclotron beam (Fig 8) (or it has been extensively usedin medical and biological may be nitrogen containing a small amount of 02. studies in the metabolism of oxygen, in pulmonary carrier). The air is bombarded in a chamber that is function and in the determination of the metabo- sufficientlydeep to absorb the energy of the lismin normal structures and in neoplasm^.^^-^^ deuterons. In this system the 015 is obtained mixed There is notmuch doubt that this isotopewill with normalair. The intense ionization producedby become extremely important in nuclear medicine the cyclotron beam in air and the recoil energy of whenagreater number of cyclotronsbecome the activated oxygen atoms results in theformation available. of anumber of chemical impurities-the most OI5 isprepared by bombardingnitrogen with important ones being ozone and nitrogen oxides. deuterons bythe N14 (dp) 015 reaction.24 This These contaminants canbe removed by filtering the reaction is exoergic, and the kinetic energy of the activated air with activated charcoal and manga- deuteronsneeds only to exceed the Coulomb nese dioxide. The radioactive impurities produced repulsion of oxygen for the reaction to take place. by this methodare not generated in sufficient Dyson and his group" have shown that the opti- amounts to interfere with most experiments.

REFERENCES 1. Newbery GR: Br J Radiol 32:633-641,1959 12.Howard FT: Cyclotrons and high energyacceleraror;. 2. Ter-PogossianMM: "Progress in Atomic Medicine," Vol. ON-2644. 1958 I, JH Lawrence, ed. Grune & Stratton, New York, London, 1956 13. Gallop JW, et al: Proceedings of the Inst. Elec. Engrs. 3. Harper PV, et al: Optimizationof a scanning method using Part B, civ. 45 1, 1957 TC~".Nucleonics 2250-54, 1964 14. Ter-Pogossian MM: Am J Roentgenol. Radium Therapy 4. Gleason, GI: Intern J Appl. Radiation & Isotopes 8:90, and Nuclear Med. 96:737-743,1966 1960 15. Nuclear Data Tables:Part 4 US Atomic Energy Comrnls- 5. Greene MW, Tucker, WD: Intern J Appl.Radiation & sion, Government Printing Office, Washington, DC, 1960 Isotopes 12162,1961 16.Radiological Health Handbook: US Dept of Health. 6. Yano Y,Anger HO: J Nuclear Med 5:484487. 1964 Education and Welfare, Government Printing Office, Washing- 7. Myers WG, Olejar M: J Nuclear Med 4:202, 1963 con, DC, 1960 8. Wagner HN Jr, Emmons H: Characteristics of an ideal 17. WolfAP: U.M.E.S.C.O./NS/lUC/202 radiopharmaceutical in radioactive pharmaceuticals(U.S. Atomic 18. Buchanan JM, Hastings AB: Physiol Revs 26:120. 1946 Energy Commission, Division of Technical Momation) 19.Ruben S, HassidWZ, Kamen MD: J Am Chem Soc 9. Silver S: Uses of radioisotopes in medicine. Nucleonics 61:661, 1939 23:106, 1965 20. Stensmom T: Scientific Report No. 5, AF EOAR Conmct 10. Whitehouse WJ, Putman JL: Radioactive Isotopes. Ox- #AF61 (052)-740.1945 ford: Clarendon Press, 1953 21, Tobias CA, et al: Am J Physiol145:253-263, 1945 11. Dyson NA, et al: Proceedings of the 2nd U.N. Intern. 22. Lawrence JH, Tobias CA: Air Surg Bull 2:452,1945 Conf. on the Peaceful Uses of Atomic Energy. UN, Geneva, 23. Newbery GR: Brit J Radiol 32~633-641,1959 September, 1958. Pergamon Press, London, New York, Paris, 24. NuclearData: Circular No.499, National Bureau of 1959 Standards, Government Printing Office, Washington, DC. 1950 212

25. Buckingham PD, Forse GR: Intern J Appl. Radiation and Isotopes 14:439-445. 1963 26. Dyson NA, Sinclair JO. West JE: J Physiol 152325-336, I960 27. DysonNAetd: BrMed1 1:231-238, 1960 18. Dollery CT, West JB: Circulation Research 8:765-771. 1960 29. Dollery CT. WestB: Nature 189:1121, 1960 ! 30. Sprart JS Jr, et al: Forum 12:7-9. 1961 31. Spratt JS Jr, et al: Surgery 50:941-946,1961