Radiochemical determination of nuclear data for theory and applications
A vast knowledge of nuclear data is available and is grouped under three headings, namely, nuclear structure, nuclear decay and nuclear reaction data. Still newer aspects are under continuous investigation. Data measurements are done using a large number of techniques, including the radiochemical method, which has been extensively worked out at Jülich. This method entails preparation of high-quality sample for irradiation, isolation of the desired radioactive product from the strong matrix activity, and preparation of thin source suitable for accurate measurement of the radioactivity. It is especially useful for fundamental studies on light complex particle emission reactions and formation of low-lying isomeric states, both of which are rather difficult to describe by nuclear model calculations. The neutron induced reaction cross section data are of practical application in fusion reactor technology, particularly for calculations on tritium breeding, gas production in structural materials and activation of reactor components. The charged particle induced reaction cross section data, on the other hand, are of significance in medicine, especially for developing new production routes of novel positron emitters and therapeutic radionuclides at a cyclotron. Both neutron and charged particle data also find application in radiation therapy. A brief overview of advances made in all those areas is given, with major emphasis on nuclear reaction cross section data.
KeywordsNuclear reactions Nuclear data Radiochemical methods Complex particle emission Isomeric cross section Fusion reactor technology Medical radionuclide production
The term “nuclear data” is very broad. It includes all data which describe either the properties of nuclei or their interactions. In general, all those data can be grouped under three headings, namely, nuclear structure, nuclear decay and nuclear reaction data. The amount of available information is extensive. The compilation, storage and dissemination of data is well organised and is generally managed via four major regional data centres located at Brookhaven (USA), Paris (France), Vienna (Austria) and Obninsk (Russia). The worldwide exchange of nuclear data for peaceful applications is effected via the IAEA. Those applications pertain to both energy-related research (fission, fusion, accelerator driven systems, etc.) and non-energy related studies (medical radionuclide production, radiation therapy, astrophysics, cosmochemistry, etc.). Despite the vast available information, some newer aspects of nuclear data research are still under investigation. Accurate experimental work in less investigated areas allows to test nuclear models; it also opens up new vistas in many applications.
A large number of physico-chemical techniques are used to determine the nuclear data, each of them having its own merits and limitations. This article deals with the radiochemical technique which, though limited in some ways, is applicable to studies of all three types of data mentioned above. It consists of irradiating a material with neutrons or charged particles, separating the activated (radioactive) product, and measuring its decay characteristics or absolute radioactivity by standard counting techniques. Thus in contrast to purely physical methods, which aim at on-line measurement of the emitted radiation spectra, the activation (or radiochemical, when combined with a chemical separation) technique deals with the residual radioactive nucleus, i.e. off-line detection and measurement of the activated reaction product. The technique has very high sensitivity, especially in the case of short-lived radioactive products. In case of stable or very long-lived radioactive products, mass spectrometry (MS) and accelerator mass spectrometry (AMS) are more useful. The radiochemically obtained data are generally application oriented; but some information can be deduced on the reaction mechanism as well. We consider briefly some of the relevant aspects below.
Nuclear structure and nuclear decay data
In general, nuclear decay data are well known [1, 2]. They encompass information on the decay of radioactive nuclei, such as half-lives, energies, intensities and angular correlations of the emitted radiation, as well as on the formation of secondary radiation like electrons, X-rays, etc. The nuclear structure data include information on the properties of the excited states of nuclei. Since many of the decaying nuclei have decay energies of up to about 3 MeV, or somewhat higher, the nuclear levels populated in the decay product are generally well characterised up to excitation energies of 2–3 MeV. More detailed information on the nuclear structure, i.e. all the discrete levels up to the continuum of the nucleus, and even beyond, is obtained via spectral studies on nuclear processes. A good example is the in-beam γ-ray spectroscopy following an (n,γ) reaction which can provide information on the level structure of the product nucleus up to about 8 MeV. Those studies are, however, beyond the scope of this article. The discussion here is limited to decay properties of radioactive nuclei and the levels of the daughter fed by the various transitions.
Radiochemical techniques have played a significant role in the study of nuclear decay and structure data. In particular, before the advent of the high-resolution solid state detectors, many new radionuclides were discovered by the author via radiochemical methods [3, 4, 5, 6]. After the discovery of the high-resolution solid state detectors, detailed decay schemes of a few radiochemically separated radionuclides were established and some systematics in level structure could be discussed . Today, extensive information is available. There are, however, still some areas where more nuclear data work is necessary and where radiochemistry can play an important role. One such area is the study of radionuclides rather far from the stability line. Another area is the search and characterisation of super heavy elements. In both cases fast radiochemical separations are mandatory to isolate the desired product from the much stronger matrix activity, and to characterise it via measurement of the radioactivity.
The radioactive decay data find applications in many areas, e.g. calculation of total radioactivity, heat generation, transmutation products, etc. However, in recent years, with the enhancing application of radioactivity in medicine, especially for in vivo diagnostic and therapeutic studies, the demands on accurate decay data have considerably increased. A higher accuracy in the data means a higher accuracy in the internal radiation dose calculation. Two areas appear to need further attention—one involves the energies and intensities of low-energy electrons (e.g. conversion or Auger electrons) and the other branching ratios in the decay of non-standard positron emitters. Since decay schemes of many of the medically interesting radionuclides were generally determined using mixtures of radionuclides, often utilising rather poor resolution counters, it appears worthwhile to reinvestigate some of the special radionuclides in more detail using radiochemical techniques. Many of those medical radionuclides can now be produced with very high purities; the use of ultrapure sources should thus provide accurate information on the decay data as well.
An example of a recent decay data measurement is provided by 64Cu (T½ = 12.7 h). This radionuclide is becoming increasingly important in Positron Emission Tomography (PET) in connection with radioimmunotherapy. It was produced via the 64Ni(p,n)64Cu reaction  on highly enriched 64Ni as well as via the 66Zn(d,α)64Cu reaction  on highly enriched 66Zn. The chemical separation was based on ion-exchange chromatography [8, 9] and the final product was obtained as a very thin source with a radionuclidic purity of >99.9%. The measurements included β counting, γγ-coincidence counting, conventional high-resolution γ-ray spectrometry and, above all, high-resolution X-ray spectrometry using a Si(Li) detector. The latter was absolutely necessary to determine the low-energy (7.47 and 8.26 keV) K α and K β X-rays of the daughter Ni which describe the electron capture (EC) component in the decay of 64Cu. There was some discrepancy in the β+ branching and consequently in the intensity of EC decay; they are now fairly well established. Thus the availability of 64Cu of high radionuclidic purity made it possible to determine the decay scheme of that radionuclide with higher accuracy .
Recently determined Iβ+ values of some non-standard PET nuclides
Nuclear reaction data
In contrast to nuclear structure and radioactive decay data, whose scope is generally limited up to excitation energies of about 10 MeV, the nuclear reaction data cover a very broad span of energies, extending from a few meV up to the region of several GeV. The lower side of the energy scale is typical of neutrons and encompasses cold, thermal and epithermal regions. The major applications of nuclear data in those regions are related to structural analysis of solids, quantitative determination of elements via activation analysis, and fission reactor technology. The neutron capture cross sections and fission yields are also useful for production of radionuclides, especially for medical applications. The energy region from about 10 keV to a few MeV can be reached both by neutrons and charged particles, and it is particularly interesting for astrophysics and fusion research, especially with respect to the interactions of light charged particles. Neutrons up to 20 MeV energy have been extensively utilised in the development work related to fast reactors and future fusion technology. With increasing energies, monoenergetic neutrons become rarer, so that work above 30 MeV is done mostly using charged particles or spectral neutrons.
The body of available nuclear reaction data is huge but well organised. Experimental data published in any part of the world are compiled within a few months in the EXFOR file, coordinated by the IAEA. The data evaluation is performed in many regions of the world and, after validation and quality assurance, the data are placed at the disposal of users in extensive evaluated nuclear data files (e.g. ENDF-B VII). In addition some special purpose files like Fission Products, Activation Data, Neutron Dosimetry, Fusion Data, Medical Applications, etc. are also prepared. The data files contain all reliable data, measured by all techniques (including the radiochemical method), and substantiated by nuclear model calculations.
Preparation of thin samples for irradiation, especially with charged particles.
Study of low-yield reactions, i.e. when the cross section of the reaction under investigation is low and the matrix activity (i.e. the radioactivity of the undesired products) is high.
Study of soft-radiation emitters. This is the case when the product decays via β− emission or EC without any accompanying γ-ray. The radionuclides decaying by EC are characterised by X-ray counting, which can be performed only when a thin source has been prepared.
Characterisation of low-lying isomeric states. The low energy transitions can be detected advantageously after chemical separation and using a high-resolution low-energy detector.
Extensive use was made of the radiochemical technique in the determination of fission yields and activation cross sections. With the increasing use of high-resolution solid state detectors in γ-ray spectrometry, the importance of radiochemical measurements has somewhat diminished. Nonetheless, there are still many interesting areas where this technique is almost ideally suited or where it has advantages over the other methods. An extensive programme of work utilising this technique has been underway at Jülich for more than 30 years. A brief description of the areas pursued is given below.
Nuclear model calculations
Low energy nuclear reactions are commonly treated in terms of the statistical model, generally using the Hauser–Feshbach formalism, which takes into account the angular momentum of the evaporated particle and the level structure of the product nucleus. An early code developed for calculations was named HELGA. Later the pre-compound effect was also introduced and the relevant codes GNASH (in USA) and STAPRE (in Europe) in several versions have been very successfully utilized over the last 30 years in evaluating excitation functions, especially of neutron induced reactions up to 20 MeV.
Above 20 MeV the pre-compound effect becomes increasingly important and, at energies above 50 MeV, it plays a dominant role. A very commonly used code in the intermediate energy region was ALICE, developed by Blann also about 30 years ago. Recently the Obninsk group introduced several modifications and termed it as the code ALICE-IPPE. The incorporated modifications include treatment of the level density in a sophisticated way and consideration of the pre-equilibrium cluster emission (d, t and 4He). The code has been successfully applied to the calculation of excitation functions of a large number of reactions.
In recent years two further calculational codes, namely EMPIRE-II and TALYS, have also been introduced. The former combines the general features of the statistical process, the pre-equilibrium excitation model and the various improvements mentioned above. The code TALYS incorporates all reaction mechanisms (including direct interactions) and appears to be very successful.
The radiochemical work at Jülich has mostly been substantiated by the model calculations using the code HELGA in early works and the codes STAPRE and ALICE-IPPE in later works. Only in a few recent studies EMPIRE and TALYS codes were also used.
Complex particle emission in interaction of neutrons with nuclei
In interactions between nuclei and neutrons mostly nucleons and electromagnetic radiation are emitted. The emission of light complex particles (d, t, 3He, α, 7Li, 7Be, etc.) may also occur with a low probability in the light mass region, but in the heavier nuclei, it is rather rare. Due to both experimental and calculational difficulties not many studies have been done (for a detailed review on this topic cf. Ref. ). First reliable results on trinucleon emission reactions in the medium and heavy mass elements were obtained via extensive use of radiochemical separations [14, 15, 16, 17, 18]. Tritium and 3He emission was studied mainly using residual product identification [14, 15, 16, 17, 18], tritium counting  and MS . The α-particle emission data were obtained both via MS and measurement of the reaction products. In studies on deuteron emission, the radiochemical data  gave a sum of (n,d + pn) processes, so that the contribution of the deuteron could be deduced only by model calculation. For 7Be detection, γ-ray spectrometry in combination with radiochemical separations was applied.
Studies with 14 MeV neutrons
In contrast to d, t and 3He-particle emission processes, the α-particle emission has been extensively investigated, both by radiochemical and spectral measurements. The reaction mechanism is known fairly well and the systematics of cross section data are comparable to those for proton emission. The radiochemical method proved to be particularly useful where the radioactive product is a soft radiation emitter [29, 30], e.g. in reactions 48Ti(n,α)45Ca and 58Ni(n,α)55Fe, or in the region of lanthanoids , where the γ-ray spectra are rather complex. The contribution of the (n,n′α) process to total α-emission could also be established :it amounts to be about 10%.
Nuclear model calculations using the code HELGA, which was based on a statistical approach, incorporating the Hauser–Feshbach formalism (see above), showed  that whereas the nucleon emission, and to some extent α-particle emission, in the interactions of 14 MeV neutrons with medium and heavy mass nuclei, are described well by statistical processes, the emission of d and 3He occurs mostly via direct interactions. The emission of a triton, on the other hand, involves contributions of both statistical and direct processes.
Effect of increasing neutron energy
The dependence of complex particle emission on the energy of the incident neutron was investigated in two ways: (a) excitation function measurements using monoenergetic neutrons in the energy range of 3–20 MeV, (b) integral cross section measurements using fast spectral neutrons.
The results of statistical model calculations on the (n,t) reaction using the HELGA code (see above) are also shown in Fig. 4. Evidently, the statistical model reproduces the (n,t) excitation function of Al very well. With the increasing mass of the target nucleus, however, the statistical contribution appears to decrease drastically. In those cases possibly direct interactions play an important role. Similar calculations on the 93Nb(n,3He)91Y and 139La(n,3He)137Cs excitation functions [36, 37] showed that the emission of 3He is dominated by direct interactions. Regarding d emission, calculations on the 58Ni(n,d)57Co reaction  were done using HELGA and on the 92Mo(n,d)91Nb reaction  using a more sophisticated program STAPRE. In both cases the statistical process was found to be less significant than the direct interactions. In contrast, the experimental excitation functions of a large number of (n,α) reactions could be satisfactorily described  by a combination of statistical, precompound and direct processes, the contribution of the latter being relatively small.
As far as the second methodology was concerned, i.e. integral measurements using fast spectral neutrons, extensive studies were performed on about 35 target nuclei distributed over the whole periodic table of the elements. Two types of spectral neutrons with different average energies were utilized. They were produced by the breakup of deuterons on a thick Be target. In one study the energy of the deuterons used was 30 MeV and in the other 53 MeV. The neutron spectra generated in the two cases are well characterised from 2 MeV up to the maximum deuteron energy, with average neutron energies of about 13 and 22.5 MeV, respectively.
In all cross section measurements, extensive use was made of radiochemical separations, and the radioactivity was determined via β− ray counting, high-resolution γ-ray spectrometry, scintillation counting or gas phase counting. On one hand the residual radioactive product of a nuclear reaction was assayed and, on the other, attempts were made to identify the emitted complex particle, accumulated in the thick target. This was extensively done for the (n,t) reaction [44, 45, 46] with 30 MeV d(Be) neutrons and the results of product identification and tritium counting were compared with those calculated using the statistical model code HELGA. The measured integral (n,t) cross sections confirmed the results of excitation function measurements. Furthermore, by performing a series of integral measurements on the accumulated tritium after irradiation of an element with spectral neutrons generated by bombarding a Be target with varying deuteron energies between 17.5 and 30 MeV, it was possible to calculate the excitation function of the (n,t) reaction on that element via an iterative unfolding technique [45, 46]. This work thus also established a new method for determining the excitation function of a reaction after irradiations with a set of standardised broad neutron spectra.
The 53 MeV d(Be) neutrons showing a hard spectrum were used to study t, 3He, 4He and 7Be emission [19, 20, 47, 48, 49]. In each case residual product identification was done as in the case of the 30 MeV d(Be) neutrons. Furthermore, the accumulated tritium was chemically separated and assayed by gas phase counting. 7Be was also chemically separated and subjected to γ-ray spectrometry. The ratio of 3He to 4He emitted particles was determined mass spectrometrically using a quadrupole mass spectrometer. The results showed that a triton is easily emitted from the light mass nuclei. In medium and heavy mass regions, on the other hand, a bound triton is emitted with a much lower probability than three nucleons; in the case of 3He emission, however, this could not be confirmed. The nuclear model calculations using the code HELGA gave integral results for (n,t), (n,3He) and (n,α) reactions again similar to those for the excitation functions described above. For the (n,7Be) reaction no model calculation could be done.
Main conclusions of the study
Among the neutron induced complex particle emission reactions, most of the experimental and theoretical work to date has been performed on 4He emission. In general, the compound and precompound models together can adequately describe the 4He emission cross section. Relatively little radiochemical work has been done on d emission, but the few available results tend to show that this reaction is dominated by direct interactions. On the other hand, radiochemical methods have contributed appreciably to the study of trinucleon emission reactions. Whereas in the case of t emission both compound and direct processes play an important role, the 3He emission appears to be dominated by direct interactions. The emission of heavier particles like 7Be has been relatively little investigated. In particular, no excitation function has been measured and the available data refer to only integral measurements with a broad neutron spectrum. In general, with the exception of d and 4He emissions, it appears that the higher the charge of the emitted complex particle, the lower is its emission probability.
Isomeric cross sections
Fusion reactor technology
Attempts to harness the energy released in the fusion of light nuclei have been underway for more than 40 years. Although, in principle, several fusion reactions could be used for power production, the fusion of deuterium and tritium, giving rise to a 14.06 MeV neutron and a 3.52 MeV α-particle, appears to be the most promising reaction due to several reasons: (a) it needs a relatively low ignition temperature, (b) the fusion cross section is high even at low energies, (c) the energy released is appreciably higher than in many other reactions. On the other hand, the neutron released has both some advantages and disadvantages. It is this neutron which will be the source of fusion energy but it is the same neutron which will cause most of the radioactivity in a fusion reactor.
An early analysis of the technological problems involved , nuclear data needs  and, above all, the role of radiochemical methods in the determination of nuclear data for fusion reactor technology , was performed at the Forschungszentrum Jülich. Regarding nuclear data, neutron reaction data from thermal energies up to about 20 MeV are of prime importance. The radiochemically determined data are of significance in estimation of nuclear transmutation products, especially hydrogen and helium gas production in first wall constituents, tritium breeding in blanket materials, and in making an inventory of total radioactivity, i.e. a summation of all activation products in various components of the reactor system. In 1970s and 1980s considerable attention was devoted to neutronics problems of a fusion system. But in 1990s those activities were considerably reduced. Some of the salient results are discussed below.
Due to high energy and intensity of the neutrons generated during the thermonuclear fusion, a large number of nuclear reactions will be induced in the constituents of the first wall separating the plasma zone from the reactor blanket. In general, neutron emission processes (i.e. (n,xn) reactions) will dominate but charged particle emission reactions ((n,xp, (n,xα), etc.) will also occur .
The (n,xn) reaction cross sections are important with respect to neutron multiplication in the reactor blanket. The (n,xp) and (n,xα) processes, on the other hand, lead to the formation of a few new elements and, if the first wall remains in the reactor for a few years, the quantity of the new elements may become appreciable. Also the decay of the (n,xn) reaction products may lead to new elements. All those processes would eventually lead to a change in the physical and mechanical properties of the structural materials. Early estimates showed that if Nb were to be used as the first wall material, in 20 years time the amount of transmuted products (Tc, Mo, Zr, Y, etc.) may reach up to 20%.
The (n,xp) and (n,xα) reactions have two further serious effects. Firstly, they cause extra nuclear heat. Secondly, they lead to the accumulation of hydrogen and helium gases. Whereas the presence of hydrogen is not a very serious problem, since it rapidly diffuses out of the structural materials at elevated temperatures, the helium gas may facilitate void formation by nucleation, having thereby adverse effects on the quality of the materials.
In view of the great significance of (n,xp) and (n,xα) processes, and considering that radiochemical separations can be very useful in identification of some special radioactive products, an extensive research programme has been underway in our institute since early 1970s. Measurements of (n,p) and (n,α) reaction cross sections were performed on a large number of potential first wall constituents, e.g. Al, V, Ti, Cr, Mn, Fe, Co, Ni, Zn, Y, Zr, Nb, Mo, W, Pb, etc. [26, 29, 30, 31, 35, 51, 53, 54, 56, 58, 63, 66, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91]. The neutron energy region studied extended from about 3 to 20 MeV. Part of the work was done in collaboration with IRMM Geel, Belgium, and IEP, Debrecen University, Hungary. In several cases, the energy regions between 3 and 13 MeV and between 15 and 20 MeV were investigated for the first time, and experimental data showed good agreement with the results of modern nuclear model calculations. At 14 MeV, the radiochemically determined data were also found to be in agreement with the values obtained by an integration of the proton and α-particle spectra [27, 28]. For the (n,α) reaction, mass spectrometrically determined data  also agreed with the radiochemical and spectrum integrated data.
Through careful radiochemical measurements using highly enriched isotopes of a few elements as target materials, it was also shown for the first time that the contributions of the [(n,d) + (n,n′p)] and (n,n′α) processes at 14 MeV are not negligible. The cross section of the (n,n′α) reaction  was found to be about 10% of the respective (n,α) reaction cross section. Similarly the cross sections of the [(n,d) + (n,n′p)] reactions [21, 71, 74] on normal nuclei were found to be about 20% of the respective (n,p) reaction cross sections. In the special case of the lightest stable nucleus of an element, with the neutron separation energy higher than the proton separation energy, the [(n,d) + (n,n′p)] cross section  may be comparable to the respective (n,p) reaction cross section (see section “Studies with 14 MeV neutrons”).
In addition to the (n, charged particle) reactions, the (n,2n) reactions were also systematically studied [66, 75, 78, 81, 82, 83, 86, 87, 88, 93, 94]. The results obtained through the above mentioned investigations considerably strengthened the data base for the (n,2n), (n,xp) and (n,xα) reactions. The available data should thus now be very reliable for various calculations related to neutron multiplication, formation of transmutation products, accumulation of hydrogen and helium gases, and generation of nuclear heat in the first wall materials of a fusion reactor.
In addition to measurements of cross sections of neutron induced reactions on structural materials leading to the formation of transmutation products, the cross sections of interactions of neutrons with an expected long-lived transmuted species, namely 99Tc (T½ = 2.1 × 105 a), were also measured [95, 96, 97]. This was possible through a skilful radiochemical handling of the radioactive target material. The data should be useful in designing strategies for the transmutation of this long-lived radionuclide.
Out of the two presently discussed plasma fuel materials (deuterium and tritium), deuterium is available in nature. Tritium will, however, have to be bred in the blanket, i.e. the zone between the first wall and the outer parts of the reactor. This will be done via the (n,t) reaction. A literature survey shows that the (n,t) excitation function has been measured from threshold up to about 20 MeV only for 13 target nuclei. Among all those cases only the (n,t) reaction on 6Li and 7Li has high cross section and is thus suitable for breeding purposes. The fusion reactor blanket will therefore consist of 6,7Li-containing materials. The calculation of tritium breeding ratios in various types of lithium-containing materials thus requires, on the one hand, accurate (n,t) reaction cross sections on 6Li and 7Li and on the other, neutron economy, i.e. minimum loss of neutrons through interactions with materials other than 6,7Li. If the breeding ratio is too low, the whole process of fuel production and recovery may be jeopardized.
Neutron scattering studies
Energy and angular distribution of the emitted tritons, followed by an integration of the spectrum
Energy and angular distribution of the associated α-particle, followed by an integration of the spectrum
Mass spectrometric determination of 4He
Mass spectrometric determination of tritium
Chemical separation and β– counting of the accumulated tritium
Concluding remarks about nuclear data for fusion
Extensive efforts have led to the establishment of several data files and the status of data with regard to the formation of transmutation products, tritium breeding and activation of materials is now fairly good. Several sophisticated nuclear model calculational codes have also been developed. Nonetheless, there are still extensive needs of high-quality data and, with the upcoming demonstration experiment ITER, the nuclear data activities relevant to fusion are expected to be rejuvenated. The interdisciplinary approach of radiochemical methods will certainly contribute appreciably to this field.
Medical radionuclide production
In nuclear medicine programmes involving radionuclides, both radioactive decay data and nuclear reaction cross section data are needed. In general, the nuclear decay data are well known, except for some special cases, as discussed above in section “Nuclear structure and nuclear decay data”. They find application in determining the imaging modality and in internal radiation dose calculations. The nuclear reaction data, on the other hand, need constant improvement. They are needed in radionuclide production, mainly for optimisation of production routes, i.e. for maximising the yield of the desired radionuclide and minimising the level of radioactive impurities. Since radionuclides are produced in reactors as well as at cyclotrons, both neutron and charged particle induced reaction cross section data are required. The energy ranges involved are rather broad. In the case of neutrons mostly thermal energies and fission neutron spectrum are important, and in charged particles, energy ranges extending from a few MeV to several hundred MeV.
In reactor production of radionuclides, the most commonly used nuclear routes include (n,γ), (n, fission) and (n, charged particle) processes. The (n,γ) reaction has generally a high cross section at thermal neutron energies, so that the yield of the product is rather high. However, a serious drawback of the process is the low specific radioactivity which makes the radionuclide less suitable for medical applications. The fission process is a very suitable method to produce a large number of radionuclides in a “no-carrier-added” form. The chemical processing involved, however, is rather extensive. The (n,p) and (n,α) reaction cross sections are generally low; these processes are therefore used to produce only a few radionuclides in the light mass element region. In general, the cross section data base for the reactor production of radionuclides is well established.
In cyclotron production of radionuclides, the reaction cross section data play a very important role (for reviews cf. [102, 103, 104, 105]). Due to rapid degradation of the projectile energy in the target material, the energy range covered within the target is relatively broad, and since the reaction cross section varies rather rapidly with energy, it is not appropriate to adopt an average cross section over the whole energy range. One needs rather the full excitation function of the nuclear process to be able to calculate the yield with a reasonable accuracy. Production of radionuclides may be carried out using protons, deuterons, 3He- or α-particles. A knowledge of all the reaction cross sections is necessary. At small-sized cyclotrons, low-energy reactions like (p,n), (p,α), (d,n), (d,α), etc. are used. At higher energies, on the other hand, (p,xn) reactions are commonly employed. In some special cases, the (p,spall) process is applied.
Some radiochemical measurements on cross sections for the production of a few radionuclides via neutron induced reactions were carried out [83, 106, 107, 108, 109, 110, 111], partly in collaboration with PINSTECH, Islamabad, Pakistan. The nuclear activities in our institute, however, have mainly concentrated on charged particle induced reactions. Here thin sample preparation plays a very special role. The determination of radioactivity, however, could be done in many cases via conventional γ-ray spectrometry (without chemical separation), though occasionally clean radiochemical separations were absolutely necessary to obtain accurate data (e.g. while investigating β– ray, X-ray and low-energy γ-ray emitters).
Considerable efforts were devoted to optimisation and standardisation of data for production of commonly used diagnostic radionuclides [112, 113, 114, 115, 116, 117, 118, 119, 120, 121]. These included photon emitters like 123I (T½ = 13.2 h), 201Tl (T½ = 73.1 h) and 81Rb/81mKr (T½ = 4.6 h/13 s) generator, which find applications in Single Photon Emission Tomography (SPECT), and standard positron emitters 11C (T½ = 20.3 min), 13N (T½ = 10.0 min), 15O (T½ = 2.0 min) and 18F (T½ = 110 min), which are used in Positron Emission Tomography (PET). Furthermore, data for the formation of 82Sr (T½ = 25.3 d), which is the parent of the positron emitter 82Rb (T½ = 1.2 min), were also standardised .
In addition to nuclear data work on standard diagnostic radionuclides mentioned above, considerable efforts were devoted in our institute over the last 25 years, partly in collaboration with ATOMKI, Debrecen, Hungary, iThemba LABS, Cape Town, South Africa, and Cyclotron Laboratory of EAEA, Cairo, Egypt to development of novel radionuclides [114, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173]. A few radionuclides like 28Mg (T½ = 21.0 h) [123, 124], 43K (T½ = 22.2 h) , 48Cr (T½ = 21.6 h) , 57Co (T½ = 271.8 d) , 88Y (T½ = 106.6 d) [167, 168], 95Ru (T½ = 1.6 h), 97Ru (T½ = 2.9 d) , 117mSm (T½ = 13.6 d) , 147Gd (T½ = 38.1 h) , etc., were developed for a few special applications. The possibility of formation of 99mTc at a cyclotron was also investigated . The emphasis was, however, on non-standard positron emitters and novel therapeutic radionuclides. A brief summary of those two areas of work is given below.
Non-standard positron emitters
The list of non-standard PET radionuclides whose production methods were developed includes 22Na (T½ = 2.6 a) , 30P (T½ = 2.5 min) , 38K (T½ = 7.5 min) [134, 135], 51Mn (T½ = 46.2 min) , 52Fe (T½ = 8.3 h) , 55Co (T½ = 17.6 h) [138, 139], 64Cu (T½ = 12.7 h) [8, 9], 72As (T½ = 26.0 h) , 73Se (T½ = 7.1 h) [141, 142], 75Br (T½ = 1.6 h), 76Br (T½ = 16.0 h) [143, 144, 145, 146, 147, 148, 149], 77Kr (T½ = 1.2 h) , 82mRb (T½ = 6.2 h) , 82Sr/82Rb (25.3 d/1.2 min) generator system [152, 153, 154], 83Sr (T½ = 32.4 h) , 86Y (T½ = 14.7 h) , 94mTc (T½ = 53 min) [157, 158, 159], 120gI (T½ = 1.3 h) [160, 161], 124I (T½ = 4.18 d) [114, 162, 163, 164]. Among them 64Cu, 86Y and 124I are finding worldwide attention.
Regarding 64Cu, the suggested nuclear route  64Ni(p,n)64Cu, using highly enriched target material, is being commercialized. The labelling of monoclonal antibodies with 64Cu has led to several new therapeutic approaches.
Production routes of 124Ia
Energy range (MeV)
Thick target yield of 124I (MBq/μA h)
14 → 10
12 → 8
21 → 15
38 → 28
22 → 13
22 → 13
35 → 13
The radionuclide 86Y was produced via the 86Sr(p,n)86Y reaction . This route is also being commercialized. The radionuclide finds application as a positron emitting analogue of the pure beta emitting therapeutic radionuclide 90Y. The uptake kinetics are measured via a PET study of 86Y and the therapeutic effect is induced by 90Y. This is a new approach and is finding considerable interest in internal open source radiotherapy.
Novel therapeutic radionuclides
Therapeutic radionuclides are generally produced in a nuclear reactor since they are mostly β– emitters. In recent years the cyclotrons have been increasingly utilized, especially for production of radionuclides emitting low-energy β– particles, Auger electrons, X-rays and α-particles. At Jülich novel production route were investigated for the therapeutic radionuclides 67Cu (T½ = 61.9 h) [165, 166], 103Pd (T½ = 16.96 d) [169, 170], 140Nd (T½ = 3.37 d) , 153Sm (T½ = 46.3 h)  and 193mPt (T½ = 4.33 d) . The radionuclides 67Cu and 153Sm emit low-energy β– particles whereas 103Pd, 140Nd and 193mPt emit Auger (or conversion) electrons and X-rays. We discuss a typical case below.
Remarks about nuclear data work for medical radionuclide production
The work has led to optimisation and standardisation of common production routes of the SPECT radionuclides 123I and 201Tl, the most important PET radionuclide 18F and the commonly used therapeutic radionuclide 103Pd. Furthermore, it has contributed to the development of about 15 non-standard positron emitters and several potentially useful Auger electron emitting therapeutic radionuclides. In particular, the production routes of three longer-lived positron emitters, viz. 64Cu, 86Y and 124I, developed at Jülich, are finding worldwide attention and are now being commercialized. Those three novel positron emitters are opening new perspectives in radioimmunotherapy and radiation dosimetry.
The internal radiotherapy using suitable radionuclides, as discussed above, is a fast developing field. However, presently radiation therapy is carried out mostly using external radiation beams. Thus, standard radiation therapy implies almost exclusively teletherapy and involves the use of low-energy electrons, X-rays, γ-rays, high-energy electrons or high-energy photons. In all those cases only atomic interactions are important, and are well understood, the significance of nuclear data being negligible. Occasionally slow neutron capture therapy is also used. This may involve the introduction of a boron-containing chemical in some specific part of the body, followed by irradiation with slow neutrons. The captured neutron interacts with the boron according to the reaction 10B(n,α)7Li. The α- and 7Li-particles emitted in this reaction are absorbed in the tissue and generate a lot of heat, thereby inducing the therapeutic effect. The cross section of the nuclear reaction at thermal energies is well known. At higher neutron energies, however, the reaction 10B(n,t)2α may also occur; its cross section was unknown. We measured the excitation function of this reaction  in connection with our studies on (n,t) reactions described in section “Complex particle emission in interaction of neutrons with nuclei”. The data may be useful in optimisation of the therapy with neutrons containing some hard components.
An important and modern modality of therapy involves the use of hadrons, i.e. neutrons and charged particles, both produced at cyclotrons. Fast neutrons are generally generated via one of the two interactions: (a) deuteron break-up at a Be target and (b) protons on Be. Very commonly about 50 MeV deuterons and 66 MeV protons are utilized for neutron production. The therapeutic effect is induced by the secondary particles, especially charged particles, generated in the interactions of neutrons with the tissue constituents. In general, an accurate knowledge of the spectral distribution of the emitted particles is required, the radiochemically determined data being only of secondary importance. Nonetheless extensive data measurements with 53 MeV d(Be) neutrons [47, 48], described in section “Complex particle emission in interaction of neutrons with nuclei”, should provide some useful information on the formation of activation products during neutron therapy.
Beams of charged particles have a unique dose distribution, exhibiting a relatively flat entrance dose region (plateau) followed by a sharp dose peak, the Bragg peak, in which the particles lose rest of their energy . Among various charged particles, protons are commonly used. Their therapeutic range of energies is between 60 and 250 MeV. The therapeutic effect is caused by secondary radiation. Thus, besides total, elastic and nonelastic scattering cross sections, the energy- and angle-dependent emission spectra of γ-rays, neutrons and secondary charged particles are needed. All those data are obtained either by physical measurements or nuclear model calculations. The radiochemical method, on the other hand, delivers useful information on the formation of activation products.
Measurement of activation products was done in the interactions of protons of energies up to 200 MeV with biologically relevant and beam collimator materials. It was found that the amount of 7Be accumulated is negligible . Similarly the medium mass activation products are also not significant . The activation of collimators, however, is appreciable [176, 177] and due precautions are necessary to protect the therapy personnel. The formation of short-lived positron emitters is, however, of considerable interest.
Comprehensive studies on the formation of the positron emitting radionuclides 11C, 13N, 15O and 18F exist in the literature (see section “Medical radionuclide production”), mainly due to their use in PET studies. The data are, however, generally limited up to 30 MeV. Since in proton therapy energies higher than 30 MeV are involved, it was considered worthwhile to strengthen the database by performing cross section measurements also in the higher energy range. It has been shown  that the activation products of major concern in proton therapy are 11C, 13N and 15O. The data for the formation of 15O are fairly well known, also in the high energy range. We therefore concentrated primarily on the proton induced reactions on nitrogen and oxygen, leading to the formation of 11C and 13N. Excitation functions were measured  using the activation technique up to 200 MeV.
It needs to be pointed out that the quantities of short-lived positron emitters formed during proton therapy are high enough to allow PET studies for dose localisation. Initially it was done after proton therapy but in recent years on-line monitoring is becoming more common. Despite this progress, an estimate of the total radioactivity was rather uncertain, mainly due to the weak database. Furthermore, the radiation dose caused by those positron emitters had not been calculated.
Using the strengthened database, the yields of the three radionuclides, viz. 11C, 13N and 15O, formed in the human tissue were calculated as a function of the incident proton energy . Under standard therapy conditions of a brain tumour (150 MeV p, 2nA, 2 min), 280 MBq of the three positron emitters are formed. This quantity appears to be sufficient for on-line or subsequent PET studies to localise the dose. The calculated dose deposited by the positron emitters in the brain amounted to 5.5 mGy . This dose turns out to be <1% of the total dose deposited in the proton therapy. It is thus negligible but has been quanified for the first time .
In summary, the standard radiation therapy involves mostly atomic interactions, the role of nuclear data being negligible. Also in proton therapy the secondary electrons cause the major therapeutic effect. Only in calculation of the activation products, nuclear reaction cross section data are needed. In fast neutron therapy, on the other hand, extensive sets of nuclear data are needed. The radiochemical technique leads to high quality data for the formation of activation products.
The radiochemical method of nuclear data determination is well established; it complements on-line physical measurements. The radiochemical technique is advantageously used in investigations on low-yield reaction products, soft-radiation emitters and low-lying isomeric states. Combined with nuclear model calculations, the measured data can lead to some mechanistic information on the nuclear reaction. The radiochemically measured data are of considerable practical importance, especially in nuclear technology, cyclotron production of medical radionuclides and radiation therapy. It should be emphasised that nuclear data research demands interdisciplinary approaches and cooperative efforts. It constitutes an interesting science and useful technology.
The article gives a brief review of the work carried out over a period of more than 35 years at the Institute of Nuclear Chemistry of the Research Centre Jülich, Germany. I am highly indebted to Professor Dr. G. Stöcklin and Prof. Dr. H. H. Coenen, the former and present directors of the Institute, for their continuous support of this field of study, and to my own research group for painstaking efforts in acquisition and analysis of data. A large number of Ph.D. students and guest scientists also contributed appreciably to our efforts. My special thanks are due to about 10 Hungarian scientists for a long-term and fruitful cooperation both in experimental studies and nuclear model calculations. The partial financial supports of some external funding agencies like DFG, DAAD, EU, IAEA, etc. are gratefully acknowledged.
- 1.Evaluated Nuclear Structure Data File (ENSDF) (2007) National Nuclear Data Center (NNDC), Brookhaven, USA, and International Atomic Energy Agency (IAEA), ViennaGoogle Scholar
- 2.Firestone RB (1996) Table of isotopes, CDROM-Edition, Version 1.0. Wiley-Interscience, New YorkGoogle Scholar
- 3.Butement FDS, Qaim SM (1964) J Inorg Nucl Chem 26:1481, 1491Google Scholar
- 7.Qaim SM (1970) Nucl. Phys. A154:145Google Scholar
- 13.Qaim SM (1995) Radiochim Acta 70/71:163Google Scholar
- 15.Qaim SM, Stöcklin G (1976) Nucl Phys A257:233Google Scholar
- 16.Qaim SM (1974) J Inorg Nucl Chem 36 (1974), 239; Erratum p. 3886Google Scholar
- 18.Qaim SM (1978) Radiochim Acta 25:13Google Scholar
- 20.Wu CH, Wölfle R, Qaim SM (1979) Nucl Phys A329:63Google Scholar
- 21.Qaim SM (1982) Nucl Phys A382:255Google Scholar
- 23.Sudár S, Csikai J (1979) Nucl Phys A319:157Google Scholar
- 26.Molla NI, Qaim SM (1977) Nucl Phys A283:269Google Scholar
- 27.Grimes SM, Haight RC, Alvar KR, Barschall HH, Borchers RR (1979) Phys Rev C19:2127Google Scholar
- 28.Haight RC, Grimes SM, Johnson RG, Barschall HH (1981) Phys Rev C23:700Google Scholar
- 29.Qaim SM, Wölfle R, Rahman MM, Ollig H (1984) Nucl Sci Eng 88:143Google Scholar
- 30.Qaim SM, Uhl M, Molla NI, Liskien H (1992) Phys Rev C46:1398Google Scholar
- 31.Qaim SM (1984) Radiochim Acta 35:5Google Scholar
- 32.Qaim SM (1986) Nucl Phys A458:237Google Scholar
- 33.Qaim SM, Klapdor HV, Reiss H (1980) Phys Rev C22:1371Google Scholar
- 34.Qaim SM, Wölfle R (1985) Phys Rev C32:305Google Scholar
- 35.Qaim SM, Wölfle R, Strohmaier B (1989) Phys Rev C40:1993Google Scholar
- 36.Qaim SM, Wölfle R, Liskien H (1986) Phys Rev C34:489Google Scholar
- 37.Qaim SM, Wölfle R (1991) Acta Phys Hung 69:137Google Scholar
- 38.Qaim SM, Wölfle R, Liskien H (1982) Phys Rev C25:203Google Scholar
- 39.Suhaimi A, Wölfle R, Qaim SM, Stöcklin G (1986) Radiochim Acta 40:113Google Scholar
- 40.Qaim SM, Wölfle R (1987) Nucl Sci Eng 96:52Google Scholar
- 41.Liskien H, Widera R, Wölfle R, Qaim SM (1988) Nucl Sci Eng 98:266Google Scholar
- 42.Suhaimi A, Wölfle R, Qaim SM, Warwick P, Stöcklin G (1988) Radiochim Acta 43:133Google Scholar
- 43.Wölfle R, Qaim SM, Liskien H, Widera R (1990) Radiochim Acta 50:5Google Scholar
- 44.Wölfle R, Khatun S, Qaim SM (1984) Nucl Phys A432:130Google Scholar
- 45.Wölfle R, Sudár S, Qaim SM (1985) Nucl Sci Eng 91:162Google Scholar
- 46.Wölfle R, Suhaimi A, Qaim SM (1993) Nucl Sci Eng 115:71Google Scholar
- 47.Qaim SM, Wölfle R (1978) Nucl Phys A295:150Google Scholar
- 48.Qaim SM, Wu CH, Wölfle R (1983) Nucl Phys A410:421Google Scholar
- 49.Scholten B, Qaim SM, Stöcklin G (1993) Radiochim Acta 62:107Google Scholar
- 50.Qaim SM (1985) Nucl Phys A438:384Google Scholar
- 51.Mannan A, Qaim SM (1988) Phys Rev C38:630Google Scholar
- 52.Qaim SM, Mushtaq A, Uhl M (1988) Phys Rev C38:645Google Scholar
- 53.Qaim SM, Majah MIbn, Wölfle R, Strohmaier B (1990) Phys Rev C42:363Google Scholar
- 54.Molla NI, Qaim SM, Uhl M (1990) Phys Rev C42:1540Google Scholar
- 55.Sudár S, Szelecsényi F, Qaim SM (1993) Phys Rev C48:3115Google Scholar
- 56.Cserpák F, Sudár S, Csikai J, Qaim SM (1994) Phys Rev C49:1525Google Scholar
- 57.Qaim SM (1994) In: Proceedings of international conference on nuclear data for science and technology, Gatlinburg, American Nuclear Society, La Grange Park, IL, p 186 (1994)Google Scholar
- 58.Birn I-G, Strohmaier B, Freiesleben H, Qaim SM (1995) Phys Rev C52:2546Google Scholar
- 59.Sudár S, Qaim SM (1996) Phys Rev C53:2885Google Scholar
- 60.Strohmaier B, Fassbender M, Qaim SM (1997) Phys Rev C56:2654Google Scholar
- 61.Dóczi R, Sudár S, Csikai J, Qaim SM (1998) Phys Rev C58:2577Google Scholar
- 63.Nesaraja CD, Sudár S, Qaim SM (2003) Phys Rev C68:024603Google Scholar
- 65.Sudár S, Qaim SM (2006) Phys Rev C73:034613Google Scholar
- 66.Al-Abyad M, Sudár S, Comsan MN, Qaim SM (2006) Phys Rev C73:064608Google Scholar
- 67.Hilgers K, Sudár S, Qaim SM (2007) Phys Rev C76:06401Google Scholar
- 69.Qaim SM, Wölfle R, Stöcklin G (1976) Atomki Közlemények 18:335Google Scholar
- 71.Qaim SM (1978) In: Proceedings of international conference on neutron physics and nuclear data for reactors and other applied purposes, Harwell, September 1978 (OECD-NEA, Paris, 1978), p 1073Google Scholar
- 72.Qaim SM, Graca C (1975) Nucl Phys A242:317Google Scholar
- 73.Qaim SM (1976) Radiochem Radioanalyt Lett 25:335Google Scholar
- 74.Qaim SM, Molla NI (1976) In: Proceedings of 9th symposium on fusion technology, Garmisch-Partenkirchen, Germany, June 1976, EUR 5602, Pergamon Press, p 589Google Scholar
- 75.Rahman MM, Qaim SM (1985) Nucl Phys A435:43Google Scholar
- 78.Majah MIbn, Qaim SM (1990) Nucl Sci Eng 104:271Google Scholar
- 80.Molla NI, Qaim SM, Kalka H (1992) Phys Rev C45:3002Google Scholar
- 81.Birn I, Qaim SM (1994) Nucl Sci Eng 116:125Google Scholar
- 82.Bostan M, Qaim SM (1994) Phys Rev C49:266Google Scholar
- 83.Klopries RM, Dóczi R, Sudár S, Csikai J, Qaim SM (1997) Radiochim Acta 76:3Google Scholar
- 84.Fessler A, Wattecamps E, Smith DL, Qaim SM (1998) Phys Rev C58:996Google Scholar
- 85.Fessler A, Qaim SM (1991) Radiochim Acta 84:1Google Scholar
- 86.Nesaraja C, Linse K-H, Spellerberg S, Sudár S, Suhaimi A, Qaim SM (1999) Radiochim Acta 86:1Google Scholar
- 88.Reimer P, Avrigeanu V, Plompen AJM, Qaim SM (2002) Phys Rev C65:014604Google Scholar
- 89.Reimer P, Hult M, Plompen AJM, Johnston PN, Avrigeanu V, Qaim SM (2002) Nucl Phys A705:265Google Scholar
- 90.Reimer P, Avrigeanu V, Chuvaev SV, Filatenkov AA, Glodariu T, Koning A, Plompen AJM, Qaim SM, Smith DL, Weigmann H (2005) Phys Rev C71:044617Google Scholar
- 91.Semkova V, Reimer P, Altzitzoglu T, Plompen AJM, Quétel C, Sudár S, Vogl J, Koning AJ, Qaim SM, Smith DL (2009) Phys Rev C80:024610Google Scholar
- 92.Kneff DW, Oliver BM, Farrar H IV, Greenwood LR (1986) Nucl Sci Eng 92:491Google Scholar
- 93.Qaim SM (1972) Nucl Phys A185:614Google Scholar
- 94.Qaim SM (1974) Nucl Phys A224:319Google Scholar
- 97.Reimer P, Koning AJ, Plompen AJM, Qaim SM, Sudár S (2009) Nucl Phys A815:1Google Scholar
- 100.Qaim SM, Spellerberg S, Cserpák F, Csikai J (1996) Radiochim Acta 73:111Google Scholar
- 102.Qaim SM (1982) Radiochim Acta 30:147Google Scholar
- 108.Zaidi JH, Karim HMA, Qaim SM (1985) Radiochim Acta 38:123Google Scholar
- 109.Zaidi JH, Karim HMA, Arif M, Qureshi IH, Qaim SM (1990) Radiochim Acta 49:107Google Scholar
- 118.Szücs Z, Hamkens W, Takács S, Tárkányi F, Coenen HH, Qaim SM (1998) Radiochim Acta 80:59Google Scholar
- 125.Qaim SM, Probst HJ (1984) Radiochim Acta 35:11Google Scholar
- 128.Comparetto G, Qaim SM (1980) Radiaochim Acta 27:177Google Scholar
- 130.Denzler F-O, Rösch F, Qaim SM (1995) Radiochim Acta 69:209Google Scholar
- 137.Fessler A, Alfassi ZB, Qaim SM (1994) Radiochim Acta 65:207Google Scholar
- 138.Zaman MR, Qaim SM (1996) Radiochim Acta 75:59Google Scholar
- 139.Reimer P, Qaim SM (1998) Radiochim Acta 80:113Google Scholar
- 142.Mushtaq A, Qaim SM (1990) Radiochim Acta 50:27Google Scholar
- 145.Qaim SM, Blessing G, Ollig H (1986) Radiochim Acta 39:57Google Scholar
- 153.Tárkányi F, Qaim SM, Stöcklin G (1988) Radiochim Acta 43:185Google Scholar
- 156.Rösch F, Qaim SM, Stöcklin G (1993) Radiochim Acta 61:1Google Scholar
- 157.Rösch F, Qaim SM, Radiochim. Acta 62 (1993) 115; Erratum 75 (1996) 227Google Scholar
- 158.Fassbender M, Novgorodov AF, Rösch F, Qaim SM (1994) Radiochim Acta 65:215Google Scholar
- 159.Denzler F-O, Rösch F, Qaim SM (1995) Radiochim Acta 68 (1995), 13; Erratum 75 (1996), 227Google Scholar
- 165.Kastleiner S, Coenen HH, Qaim SM (1999) Radiochim Acta 84:107Google Scholar
- 174.Fassbender M, Scholten B, Qaim SM (1998) Radiochim Acta 81:1Google Scholar
- 175.Fassbender M, Shubin YN, Qaim SM (1998) Radiochim Acta 84:59Google Scholar
- 177.Qaim SM, Kettern K, Shubin YN, Sudár S, Coenen HH (in press) Radiochim ActaGoogle Scholar