Validation and upgrading of the recommended cross section data of charged particle reactions used for production of PET radioisotopes

https://doi.org/10.1016/S0168-583X(03)01264-3Get rights and content

Abstract

A validation test and upgrading of the recommended cross section database for production of PET radioisotopes in charged particle induced reactions, published by the IAEA in 2001, was performed. Experimental microscopic cross section data published earlier or measured recently and not yet included in the evaluation work of IAEA were collected and added to the primary database in order to improve the quality of the recommended data. The newly compiled experimental data supported the previous recommended data in most cases, but in a few cases this influenced the decision made earlier and resulted in new selected cross section data sets. A spline fitting method was applied to the selected data sets and updated recommended data were produced in these cases. The integral thick target yields deduced from the new recommended cross sections were critically compared with experimental yield data available in the literature.

Introduction

Production of medical radioisotopes, used for diagnostic or therapeutic purposes, is a very important non-energy related application of nuclear physics and nuclear technology. In practice, they are produced in neutron or charged particle induced nuclear reactions. The status of the cross section data for neutron induced reactions is appropriate, considering the intensive experimental and evaluation work required by different other applications of neutron activation cross sections. The application of the available charged particle nuclear data is still in a developing phase, having an especially promising future in medical radioisotope production. Dedicated compilations and evaluations of charged particle cross section data were started just a few years ago. They dealt with thermonuclear reactions for astrophysics, medium energy data for cosmochemistry, waste transmutation, beam monitoring, medical radioisotope production and activation analysis. The first pioneer compilation work was organized and coordinated by the International Atomic Energy Agency (IAEA). The list of nuclear reactions nowadays used for production of diagnostic radioisotopes and of the commonly employed monitor reactions is long. For the recent IAEA compilation concerned those nuclear reactions used for production of diagnostic radioisotopes and for the reactions used to monitor beam parameters during irradiation [1]. A second Coordinated Research Project (CRP) concerning production of therapeutic radioisotopes is in preparation at the Agency.

In spite of the fact that the status of available experimental database was poor in the IAEA work, an ambitious compilation and evaluation was done including 22 monitor reactions and 26 reactions to produce diagnostic gamma and positron emitters. A database including recommended cross section data and the corresponding deduced yields was published in IAEA-TECDOC-1211 and also was placed on the world-wide-web [1]. From statistics of retrievals and from analysis of references in recently published papers, it can be concluded that a significant interest is shown for this database. Soon after that the compilation project was published, it was recognized that a validation or benchmarking of the recommended data would be necessary that could be performed by precise experimental yield data measured on thick or partially stopping targets. Some of the integral or thick target yields published elsewhere were already used in the selection process during the initial evaluation helping to deselect those cross section data sets which were published accompanied by very contradicting yield values.

Different reasons made the benchmarking for all the involved reactions impossible during the project:

  • the limited time available to complete the project,

  • the large number of reactions to be validated,

  • the required highly enriched and very expensive target material (in high amounts) for performing new experiments,

  • the lack of availability of irradiation time at the required irradiation facilities.


After the initial work was completed and the results were published, it became also clear that not all the available literature data were included in the CRP database. The results of experiments that were going on to determine cross sections of reactions involved in the CRP became also available. Some computational error and/or sub-optimal fitting procedures were also detected among the published data.

All these facts made necessary to start a smaller project to upgrade and to correct a limited part of the database and in the same time to start a validation of the database. This validation effort is here handled with a twofold aim.

  • Recognizing that an independent experimental benchmark test for all the evaluated reactions was unrealistic we propose to validate the recommended cross section data by comparing the yields derived in the CRP with experimental thick target and integral yields data reported in literature.

  • An extensive compilation work of integral yield data is available and is frequently used at production facilities [2]. The reliability of this database however never was checked although it contains large internal disagreements and contradictions, which were gradually discovered during its application. Comparing the yields available in this publication with the ones given in the CRP report and/or with other calculations and experimental values, allows us to evaluate the quality and reliability of this yield database and to validate the CRP database.


In this paper we discuss only the reactions corresponding to Chapter 5.2 of the CRP report leading to the production of positron emitter radioisotopes used in medical diagnostics. We evaluate the reactions here one by one, present additional and new cross section values with their literature references, upgrade the fitting procedure where it is necessary and derive new recommended data and corresponding yields. Finally, the calculated thick target yields are compared with Dmitriev’s data [2] and different types of calculated or independently measured thick target yield information. Obviously, the misprinting and computational errors of the database are also corrected.

Since the targets used for production of positron emitter radioisotopes are mostly gaseous or highly enriched, validation through experimental determination of thick target yields is not always applicable. On the other hand, as the production is carried out routinely in laboratories connected to universities or research centres, it is obvious to suppose that results of systematic investigations of the production and integral yields are available at those places.

The new status and the changes proposed here to the positron emitter part of the database are summarised in Table 1. The upgrade of reactions presented in other chapters of the CRP – monitor reactions and gamma emitters – will be prepared later in separate publications.

Section snippets

Yield definitions and practical aspects of yield comparisons

A major drawback of the use of integral yield data for validation of microscopic cross section values is the confusion in definitions of yields that exist, especially when production data are used. The calculated and directly measured integral yields values are dependent on the external circumstances. The computation methods used strongly influence the final result. To clarify the situation we shortly summarise the definitions of different integral data, which are used in the comparison. We

Compilation and analysis methodology of experimental yield data

During our survey of measured experimental yields all available literature sources were used (primary journals, reports, conference abstracts and proceedings, yield compilations, reference databases, nuclear reaction databases as NSR and EXFOR, Ph.D. thesis, etc.). The yield values were taken from the original publications, when they were available. Data given in graphical form were digitised. Compilation works containing yield data deduced from published experimental cross section were not

Results

Upgrading of the CRP database by adding cross section data, the need for a new fit, determination of new recommended cross sections and associated yield values and the final validation through comparison with experimental thick target saturation yields will be discussed by reaction. The numerical values of the updated recommended cross sections and yield values will be included in a revised web version of the CRP database [1] and in the EXFOR database (in EXFOR entry: D4111). In this paper only

Conclusion

In this paper we have investigated 11 reactions in which positron emitter isotopes can be produced for medical applications. The investigated reactions were evaluated earlier by an international team coordinated by the IAEA. We found additional works to update the database available in the CRP [1]. New evaluation and fit was performed for 8 cases to produce improved recommended cross section data for the investigated reactions. Data validation was performed by collecting and comparing

Acknowledgements

The work was partly supported by the Nuclear Data Section of the International Atomic Energy Agency. Contract No. 11944/R0.

References (60)

  • C. Vandecasteele et al.

    Nucl. Instr. and Meth. A

    (1985)
  • T. Vandewalle et al.

    Int. J. Appl. Radiat. Isot.

    (1983)
  • M.S. Berridge et al.

    Int. J. App. Radiat. Isot.

    (2002)
  • M. Vogt et al.

    Int. J. Appl. Radiat. Isot.

    (1986)
  • R.J. Nickles et al.

    J. Appl. Radiat. Isot.

    (1983)
  • R.J. Nickles et al.

    J. Appl. Radiat. Isot.

    (1984)
  • T. Nozaki et al.

    Int. J. Appl. Radiat. Isot.

    (1974)
  • R.M. Lambrecht et al.

    Int. J. Appl. Radiat. Isot.

    (1978)
  • M. Guillaume

    Nucl. Instr. and Meth. B

    (1976)
  • S. Takács et al.

    Nucl. Instr. and Meth. B

    (2002)
  • C. Loc’h et al.

    Int. J. Appl. Radiat. Isot.

    (1982)
  • S. Kastleiner et al.

    Int. J. Appl. Radiat. Isot.

    (2002)
  • T. Ido et al.

    Nucl. Instr. and Meth. B

    (2002)
  • M.R. Cackette et al.

    Int. J. Appl. Radiat. Isot.

    (1993)
  • L.F. Mausner et al.

    Int. J. Appl. Radiat. Isot.

    (1987)
  • Yu.I. Vereshchagin et al.

    Nucl. Instr. and Meth. A

    (1993)
  • M.C. Lagunas-Solar

    Nucl. Instr. and Meth. B

    (1992)
  • E. Gilabert et al.

    Nucl. Instr. and Meth. B

    (1998)
  • D.R. Phillips et al.

    Radiochim. Acta

    (2000)
  • K. Kondo et al.

    Int. J. Appl. Radiat. Isot.

    (1977)
  • S.M. Qaim et al.

    Int. J. Appl. Radiat. Isot.

    (2003)
  • B. Scholten et al.

    Int. J. Appl. Radiat. Isot.

    (1995)
  • F. Tárkányi, S. Takács, K. Gul, A. Hermanne, M.G. Mustafa, M. Nortier, P. Oblozinsky, S.M. Qaim, B. Scholten, Yu.N....
  • P.P. Dmitriev

    Radionuclide Yield in Reactions with Protons, Deuterons, Alpha Particles and Helium-3

    (1986)
  • V.A. Muminov et al.

    Izv. An. Uzbeg SSR Ser. Phys.-Mat.

    (1974)
  • Z. Kovács et al.

    Radiochim. Acta

    (2003)
  • V.A. Kuzmenko et al.

    J. Radioanal. Nucl. Chem.

    (1985)
  • C.W. Alword et al.
  • P.P. Dmitriev

    Radiochimiya

    (1983)
  • R.J. Nickles

    Acta Radiol.

    (1991)
  • Cited by (53)

    • Estimation of neutron production yields from H<inf>2</inf><sup>18</sup>O as the <sup>18</sup>F-production target bombarded by 18-MeV protons

      2020, Radiation Physics and Chemistry
      Citation Excerpt :

      In the present work, INCL-4.6, Bertini, and TENDL-2017 (Rochman et al., 2017) were used for protons as shown in Table 1 and the TENDL library was used for neutron interactions. The TENDL cross sections of the 18O(p,n)18F reaction is shown in Fig. 1 together with experimental data (Hess et al., 2001; Kitwanga et al., 1990; Bair, 1973; Amsel and Samuel, 1967; Blair and Leigh, 1960; Blaser et al., 1952; Takács et al., 2003; Ruth and Wolf, 1979). It is seen that the TENDL library reproduces the trend of the excitation function well.

    • Radionuclide candidates for β+γ coincidence PET: An overview

      2020, Applied Radiation and Isotopes
      Citation Excerpt :

      The method of choice for 82Sr/82gRb generator production is the (p,4n) reaction on 85Rb which benefits from high natural abundance. This method was studied multiple times and is most often employed for the large-scale production (Mausner et al., 1987; Huszár et al., 1989; Deptula et al., 1990; Lagunas-Solar, 1992; Cackette et al., 1993; Gilabert et al., 1998; Ido et al., 2002; Buthelezi et al., 2006), although, as summarized by Takács et al. (2003), significant discrepancies still exist. For the proton energy of 70-60 MeV (with 40 MeV exiting from the thick Rb or RbCl target), the 82Sr/82gRb production yields of 8-13 MBq/μAh were reported.

    • Investigation of the production of <sup>68</sup>Ga using pre-equilibrium models

      2016, Applied Radiation and Isotopes
      Citation Excerpt :

      The calculated excitation functions for 69Ga(p,2n)68Ge reaction using pre-equilibrium reaction models are presented in Fig. 7. The obtained results using two–component exciton model up to 16 MeV proton incident energy are close to experimental data of Takács et al. (2003), Levkovskij (1991), Porile et al. (1963). In the range of 16–27 MeV proton incident energies, there is a difference between theoretical calculations and experimental data of Takacs et al., and Levkovskij.

    View all citing articles on Scopus
    View full text