White Paper on Emerging Facilities for Production of Novel Radionuclides for Use in Nuclear Medicine

Deliverable D8.3 White Paper on Emerging Facilities for Production of Novel Radionuclides for Use in Nuclear Medicine This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 101008571 (PRISMAP). This document reflects only the view of the author(s). The Agency is not responsible for any use that may be made of the information it contains. Deliverable D8.3 ii Project Acronym PRISMAP Project Title The European medical isotope programme: Production of high purity isotopes by mass separation Grant Agreement No. 101008571 Topic INFRAIA-02-2020: Integrating Activities for Starting Communities Project start date 01 May 2021 Nature Report Dissemination level Public Due date M48 Date of delivery M55 Lead partner ESS Contributing partners ARRONAX, DTU, CERN, ILL, NCBJ, PSI, SCK CEN, CEA Authors Luca Zanini, Elena Donegani (ESS); Lucia Popescu, Marc Dierckx, Donald Houngbo, João Pedro Ramos, Alexey Stankovskiy (SCK CEN) Renata Mikolajczak (NCBJ); Nicholas van der Meulen, Aleksandar Ivanov, Robert Eichler, Daniela Kiselev (PSI); Thierry Stora, Charlotte Duchemin (CERN); Ferid Haddad (ARRONAX); Mikael Jensen (DTU); Ulli Köster (ILL); Marion Libessart, David Blanchet (CEA), M. Manzolaro, Gaia Pupillo (INFN), Xavier Ledoux, Gilles Defrance (GANIL) , Elena Lopez-Melero, Javier Praena (University of Granada, IFMIF-DONES) Reviewers Charlotte Duchemin (CERN), Peter Ulrich (SCIPROM) Point of Contact Luca Zanini Institution ESS E-mail [email protected] Phone +46 72 179 2064 © PRISMAP 2021. This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. Deliverable D8.3 iii Revision History Version Date Author Comment 0.1 2.1.2025 Luca Zanini (ESS), Lucia Popescu (SCK CEN) and workshop participants Table of contents discussed at the SCK CEN workshop 0.2 1.6.2025 Luca Zanini (ESS) First sketch 0.3 22.6.2025 Luca Zanini (ESS), Lucia Popescu (SCK CEN), Renata Mikolajczak (NCBJ), Mikael Jenssen (DTU), João Pedro Ramos (SCK CEN), Thierry Stora (CERN) First draft including introductory chapters, chapter on emerging facilities, and first results on Sc-47, Cu-67, Tb152-155, Ac-225 0.4 27.6.2025 Luca Zanini (ESS), Joāo Pedro Ramos (SCK CEN), Renata Mikolajczak (NCBJ) Text revision. First reviews from Joāo Pedro Ramos (SCK CEN), Renata Mikolajczak (NCBJ) 0.5 12.7.2025 Luca Zanini (ESS) Text revision. Draft in google drive 0.6 30.7.2025 Luca Zanini (ESS), Ulli Köster (ILL), Lucia Popescu (SCK CEN), Elena Donegani (ESS), Renata Mikolajczak (NCBJ) Text revision. Review from Ulli Köster (ILL), Renata Mikolajczak (NCBJ), Elena Donegani (ESS), Lucia Popescu (SCK CEN). Added text on Pt-195m 0.7 12.8.2025 Luca Zanini (ESS) Draft revision with comments from Ferid Haddad (ARRONAX) 0.8 21.9.2025 Charlotte Duchemin (CERN) Review by the PRISMAP technical manager 0.9 28.9.2025 Ulli Köster (ILL), Lucia Popescu (SCK CEN) Review by sustainability manager and WP8 leader 0.95 28.9.2025 Thierry Stora (CERN) Version for review and approval by the general assembly 1.0 4.10.2025 Luca Zanini (ESS) Draft version for Zenodo 1.1.1 8.10.2025 Lucia Popescu (SCK CEN), Peter Ulrich, Kirsten Leufgen (SCIPROM) Review. Revised version for final review by main authors 1.1.2 11.10.2025 Lucia Popescu (SCK CEN) Final review 1.1.3 17.10.2025 Peter Ulrich (SCIPROM) Review. Corrections with contributions from main authors 3.11.2025 Ulli Köster (ILL), Charlotte Duchemin (CERN) Revision of Sections 5.9, 6.4 and 7.4 1.1.4 5.11.2025 Peter Ulrich, Kirsten Leufgen (SCIPROM) Final review and formatting 1.0 7.11.2025 Kirsten Leufgen (SCIPROM) Final version, approved by the general assembly Deliverable D8.3 iv Contents Abbreviations, Participant short names viii! Abbreviations viii! Participant short names xi! List of Figures xii! List of Tables xiii! Summary 1! 1. Introduction 2! 2. The PRISMAP portfolio of radionuclides 4! 3. Radionuclides selected for this study 6! 3.1 Criteria for the selection 6! 3.2 Sc-47 8! 3.2.1 Applications 8! 3.2.2 Production modes 8! 3.3 Cu-67 8! 3.3.1 Applications 8! 3.3.2 Production modes 8! 3.4 Tb-152, Tb-155 8! 3.4.1 Application 8! 3.4.2 Production modes 9! 3.5 Pt-195m 9! 3.5.1 Application 9! 3.5.2 Production modes 9! 3.6 Ac-225 9! 3.6.1 Application 9! 3.6.2 Production modes 9! 4. Facilities description 10! 4.2 GIP ARRONAX 11! 4.2.1 ARRONAX cyclotrons 11! 4.2.2 Radionuclides of interest 13! 4.2.3 Radiopharmaceutical production in our radiopharmacy (APUI) 13! 4.3 LARAMED 13! 4.4 CERAD 14! 4.5 CERN-MEDICIS 16! 4.6 SPES 18! 4.7 MYRRHA and ISOL@MYRRHA 20! Deliverable D8.3 v 4.8 TATTOOS 21! 4.9 SMILES 23! 4.10 JHR 24! 4.11 GANIL-NFS 27! 4.12 IFMIF-DONES 28! 5. Methods 30! 5.1 Methodology followed by ARRONAX 30! 5.2 Methodology followed by TATTOOS 31! 5.3 Methodology followed by SPES and LARAMED 31! 5.3.1 Production of Sc-47 at the SPES ISOL facility 31! 5.3.2 Direct production of Sc-47, Cu-64, Cu-67 and Tb-155 at the LARAMED facility 32! 5.4 Methodology followed by SCK CEN – MYRRHA 33! 5.4.1 Pt-195m production 34! 5.4.2 Ac-225 production 34! 5.4.3 Heating estimations 34! 5.5 Methodology followed by SCK CEN – ISOL@MYRRHA 35! 5.5.1 Ac-225 and Tb-152,155 production 35! 5.5.2 Sc-47 production 35! 5.6 Methodology followed by JHR 35! 5.6.1 Irradiation scenario 36! 5.6.2 Neutron flux performances 37! 5.7 Methodology followed by ESS 37! 5.8 Methodology followed by MEDICIS 40! 5.8.1 Sc-47 production 40! 5.8.2 Cu-67 production 40! 5.8.3 Tb-152, Tb-155 production 41! 5.8.4 Ac-225 production 41! 5.8.5 Pt-195m at the PAEC reactor 41! 5.9 Methodology followed by GANIL for ions 41! 5.9.1 Cu-67 – Studied reactions 42! 5.9.2 Sc-47 – Studied reactions 42! 5.10 Methodology followed by GANIL for neutrons 43! 5.10.1 Neutron flux and spectra 43! 6. Sc-47 production from emerging facilities 44! 6.1 Physical and chemical properties of Sc-47 44! 6.2 Sc-47 conventional production in PRISMAP 44! 6.3 Non-conventional production from emerging facilities and new routes 44! 6.4 Summary of expected Sc-47 activity and quality 45! 6.4.1 Ti-50(p,x) + mass separation (CERN-MEDICIS) 45! 6.4.2 Ti-50(p,x) + mass separation (SPES) 45! 6.4.3 Ti(p,x) + mass separation (ISOL@MYRRHA) 45! Deliverable D8.3 vi 6.4.4 Ti-nat(p,x)Sc-47 + mass separation (SPES) 46! 6.4.5 Ti-50(p,a)Sc-47 (GANIL) 46! 6.4.6 Ti-48(p,2p)Sc-47 (GANIL) 46! 6.4.7 Ti-nat(p,X)Sc-47 (GANIL) 47! 6.4.8 V-nat(p,x)Sc-47 (LARAMED) 47! 6.4.9 Ca-44(α,p)Sc-47 (ARRONAX) 47! 6.4.10 Ti-47(n,p)Sc-47 with fast neutrons (ESS) 48! 6.4.11 Ti-47(n,p)Sc-47 with fast neutrons (JHR) 48! 6.4.12 Ti-47(n,p)Sc-47 with fast neutrons (MYRRHA) 48! 7. Cu-67 production from emerging facilities 49! 7.1 Physical and chemical properties of Cu-67 49! 7.2 Conventional production in PRISMAP 49! 7.3 Non-conventional production in the future from emerging facilities and new routes 49! 7.4 Summary of expected Cu-67 activity and quality 50! 7.4.1 Zn-70(p,α)Cu-67 (Hevesy Lab, DTU) 50! 7.4.2 Zn-70(d,x)Cu-67 (ARRONAX) 50! 7.4.3 Zn-67(n,p)Cu-67 with fast neutrons (MYRRHA) 51! 7.4.4 Zn-67(n,p)Cu-67 (JHR) 51! 7.4.5 Zn-67(n,p)Cu-67 with fast neutrons (GANIL) 51! 7.4.6 Zn-67(n,p)Cu-67 with fast neutrons (ESS) 52! 7.4.7 Zn-70(p,alpha) Cu-67 (LARAMED) 52! 7.4.8 Zn-68(p,x)Cu-67 (LARAMED) 53! 7.4.9 Multiple reactions with charged particles (GANIL) 53! 7.4.10 Spallation in ThCx or UCx + mass separation (CERN-MEDICIS) 54! 8. Tb-155 and Tb-152 production overview and main challenges 54! 8.1 Efficiencies using ISOL mass separation 55! 8.2 Tb-155 and Tb-152 production pathways: results 57! 8.2.1 High Z target facilities – all Ta(p,X)spallation 57! 8.2.2 Literature case on Eu-151(α,3n)Tb-152 + Gd-155(α,4n)Dy-155 àTb-155 66! 8.3 Discussion and conclusions 66! 9. Pt-195m production from emerging facilities 68! 9.1 Physical and chemical properties of Pt-195m 68! 9.2 Nuclear reactions leading to Pt-195m 69! 9.2.1 Carrier-added versus non-carrier-added production routes 69! 9.2.2 Pt-194(n,g)Pt-195m 69! 9.2.3 Pt-195(n,n’)Pt-195m 71! 9.2.4 Pt-196(n,2n)Pt-195m 71! 9.2.5 Pt-196(g,n)Pt-195m 71! 9.2.6 Pt-195(g,g’)Pt-195m 71! 9.2.7 Pt-195(p,p’)Pt-195m, Pt-195(d,d’)Pt-195m or Pt-195(a,a’)Pt-195m 72! Deliverable D8.3 vii 9.2.8 Au-197(g,np)Pt-195m and Hg-199(g,a)Pt-195m 72! 9.2.9 Os-192(a,n)Pt-195m 72! 9.2.10 Ir-193(a,pn)Pt-195m 73! 9.2.11 Ir-193(d,2n)Pt-193m 73! 9.2.12 Ir-193(n,g)Ir-194(n,g)Ir-195m(β-)Pt-195m 74! 9.3 Production at existing and emerging PRISMAP facilities 74! 9.3.1 Production at JHR Cadarache 74! 9.3.2 Production at MYRRHA 76! 9.3.3 Production at IFMIF-DONES or SPIRAL2 76! 9.3.4 Mass separation at MEDICIS – SPES – ISOL@MYRRHA – SMART – TATTOOS 77! 9.4 Recommendations 77! 10. Ac-225 production from emerging facilities 78! 10.1 Overview 78! 10.2 Ac-225 production via Ra-226(p,n) reaction at a cyclotron 79! 10.2.1 Production 79! 10.2.2 Discussion 80! 10.3 Production of Ac-225 via the Th-232(p,x) reaction with 100 MeV protons at ISOL@MYRRHA 81! 10.3.1 Production 81! 10.3.2 Discussion 82! 10.4 Production of Ac-225 with epithermal and fast neutrons on Ra-226 at ESS 83! 10.4.1 Production 83! 10.4.2 Discussion 84! 10.5 Production of Ac-225 with epithermal and fast neutrons on Ra-226 at JHR (CEA) 84! 10.5.1 Production 84! 10.5.2 Discussion 85! 10.6 Production of Ac-225 via high-energy proton (1.4 GeV) irradiation of Th-232 at CERN-MEDICIS 86! 10.6.1 Production 86! 10.6.2 Discussion 87! 11. Conclusions 88! 11.1 Overview 88! 11.2 Conclusion from scandium studies 88! 11.3 Conclusions from copper studies 88! 11.4 Conclusions from terbium studies 89! 11.5 Conclusions from platinum studies 89! 11.6 Conclusions from actinium studies 89! Deliverable D8.3 viii Abbreviations, Participant short names Abbreviations AARE ABLA ADS ANL ANSYS APUI AVLIS AX BNL BR2 C70XP CEM03 CERAD CNRS CRS CT DART DEMO DGA DNA DOE DONES DOTA EAF EFPD ENDF EOB EOC EOI EOP EOS ERIC EU EXFOR FAIR FLUKA GANIL GIP GMP Activation in Accelerator Radiation Environments Evaporation code (GSI) Accelerator Driven System Argonne National Laboratory Analysis System (commercial simulation code) Radiopharmaceutical production in our radiopharmacy Atomic Vapour Laser Isotope Separation Experimental vault at ARRONAX Brookhaven National Laboratory Belgian Reactor 2 Cyclotron at the GIP ARRONAX facility Cascade-Exciton Model 03 Centre for Design and Synthesis of Molecularly Targeted Radiopharmaceuticals Centre National de la Recherche Scientifique Center of Radiopharmaceutical Sciences Computed Tomography Diffusing alpha-emitters Radiation Therapy Demonstration Power Plant (nuclear fusion experimental reactor) Diglycolamide Deoxyribonucleic acid Department Of Energy Demo Oriented NEutron Source 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid European Activation File Equivalent Full Power Days Evaluated Nuclear Data File End Of Bombardment (synonymous to EOI) End Of Cycle End Of Irradiation (synonymous to EOB) End Of Purification (synonymous to EOS) End Of Separation (synonymous to EOP) European Research Infrastructure Consortium European Union EXchange FORmat (database assembling nuclear reaction data) Facility for Antiproton and Ion Research Fluctuierende Kaskade (Monte Carlo code for radiation interactions) Grand Accélérateur National d'Ions Lourds Groupement d'Intérêt Public (Public Interest Group) Good Manufacturing Practice Deliverable D8.3 ix GPS HALEU HF HFIR HFR HFTM HIMB HIPA HPGe HRS HUG IBA IBR-2 ICO IST-ID IFMIF IMPACT IMT INCL4 INFRAIA INSERM IPS IRA ISAC ISOL ISOL@MYRRHA ISOLDE ITER JEFF JENDL JHR LARAMED LARISSA LBE LINAC LNL LS2 LS3 MCNP MCNPX MDA MEDICIS MELISSA General Purpose Separator (mass separator at ISOLDE) High Assay Low Enriched Uranium High Flux High Flux Isotope Reactor High Flux Reactor High Flux Test Module High Intensity Muon Beams High Intensity Proton Accelerator High Purity Germanium High Resolution Separator (mass separator at ISOLDE) Hôpitaux Universitaires de Genève Ion Beam Application SA Impulse Fast Neutron Reactor Institut de Cancérologie de l'Ouest Instituto Superior Técnico para a Investigação e Desenvolvimento International Fusion Materials Irradiation Facility Isotope and Muon Production using Advanced Cyclotron and Target technologies Institut Mines-Télécom IntraNuCLear Cascade Integrating Activities for Advanced Communities French National Institute of Health and Medical Research InPile Section Institute of Radiation Physics Isotope Separator and Accelerator Isotope Separation On-Line Isotope Separator On-Line facility at MYRRHA Isotope Separator On-Line facility International Thermonuclear Experimental Reactor Joint Evaluated Fission and Fusion File Japanese Evaluated Nuclear Data Library Jules Horowitz Reactor LAboratory of RAdionuclides for MEDicine Laser Resonance Ionization for Spectroscopy in Selective Applications Lead-Bismuth Eutectic Linear Accelerator Laboratori Nazionali Legnaro CERN Long Shutdown 2 CERN Long Shutdown 3 Monte Carlo N-Particle Monte Carlo N-Particle eXtended transport code Minimum Detectable Activity Medical Isotopes Collected from ISOLDE MEDICIS Laser Ion Source Setup At CERN Deliverable D8.3 3 The PRISMAP consortium includes 23 academic and research institutions across Europe (Figure 1), subdivided into production facilities, biomedical facilities for translational research, and emerging and future facilities. PRISMAP aims to improve the access to and increase the production of novel radionuclides. It also aims to conduct proof-of-concept investigations and joint research activities, and to research new production technologies and purification methods. The consortium has always been seeking the increase of the production capabilities and the inclusion of novel European facilities, such as: ISOL facilities (ISOL@MYRRHA at SCK CEN, Belgium, and the SPES accelerator complex in the INFN’s Legnaro National Laboratories, Italy), neutron facilities (the Jules Horowitz Reactor in CEA Cadarache, France, and the European Spallation Source in Lund, Sweden) and multi-particle accelerators (the CERAD cyclotron-based emerging infrastructure in Otwock, Poland, the SPIRAL2 facility at GANIL, France and the FAIR facility in GSI, Germany). Considering that the PRISMAP project aims for long-term sustainability, this white paper deals with the strategies for developing access services and preparing for the subsequent PRISMAP+ programme, i.e. a more mature European medical radionuclide programme. Section 2 lists the radionuclides in the PRISMAP portfolio; Section 3 provides details for each selected radionuclide. In Section 4 the emerging facilities within PRISMAP are described. Section 5 describes the methodology used by the participants of the study to determine the production and the results presented. The next five sections, from Section 6 to Section 10, contain the results of the studies for each of the radionuclides: Sc-47, Cu-67, Tb-152, Tb-155, Pt-195m, and Ac-225. The Conclusions are in Section 11. Figure 1. Map of the PRISMAP consortium including European facilities providing access to radionuclides, emerging facilities and the collaborating facility IFMIF-DONES. Deliverable D8.3 4 2. The PRISMAP portfolio of radionuclides The PRISMAP portfolio (Figure 2 and Table 1) consists of 28 radionuclides that have been identified for their potential in diagnostic applications or targeted radionuclide therapy, either from the day-1 portfolio or arising from user requests. It is also the partial inclusion of a survey performed in 20224,5, which involved stakeholders in the industry, research, pre-clinical and clinical sectors. Figure 2. The PRISMAP portfolio of radionuclides6. 4 M. Radzina et al., EJNMMI Radiopharmacy and Chemistry (2023) 8:27 https://doi.org/10.1186/s41181-023-00211-5 5 M. Radzina et al., Europe EJNMMI Radiopharmacy and Chemistry (2024) 9:85 https://doi.org/10.1186/s41181-024-00318-3 6 https://www.prismap.eu/radionuclides/portfolio/ Deliverable D8.3 5 Table 1. The radionuclides in the PRISMAP portfolio, their potential application and innovative aspects. Isotope Application Innovative potentials Sc-43 PET Moderate γ–ray emission and lower dose, compared to Sc-44 Sc-44 PET Prime candidate for 3γ-imaging Prime candidate for ortho-positronium mean lifetime imaging Sc-47 β-therapy, SPECT Treatment radionuclide matched with Sc-43 or Sc-44 in theranostic pairs Mn-52 PET Imaging of labelled cells and nanomedicines for long periods, candidate for 3γ-imaging or for ortho-positronium mean lifetime imaging Cu-64 β-therapy, PET Imaging radionuclide in a theranostic pair together with Cu-67 Cu-67 β-therapy, SPECT Treatment radionuclide in a theranostic pair together with Cu-64 Pd-103 Generator In-vivo or ex-vivo generator of Auger electron emitter Rh-103m Ag-111 β-therapy, SPECT Imaging and TRT with silver-based drugs Ba-128/ Cs-128 PET PET imaging with radium surrogate La-135 Auger Auger-TRT Sm-153 β-therapy, SPECT High-specific activity Sm-153 by mass separation for β-TRT and theranostic imaging properties Tb-149 α-therapy, PET α-emitter suitable for PET imaging, and part of Tb matched quadruplet Tb-152 PET Theranostic pair together with Tb-161, dosimetric and imaging studies over several days Tb-155 SPECT Theranostic pair together with Tb-161, imaging over extended periods Tb-161 β-therapy, SPECT Theranostic pair together with Tb-152/Tb-155, combined β/conversion electron/Auger electron-TRT Tm-165 Generator Generator of Er-165 for Auger-TRT Er-165 Auger Auger-TRT Er-169 β-therapy β-TRT with pure β emission Yb-175 β-therapy β-TRT Au-199 β-therapy, SPECT Imaging and TRT with Au-based drugs, Au nanoparticles, etc. Pb-203 SPECT Imaging in a matched theranostic pair together with Pb-212 At-211 α-therapy α-TRT Bi-213 SPECT α-TRT Ra-223 α-therapy α -TRT Ra-224 α-therapy, generator Generator for Pb-212 matched in a Pb theranostic pair and diffusing α-Emitter Radiation Therapy (DART) Ac-225 α-therapy, generator α-TRT Th-227 α-therapy α-TRT Deliverable D8.3 6 3. Radionuclides selected for this study 3.1 Criteria for the selection Among the radionuclides in the PRISMAP portfolio, a subsection was selected for the detailed study presented in this white paper. Different criteria were adopted for the selection. First, the aim was to define a sample of radionuclides representative of the complete PRISMAP portfolio. From the point of view of production cross-sections, reactions such as (n,p), (n,2n), and others have comparable excitation functions across the chart of nuclides. Therefore, the yield of other radionuclides produced by the same reactions can, to first order, be scaled from that of a chosen representative isotope. Another criterion is to leave out radionuclides where good or optimum production paths are already known, e.g. at small cyclotrons or by thermal neutron irradiations. To the first category belong e.g. the following radionuclides: Sc-43, Sc-44, Mn-52, Cu-64, La-135, Er-165 and Pb-203 and to the second category Ag-111, Sm-153, Tb-161, Yb-175, and Au-199. Therefore, these were not considered for this study. Other isotopes from the portfolio were left out of the study, either because production can be scaled from the selected ones or because no novel production routes are expected. To assure diversity in the chemical properties and radiochemical separation schemes, radionuclides belonging to different chemical groups were selected. Using these criteria, a list of six isotopes from five elements was selected. Sc-47, Cu-67, Tb-152, Tb-155, Pt-195m, and Ac-225 are shown in Table 2 together with the possible production methods and the corresponding emerging facilities where they could be produced or PRISMAP facilities with emerging processes. Deliverable D8.3 7 Table 2. The selected radionuclides for this study, production modes, and candidate facilities for production7. Isotope Production method Facility Sc-47 Ti(p,X)+ mass separation V(p,X) (+ mass separation) Ti-47(n,p) with fast neutrons V-51(γ, α) Ca-44(α,p) SPES, ISOL@MYRRHA, ARRONAX*+SMILES SPES, ISOL@MYRRHA, ARRONAX*+SMILES GANIL, IFMIF-DONES, ESS, MYRRHA, JHR Bremsstrahlung facility ARRONAX*, CERAD, GANIL Cu-67 Zn-68(p,2p) Zn-68(γ,p) Zn-67(n,p) with fast neutrons Zn-68(n,np) with fast neutrons Ni-64(α,p) ARRONAX, LARAMED Bremsstrahlung facility GANIL, IFMIF-DONES, ESS, MYRRHA, JHR GANIL, IFMIF-DONES ARRONAX, CERAD, GANIL Tb-152, Tb-155 Gd(p,X) + mass separation Gd(p,X) or Tb(p,X) Ta-181(p,X)+mass separation Eu(α,X) SPES, ISOL@MYRRHA, ARRONAX*+SMILES ARRONAX, LARAMED TATTOOS, ISOLDE, ISAC, MEDICIS ARRONAX*, CERAD, GANIL Pt-195m Off-line mass separation of Pt Ir-193(n,γ)(n,γ)bwith thermal n Pt-196(n,2n) with fast neutrons Pt-195(n,n’) with fast neutrons Pt-194(n,γ) with epithermal neutrons Pt-195(γ, γ’) Au-197(γ,np) Pt-194(d,p) Os-192(α,n) MEDICIS*, SPES, ISOL@MYRRHA, ARRONAX*+SMILES ILL* GANIL, IFMIF-DONES, ESS, MYRRHA, JHR GANIL, IFMIF-DONES, ESS, MYRRHA, JHR ESS, MYRRHA, JHR Bremsstrahlung facility Bremsstrahlung facility GANIL, IFMIF-DONES ARRONAX, CERAD, GANIL Ac-225 Ra-226(n,2n) with fast neutrons Ra-226(γ,n) Th-232(p,X)+mass separation GANIL, IFMIF-DONES, ESS, MYRRHA, JHR Bremsstrahlung facility ISOL@MYRRHA, TATTOOS, MEDICIS* 7 * PRISMAP facilities with candidate production pathways Deliverable D8.3 8 3.2 Sc-47 3.2.1 Applications Sc-47 (T1/2=3.35 d, b - =100%, <E b >=162 keV) has theranostic applications; it is the therapeutic match to either Sc-43 or Sc-44, which can both be used for PET. Sc-47, with a g energy of 159 keV (I = 68%) can be imaged by SPECT. The development of Sc-47 is still in its infancy, but it has promising applications8,9,10. 3.2.2 Production modes The possible production modes for Sc-47 are indicated in Table 2. We note the following: ¡ For Ca-44(α,p) there can be co-production of Sc-46. ¡ Ti(p,X) will lead to co-production of Sc-46, Sc-48 and V-48. This will require mass separation for clinical applications. If an enriched Ti-50 target is used, Sc-47 will be produced via (p,α). However, due to the co-production of Sc-46 and Sc-48, mass separation might still be required. ¡ Also, V-51(γ, α) will lead to the co-production of Sc-46 and Sc-48, likely requiring mass separation for clinical applications. ¡ Production via (n,p) will lead to strong co-production of Sc-46 and Sc-48, requiring the use of an enriched Ti-47 target. Even with an enriched target, mass separation might be required for clinical applications. 3.3 Cu-67 3.3.1 Applications Cu-67 (T1/2=61.83 h, b - =100%, <Eb>=145 keV) is a promising b - emitter radioisotope, with g emission usable for SPECT/CT, that forms a theranostic pair with Cu-64. In addition to b - emission, it also emits low-energy conversion and Auger electrons. 3.3.2 Production modes Currently, within PRISMAP, Cu-67 is produced by proton irradiation of enriched Zn-70 targets at Hevesy Lab, DTU, Denmark, and it could be produced by deuteron irradiation of enriched Zn-70 targets at the cyclotron ARRONAX, Nantes, France. This production method has the drawback of limited yield and the cost of the enriched target. The possible alternative production modes for Cu-67 are indicated in Table 2. We note the following: – Zn-68(p,2p) will lead to the co-production of Cu-64, Ga-66, Ga-67 and Zn-65, which must be considered for the handling of the irradiated target. – Zn-68(g,p) will not lead to co-production of Cu-64. 3.4 Tb-152, Tb-155 3.4.1 Application Tb-152 (T1/2=17.5 h, β+=20.3%, <Eβ>=1140 keV) has promising applications in PET11,12. 8 C. Müller et al. Br J Radiol. 2018;91:20180074, doi: 10.1259/bjr.20180074. 9 C. Müller et al., Journal of Nuclear Medicine, Vol. 55, No. 10, 2014, doi: 10.2967/jnumed.114.141614 10 K. Siwowska et al., Pharmaceutics 2019, 11, 424; doi: 10.3390/pharmaceutics11080424 11 R. P. Baum et al., Dalton Trans. 2017;46(42):14638–46, doi: 10.1039/c7dt01936j 12 Müller et al., EJNMMI Res. 2019;9:68, doi: 10.1186/s13550-019-0538-1. Deliverable D8.3 9 Tb-155 (T1/2=5.32 d, Eγ and X-ray ≈ 45 keV (107%), Eγ=86.6 keV (32 %), Eγ=105.3 keV (25.1 %) may be of interest for low-dose SPECT prior to therapy with a therapeutic match13,14. 3.4.2 Production modes Tb-152 and Tb-155 are produced by high-energy proton-induced spallation of tantalum targets at ISOLDE at CERN, where online mass separation is performed. Similarly, they can be produced by spallation of tantalum targets at MEDICIS at CERN, but here the mass separation is performed offline. 3.5 Pt-195m 3.5.1 Application Pt-195m (T1/2=4.01 d, IT=100%) is the most prolific Auger electron emitter known, with a half-life suitable for clinical applications 15 . Moreover, various DNA-targeting Pt-compounds (cisplatinum, carboplatinum, oxaliplatinum, etc.) are already known and even clinically used (albeit in cold, non-radioactive form). Provided the production issues could be solved, Pt-195m could revolutionize Auger electron therapy. Pt-195m emits also g-rays with 99 keV (11.7%) suitable for SPECT imaging. 3.5.2 Production modes The possible alternative production modes for Pt-195m are indicated in Table 2. We note the following: ¡ Pt-195m has most likely a very high (≈ 13000 b) thermal capture cross-section, which may limit the production of Pt-195m in a high thermal neutron flux. 3.6 Ac-225 3.6.1 Application Ac-225 (T1/2=9.92 d, αcumulative=400%, <Eα, cumulative>=6.88 MeV) is a promising radioisotope for targeted alpha therapy16,17,18. 3.6.2 Production modes Ac-225 is produced at JRC Karlsruhe by elution from Th-229 generators. Ac-225 is also produced by highenergy proton-induced spallation of thorium targets at MEDICIS at CERN. The possible alternative production modes for Ac-225 are indicated in Table 2. We note the following: ¡ Ra-226(n,2n) followed by b decay to Ac-225 will lead to co-production of Ac-227 via Ra-226(n,g). Therefore, shielding of thermal neutrons is recommended. ¡ Significant production of Rn volatile daughters is expected and should be calculated to estimate the mass of Ra-226 that can be realistically handled. 13 C. Müller et al., Nucl. Med Biol. 2014a;41(Suppl):e58–65, doi: 10.1016/j.nucmedbio.2013.11.002. 14 C. Favaretto et al., EJNMMI radiopharm. chem. 6, 37 (2021) , doi: 10.1186/s41181-021-00153-w 15 J. Bolcaen et al., J Nucl Med 2023; 64:1344–1351 16 G. J. Beyer et al. Nucl Med Biol. 1997;24(5):367–72, doi: 10.1016/s0969-8051(97)00011-5. A. Morgenstern et al., Semin Nucl Med. 2020;50(2):119–23, doi: 10.1053/j.semnuclmed.2020.02.003. 17 A. Morgenstern et al., Semin Nucl Med. 2020;50(2):119–23, doi: 10.1053/j.semnuclmed.2020.02.003 18 W. P. Fendler and C. Cutler, J. Nucl Med. 2017;58(11):1709–10, doi: 10.2967/jnumed.117.198333. Deliverable D8.3 10 4. Facilities description This chapter introduces operational, emerging and projected infrastructures in PRISMAP, categorized into three main types based on the employed technology and applications. 4.1.1.1 Cyclotron-based Facilities ¡ ARRONAX: Located in Nantes, France, ARRONAX is an operational cyclotron-based facility dedicated to the production of radionuclides for medical and research purposes. ¡ LARAMED: Implemented at INFN-LNL in Legnaro (Padova), Italy, this emerging cyclotron facility is dedicated to the production of radionuclides for medical research purposes. ¡ CERAD: Based in Otwock, Poland, at the National Centre for Nuclear Research (NCBJ), the emerging CERAD infrastructure with cyclotron technology enables research on new diagnostic/therapy radionuclides and radiopharmaceuticals based on ligands biologically active on the cell/molecular level. 4.1.1.2 Mass-separator Facilities ¡ MEDICIS: Implemented at CERN, in Switzerland, MEDICIS is dedicated to the production of high-purity radionuclides for medical research and applications. ¡ SPES: An emerging mass-separator facility at INFN-LNL in Legnaro (Padova), Italy, is dedicated to the production of beams of radionuclides for fundamental science and medical research. ¡ ISOL@MYRRHA: Under construction at SCK CEN, in Belgium, the ISOL facility at MYRRHA will produce high-intensity radioactive ion beams for fundamental research in nuclear physics, condensed-matter research and medical applications. ¡ TATTOOS: Approved to be constructed at the Paul Scherrer Institute (PSI) in Switzerland, TATTOOS (Targeted Alpha Tumour Therapy and Other Oncological Solutions) utilizes high-intensity proton beams from the HIPA facility to produce radionuclides for advanced cancer treatment and diagnostics. ¡ SMILES: In France, the SUBATECH laboratory is developing mass separation devices using electrostatic or magnetic fields coupled to laser ionization. It is devoted to environmental and fundamental research measurements. This work is a first step toward a larger project, which aims to develop an offline mass separator for radionuclide production in association with ARRONAX. 4.1.1.3 Reactor-based and fast-neutron facilities ¡ ILL: Located in Grenoble, France, the Institute Laue-Langevin (ILL) High-Flux Reactor produces the most intense continuous neutron flux in Europe. The primary application is for fundamental and applied research with neutron beams, but in-pile irradiations for the production of medical or R&D radionuclides are also performed. ¡ BR2: Located at SCK CEN in Mol, Belgium, the Belgian Reactor 2 (BR2) is one of the world's most powerful research reactors. It produces neutron fluxes for testing nuclear materials and fuels. It is a key player in the global supply of radionuclides used in nuclear medicine for diagnostics and treatment. ¡ JHR: Under construction at the CEA Cadarache research centre in the south of France, the Jules Horowitz Reactor (JHR) is a state-of-the-art Material Testing Reactor (MTR). It is designed to perform advanced research on material and fuel behaviour under irradiation, support the development of next-generation nuclear power plants, and produce medical radionuclides. ¡ MYRRHA: Under design and construction at SCK CEN in Mol, Belgium, MYRRHA (Multi-purpose hYbrid Research Reactor for High-tech Applications) is projected to be the world's first large-scale Accelerator Driven System (ADS). It combines a subcritical nuclear reactor with a high-power linear accelerator and aims to demonstrate efficient transmutation of high-level nuclear waste and to produce neutrons for advanced research in reactor development, materials science, and medical applications. ¡ ESS: Located in Lund, Sweden, the European Spallation Source (ESS) is an emerging research facility designed to be the world's most powerful accelerator-based neutron source. It utilizes a high-power Deliverable D8.3 11 linear proton accelerator to produce intense neutron beams for advanced research in materials science, physics, chemistry, and biology. The feasibility of radionuclide production is being investigated. ¡ GANIL-NFS: Located at GANIL in Caen, France, the Neutrons for Science (NFS) facility operates as part of the SPIRAL-2 complex. It provides high-flux pulsed neutron beams for fundamental research in nuclear physics, as well as applied studies in nuclear medicine. NFS enables precise measurements of neutroninduced reactions, crucial for advancing nuclear data and technologies such as radionuclide production. ¡ IFMIF-DONES: Envisaged in Granada, Spain, the International Fusion Materials Irradiation Facility – Demo Oriented NEutron Source (IFMIF-DONES) is projected as a future research infrastructure. It will use a high-current deuteron accelerator to produce intense neutron beams for testing and qualifying materials under extreme conditions, simulating those in future fusion reactors. This facility will support various scientific and industrial applications. The feasibility of radionuclide production is being investigated at the moment. The following sub-sections give more details on cyclotron-based and emerging PRISMAP infrastructures. 4.2 GIP ARRONAX GIP ARRONAX is a public research institute grouping several public members: the French State and the Region "Pays de la Loire", large national organizations of research (CNRS and INSERM), higher education and research institutions (Nantes Université and IMT Atlantique), hospitals (Nantes University and ICO). ARRONAX aims at producing innovative radionuclides and radiopharmaceuticals for research in nuclear medicine and at performing research using its beams in physics, radiochemistry and biology. The GIP ARRONAX is equipped with a high-energy, high intensity multiparticle cyclotron, C70XP installed in the vault CC (see Figure 3), which is operational since 2010 and has recently installed a biomedical cyclotron named IK18 in the vault P1 (see Figure 3). GIP ARRONAX has also set up a radiopharmacy in collaboration with Nantes University Hospital to prepare radiopharmaceuticals for clinical trials. Figure 3. Layout of the GIP ARRONAX Facility. The C70XP cyclotron is installed in vault CC and can provide beams in all vaults, but P1 where a Kiube 180 is installed with its beam line. Production vaults are equipped with a rabbit system that allow sending irradiated targets to the hot cell lines as depicted on the figure. 4.2.1 ARRONAX cyclotrons The C70XP machine is very versatile and well-adapted for proof-of-concept (PoC) studies. Its wide capabilities allow for enlarging the scope of nuclear reactions that can be used to produce a given radionuclide by looking Deliverable D8.3 12 to exotic nuclear reactions, high-energy reactions and even low cross-section reactions. The new biomedical machine allows translation of PoC to conditions closer to industrial methods. 4.2.1.1 C70XP cyclotron The C70XP can accelerate both positive (HH+, He-4++) and negative ions (H-, D-) up to 70 MeV. It delivers up to 750 µAe (two beams of 375 µAe) of protons. The capabilities of C70XP are summarized in Table 3. Negative ions are extracted using the stripper foil technique that allows access to a wide range of incident energy by changing the radial position of the foil. As an example, using this technique, our proton beam can be extracted from 30 MeV up to 70 MeV. ARRONAX is equipped with two such devices, opposed, giving the ability for negative ions to extract two beams simultaneously. Positive ions are extracted using an electromagnetic septum, limiting the output beam at a fixed energy, 68 MeV for alpha (a) particles. Beams extracted from the C70XP cyclotron can be delivered in 5 experimental vaults named AX, A1, A2, P2, and P3 (see Figure 3). Due to the extraction method, which is different for negative and positive ions, protons and deuterons are available in every experimental vault, whereas α-particles and HH+ beams are available only in vaults A1, A2 and AX. Vaults A1, A2, P2 and P3 are equipped with target systems and remote pneumatic transfers (rabbit system) of the irradiated materials from the beam lines into the hot cells. This design allows for flexibility in production. The largest Vault, AX, is devoted to research using our beams. Researchers can benefit from up to 3 consecutive days of beam time to collect their data. They have to install their apparatus prior to the experiment and dismantle it at the end of the experiment. Any of the 3 beam lines available on AX can be used. Several laboratories are also available to prepare samples or perform measurements. For external radiotherapy experiments, we have set up an animal housing facility on site. An upgrade of our facility is ongoing to be able to have access to an a beam at 29 MeV for At-211 production. Indeed, the GIP ARRONAX is one of the two European facilities producing At-211 regularly, but our production method is far from optimal. As we currently extract the a beam at a fixed 68 MeV, we need to degrade the beam energy by almost 40 MeV to produce At-211 with the proper quality, i.e. with a minimal amount of At210. This limits the production yield. To overcome this limitation, we are installing an internal target system that will allow to enter the target inside the cyclotron to irradiate it at the right energy without longer need for a beam energy degrader. This should allow to increase our production yield and beam intensity on target. 4.2.1.2 IK18 cyclotron The IK18 cyclotron is a Kiube 180 from IBA. It is installed in the vault P1 and allows delivery of proton beams at 18 MeV up to 180 µA (see Table 3). It can be equipped with up to 8 target ports placed on the periphery of the machine and can be upgraded in the future to 300 µA if needed. In our case, we have used one target port to install a short beam line connected to a high-power solid target system of the same type as of the C70XP. This target system is connected to our rabbit system, allowing solid targets to be sent to any of our hot cells for chemical processing. Table 3. Characteristics of the available beams at ARRONAX Beam Accelerated ions Energy Range (MeV) Intensity (µAe) C70XP Protons H30-70 <375 (x2) HH+ 17 <50 Deuterons D15-35 <80 α-particles He-4++ 68 <70 Kiube 180 Protons H18 150 Dual beam Deliverable D8.3 19 positioned to play a significant role in addressing the increasing global demand for radionuclides used in medical diagnostics and radiotherapy, particularly those difficult or impossible to obtain using conventional reactor-based or cyclotron-based systems. The heart of the facility is the SPES high-temperature target, specifically designed to withstand intense irradiation from proton beams with energies up to 40 MeV and currents up to 200 μA, delivered by a highperformance cyclotron. This configuration triggers nuclear reactions necessary for the production of exotic nuclei. These nuclei are then extracted, ionized, mass-separated, and directed to experimental areas for a variety of applications. The modular design of SPES allows for a high degree of flexibility in target and ion source combinations, opening the door not only to a wide range of radionuclides for fundamental research but also to medically relevant radionuclides, including both diagnostic and therapeutic candidates. In this context, SPES is conceived as a multi-purpose platform that could foster collaborations between nuclear physicists, medical researchers, and pharmaceutical developers. The commissioning phase of the SPES facility was successfully completed between the end of 2024 and the beginning of 2025, marking a major milestone in the project’s development. During this phase, a lowintensity proton beam was used to irradiate a high-temperature silicon carbide (SiC) target under operational conditions. This test allowed for the first online production and extraction of radionuclides through the full SPES beamline. In particular, the radionuclides Al-28 and Al-29 were monitored at the tape station, providing key validation of the target performance, ion source operation, mass separation system, and radioactive beam diagnostics. These successful operations confirmed the functionality of the critical components and established a solid foundation for the next stages of activity. Today, SPES is recognised as an emerging ISOL facility on the European and international scene, and its roadmap for the coming years includes a progressive increase in the operational proton beam intensity, which will significantly enhance the yield and diversity of radioactive species that can be delivered. Future irradiations will employ more advanced target materials, such as titanium carbide (TiC) and uranium carbide (UCx), to explore new regions of the nuclear chart and expand the range of available radionuclides for both scientific and applied purposes. Through these developments, SPES aims to become not only a reference infrastructure for the nuclear physics community but also a key enabler of medical and industrial applications of radionuclides, reinforcing the strategic mission of INFN to bridge fundamental research and societal needs. Figure 9. The SPES ISOL facility at Legnaro National Laboratories. Deliverable D8.3 20 4.7 MYRRHA and ISOL@MYRRHA The MYRRHA (Multi-purpose hYbrid Research Reactor for High-tech Applications) project has the ambitious goal of developing a liquid Lead-Bismuth Eutectic (LBE) spallation target and an LBE-cooled, sub-critical fast core driven by a high-intensity proton beam19. The maximum proton beam current is foreseen to reach 4 mA at 600 MeV. This project will be realised at SCK CEN, in Belgium. One of the milestones of MYRRHA is the demonstration of the Accelerator Driven Systems (ADS) concept to tackle the pressing issue of the longevity of nuclear waste. This will be achieved by the transmutation of heavy nuclear waste, thus effectively reducing the risks these radionuclides impose on the environment. MYRRHA is implemented in 3 phases, starting with the construction of a 100-MeV linear accelerator the Proton Target Facility hosting ISOL@MYRRHA20 and the Full Power Facility, dedicated to materials irradiation for fusion research. Phase two is focused on the implementation of the 600 MeV linear accelerator, which will be coupled to the subcritical nuclear reactor in phase 3. The ISOL facility at MYRRHA is being constructed and commissioned within phase 1 of the MYRRHA programme scheduled for commissioning starting in 2030. The MYRRHA accelerator will provide this radioactive ion beam laboratory with a high-intensity 100-MeV proton beam, up to 0.5 mA at 250 Hz. A wide variety of target materials from light (eg. SiC) to heavy compounds (e.g. ThCx), will be used for producing the radioactive ion beams to be delivered to the users. These have been developed over the years at established ISOL facilities, and together with the ion source they will be further optimised at ISOL@MYRRHA to achieve the best performance under high-intensity irradiation. The delivery of a purified ion beam is envisioned by using a pre-separator followed by high-resolution mass separators (potentially, in combination with a gasfilled linear Paul trap). ISOL@MYRRHA will house one target station, an experimental area as well as user laboratories for supporting the activities of different research groups (see Figure 10). Figure 10. Horizontal cut at beam-level through the MYRRHA-phase1 buildings. 19 MYRRHA: https://myrrha.be 20 ISOL@MYRRHA: https://myrrha.be/myrrha-applications/nuclear-science/isolmyrrha Deliverable D8.3 21 The ISOL@MYRRHA facility will provide opportunities to perform measurements for extended periods, considerably longer than in other currently operational laboratories. Moreover, various medical radionuclides will be produced, which can be exploited for research studies with radiopharmaceuticals or searches for innovative medical radionuclides. Radionuclides for medical applications will be collected in a dedicated collector station, from where the samples can be packaged and shipped either to the users or to other specialised facilities at SCK CEN, for undergoing chemical purification. SCK CEN has dedicated radiopharmacy labs for the development of new radiopharmaceuticals. These facilities cover domains as radiolabelling, organic synthesis, or chelator design. To these, one must add the R&D radioisotope laboratory for the development/upscaling of radiochemical processes towards GMP radionuclide production21. 4.8 TATTOOS IMPACT (Isotope and Muon Production using Advanced Cyclotron and Target technologies) is a new Swiss Research Infrastructure to be constructed at the Paul Scherrer Institute (PSI), in collaboration with the University of Zurich (UZ) and the University Hospital Zurich (USZ)22. It entails the construction of two new target stations and beamlines at HIPA, thereby extending the existing infrastructure. The first aims to increase currently available muon intensities by up to 100-fold for experiments in particle physics and materials science (High Intensity Muon Beams - HIMB), while the second (Targeted Alpha Tumour Therapy and Other Oncological Solutions - TATTOOS) aims to provide a wide range of previously elusive radionuclides suitable for advanced cancer treatments in desired quantities towards potential clinical studies. Terbium is a unique element in that it presents four interesting radioisotopes for all modalities in diagnostic and therapeutic nuclear medicine: Tb-155 emits γ-radiation suitable for Single Photon Emission Computed Tomography (SPECT), while Tb-152 decays by the emission of β+ particles useful for Positron Emission Tomography (PET). Tb-161 emits low-energy βparticles, but also a significant number of Auger/conversion electrons, which present a powerful combination for therapeutic purposes. Finally, there is Tb-149, the “flagship” nuclide for the TATTOOS project, which has promising properties for α-therapy, even possibly combined with PET23. The existence of four radioisotopes that allow the preparation of chemically identical radiopharmaceuticals, with equal pharmacokinetic profiles, is a unique situation that could serve the concept of personalized medicine, thereby offering diagnostic tools and various therapy options for cancer patients in future. While Tb-161 can be produced via a research reactor24, the others are more challenging to obtain and spallation, followed by mass separation and collection, is necessary (also known as Isotope Separation OnLine, or ISOL). It is possible to utilize the mass-separation technique at PSI’s HIPA infrastructure with the construction of an ISOL facility (TATTOOS) to foster this new research area based on novel, emerging medical radionuclides. TATTOOS will take advantage of the 590 MeV, high-intensity proton beam at HIPA to induce spallation reactions on heavy-element targets (see Figure 11). The spallation process yields radionuclides that are currently not accessible or only very difficult to obtain via lowand medium-energy proton accelerators and neutron sources. The 50-fold higher proton beam intensity and the almost unrestricted irradiation time at TATTOOS (100 µA) compared to ISOLDE/CERN (2 µA) will yield multi-GBq quantities of promising and, possibly, unexplored radionuclides of medical relevance or interest. Online isotope mass separation, as part of the TATTOOS facility, will provide a smorgasbord of interesting radionuclides at an unprecedented quality and quantity, thereby, potentially facilitating clinical studies using these novel radionuclides. 21 https://www.sckcen.be/en/infrastructure/crf-centralized-radiochemical-facility 22 IMPACT, https://www.psi.ch/en/impact 23 Müller et al, J Nucl Med, 2012, 53: 1951. 24 Gracheva et al., EJNMMI Radiopharmacy and Chemistry, 2019, 4: 12. Deliverable D8.3 22 This venture aims at enabling the collection, as well as chemical separation of novel, promising radionuclides in unprecedented activities, e.g. Tb-149 and Tb-152. These can be introduced into the GMP facility on site (at Center of Radiopharmaceutical Sciences – CRS) and at the University Hospital Zurich (USZ), used for labelling of biomolecules, and the resultant radiopharmaceutical can be transported to medical imaging and therapy centres for patient application. The concept will cover basic research and development thereof as part of a “bench-to-bedside” principle. Figure 11. The IMPACT layout concept, with HIMB highlighted in green and TATTOOS in blue. 100 µA of proton beam will be split from the available 2.4 mA 590-MeV proton beam from the “Ring” cyclotron to induce spallation reactions using a constructed beamline and target station at TATTOOS. The TATTOOS concept is based on what is currently operated at ISOLDE/CERN25 . Some radionuclides attractive for medical purposes but are difficult to produce have been mass-separated and collected following a high-energy spallation reaction (ISOL) using a tantalum target. This was demonstrated when terbium radioisotopes were extracted and, subsequently, chemically separated from their zinc collection material, as well as from their isobars, such that it could be utilised for radiolabelling, preclinical and even a first-in-human application26,27,28,29. In the desire to capitalise on PSI/ISOLDE’s successes with these fundamental research experiments, to introduce them into the clinics, the TATTOOS concept germinated. 25 ISOLDE-CERN, https://isolde.cern/isolde-facility 26 Müller et al., EJNMMI Radiopharmacy and Chemistry, 2016, 1: 5. 27 Umbricht et al., Scientific Reports, 2019, 9: 17800. 28 Baum et al., Dalton Trans., 2017, 46: 14638. 29 Müller et al., EJNMMI Res., 2019, 9: 68. Deliverable D8.3 23 4.9 SMILES The SMILES project, funded by the French State and the Region "Pays de la Loire" is led by the SUBATECH laboratory in Nantes. It consists in developing mass separation devices using electrostatic or magnetic fields coupled to laser ionization. These facilities will meet the needs for isotopic analysis of trace elements in the environment - a major challenge for determining their origin, assessing their impact and monitoring their migration (copper, radium ...) - and for isotopic analysis, and research on separation and purification for the production of innovative radionuclides for nuclear medicine (e.g. Sc-44, Sc-47, Cu-67). Finally, it will also allow purification and enrichment of rare isotopes. The new facilities will be located in a dedicated building within the SUBATECH laboratory on the IMT Atlantique campus (see Figure 12). Figure 12. SMILES building with Laser room, magnetic mass separator room. The design and construction, which began in 2023, were divided into three parts: 1. Calculations and simulations of charged particle optics using SIMION software30. 2. Construction of the analytical mass separator, a time-of-flight (TOF) mass separator using electric fields and a reflectron to enhance its isotope separation power. This TOF mass separator can be combined with a desorption ionization laser source for sample mapping or with a thermal source for sample volume analysis. The laser ionization system with four Ti:Sa cavities is based on a model developed by the LARISSA group at the Johannes Gutenberg University in Mainz, Germany. The assembly of the first TOF prototype is now in progress (see Figure 13). A Synchronization system between lasers, electrode voltage and detection has been implemented. A prototype of the thermal source is under design (see Figure 14). The first experiments on copper are scheduled before June 2025, and the results will be compared with simulations already carried out. 3. The combination of the magnetic mass separation with laser ionization and a thermal source. 30 https://simion.com/references.html Figure 13. Desorption-ionization source in TOF-L with target holder (1), acceleration electrode (2), Einzel Lenses (3) and mirrors. Deliverable D8.3 24 Figure 14. Prototype design of thermal source (a) and ANSYS simulation of the tantalum oven (b). The entire installation is scheduled for completion by the end of 2028. It will be followed by a second project including an off-line mass separator dedicated to radionuclide production close to ARRONAX. 4.10 JHR The Jules Horowitz Reactor (JHR) is a project led by the CEA (see Figures 15 and 16) and sponsored by several international partners (15 members). It is currently under construction in France at the CEA Cadarache centre with a start-up foreseen for the end of 2032. JHR has been designed and developed to: ¡ support the nuclear industry (current and future nuclear reactors), providing a large range of experiments to test the behaviour of materials and fuels under irradiation (from selection to qualification needs), ¡ ensure between 25% to 50% of the European supply of the Mo-99; and other radionuclides needed in nuclear medicine and for other applications. To be able to be a significant player in the supply of medical radionuclides, it will feature, among other things, the following items: ¡ 20 locations suitable for these irradiations: ¡ 3 in core locations (for long-duration irradiations) ¡ 4 displacement systems dedicated to Mo-99 production located in the reflector ¡ 13 reflector locations (for short irradiations) ¡ specific irradiation devices and containers ¡ One dedicated hot cell for the extraction of the Mo-99 targets to the transfer cask ¡ One shared hot cell for the extraction of other highly radioactive targets (i.e. Co-60, Ir-192) ¡ One shielded box for the extraction of less radioactive targets (i.e. Lu-177, Ho-166, …) Deliverable D8.3 25 Figure 15. View of the core and reflector locations (in yellow locations dedicated to radioisotope production – in blue shared locations with experimental irradiations). Figure 16. Left: 3D view of the reactor and auxiliary JHR buildings with the location of the hot cells (light blue) and the pool channels (dark blue). Right: 3D view of the 7 JHR hot cells. Additional information is available via email or at the JHR website31. The neutron flux and spectrum for some irradiation positions are shown in Figure 22. 31 https://jhrreactor.com/en. email: [email protected] Deliverable D8.3 26 The European Spallation Source (ESS) is being constructed in Lund (Sweden) to deliver neutron beams of high brightness to a new generation of scientific instruments32,33. The aerial view of the ESS facility is shown in Figure 17. Figure 17. Aerial view of the ESS site in Lund (Sweden) highlighting the 540-meter-long high-power proton linac and the target building. A simplified model of the spallation target is included to show the rotating tungsten wheel with a diameter of 2.5 m and the two moderators/reflector systems (in green) to moderate the neutrons to thermal and cold energies for neutron scattering experiments. At present, only one moderator/reflector system, placed above the target, serves all the instruments. This gives the option to place an irradiation facility below the target during future upgrades of the ESS target. The ESS will become the world’s most powerful accelerator-based neutron source, once a 2-GeV proton beam will be accelerated with a current of 62.5 mA in 2.86-ms long pulses. The installation and commissioning of the ESS proton accelerator started in 2018. During the first half of 2025, for the first time, 800 MeV protons were accelerated with a current 10 times lower than the nominal one and in test pulses of 5 µs. The facility is expected to start commissioning in early 2026 with the first beam on target, and the construction is expected to be completed by 2027. After an initial phase of ramp-up of the proton current (which can last a few years), ESS will operate for several years with a 2 MW proton beam with 800 MeV energy. Spallation neutrons will be produced by the impact of the proton beam with a rotating tungsten target; neutrons will be slowed down by a moderator system placed right above the spallation target; thermal and cold neutrons will be used in a variety of neutron scattering experiments performed in dedicated instruments. Currently, 15 instruments are in construction, with 22 instruments foreseen for the full scope of ESS, but more instruments are possible, thanks to an available grid of 42 beam ports for neutron extraction. The 15-instrument suite consists of small-angle instruments, reflectometers, imaging beamlines, singlecrystal diffractometers, powder diffractometers, as well as an array of five inelastic instruments. The instruments are designed to provide new scientific capabilities, not currently available at existing facilities, 32 R. Garoby et al., The European Spallation Source Design, 2018 Phys. Scr. 93 014001 33 K. Andersen et al., The instrument suite of the European Spallation Source, Nucl. Instrum. Methods A, Volume 957, 21 March 2020, 163402, https://doi.org/10.1016/j.nima.2020.163402 Deliverable D8.3 27 building on the inherent strengths of the ESS long-pulse neutron source of high flux, the flexible resolution and large bandwidth. The main parameters of ESS are listed in Table 5. Table 5. ESS high-level parameters. The source is pulsed with a repetition rate of 14 Hz and pulse width of 2.86 ms. Parameter Units Figure day 1 (nominal) Average beam power MW 2(5) Proton kinetic energy GeV 0.8 (2) Pulse repetition rate Hz 14 Average pulse current mA 62.5 Macro-pulse length ms 2.86 Number of instruments 15(22) Number of neutron beam ports 42 Number of moderators 1(2) Although not in the scope of ESS, the facility can offer various opportunities for sample irradiation with fast neutrons. The design optimisation of the source has resulted in a single moderator system placed above the spallation target. Combined with the flexibility offered by the large number of beamports, many of which are unused, one could design and build a facility for radioisotope production with fast neutrons placed below the spallation target, with a dedicated insertion and extraction via e.g. a “rabbit system”. It is important to place the samples close to the target, giving the possibility to use the unmoderated high-energy spectrum of spallation neutrons. Below the spallation target, a fast neutron flux of about 6 × 1014 cm-2 s-1 at 5 MW power, integrated for neutrons above 100 keV, could be reached. In case a second moderator would be installed in the future below the spallation target (see Section 5.7), this would still allow for the placement of such an irradiation facility, with a resulting lower high-energy flux, and an increased thermal component, which could also be exploited for radionuclide production. 4.11 GANIL-NFS The Grand Accélérateur National d’Ions Lourds (GANIL) located in Caen, France, has operated cyclotrons for more than 40 years and a superconducting LINAC since 202034. The LINAC accelerates light charged particles (p, d, He-4) as well as heavy ions via an A/q=3 injector. The main characteristics of the beams accelerated by the LINAC are as follows: ¡ Up to 33 MeV for protons; ¡ Up to 40 MeV for deuterons; ¡ < 14.5 MeV/A for heavy ions. The intensities can be as high as 5 mA for light charged particles (p, d, He-4) and 1mA for heavier ions. Neutron beams are produced using two methods at the NFS (Neutron For Science) facility35,36 , which is composed of a converter room and a Time-of-Flight (ToF) hall (see Figure 18). An irradiation station is installed in the converter room. This station allows for putting a sample under vacuum in order to be 34 https://www.ganil-spiral2.eu/scientists/ganil-spiral-2-facilities/accelerators/ 35 X. Ledoux et al., Eur. Phys. J. A 57, 257 (2021). 10.1140/epja/s10050-021-00565-x 36 X. Ledoux et al., EPJWeb Conf. 146, 03003 (2017). 10.1051/epjconf/201714603003 Deliverable D8.3 28 irradiated by protons, deuterons or ions. It is connected to a pneumatic transfer (rabbit) system. After irradiation, the irradiated samples are sent to the ToF area to measure the activity and to deduce the crosssections. Figure 18. The two areas of the NFS facility at GANIL. Left: the converter room hosting a rabbit system for activation studies, the converters and a bending magnet to remove protons crossing the thin converter from the neutron beam. Right: the ToF area. A first collimator separates the converter and ToF rooms and a second one (not shown on the picture) can be installed in order to shape the neutron beam and remove the halo. Neutrons are produced either by deuteron breakup reaction on a thick beryllium converter or by (p,n) reactions on a thin lithium converter. This will produce a continuous energy distribution extending up to 40 MeV in the first case and a quasi-mono energetic spectrum in the second one. While the projectiles are stopped in the thick converter, the protons are swept out from the neutron beam by a bending magnet after crossing the thin converter. The LINAC operates with a beam frequency of 88 MHz. For ToF measurement, a fast chopper is used to reduce the frequency by a factor of 100 or more. 4.12 IFMIF-DONES The main objective of the International Fusion Materials Irradiation Facility Demo Oriented NEutron Source (IFMIF-DONES) will be to test materials under similar neutron irradiation conditions as expected in nuclear fusion reactors37,38 (see Figure 19). Together with the International Thermonuclear Experimental Reactor (ITER), IFMIF-DONES is the key facility to design the DEMO fusion reactor. DONES, a first step to IFMIF, will consist of a 40 MeV deuteron accelerator at 125 mA, meanwhile, IFMIF will consist of two accelerators and one neutron production target. In DONES, the deuteron beam with a cross-section area of 20 × 5 cm2 will be intercepting a lithium jet target in the Test Cell (TC). The fast neutrons (»14 MeV) will be produced by means of the stripping reaction 2H + Li, and they will be contained inside the TC39. The samples for fusion studies will be located inside the High Flux Test Module (HFTM) just behind the lithium jet, where a neutron flux of 1014 cm-2 s-1 is expected. In parasitic mode to the irradiation of materials, the facility will be available to produce radionuclides with neutrons in a large area just behind the HFTM. In addition, the deflection of a part of the deuteron beam, 1% or 0.1%, to a new experimental area (R026) located below the accelerator is being studied to provide a large variety of applications. One of them would be the production of radionuclides with deuterons40. The next figure shows the schematic of the accelerator, target assembly, TC, HFTM, and deflection of the beam to the R026 hall. 37 F. Mota et al., Nucl. Fusion 55 (2015) 123024. 38 D. Bernardi et al., J. Fusion Energy 41 (2022) 24. 39 W. Królas, et al., Nucl. Fusion 61 (2021) 125002. 40 J. Praena, et al., EPJ Web Conf. 239 (2020) 23001. Deliverable D8.3 35 considering the entire capsule. The highest heating values were observed for the platinum sample due to its high density and strong photon attenuation properties. Despite this, photons remained the primary contributors to heating of the capsule. In general, target cooling is not a concern for heavy-metal-cooled fast reactors, making them promising candidates for medical isotope production. 5.5 Methodology followed by SCK CEN – ISOL@MYRRHA 5.5.1 Ac-225 and Tb-152,155 production This study investigated the production capabilities of Ac-225 from a 100 MeV proton beam with a current of 200 µA on thorium-based targets. For the production of Ac-225 the Th-232(p,x)Ac-225 and Th-232(p,x)Ra225 routes were considered. A 10 g cm-2 thorium-carbide target at 200 µA was used in the study. For this configuration, 50 MeV is lost in the target, corresponding to 10 kW in target heat deposition, which is assumed to be within reach with current technologies. FLUKA 73 was used to calculate the in-target production rates of the radionuclides74, and a custom script was used to further process the scenarios by simulating the evolution of the isotopic inventory during irradiation, decay, chemical purification and mass separation. For the routes via radium, three milking periods for the extraction of actinium are considered. This study also investigated the production capabilities of Tb-152 and Tb-155 through the irradiation with a 100 MeV proton beam on natural gadolinium targets, followed by chemical purification and subsequent mass separation. TENDL-based cross-sections were used to estimate the in-target production rates. In a later phase, FLUKA was used to estimate the purity of the samples. The results from these investigated scenarios are in Section 10.3 for actinium and in Section 8.2.1.5 for terbium. 5.5.2 Sc-47 production This work investigates proton irradiation of a Ti target as a potential pathway to address the limited availability of Sc-47 for clinical applications. The aim is to assess the production when operating a 100 MeV proton beam impacting on a thick target and explore the feasibility of mass separation post-chemical purification as a mean to improve the quality of the final product. The considered production path involves a chemical process for which the dissolution of the target is required. Therefore, the oxide and the metal forms of Ti are studied. TiO2 can be dissolved in concentrated HF or in H2SO4 or NH4HF2 and HCl, while Ti metal can be dissolved in NH4HF2 and HCl. Besides, 50TiO2 is available from Isoflex USA (83 - 95%), from Trace Sciences International Inc., Delaware, USA (90 – 94+%). Both natural and enriched targets have therefore been studied. The Monte Carlo code FLUKA was used to simulate the production rate of radionuclides in the targets. These production rates are subsequently processed with a script that calculates the radioactive decay and buildup during decay, along with the recovered nuclides after chemical purification and mass separation. Owing to the thermal challenges of a target design at higher current, the results given in this study correspond to a low-power target and a high-power target (current values of 10 and 100 µA). 5.6 Methodology followed by JHR The present study is based on simulations executed with the TRIPOLI-4® code version 12.175, a probabilistic 3D simulation tool, based on the Monte Carlo method, using the JEFF3.1.1 nuclear data library76. The TRIPOLI73 T. Bohlen et al., Nuclear Data Sheets 120, 211-214 (2014) 74 Calculations performed at SCK CEN in 2022 were used in this study 75 F.-X. Hugot et al., EPJ Nucl. Sci. Technol. 10, 17 (2024). 76 A. Santamarina et al., JEFF report, 22, OECD-NEA data bank, 2009. Deliverable D8.3 36 4® code has been selected to accurately solve the Boltzmann neutron transport equation for a detailed 3D model of the JHR. The modelling of depletion, activation, transmutation and burn-up induced by neutrons is based on the inventory code DARWIN77 version 2.4.8, using the EAF2010 nuclear data libraries78 The modelling of the reactor core comprises 34 fuel elements of HALEU-type (High Assay Low Enriched Uranium), based on U3Si2/Al with a density of uranium of 4.8 g cm-3, and takes into account accurate burnup distributions corresponding to mid-cycle conditions. The control rods are withdrawn in a configuration that approaches a critical state. A cross-sectional view of the JHR core is presented in Figure 21 (left), where Mo-99 will be produced in the so-called MOLFI movable systems. The figure also illustrates two locations specifically reserved for the production of radionuclides: one embedded in a fuel element at the centre of the core, and the other located in the innermost part of the reflector, in a rotating aluminium plug. The targets are hermetically sealed in quartz ampoules (Suprasil, 8 x 36 mm), which are placed in welded aluminium containers (or "rabbits") for irradiation. The containers are axially stacked on top of each other and cooled by water in forced convection. In the present study, two-irradiation positions are being considered for comparison: one is located in the centre of fuel element #105 in order to get the highest fast neutron flux, the other is placed in a rotative aluminium plug located in the first ring of reflector elements, as indicated in Figure 21 (right side). Figure 21. Cross-sectional view of the JHR core (left) and the irradiation rig loading with the target material (right). 5.6.1 Irradiation scenario The reactor is designed to operate for 25 Equivalent Full Power Days (EFPD) at 100 MW rated power, followed by 25 days of outage. This corresponds to a cycle length of approximately 36 days at 70 MW, the effective power level assumed for radionuclide production in this study. It is important to note that, in the current design, the irradiation positions within the core are only accessible during the outage period of refuelling. Conversely, the irradiation positions situated in the reflector can be readily discharged at any time during the reactor operation. For radionuclides characterised by very short half-lives, it is imperative to optimise the duration of irradiation and the time for the extraction and transfer 77 A. Tsilanizara et al., Journal of Nuclear Science and Technology, Supplement 1, 845 (2000). 78 J.-Ch. Sublet et al., The European Activation File: EAF-2010 neutron induced cross-section library (CCFE-R(10) 05, 2010)- Deliverable D8.3 37 of the radionuclides. In light of this, the JHR operator is currently considering the design of a transfer system that would facilitate online access to the core while the reactor is operating. The implementation of such a system would ensure the most favourable irradiation duration for the radionuclide production. The JHR facility is equipped with hot cells for the handling of the irradiated samples and radionuclide targets. The hot cells can accommodate a gamma source equivalent to 3.7 1012 Bq of Co-60 and a neutron source above 5 108 n·s-1. 5.6.2 Neutron flux performances The "unperturbed" neutron flux is calculated at the core axial mid-plane in a volume occupied by a fictitious target material. This modelling hypothesis is deemed acceptable since the reactions of interest in this study are in the epithermal and fast energy range of the neutron flux, and the small amount of material involved in the targets is unlikely to cause a significant flux depression. A cadmium screen is employed as a neutron filter to eliminate the thermal component of the neutron spectrum. The results are reported in Table 7. As observed in Figure 22, the implementation of a 2 mm thick cadmium screen results in a substantial reduction in the thermal flux below the characteristic cadmium energy cut-off, which is situated at 0.5 eV. Conversely, this screen has a negligible impact (- 4 % only) on the epithermal and fast components of the flux, as expected. It is also important to note that the fast flux level in the core is almost 10 times as high as in the reflector. Consequently, the remainder of the study will focus on the evaluation of production capacities within the core exclusively. Table 7. A comparison of the Cd-filtered and unfiltered neutron flux (cm-2·s-1) obtained in the two irradiation positions (in-core and in-reflector) at the core axial mid-plane, at operating power level of 70 MW. Irradiation position / Container Thermal (E ≤ 0.625 eV) Epithermal (0.625 eV ≤ E ≤ 1 MeV) Fast (1 MeV ≤ E) Total Core / quartz 3.48E+14 1.07E+15 4.07E+14 1.82E+15 Core / cadmium 3.94E+12 1.02E+15 3.89E+14 1.42E+15 Reflector / cadmium 2.01E+12 1.73E+14 4.01E+13 2.15E+14 5.7 Methodology followed by ESS The calculations presented in this report were performed for two models of the ESS spallation target. The models are shown in Figure 23. The essential components of the ESS source are displayed: the tungsten spallation target and the upper moderator/reflector system. The two models differ in the components installed below the spallation target. Figure 22. Neutron flux using a 281-group energy structure at the in-core position #105 of the JHR at operating power level of 70 MW. Deliverable D8.3 38 The “one-moderator” model (Figure 23, left), corresponds to the source with one moderator system placed above the target, while the space below the target is occupied by a steel plug, which also acts as a reflector for the fast neutrons. This is the present configuration of the ESS source, as all the 15 instruments under construction are pointing to the upper moderator. In the “two-moderators” model (Figure 23, right), the space below the target is occupied by a large liquid deuterium moderator, which complements the upper moderator to provide a cold neutron beam from a larger emission surface, intended for additional future instruments. As of 2025, it is unclear if a second moderator system will be installed, but it is likely that in the future a second moderator system will be placed below the target, to fully profit from the potential of ESS. This would not prevent the construction of an irradiation facility in the same location, but it is important to perform calculations with and without a second moderator system, as this will influence the neutron spectrum, as shown in Figure 24. In both models, the irradiation samples are located a few centimetres below the spallation target and correspond to the location with the highest flux of fast neutrons. In the one-moderator model, this place is currently empty, due to the high heat load in the area, which makes the cooling challenging. Such a challenge will also exist for an irradiation facility and must of course be addressed in the design stage. In the twomoderators model, the irradiation location is inside the water pre-moderator. In this case, in order to have the irradiation facility and the lower moderator coexisting, a new design of the lower moderator would be necessary to be able to insert and extract samples without significantly reducing the performance of the lower moderator. Figure 23. MCNP models of the ESS target-moderator-reflector system. Left: one-moderator model. Right: twomoderators model. For both ESS models it was assumed that the ESS linear accelerator will provide a proton beam of 2 GeV, 62.5 mA peak current, repetition rate of 14 Hz and 2.86 ms pulse length, corresponding to a time average current of 2.5 mA and a beam power of 5 MW. Calculations of the neutron fluxes were performed in MCNPX79 and are shown in Figure 24. In both cases, a peak of fast neutrons is present in the 1-MeV range, with a highenergy tail extending to the energy of the incoming proton beam. In addition, in the two-moderators model, a second peak of low-energy neutrons is present below 1 eV, due to thermalization in the large liquid deuterium moderator. Integrated fluxes are listed in Table 8. 79 L.S. Waters, et al., The MCNPX Monte Carlo radiation transport code, AIP Conf. Proc. 896 (2007) 81–90. Deliverable D8.3 39 Figure 24. Calculated neutron fluxes for the ESS one-moderator model (black) and two-moderators model (red). Table 8. Calculated integral fluxes in the irradiation positions at ESS, for oneand two-moderator options. Fluxes are calculated assuming 5 MW beam power. Energy Range One moderator [cm-2 s-1] Two moderators [cm-2 s-1] Thermal (E < 0.5 eV) 2.2 1013 3.6 1014 Epithermal (0.5 eV < E < 100 keV) 2.5 1014 4.7 1014 Fast (E < 100 keV) 5.9 1014 4.1 1014 Activation calculations were performed using CINDER9080 for three samples with a mass of 10 g: enriched Zn, enriched Ti and natural Pt with a cadmium shield to produce the radionuclides of interest as listed in Table 1. For the enriched samples, these are the compositions implemented in the MCNPX input file: ¡ Enriched Zn: 88.6% Zn-67, 0.05% Zn-64, 5.9% Zn-66, 5.3% Zn-68, and 0.11% Zn-70. ¡ Enriched Ti: 95.7% Ti-47, 0.41% Ti-46, 3.61% Ti-48, 0.15% Ti-49, 0.13% Ti-50. In the case of Ra-226, it was not possible to perform MCNPX/CINDER’90 simulations, as Ra-226 is not included in the list of implementable materials. Therefore, activation calculations were performed analytically for a sample of 10 g of Ra-226. 80 F.X. Gallmeier et al., The CINDER transmutation code package for use in accelerator applications in combination with MCNPX, in: Proc. ICANS XIX Conference, 2010 Deliverable D8.3 40 5.8 Methodology followed by MEDICIS 5.8.1 Sc-47 production This study investigates the production capabilities of Sc-47 under the following conditions: a 1.4 GeV proton beam at 2 µA on TiC-based targets81. FLUKA-CERN 4-3-0 was used to calculate the in-target production, followed by a decay calculation performed with the CERN ActiWiz software version 3.7.282. An irradiation time of 84 hours and a cooling time of 24 hours have been considered, together with 1.4 GeV protons delivered by the CERN PSB impinging on a TiC target of 29.2 g with a density of 2.7 g cm-3 (volume of 10.7 cm3). The separation efficiency is considered to be 2% as experimentally achieved in 2024 at CERNMEDICIS83. Chemical separation was done on mass-separated samples, implanted in Zn or Al metallic layers. Chemical separation efficiency of 81 ± 9 % with a semi-automated system, 3 mm column, 0.1 g N,N,N’,N’-tetrakis-2ethylhexyldiglycolamide (DGA Resin, Branched, Triskem international) resin was achieved. The final product in 2.5-5 mL 0.1 M HCl. The gravity column method and branched DGA resin yielded a chemical separation efficiency of 91-95 %. Activity at the end of production (including chemical separation after collection) was calculated from experimentally obtained efficiency data and decay-corrected starting activity, leading to a combination of 2 % mass separation and 90 % chemical separation efficiencies. The production of Sc-47 at CERN-MEDICIS undergoes a mass separation step. Therefore, no stable Sc-45 is present in the sample. Any other radioactive Sc isotope impurities were not observed, yielding the highest possible molar and specific activity of the product. The Sc isotope impurity ratio was given from massseparated sample gamma-spectroscopy analysis MDA values, as no impurities were found. The radionuclidic purity was calculated from a mass-separated sample gamma spectroscopy analysis of possible radioactive impurities, such as V-48 and Sr-85 (appearing as atomic mass when Sc-47 is separated in the difluoride sideband). 5.8.2 Cu-67 production This study investigates the production capabilities of Cu-67 under the following conditions: a 1.4 GeV proton beam at 2 µA on ThCx and UCx-based targets. FLUKA-CERN 4-3-0 was used to calculate the in-target production, followed by a decay calculation performed with the CERN ActiWiz software version 3.7.2. 5.8.2.1 Definition of the scenarios A first scenario utilizes an irradiation time of 72 hours and a cooling time of 24 hours with the 1.4 GeV protons delivered by the CERN PSB impinging on a Th-232 target of 100 g of 30.3 cm3 volume with a density of 3.3 g cm-3. The separation efficiency is considered to be 10%84. The second scenario utilizes the same beam and irradiation parameters previously presented, considering Unat target of 105 g of 30 cm3 volume with a density of 3.5 g cm-3. 81 Alfa Aesar (CAS: 12070-08-5), 1-2 µm particle size powder. Target material purity of 99.5 %. Main impurities: Al 150 ppm, Cr 200 ppm, Fe 600 ppm and Si 500 ppm. Target material is pressed in 13mm (diameter) x 1mm pellets, giving density of 2.73 g cm-3. 82 Vincke, H.; Theis, C. ActiWiz–optimizing your nuclide inventory at proton accelerators with a computer code. Prog. Nucl. Sci. Tech. 2014, 4, 228–232 83 Mamis et al, Target Development towards First Production of High-MolarActivity 44gSc and 47Sc by Mass Separation at CERNMEDICIS, Pharmaceuticals 2024, 17(3), 390, https://www.mdpi.com/1424-8247/17/3/390 84 Fedosseev et al., Ion beam production and study of radioactive isotopes with the laser ion source at ISOLDE, J. Phys. G: Nucl. Part. Phys. 44 (2017) 084006 (28pp),doi: 10.1088/1361-6471/aa78e0. Deliverable D8.3 41 5.8.3 Tb-152, Tb-155 production This study investigates the production capabilities of Tb-152, Tb-155 under the following conditions: a 1.4 GeV proton beam at 2 µA on Ta-based targets. FLUKA-CERN 4-3-0 was used to calculate the in-target production followed by a decay calculation performed with the CERN ActiWiz software version 3.7.2. 5.8.3.1 Definition of the scenarios The first scenario utilizes an irradiation time of 24 hours and a cooling time of 24 hours. 1.4 GeV proton delivered by the CERN PSB impinges on a natural Ta metallic foil target. The mass of Ta considered is 180 g for a 30 cm3 volume with a density of 6 g cm-3. The separation efficiency is considered to be ranging from 1 to 10% for Tb following the previous experience with mass separation performed at CERN-MEDICIS. The second scenario considers an irradiation time of 120 hours and a cooling time of 120 hours, with the same target criteria shown above. The third scenario implies an irradiation time of 24 hours and a cooling time of 1 hour, with the same target criteria shown above. 5.8.4 Ac-225 production This study investigates the production capabilities of Ac-225 and its precursor Ra-225 under the following conditions: a 1.4 GeV proton beam at 2 µA on Th-based targets. FLUKA-CERN 4-3-0 was used to calculate the in-target production, followed by a decay calculation performed with the CERN ActiWiz software version 3.7.2. 5.8.4.1 Definition of the scenarios Scenarios #1 and #2 utilize an irradiation time of 240 hours and a cooling time of 72 hours. 1.4 GeV proton delivered by the CERN PSB impinges on Th-232 (ThCx or ThOx) targets. The mass of ThCx is considered as of 100 g for a 30 cm3 volume with a density of 3.3 g cm-3. The separation efficiency is considered to be 72% for Ra-225 as reached in 2024 at CERN-MEDICIS and 12% for ThO2 as reached in 202085. Scenarios #3 and #4 consider an irradiation time of 24 hours and a cooling time of 72 hours and the same parameters as described above. 5.8.5 Pt-195m at the PAEC reactor From its collaboration with PAEC (Pakistan), MEDICIS also contributed data on the production capabilities of Pt-195m under the following conditions: an enriched platinum metal (Pt-194, 96%) target was irradiated at the Pakistan Atomic Research reactor (PARR-I). It was chemically processed to be delivered in the form of PtCl2. Radionuclidic analysis was performed with a BSI HPGe gamma detector with 60 % relative efficiency. 5.8.5.1 Definition of the scenario An enriched Platinum (Pt-194, 96 %) target of 245 mg sealed in a quartz ampule and aluminium can was irradiated at thermal flux 1 1014 cm-2 s-1 inside core for 18 hr at PARR-I. After 24 hrs cooling time, it was chemically processed to get Pt-195m in the form of Pt(II) Cl2. 5.9 Methodology followed by GANIL for ions The aim of this study is to calculate the activities of Sc-47 and Cu-67 that could be produced at the NFS facility at GANIL. The calculations were performed with the FISPACT code. Several parameters, such as target masses or irradiation times, can be adjusted in future studies. 85 Johnson, J.D., Heines, M., Bruchertseifer, F. et al. Resonant laser ionization and mass separation of 225Ac. Sci Rep 13, 1347 (2023), doi: 10.1038/s41598-023-28299-4 Deliverable D8.3 42 The LINAC accelerator can deliver many different beams. The beams of interest and their characteristics are shown in Table 9. Table 9. Characteristics of the beams of interest. Ion Emax (MeV) Imax (mA) Proton 33 5 Deuteron 40 5 Helium-4 80 1 For each calculation, the charged particle flux is calculated in the sample by using the MCNP6 code. The activation calculations have been performed with the FISPACT-II code using the EAF-2010 data and the flux and the sample mass used in the MCNP6 calculation. The thickness of the sample depends on the material and on the beam energy. In each case (Sc-47, Cu-67), several reactions are investigated, and for each of them, the target composition, the irradiation time and the activity of the main radionuclides are given. The calculations are normalized to 100 µA of ion beam. As indicated in Table 9, the LINAC can produce much more intense beams, the limiting factor is probably the maximum power density the sample can sustain. Consequently larger beam intensities are possible, but would require that the beam is spread over a larger area target. 5.9.1 Cu-67 – Studied reactions Table 10. Studied reactions for Cu-67. Reaction Target thickness Main production reaction 30 MeV p + Zn-68 100 µm » 70 mg cm-2 Zn-68(p,2p)Cu-67 20 MeV d + Zn-70 200 µm » 140 mg cm-2 Zn-70(d,αn)Cu-67 16 MeV p + Zn-70 100 µm » 70 mg cm-2 Zn-70(p,α)Cu-67 20 MeV α + Ni-64 20 µm » 18 mg cm-2 Ni-64(α,p)Cu-67 5.9.2 Sc-47 – Studied reactions Table 11. Studied reactions for Sc-47. Reaction Target thickness Main production reaction 18 MeV p + Ti-50 200 µm » 90 mg cm-2 Ti-50(p,α)Sc-47 30 MeV p + Ti-nat 200 µm » 90 mg cm-2 Ti-nat(p,α)Sc-47 30 MeV p + Ti-48 200 µm » 90 mg cm-2 Ti-48(p,α)Sc-47 15 MeV α + Ca-44 50 µm » 8 mg cm-2 Ca-44(α,p)Sc-47 15 MeV α + Ca-nat 50 µm » 8 mg cm-2 Ca-nat(α,p)Sc-47 Deliverable D8.3 43 5.10 Methodology followed by GANIL for neutrons 5.10.1 Neutron flux and spectra At GANIL, the neutrons are produced from the interaction of 40 MeV deuterons with a thick beryllium target (8mm). The neutron flux and energy distribution are calculated with the MCNP6 code. A double differential neutron source was built and used. The whole NFS converter room is included in MCNP6 geometry in order to take into account neutron scattering on walls and equipment. The neutron flux and energy distribution is calculated at 35 cm from the converter in a direction with an angle of 0, 20 and 90 degrees with respect to the beam direction, and using a deuteron beam intensity of 1 µA (see Figure 25). Figure 25. Calculated neutron energy spectra at GANIL. In the present work we study the irradiation of samples located 20 cm from the converter in a direction with an angle of 0, 20 and 90 degrees with respect to the beam direction. The deuteron beam intensity is 40 µA, leading to the production of neutrons at a rate of approximately 1.5 × 1013 s-1. The integrated neutron fluxes for the 3 positions are presented in Table 12. In the following of this study only the position 1 is used. Table 12. Integrated neutron fluxes for the three irradiation positions for a deuteron beam intensity of 40 µA. Position 1 2 3 Angle 0 20 90 Neutron flux (cm-2 s-1) 3.2 1010 1.1 1010 1.3 109 The activation calculation has been performed with the FISPACT-II code using the EAF-2010 data. Deliverable D8.3 44 6. Sc-47 production from emerging facilities Sc-47, a beta-radiation emitter, and Sc-43 and Sc-44 as positron emitters, seem to fit ideally into the concept of a theranostic pair. Sc-47 is also a low-energy g-emitter and allows SPECT and planar imaging. Similarities in chemistry between Ga-68 and Sc-44 and their different physical properties have opened a wider avenue for applications of other Sc radionuclides. For these reasons, Sc-47 was included in the portfolio of PRISMAP radionuclides. 6.1 Physical and chemical properties of Sc-47 Sc-47 (T1/2 = 3.35 d) emits a low-energy βspectrum with Eβav= 162 keV and Eβmax = 600 keV. The emission of conversion and Auger electrons is negligible (<1%). Sc-47 emits g-rays at 159.4 keV (68.3%) suitable for SPECT imaging. Sc-47 decays to stable Ti-47. Scandium is chemically similar to yttrium by stabilising Y3+, similarly to the heaviest lanthanides. Scandium is almost exclusively present in its compounds in the trivalent state. Therefore, ligands developed for these cations are also suitable for chelating Sc. Since DOTA is the most common chelator used for labelling with Sc-47, it can be assumed that the content of chemical impurities such as copper, iron, zinc and lead ions may have an impact on Sc-47 quality. 6.2 Sc-47 conventional production in PRISMAP Sc-47 can be produced with thermal neutrons, which are available in most of the research reactors. Neutron capture on Ca-46 produces Ca-47 (T1/2 = 4.5 d), which decays into Sc-47 by β− emission: Ca-46(n,γ)47Ca→Sc-47, and the obtained Ca-47 can be further exploited as the Ca-47/Sc-47 generator system. This method suffers mainly from the low natural abundance of Ca-46 (0.004%). To date, the Sc-47 produced by thermal neutron activation is available for PRISMAP users’ projects at PSI and NCBJ. To obtain significant activity of Ca-47, the target enriched in Ca-46 must be used (presently, Ca-46 is available with a maximum 30% enrichment), which is rather expensive, and recycling is a must to reduce that cost. For example, when 0.97 mg of Ca-46 (48.5 mg of 5% enriched [Ca-46]CaCO3) was irradiated in a thermal neutron flux of 1.2 1014 cm−2 s−1 for 6 days, around 700 MBq of Ca-47 and 350 MBq of Sc-47 were produced at EOB86. Shielding of 1.3 MeV g-rays of Ca-47 is a concern. The produced Ca-47 decays to Sc-47 with a half-life of 4.5 days, which is longer than the half-life of Sc-47, enabling multiple separations of ingrown Sc-47 in the generator-like system. Using this approach, Domnanich et al.87 demonstrated that up to 2 GBq Sc-47 can be produced by thermal neutron irradiation of enriched Ca46 targets. The chemical isolation of Sc-47 from the target material allowed the formulation of up to 1.5 GBq Sc-47 with high radionuclidic purity (> 99.99%) in a small volume (∼700 μL), which was useful for labelling purposes. Up to four consecutive separations within 1 week were possible by isolating the ingrown Sc4762,63. 6.3 Non-conventional production from emerging facilities and new routes Proton irradiations Ti(p,X) followed by mass separation were explored at CERN, INFN and SCK CEN. Proton irradiation of natural vanadium V-nat(p,x)Sc-47 and Ti-50(p,x)Sc-47 was evaluated at INFN and a irradiation of Ca enriched in Ca-44 in nuclear reaction Ca-44(a,p)Sc-47 was studied at ARRONAX. Alternative route using fast neutrons in Ti-47(n,p)Sc-47 nuclear reaction was evaluated at ESS, SCK CEN and JHR. 86 D. Pawlak et al., Appl Radiat Isotopes (2019) 151:140 87 K. Domnanich et al., EJNMMI Radiopharm Chem (2017) 2:5 Deliverable D8.3 51 EOS (end of chemistry) act. 18 GBq Molar activity Unknown, but expected high Radionuclidic impurities Low amounts of Cu-64 from Zn impurities Comments 7.4.3 Zn-67(n,p)Cu-67 with fast neutrons (MYRRHA) Production status and data type Calculated data Nuclear process Zn-67(n,p)Cu-67 Target mass and enrichment >1 g, enriched Irradiation time 3 days Cooling time 1 day Beam current or flux Not specified Beam energy or n-spectrum Not specified EOB activity of Cu-67 1 GBq EOS (end of chemistry) act. 0.8 GBq Molar activity 15 GBq/µmol Cu Radionuclidic impurities Not specified Comments Not yet established, can be scaled up x10 by larger targets 7.4.4 Zn-67(n,p)Cu-67 (JHR) Production status and data type Calculated data, expected neutron spectrum and flux Nuclear process Zn-67(n,p)Cu-67 Target mass and enrichment 1 g, 88.6% enriched in Zn-67 Irradiation time 3 days Cooling time 1 day Beam current or flux 4 × 1014 cm-2 s-1 above 1 MeV Beam energy or n-spectrum In-core irradiation position, with Cd screen EOB activity of Cu-67, 1 d cool 2.4 GBq (based on 90% chemistry yield) EOS (end of chemistry) act. 2.15 GBq Molar activity Not specified Radionuclidic purity 99.18%, 0.82% Cu-64 Comments Not yet established, can be scaled up x10 by larger targets 7.4.5 Zn-67(n,p)Cu-67 with fast neutrons (GANIL) Production status and data type Calculated data, expected neutron spectrum and flux Nuclear process Zn-67(n,p)Cu-67 Target mass and enrichment 1 g, 88.6% Zn-67 Irradiation time 3 days Deliverable D8.3 52 Cooling time 1 day Beam current or flux 40 mA d on Be at 35 cm, 3.2 × 1010 cm-2 s-1 Beam energy or n-spectrum Up to 40 MeV from 40 MeV d on Be EOB activity of Cu-67, 1 d cool 3.4 MBq EOS (end of chemistry) act. Not specified Molar activity Not specified Radionuclidic purity 96.7% Comments Not yet tested 7.4.6 Zn-67(n,p)Cu-67 with fast neutrons (ESS) Production status and data type Calculated data, expected neutron spectrum and flux Nuclear process Zn-67(n,p)Cu-67 Target mass and enrichment 1 g, 88% Zn-67 Irradiation time 3 days Cooling time 1 day Beam current or flux 2.5 mA protons Beam energy or n-spectrum 2 GeV protons; configuration “below wheel, 1 moderator” EOB activity of Cu-67, 1 d cool 24 GBq EOS (end of chemistry) act. Not specified Molar activity 429 GBq/µmol Cu (enriched Zn-67, one moderator) Radionuclidic purity Radionuclidic impurities 73.4% (one moderator) 25.1% Cu-64; 1.1% Cu-66 (one moderator) Comments Not yet established, can be scaled up x10 by larger targets 7.4.7 Zn-70(p,alpha) Cu-67 (LARAMED) Production status and data type Calculated data, from measured cross-sections Nuclear process Zn-70(p,alpha)Cu-67 Target mass and enrichment 2.18 g, 100% Zn-70 Irradiation time 3 days Cooling time 1 day Beam current or flux 1 µA Beam energy or n-spectrum 70 MeV to 56 MeV EOB activity of Cu-67, 1 d cool 1.6 GBq EOS (end of chemistry) act. 1.1 GBq Molar activity Not specified Radionuclidic impurities Not specified Comments Numbers are calculated considering a radiochemical efficiency of 90% Deliverable D8.3 53 7.4.8 Zn-68(p,x)Cu-67 (LARAMED) Production status and data type Calculated data, from measured cross-sections Nuclear process Zn-68(p,2p)Cu-67 Target mass and enrichment 2.74 g, 100% Zn-68 Irradiation time 3 days Cooling time 1 day Beam current or flux 1 µA Beam energy or n-spectrum 56 MeV target 35 MeV EOB activity of Cu-67 564 MBq EOS (end of chemistry) act. 388 MBq Molar activity Not specified Radionuclidic impurities Not specified Comments Numbers are calculated considering a radiochemical efficiency of 90% 7.4.9 Multiple reactions with charged particles (GANIL) Production status and data type Calculated data, high currents of p,d and a available Nuclear process See Table 13 Target mass and enrichment Not specified Irradiation time 3 days Cooling time 1 day Beam current or flux 100 µA Beam energy Comments 20 MeV d 30 MeV p (Zn-68), 16 MeV p (Zn-70) 20 MeV a Numbers are calculated considering a radiochemical efficiency of 90% Table 13. The obtainable Cu-67 activity and purity from projected charged particle reactions at GANIL At EOI After 1 day cooling Reaction Activity (GBq) % Atot Activity (GBq) % Atot Radionuclidic purity after radiochemical separation d+Zn-70 1.20 2.2 0.83 31 99.99% p+Zn-68 0.93 0.75 0.64 1.6 13.9% p+Zn-70 0.79 0.24 0.54 1.6 99.99% a+Ni-64 0.59 57.2 0.41 99.99 100.00% Deliverable D8.3 54 7.4.10 Spallation in ThCx or UCx + mass separation (CERN-MEDICIS) Production status and data type Extrapolated data from measured cross-sections Nuclear process Spallation in ThCx or UCx Target mass and enrichment Not specified Irradiation time 3 days Cooling time 1 day Beam current or flux 2 mA Beam energy or n-spectrum 1.4-GeV to1.2 GeV EOB activity of Cu-67, 1 d cool 1.1 to 1.4 GBq EOS (end of collection) 110 to 140 MBq with 10% collection efficiency Molar activity Not specified Radionuclidic impurities Not specified Comments Some post collection chemistry may be needed to remove Ga-67 and to recover the Cu-67 from collection foils into a suitable medium for labelling. 8. Tb-155 and Tb-152 production overview and main challenges The production of terbium isotopes has gained significant interest due to their potential applications in nuclear medicine and scientific research90. Terbium is a unique element in that it presents four interesting radioisotopes for all modalities in diagnostic and therapeutic nuclear medicine: Tb-155 emits γ-radiation suitable for Single Photon Emission Computed Tomography (SPECT), while Tb-152 decays by the emission of β+-particles useful for Positron Emission Tomography (PET). Tb-161 emits low-energy β-particles, but also a significant number of conversion and Auger electrons, which presents a powerful combination for therapeutic purposes. Finally, there is Tb-149, the “flagship” nuclide for the TATTOOS (Targeted Alpha Tumour Therapy and Other Oncological Solutions) project, which has promising properties for α-therapy and is suited for simultaneous PET imaging90. The existence of four radioisotopes that allow the preparation of chemically identical radiopharmaceuticals, with equal pharmacokinetic profiles, is a unique situation that could serve the concept of personalized medicine thereby offering diagnostic tools and various therapy options for cancer patients in future. However, except for the reactor produced Tb-161, the other Tb isotopes of interest -Tb-149, Tb-152 and Tb155 - current production methods face substantial limitations. Neither light charged particle activation nor heavy ion activation techniques are presently suitable for large-scale production of these radionuclides. This is mainly due to three key technological constraints: (i) the need for substantial enrichment of stable isotopes, (ii) the limited availability of high-energy driver beams in commercial cyclotrons, and (iii) the absence of integrated isotope separation techniques in conjunction with commercial accelerators91 . In contrast, spallation reactions, when coupled with Isotope Separation OnLine (ISOL) systems, have demonstrated great promise for the production of the mentioned terbium nuclides. The latter can produce isobarically pure Tb isotopes, almost eliminating any trace of other Tb contamination, which is otherwise impossible to chemically separate. Nevertheless, even with the ISOL technique, radiochemical separation remains essential to achieve high radioisotopic purity, because other elements (isobars) are typically present as contaminants. Nonetheless, mass separation of Tb isotopes is challenging and it still presents quite some scattering (see below). 90 C. Müller et al., 2012, vol. 53, no. 12: Society of Nuclear Medicine, pp. 1951––1959, doi: 10.2967/jnumed.112.107540. 91 S. Lahiri and N. Naskar, 2021, doi: 10.3389/fmed.2021.675014. Deliverable D8.3 55 Provided that there are enriched targets at sufficiently high purity, lower energy cyclotrons/linear accelerators - impinging the beam in targets with an atomic mass closer to the isotope of interest - can also be a suitable alternative for the production of Tb isotopes. The main challenges are the limited availability of enriched targets, where otherwise high contamination with other Tb isotopes arises, and the typical highpower densities, limiting the production rates in these targets. In those cases, by adding mass separation as an additional step can drastically improve the purity of the obtained product, but at the same time, considerably reduce the total activity. However, in the case of Tb, the targets are typically made of Gd, whose boiling point is only about 50 °C lower than the one of Tb. Since, for mass separation to happen, the atoms have to vaporize to be mass separated, the contribution from the Gd vaporization overwhelms the system since they are several orders of magnitude above the ones of radionuclides to extract. This typically causes vacuum issues and ion source quenching, drastically reducing the mass separation efficiency. This has been solved by doing a radiochemistry step to the target before mass separation to bring the quantity of target material (Gd) closer to the quantity of Tb to separate, which allows mass separation at reasonable efficiencies without the mentioned issues92. Mass separation of previously radiochemically separated targets are also sensitive to the radiochemical process itself. The Tb mass separation efficiencies state of the art is discussed in the next section, in order to better frame the calculations done for the different production pathways (see 10.1). But to summarize, it is known, that the mass separation of the Tb radioisotopes is currently plagued by scattering, especially if one takes the atomic form of Tb. By focusing the R&D in more volatile venues to get the Tb (extracting volatile molecules, or less refractory parent radionuclides, such as Er, Dy, etc.), it is expected that more reliable numbers can be reached. For this reason, a range was given for the mass separation efficiency between 1 – 5 %. Regarding the radionuclide half-life, the relatively short half-life of Tb-152 (17.88 h) can be challenging, especially when the production and separation facilities are geographically distant. In such cases, decay losses during transport can significantly impact the final activity available for application. This radionuclide would require either local or nearby mass separation. Another challenge connected to mass separation facilities is the ion source throughput, where it can very easily get quenched above a few tens nA of ion beam current, reducing the efficiency of mass separation. This becomes even more important when using targets such as Gd ones from lower driver beam energies, even if radiochemically separated before. For this ion beam current, one has to consider as well the target and stable contaminants vaporizing from the target itself and structural materials surrounding it. In contrast to mass separation, chemical separation techniques consistently yield much higher efficiencies. Studies report values between 80–90%65, with 60% considered a lower bound in less optimized scenarios. These figures apply to both target dissolution and foil processing, although precise efficiencies may vary depending on the specific chemical protocol and radionuclides to be extracted. Additionally, efficient target recycling becomes essential, especially when using enriched materials, to have more cost-efficient facilities. This part of the white paper focuses on the production of Tb-152 and Tb-155 isotopes with both high Z targets (requiring high energy driver beams for spallation reactions) and lower Z targets (typically used in low energy cyclotrons or facilities), closer in atomic mass to the radionuclide of interest which require typically enriched targets and almost always chemical separation of the production target. 8.1 Efficiencies using ISOL mass separation In order to mass separate terbium isotopes using the ISOL technique, the element must first be released from the target material. However, terbium's low volatility—reflected in its high boiling point of 3230 °C—makes it one of the most challenging lanthanides to extract. This requires operating target materials, typically tantalum foils, and ion sources at extreme temperatures, typically exceeding 2200 °C or even beyond. Such conditions often lead to degradation of the target and ion source system, resulting in low or decreasing yields 92 C. Duchemin et al., 2021, vol. 8: Frontiers Media SA, doi: 10.3389/fmed.2021.693682. Deliverable D8.3 56 over time. Tb is known to diffuse well in the Ta foils93, however recent studies point to long effusion times (diffusion through the vacuum of the target to the ion source) in the order of hours94. CERN-MEDICIS has reported offline mass separation efficiencies of the order of 1.2% for Tb-15595 while benchmarking calculations from the ISOLDE historical yields96 from the 80s-90s show an efficiency close to 1.5% for Tb-14997. To benchmark these numbers with another facility, ISAC efficiencies were calculated from the ISAC database for Tb-154, which can’t be fed from Dy-154 (3E6 y) due to its very long half-life and small branching ratio to Tb-154. TRIUMF reported a yield of 6 × 107 ions s-1 for this radionuclide (40 µA Ta target, typically98 0.14 mol cm-2), which when compared to its in-target production of 6 × 105 ions s-1 (mmol cm-2) µA-1, gives an efficiency of 1.8%, which is close to the values of ISOLDE and MEDICIS. To circumvent the volatility issue of Tb and avoid extreme operation conditions, an alternative approach involves extracting dysprosium (Dy), the parent radionuclide of terbium, via laser ionization. Dy has a significantly lower boiling point (2567 °C), making its extraction more feasible. Reported efficiencies for terbium production via Dy decay include 1.2% (Dy-152) to 3.4% (Dy-155), with 0.1% (Dy-149)99. However, recent developments using an optimized target container oven design, which minimizes cold spots100, point to efficiencies of about 24% of Dy-152 extraction (number extracted from the collection101 and ISOLDE operational data). At ISAC (TRIUMF), a broader strategy is employed where all isobars in A=155 (Er, Ho, Dy, Tb) are collected and allowed to decay into terbium98,102. In the 500 MeV p+ range, the production peaks around Er-155. It allows ISAC to benefit from the higher production zones for the Tb-155 activity build-up. Another promising route involves extracting terbium in volatile molecular forms, such as TbF, using electron impact ion sources. This has been tested at both ISOLDE and MEDICIS. Recent studies94,103 have shown that at efficiencies of 10’s of a percent up to above a percent level can be obtained: (i) online mass separation at ISOLDE, using Ta metal targets, shows efficiencies of 0.5% for Tb-149F₂, 0.1% for Tb-149F93,94, 0.4% for atomic Tb-149, and 3.5% for Tb-155F281, while in offline mass separation at MEDICIS, which used a tantalum carbide target, efficiencies ~1% were seen for TbF103. In both cases, the typical Ce-139 contamination from the oxide side band was not seen, since the mass is moved to a much higher one (A=174 or higher, depending on molecule, A=19 for F). Attempts have also been made to mass separate Tb from lower Z targets (Gd) produced at ARRONAX, in a 30 MeV cyclotron. In this case Tb was produced in a natural Gd target, which was then chemically separated to reduce the Gd (target material) to Tb ratio to have a more efficient mass separation at MEDICIS. In this case, the achieved efficiencies reached were 6%952. Laser ionization of natural terbium has shown promising results, with efficiencies reaching up to 53%104. However, this figure has yet to be validated under operational ISOL conditions. 93 G. J. Beyer et al., Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions with Materials and Atoms, doi: 10.1016/s0168-583x(02)01913-4 vol. 204, pp. 225–234–225–234, 2003. 94 W. Wojtaczka, PhD Thesis, KU Leuven, 2025. 95 C. Duchemin et al., 2020, vol. IPAC2020: JACoW Publishing, Geneva, Switzerland, p. France, doi: 10.18429/JACOW-IPAC2020THVIR13. 96 "The ISOLDE Yield database (Online: Accessed 9th December 2021)," ed, 2018. 97 J. P. Ramos, "The MEDICIS Facility - Overview, 2018 operation report and plans for CERN long shutdown 2," ed, 2018. 98 D. E. Fiaccabrino, P. Kunz, and V. Radchenko, "Potential for production of medical radionuclides with on-line isotope separation at the ISAC facility at TRIUMF and particular discussion of the examples of 165Er and 155Tb," vol. 94–95, ed: Elsevier BV, 2021, pp. 81––91. 99 K. Chrysalidis et al.," Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions with Materials and Atoms, doi: 10.1016/j.nimb.2019.04.021 vol. 463, pp. 472–475–472–475, 2020. 100 S. Rothe et al., Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions with Materials and Atoms, vol. 542, pp. 38–44, 2023, doi: 10.1016/j.nimb.2023.05.058. 101 S. M. Collins et al., Applied Radiation and Radioisotopes, Vol 202, 2023, 111044, doi: 10.1016/j.apradiso.2023.111044 102P. Kunz et al., vol. 541, ed: Elsevier BV, 2023, pp. 117––120. 103 "MEDICIS operation report Tb-155 Development run #2 2025," CERN, 2025. 104 V. M. Gadelshin et al., Front Med (Lausanne), vol. 8, p. 727557, 2021, doi: 10.3389/fmed.2021.727557. Deliverable D8.3 57 It should be highlighted that there is a strong development of Tb extraction as shown above. As such it is reasonable to assume that Tb efficiencies are to be between 1 and 5%, where the latter number is an upper limit accounting for future developments. All calculated activities throughout this report using mass separation, will be given with this range, except for the cases of an experimentally obtained value. 8.2 Tb-155 and Tb-152 production pathways: results The calculated activities reports will be divided in two big sections: (i) those that use high Z targets, typically high-energy facilities (0.5 to 1.4 GeV) using tantalum target spallation to produce the Tb isotopes, (ii) and those using lower Z targets in the vicinity of the atomic mass of the Tb isotopes to produce, typically using lower energy particle beams around 100 Mev or lower. Results for both Tb-152 and Tb-155 are reported in each section. 8.2.1 High Z target facilities – all Ta(p,X)spallation In this section the activities calculated by MEDICIS at CERN and ISAC at TRIUMF are shown based on operational experience. Calculated values for the future TATTOOS facility at PSI are presented. ISOLDE experimental values are also included in this section for reference. 8.2.1.1 ISOLDE at CERN ISOLDE traditionally uses Ta targets to make Tb-152 and Tb-155 collections. These collections used to be made on the Tb isotopes directly but were then shifted to collect on the Dy isotopes73. This allowed to keep the targets at reasonable operation conditions since Dy is more volatile than Tb, unlike with the case of Tb where temperatures in the vicinity of 2200°C had to be used to extract the Tb. However, recent developments in the design of target containers with a more homogeneous temperature105 have increased the Dy extraction efficiency, in particular for short-lived (T1/2=4.2 min) Dy-149 used to populate Tb-149. A 2.12-hour collection of Tb-152 (296.3 g Ta, mix of 6 µm and 25 µm foil rolls) with resonantly laser ionized Dy-152 at an average 1.4 GeV proton current of 0.74 µA provided an EOI-equivalent Tb-152 activity of 0.75 GBq106. Comparing with the calculated cumulative in-target production of Dy-152 indicates an overall (release and ionisation) efficiency of »24%. Scaling the results of this actual collection to a hypothetical collection for 12 hours at 2 µA proton current followed by 12 hours decay would result in »5 GBq Tb-152. A detailed gamma-ray spectrometry analysis of Tb-152 samples from ISOLDE has been described by S. Collins et al106. Observed isobaric and quasi-isobaric impurities were Eu-152m (at EOI 4.96(64) × 10-4 Bq per Bq Tb152), Eu-152 (at EOI 2.33(14) × 10-7 Bq per Bq Tb-152) and Ce-134 (in equilibrium with its La-134 daughter): at EOI 2.04(39) × 10-4 Bq per Bq Tb-152. Here, Ce-134 is a pseudo-isobaric impurity separated as molecular ion 134Ce18O+ on mass A=152. Also, a small admixture of the neighbouring mass Tb-151 was observed: at EOI 4.96(64)×10-4 Bq per Bq Tb-152. We note that this particular admixture may vary depending on beam tuning and slit settings in the mass separator. Eu and Ce impurities are eliminated by radiochemical postprocessing107 to that only traces of Tb-151 remain. Consequently, the radionuclidic purity of chemically purified Tb-152 collections is about 99.95%. The tiny admixture of Tb-151 is not disturbing in practice. A »5-hour collection of Tb-155, done in September 2017, with resonantly laser ionized Dy-155 at an average 1.4 GeV proton current of »1.6 µA provided an EOI-equivalent Tb-155 activity of 0.165 GBq108 (note: this collection had been performed under non-optimized conditions and with some interruptions). Scaling the results of this actual collection to a hypothetical collection for 24 hours at 2 µA proton current followed by 105 S. Rothe et al., "Targets and ion sources at CERN-ISOLDE — Facilities and developments," Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions with Materials and Atoms, vol. 542, pp. 38–44, 2023, doi: 10.1016/j.nimb.2023.05.058. 106 Sample 1 of Collins 2023, doi: 10.1016/j.apradiso.2023.111044 107 B. Webster et al., Scientific Reports 2019;9:10884, doi: https://doi.org/10.1038/s41598-019-47463-3 108 Sample of Webster 2019, doi: 10.1038/s41598-019-47463-3 Deliverable D8.3 58 24 hours decay would result in »0.7 GBq Tb-155. Given that Tb-155 is mainly collected indirectly via its Dy155 precursor, a somewhat longer decay period (36 to 48 hours) before chemical separation could slightly increase the Tb-155 activity. For Tb-155 samples again the pseudo-isobaric CeO+ admixture dominates at late times. Ce-139/Tb-155 ratios between 1.9% and 3.7% have been reported. Again, Ce and residual Dy are removed radiochemically107,109. Due to the indirect collection via Dy precursor isotopes, the admixture of neighbouring isotopes Tb-154 and Tb-156 is negligible since the corresponding Dy precursor isotopes are quasi-stable or stable. Consequently, the radionuclidic purity of chemically purified Tb-155 collections can easily exceed 99.98%110. 8.2.1.2 MEDICIS at CERN This study investigates the production capabilities of Tb-152,155 under the following conditions: a 1.4 GeV proton beam at 2 µA on Ta based targets. FLUKA-CERN 4-3-0 was used to calculate the in-target production followed by a decay calculation performed with the CERN ActiWiz software version 3.7.2. The following scenarios were considered and results are shown in Table 14: ¡ Scenario #1 utilizes an irradiation time of 24 hours and a cooling time of 24 hours. 1.4 GeV proton delivered by the CERN PSB impinges on natural Ta metallic foil targets. The mass of Ta is considered as of 180 g for a 30 cm3 volume with a density of 6 g cm-3. Separation efficiency is considered to be 1-5% for Tb following the conditions above defined in this report. ¡ Scenario #2 utilizes an irradiation time of 120 hours and a cooling time of 120 hours, with the same target criteria shown above. ¡ Scenario #3 utilizes an irradiation time of 24 hours and a cooling time of 1 hour, with the same target criteria shown above. Table 14. Calculation results of the MEDICIS Tb-152 and Tb-155 activities. Scenario Target Nuclide Irradiation Activity (MBq) Eff. (%) Activity EOC (MBq) Impurities #1 Ta foils Tb-152 Direct p 1.4 GeV 14800 1-5 148-740 Tb-153 Tb-155 Direct p 1.4 GeV 7660 1-5 77-383 Ce-139 #2 Ta foils Tb-155 Direct p 1.4 GeV 20200 1-5 202-1010 Ce-139 #3 Ta foils Tb-152 Direct p 1.4 GeV 31000 1-5 310-1550 Tb-153 Tb-155 Direct p 1.4 GeV 4780 1-5 48-239 Ce-139 8.2.1.3 ISAC at TRIUMF ISAC at TRIUMF can produce Tb isotopes using Tantalum metal foil targets. They made a ~130 min test collection98 to obtain Tb-155 by collecting, all isobars relatively well produced on A/q=155 (Er-155, Ho-155, Dy-155, Tb-155, where the in-target production peaks at Tb-155, see Figure 26). After a few days of cooldown the decay chain of all isobars leads into building up of the relatively long-lived Tb-155. After 8-day cooldown period a sample containing 37 MBq (97% pure) was obtained for radiolabelling. 109 U. Jakobsson et al., Nuclear Medicine and Biology 84–85 (2020) 102–110. https://doi.org/10.1016/j.nucmedbio.2020.04.001 110 This was measured by Webster 2019, https://doi.org/10.1038/s41598-019-47463-3 but with an “old” Tb-155 source after 4.1 Tb155 half-lives. With a Ce/Tb separation performed few days after EOI and at equal Ce/Tb separation factor the RNP should be an order of magnitude better, i.e. >99.998%. Deliverable D8.3 59 Figure 26. Excerpt from chart of nuclides with lanthanides region and calculated production rates with ISAC facility using a tantalum target at 70 µA111. Highlights show A/q=155 (for the production of Tb-155) and A/q=165 (for the production of Er-165). Another study based also on the same production pathway for Tb has calculated that if one collects the A/q=155 for 5 h, a total activity of Er-155, Ho-155, Dy-155, Tb-155 of about 10 GBq are obtained which after 5 days of cooldown can give origin to 370 MBq of Tb-155102. In principle the isobar collection can also be done on A=152 for Tb-152, however one has to account for the lower branching ratios in the decay chain (e.g. only 10% of the Er-152 decays to Ho-152) and the lower production rates and lower half-life’s which will both result in lower yields in the A=152 comparing to the A=155. 8.2.1.4 TATTOOS at PSI It is possible to utilize the mass-separation technique at PSI’s HIPA infrastructure with the construction of an ISOL facility (TATTOOS) to foster this new research area based on novel, emerging medical radionuclides. The aim is to provide improved access to new medical radionuclides, which are difficult to obtain via more conventional methods. The use of online mass separation ensures collection of radionuclides of specific mass that can, subsequently, be chemically separated/purified. The production route over nuclear spallation reactions using the world most powerful proton accelerator (PSI’s Ring cyclotron – 590 MeV) allows for unprecedented activities of a broad spectrum of radionuclides, thereby, potentially facilitating clinical studies with novel radionuclides. The aim of this venture is to utilize and expand production facilities at PSI, such that an ISOL facility is built and installed to enable collection, as well as chemical separation of novel promising radionuclides in unprecedented activities, e.g. Tb-149 and Tb152. These can be introduced into the GMP facility on site (at Center of Radiopharmaceutical Sciences – CRS), used for labelling of biomolecules, and the resultant radiopharmaceutical can be transported to imaging and therapy centres for patient application. The concept will cover basic research and development thereof as part of a “bench-to-bedside” principle. 111 D. E. Fiaccabrino et al., " vol. 94–95, ed: Elsevier BV, 2021, pp. 81––91. Deliverable D8.3 60 The unique option at PSI to produce and process radionuclides at the same site, will be particularly important for the relatively short-lived Tb-149 (T1/2 = 4.1 h). The details on the codes and calculation procedure are given in section 5.2. Low energy beam facilities In this section the activities calculated by SCK CEN, ARRONAX and LARAMED are presented. 8.2.1.5 ISOL@MYRRHA at SCK CEN This study investigates the production capabilities of Tb-152 and Tb-155 through the irradiation with a 100 MeV proton beam on natural gadolinium targets followed by chemical purification and then mass separation. Offline mass separation was selected due to the similarity of Tb and Gb chemical and physical characteristics which makes online mass separation very challenging112. TENDL-based cross-sections were used to estimate the in-target production rates113. In a later phase FLUKA was used to estimate the purity of the samples. The following scenarios were considered: ¡ For Tb-152: ¡ 1 day irradiation of a 40 g natural gadolinium target at 20 µA ¡ 1 day of decay for target transfer and chemical processing ¡ Chemical extraction of terbium (90% efficiency) ¡ 12 hours for mass separation (5% efficiency) ¡ For Tb-155: ¡ 5 days irradiation of a 45 g natural gadolinium target at 20 µA ¡ 1 day of decay for target transfer and chemical processing ¡ Chemical extraction of terbium (90% efficiency) ¡ 12 hours for mass separation (5% efficiency) This results in following activities of the collected sample: 1.3 GBq Tb-152 and 12 GBq Tb-155. To estimate the achievable purity of the samples, FLUKA and a custom script were used to process the scenarios by simulating the evolution of the isotopic inventory during irradiation, decay, chemical purification and mass separation. The targets considered here have a mass between 40 and 45 g, this is still be addressed if chemical separation of such masses is feasible. The results are summarized in the table below. Table 15. Calculation results of the Tb-152 and Tb-155 activities available at SCK CEN Scenario Target Nuclide Irradiation Activity (MBq) Tb-152|155 / Tb (atoms/atoms) Tb-152|155 / Tb (Bq/Bq) Tb-152|155 / all (Bq/Bq) Irradiation + chem. sep. + mass sep. Gd-nat Tb-152 1 day 1300 99.44% 99.75% 99.75% Irradiation + chem. sep. + mass sep. Gd-nat Tb-155 5 days 12000 99.92% 99.85% 99.85% 112 B. Leenders, A. Aerts, T. E. Cocolios, S. Cottenier, D. Houngbo, and L. Popescu, "On the feasibility of online terbium extraction at ISOL@MYRRHA," Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions with Materials and Atoms, vol. 541, pp. 249–252, 2023, doi: 10.1016/j.nimb.2023.05.034 113 https://tendl.web.psi.ch/tendl_2023/tendl2023.html Deliverable D8.3 67 155). However, in the case of Tb-155 even if 100% enrichment is possible (currently about 93% is possible), it is still expected that only 95% purity can be reached. In the Tb-155 production, Tb-156 is a typical impurity (similar half-life and also a gamma emitter) and is not very clear, for now, what is the threshold allowed for medical applications. In the latter case, mass separation could be considered to separate the Tb-156 and also Tb-154 to considerably increase the Tb-155 specific activity. INFN also shows a very promising venue to produce pure Tb-155 with quite high activities (1.2 GBq) without relying on target enrichment. Similarly high activities for Tb-155 have been found in the literature, using α particles, with also quite high purity (99.7%), while co-producing Tb-152 (5.8 GBq with 89% purity) in the tandem target. However, the latter relies also on enriched targets for both production and co-production. A meaningful comparison of the (α,4n) route with the (p,n) route using equal target thickness and equal power deposited in the target. 11 MeV to 8 MeV protons would require 137 mg cm-2 Gd (scaling to oxide can be done if needed) and 30 µA protons would deposit 90 W in the target layer. The same Gd-155 target thickness would cover the energy range 55 MeV to 45 MeV and 9 eµA a would deposit 88 W in the target. With 24 h irradiation and 24h decay, 4 GBq Tb-155 would be obtained to be compared with 3.1 GBq with 30 µA protons. The indirect way produces about 2.2x less, but obviously reduces Tb-154/156 admixtures. The feasibility of high-power ISOL targets is another critical factor. E.g., at TATTOOS, up to 22 kW of deposited power in a target are suggested for operation, however this has never been done. The large target volume (10 cm diameter, 20 cm in length – compared with the typical 2 cm diameter, 20 cm length), can also play a role, since the diffusion volume is much higher and much more susceptible to cold spots as larger bodies are harder to ensure homogeneous temperature, which if not tackled may overall reduce the overall system efficiency by trapping less volatile elements (e.g. Tb). The activities presented from TATTOOS can be further worked out in a more detailed study to take some of these factors in consideration, and these high-level calculations can serve as a first view on the potential of this facility. Relatively high deposited power is also presented by the ISOL@MYRRHA facility at SCK CEN (1.6 kW) but those are within typical powers already operated in facilities like ISAC at TRIUMF. This facility also presents quite high activities in both Tb-155 and Tb-152, however the current suggested high mass of the targets (4045 g) can make the intermediate target radiochemistry quite challenging. Similarly to TATTOOS these open points can be further optimized in the future in a more detailed assessment. Currently operating mass separation facilities like MEDICIS, ISOLDE and ISAC present quite good yields on both Tb-152 and Tb-155 however always need a radiochemistry separation step, especially in the case of Tb155 where the isobar Ce-139 oxide (A=155) is expected in quite high activities in case of elemental separation. In any case, the radiochemistry step is essential in all the productions described, no matter the pathway. For the facilities working with enriched targets, recycling of the target material should be taken into account in the radiochemistry step in order to justify the high cost of enriched materials. Additionally, the return of experience shows quite some scattering in the Tb separation efficiency numbers, requiring a very good control of the entire process. Facilities like CERN-ISOLDE and CERN-MEDICIS can even collect on two isobars at the same time (i.e. collecting on Tb-152 and Tb-155), making efficient use of the irradiated targets. Overall, all facilities have complementary on-going R&D programs to improve the overall Tb activities and their reliability which can benefit all the PRISMAP partners, especially the emerging facilities, making programs like PRISMAP very successful for knowledge transfer. Deliverable D8.3 68 9. Pt-195m production from emerging facilities 9.1 Physical and chemical properties of Pt-195m Among all known Auger electron emitters with half-lives suitable for clinical use, Pt-195m (T1/2 = 4.010(5) d) is the Auger electron emitter with the highest multiplicity of Auger electrons per decay: »36.5118. This high Auger electron emission is the result of the high spin difference between 13/2+ isomer and 1/2ground state that has to be bridged in an internal transition decay. The decay proceeds by highly converted M4 transitions from the 13/2+ high-spin isomer, followed by a cascade of partially converted M1/E2 transitions. On average this leads to 0.8 K-shell vacancies, 1.44 primary L-shell vacancies and 0.39 primary M-shell vacancies per decay. This high number of initial vacancies followed by atomic cascades in a high-Z atom is responsible for the exceptionally high Auger electron multiplicity. In addition to the abundant electron radiation, also 99 keV g-rays with Ig=11.7(8)% and K X-rays of »66 keV (IX » 60%) and »76 keV (IX » 16%) are emitted. This photon radiation is well suited for SPECT imaging. Other long-lived platinum radioisotopes are Pt-193m (T1/2 = 4.33(3) d) and its very long-lived daughter Pt-193g (T1/2 = 50(6) y), Pt-191 (T1/2 = 2.83(2) d) and Pt-188 (T1/2 =10.16(18) d). All other platinum radioisotopes and isomers (in particular Pt-189, Pt-197, Pt-199) have half-lives below 20 hours and can be reduced as needed by an appropriate decay time between production and use. Co-production of the neutron-deficient Pt-188 can usually be avoided by a proper choice of projectile energies. Alike Pt-195m, also Pt-193m is a 13/2+ isomer that decays by IT with prolific emission of conversion and Auger electrons: »27.4118 per decay. Photon emission is much weaker compared to Pt-195: only 0.1% 135 keV grays and five times less K X-rays per decay than Pt-195m. Due to similar half-lives, ample emission of “useful” Auger electrons and low photon dose Pt-193m is not really a disturbing radionuclidic impurity, but in terms of patient dosimetry and therapeutic effect high admixtures of Pt-193m (even exceeding the Pt-195m activity) are well acceptable. The long-lived daughter activity Pt-193g (240 kBq Pt-193g resulting from decay of 1 GBq Pt-193m) is of no radiological concern since no radiation above 14 keV is emitted in its low-energy EC decay. Radioactive waste issues have to be considered in more detail, but compared to the known Lu-177m problem119 that appears with use of carrieradded Lu-177g, Pt-193g is of less concern since its exemption limits are an order of magnitude higher compared to Lu-177m. Pt-191 decays by EC, emitting relatively energetic g-rays of 539 keV (Ig=14.4(22)%), 409 keV (Ig=7.9(12)%), etc. A significant admixture of Pt-191 would be disturbing for SPECT imaging due to collimator punch-through and add distal photon dose for patients and caretakers. We summarize that in terms of radioisotopic purity Pt-195m and Pt-193m should be considered as “sisters”, and at virtually any activity ratio this mixture can be useful for Auger-TRT, similar to Sc-43/Sc-44 mixtures for PET imaging120. Pt-191 admixtures should be minimized and other Pt radioisotopes can be eliminated by decay. Several DNA-targeting Pt compounds are known: cisplatin, carboplatin, oxaliplatin, etc. and in cold (nonradioactive) form these are currently in clinical use as chemotherapeutic agents. Cisplatin radiolabelled with Pt-195m has been used in diagnostic studies to monitor the biodistribution of cisplatin121. However, an unambiguous demonstration of therapeutic effects of Pt-195m-cisplatin remains challenging. When the specific activity is too low, chemotherapeutic effects due to the high chemical toxicity of the compound are 118 J. Bolcaen et al., J Nucl Med 2023; 64:1344–1351. 119 S. Prevot et al. EJNMMI Physics (2023) 10:3, doi: 10.1186/s40658-023-00524-z 120 T.V.M. Lima et al., Diagnostics 2021, 11, 1826, doi: 10.3390/diagnostics11101826 121 R.C. Lange et al., J Nucl Med 1973;14:191.J. Shani et al., Cancer Res 1989;49:1877. M. Sathegke et al., Nuklearmedizin 2013;52:222. D.S. Hoogenkamp et al., EJNMMI Res 2025;15:87. Deliverable D8.3 69 dominating and make radioactive dose escalation studies challenging or impossible. Therefore the main objective is the identification of production paths that provide Pt-195m (or Pt-195m/Pt-193m mixtures) at high specific activity / high molar activity. 9.2 Nuclear reactions leading to Pt-195m The high spin of Pt-195m that on one hand is so beneficial for the abundant emission of Auger electrons in its decay represents on the other hand a fundamental hindrance to reach high cross-sections and yields in most production reactions. Potential target nuclides surrounding Pt-195m have much lower spin. Consequently, most production reactions populate preferentially the stable 1/2ground state Pt-195g and much less the 13/2+ high spin isomer. 9.2.1 Carrier-added versus non-carrier-added production routes Pt-195m is the only radionuclide in this report that can be produced by irradiating either targets of the same chemical element, i.e. platinum, or neighbouring elements, respectively. When irradiating platinum targets, only a tiny fraction of the target material gets converted into other elements. Even a high fluence irradiation of 7 days in a high thermal neutron flux of 1015 cm-2 s-1 would transmute less than <0.03 % of the Pt mass into other elements. Thus, the yield, expressed in GBq Pt-195m produced per mg initial Pt target material, is basically identical to the specific activity, expressed in GBq Pt-195m per mg remaining Pt target material. However, due to the small transmutation yield into Pt-195m the specific activity and molar activity are very limited, i.e. the radionuclide Pt-195m is accompanied by a large excess of stable Pt isotopes that cannot be separated by chemical means. This quality is called carrier-added. Alternatively, targets made from neighbouring elements are irradiated to produce Pt-195m. In this case a radiochemical separation can isolate the produced Pt fraction from the unreacted target elements. Consequently, the specific activity (Pt-195m activity per Pt mass) can be significantly higher compared to the yield (Pt-195m activity per target element mass) and definitively higher compared to the carrier-added case. This quality is called non-carrier-added. The specific activity is theoretically limited by coproduction of other Pt isotopes in the production reaction that would “dilute” the produced Pt-195m. It is also practically limited if the target material contains stable Pt as chemical impurity because this stable Pt would further “dilute” the produced Pt-195m. 9.2.2 Pt-194(n, g )Pt-195m Neutron capture on enriched Pt-194 targets is the most straightforward way to produce Pt-195m. Thermal neutron capture on the Pt-194 0+ ground state populates 1/2+ capture states at 6105 keV in Pt-195. From there g-ray cascades decay dominantly to Pt-195g, but due to the large spin difference of 6ħ only a tiny fraction decays to Pt-195m122. Partial Pt-194(n,g)Pt-195m cross-sections of 87(13) mb, 36(4) mb, 42(8) mb and 117(9) mb have been reported in EXFOR. The high non-statistical scatter of these thermal cross-section values can be explained by the substantial contribution of epithermal neutron capture: 1.0(1) b resonance integral, i.e. 8 to 30 times higher than the thermal capture cross-section. Hence, irradiations in different neutron spectra may lead to significantly different Pt-195m yields and thus to different “apparent” cross-sections. Experimental data on Pt-195(n,g)Pt-195m production yields were collected from literature and provided by PRISMAP partners. In the frame of the SECURE project (funded by the European Union under grant agreement No 101061230), ILL had performed a 7-day irradiation of a natural Pt sample in the V4 irradiation tube of its high-flux reactor in a well thermalized flux of 8 × 1014 cm-2 s-1 to produce Au-199. As a side result, 122 D.D. Warner et al., Phys. Rev. C26 (1982) 1921. https://doi.org/10.1103/PhysRevC.26.1921 Deliverable D8.3 70 a Pt-195m yield of 14 MBq/mg has been measured123. Considering dominant production by the quite pure thermal neutron flux, this value can be extrapolated to 42 MBq/mg expected with pure Pt-194. MEDICIS reported results from its partner lab PAEC. For a 245 mg sample irradiated for 18 hours, the measured activity was 311 MBq at EOI with a radionuclidic purity of 44.8%, rising to 88% after 24 h cooling time. This measured specific activity of 1.27 MBq/mg corresponds to 1.32 MBq/mg expected for pure Pt-194. For the purpose of plotting in Figure 33 this number was extrapolated to the saturation activity (after a very long irradiation) of 10.9 MBq/mg. Figure 33 shows that experimentally achieved specific activities tend to saturate around 50-100 MBq/mg, even for the highest thermal neutron fluxes. Hoeschele et al.124 explained this phenomenon by a very large thermal neutron capture cross-section on Pt-195m of about 13,000 b that destroys already produced Pt195m (“product burnup”) and hence limits the achievable specific activity. The red curve in Figure 33 shows expected specific activities with 13,000 b destruction cross-section. However, we note that the value of 13,000 b had been deduced assuming a cross-section of about 90 mb for the production reaction Pt194(n,g)Pt-195m. With lower production cross-sections also the deduced destruction cross-section would be lower. There is no unique analytical solution with the experimental data shown in Figure 33 because this two-dimensional plot does not account for the contribution of epithermal neutron capture that may vary considerably from one reactor to another. Only a three-dimensional plot (thermal flux – epithermal flux – specific activity), ideally with activations performed in neutron spectra providing extremely high or low ratios of thermal to epithermal flux, respectively and very different irradiation durations would allow to better constrain the individual Pt-195m production and destruction cross-sections. Figure 33: Experimental specific activities at EOI from literature and are plotted normalized to 100% Pt-194 target enrichment.124,125,126,127,128,129 Irradiation durations varied from 1.5 days to 8.3 days, but the given specific activities were NOT extrapolated to saturation activity because this extrapolation would depend on the not accurately known burnup cross-sections. Only the very short 0.75 day irradiation at PAEC was extrapolated to saturation activity for the purpose of plotting. The counter-productive effect of Pt-195m destruction by thermal neutron capture can be avoided or attenuated when shielding thermal neutrons and mainly exploiting epithermal neutrons for the 123 L. Rovan-Stiplošek et al., submitted to Appl Radiat Isot. 124 J.D. Hoeschele et al., Radiochim Acta 31 (1982) 27. 125 M. Akaboshi et al., Nucl.Med. Biol 20 (1993) 389. 126 S.E. Buckley et al., Phys. Med. Biol. 51 (2006) 1325. 127 U. Köster et al., Radiother. Oncol. 102 (2012) S170. 10.1016/s0167-8140(12)70279-7. 128 J.R. Zeevaart et al., J. Labelled Comp. Radiopharm. 56 (2013) 495. 129 R.H. de Roest et al. EJNMMI Res. 14 (2024) 22. Deliverable D8.3 71 Pt-194(n,g)Pt-195m production reaction. Thus, a factor 2 to 3 in specific activity can be gained. Irradiations at the HFR reactor (Petten, NL) were reported with specific activity at EOI up to 136.5 MBq/mg130. This production method is currently used to produce Pt-195m-cisplatin that is commercially available under the trade name Cisspect®. As the name says this product at limited specific activity is mainly employed for SPECT imaging, while therapeutic applications with selective vectors would require significantly higher specific activity. 9.2.3 Pt-195(n,n’)Pt-195m Energetic epithermal (>260 keV) or fast neutrons can serve for inelastic neutron scattering reactions on Pt-195g that populate Pt-195m. Test irradiations at HFIR (Oak Ridge, US) reached specific activities up to »50 MBq/mg, i.e. barely higher than the Pt-194(n,g) route131. 9.2.4 Pt-196(n,2n)Pt-195m Very energetic neutrons (>8.2 MeV) can induce (n,2n) reactions on Pt-196. Also this target nuclide has spin/parity 0+, but due to the higher projectile energy and the emission of two energetic neutrons more angular momentum can be transferred in such a reaction compared to the low-energy Pt-194(n,g) reaction. Consequently the probability to populate the Pt-195m high-spin isomer is significantly higher and the crosssection for the Pt-196(n,2n)Pt-195m is predicted to reach »1.3 b with 15 MeV neutrons. Experimental elemental cross-sections measured with Pt samples of natural isotopic composition (25% Pt-196, 34% Pt-195) are around 0.4 b with 13-18 MeV neutrons for the abundance-weighted sum of the Pt-196(n,2n)Pt-195m and Pt-195(n,n’)Pt-195m reactions. 9.2.5 Pt-196( g ,n)Pt-195m The Pt-196(g,n)Pt-195m reaction uses the same target as the (n,2n) reaction, but cross-sections of photonuclear reactions tend to be significantly lower compared to those of particle-induced reactions and less angular momentum is transmitted in photonuclear reactions. TENDL predicts that less than 10% of the (g,n) reaction populate the high-spin isomer with the partial cross-section peaking at »0.04 b for 14 MeV g-rays. Dikiy et al. predicted a specific activity of 250 MBq/mg at saturation when irradiating enriched Pt-196 targets with a 200 µA beam of 50 MeV electrons132. 9.2.6 Pt-195( g , g ’)Pt-195m Similarly, the Pt-195(g,g’)Pt-195m inelastic photon scattering reaction uses the same target as the Pt-195(n,n’)Pt-195m reaction, but has orders of magnitude lower cross-section. It has been postulated that specific “doorway states” could exist in Pt-195 that are connected by g-ray cascades with both, the ground state and isomer. If these doorway states could be resonantly excited by absorption of quasi-monochromatic g-rays on Pt-195g, then Pt-195m could be produced by “pumping” some nuclei to the isomeric level. With very optimistic assumptions of a “doorway state” populated in a very strong resonance and the availability of an extremely intense highly brilliant g-beam facility one might theoretically reach specific activities of several ten GBq/mg133. However, the postulated performance of the extremely intense g-beam facility remains many orders of magnitude above presently realized g-beam facilities and a first experimental search for suitable doorway states did not observe strong resonances above the expected continuum134. 130 R.H. de Roest et al. EJNMMI Res. 14 (2024) 22 131 F.F. (Russ) Knapp Jr. et al., J Radioanal Nucl Chem 263 (2005) 503. 132 M.P. Dykyi et al., J Label Compd Radiopharm 50 (2007) 480. 133 D. Habs, U. Köster, Appl Phys B (`2011) 103: 501 134 Dennis Wilmsen, PhD thesis, Univ. Caen (2017), https://theses.hal.science/tel-01768580 Deliverable D8.3 72 9.2.7 Pt-195(p,p’)Pt-195m, Pt-195(d,d’)Pt-195m or Pt-195( a , a ’)Pt-195m Inelastic scattering from the Pt-195 ground state to the Pt-195m isomer can also be induced by charged particles such as protons, deuterons or alpha particles. Predicted cross-sections and specific activities close to the peak of the excitation curve could be comparable or better to neutron-induced reactions, but only in a very thin layer, where the charged particles have optimum energy. However, the dominant reaction channels (p,xn), (d,xn) or (a,xn) would produce ample amounts of disturbing Au or Hg radioisotopes. Therefore, these reactions are not considered for practical use. While the presence of residual stable Pt target isotopes from the target limits the achievable specific activity in all production routes based on Pt targets, this limitation is not present when other elements are used as production target. Such production routes are discussed in the following sections 9.2.8–9.2.12. Often no experimental specific activities or molar activities have been reported yet, therefore calculated values of theoretical specific activity are given instead. 9.2.8 Au-197( g ,np)Pt-195m and Hg-199( g , a )Pt-195m The photonuclear reactions Au-197(g,np)Pt-195m and Hg-199(g,a)Pt-195m have been discussed by Dikiy et al135. When irradiating a Au target with a 34 MeV 500 µA electron beam they estimated activities of »3 GBq/day and specific activities of several 10 GBq/mg, depending on the delay time before Au/Pt separation136. When irradiating a Hg target with natural isotopic composition with a 26 MeV 200 µA electron beam, Dikiy et al. reported activities of about 1 GBq/day and specific activities up to »37 GBq/mg. 9.2.9 Os-192( a ,n)Pt-195m Hilgers et al.137 and Uddin et al. studied the Os-192(a,n)Pt-195m and Os-192(a,n)Pt-193m reactions. In the overlapping energy range the cross-section values of Hilgers et al. are twoto threefold lower compared to those achieved by Uddin et al. and to theoretical descriptions. We will therefore base the following conclusions on the TENDL cross-sections that follow more closely the experimental data by Uddin et al. A very long irradiation of 4 days with a 50 eµA a beam, followed by a decay period of 1 day for radiochemical separation and labelling would provide the following activities and theoretical/molar specific activities (not considering stable Pt contamination of the Os targets) as function of kinetic energy: An a energy range 24®18 MeV would produce »0.1 GBq Pt-195m with a radionuclidic purity >99.9% after chemical separation (assuming a 100% enriched Os-192 target), a calculated specific activity of »130 GBq/mg and a calculated molar activity of »26 GBq/µmol. Hilgers et al. reported an experimentally determined specific activity of only 1 GBq/mg, probably limited due to use of a Pt electrode during target preparation. A wider a energy range 28®18 MeV would increase the Pt-195m activity to »0.15 GBq, but with significant co-production of »0.8 GBq Pt-193m created in Os-192(a,3n) reactions138, i.e. the latter becomes now the dominant Pt activity. The combined radionuclidic purity of Pt-193m plus Pt-195m is still >99.9% after chemical separation (assuming a 100% enriched Os-192 target), the calculated combined specific activity is »400 GBq/mg and the calculated combined molar activity of »80 GBq/µmol. This production scenario could be realized with a-beams at ARRONAX, CERAD or SPIRAL2, respectively. However, it would monopolize the facility for four days of beam time to reach the mentioned yields. It is evident that this reaction is not scalable to therapeutic activities for clinical use of Auger-electron-TRT, while 135 N.P. Dikiy et al., Probl Atom Sci Technol 2007;5:118. 136 E.N. Bodnar et al., J Radioanal Nucl Chem 2015;305:133. 137 K. Hilgers et al., Appl Radiat Isotop 66 (2008) 545 138 M.S. Uddin et al., Appl Radiat Isotop 68 (2010) 2001 Deliverable D8.3 73 the same machines could produce with the same a beam during the same time (in several consecutive irradiations) over 100 patient doses of At-211 for a-TRT. 9.2.10 Ir-193( a ,pn)Pt-195m Takasz et al.139 studied the Ir-nat(a,X)Pt-195m reaction and measured cross-sections up to 9 mb peaking around 42 MeV a energy. Normalized to pure Ir-193 this would correspond to 14 mb, while TENDL predicts twice higher cross-sections. Therefore, we downscaled the TENDL predictions of Pt-195m and Pt-193m production by a factor 2. A very long irradiation of 4 days with a 50 eµA a beam, followed by a decay period of 1 day for radiochemical separation and labelling would provide the following activities and theoretical specific/molar activities (not considering stable Pt contamination of the Ir targets). An a energy range 50®35 MeV would produce »2.4 GBq Pt-195m and »6.5 GBq Pt-193m with a radionuclidic purity >99.9% after chemical separation (assuming a 100% enriched Ir-193 target), a calculated combined specific activity of »700 GBq/mg and a calculated combined molar activity of »130 GBq/µmol. Compared to the Os-192(a,X) reaction with 28 MeV a beams this reaction appears more favourable in terms of predicted activity and specific/molar activity. However, instead of pure Pt-195m, a mixture of Pt-193m and Pt-195m will be produced. Depending on the intended application this significant admixture of Pt-193m can be considered as useful or as radionuclidic impurity. We note significant co-production of Au radioisotopes with calculated activities of 7 GBq Au-193 and 5.5 GBq Au-194. The latter (T1/2 = 38 h) emits hard g-rays up to 2 MeV which makes efficient shielding during transport and radiochemical processing more challenging. ARRONAX and SPIRAL2 could provide appropriate a beam energies for this scenario. 9.2.11 Ir-193(d,2n)Pt-193m Considering that the highest yields of Pt-195m were obtained in reactions with dominant co-production of Pt-193m, it is instructive to compare with a reaction that produces only Pt-193m. Tarkanyi et al. have studied deuteron-induced reactions on Ir-nat140 and found that the cross-section peaks at 250 mb around 12-13 MeV deuteron energy. The corresponding cross-section on enriched Ir-193 would be 400 mb. A very long irradiation of 4 days with a 50 µA d beam, followed by a decay period of 1 day for radiochemical separation and labelling would provide the following activities and theoretical specific/molar activities (not considering stable Pt contamination of the Ir targets). A deuteron energy range 15®11 MeV could produce »20 GBq Pt-193m with a radionuclidic purity >99.9% after chemical separation (assuming a 100% enriched Ir-193 target), a calculated specific activity of »1.6 TBq/mg and a calculated molar activity of »0.3 TBq/µmol. Increasing the incident deuteron energy to cover a range 18®11 MeV could produce »34 GBq Pt-193m with a radionuclidic purity >99.9% after chemical separation (assuming a 100% enriched Ir-193 target), a calculated specific activity of »1.2 TBq/mg and a calculated molar activity of »0.22 TBq/µmol. Yet higher deuteron energies are not beneficial since co-production of disturbing Pt-191 would set in, considerably reducing the radionuclidic purity. In contrast to the a-beam irradiations discussed above, with deuteron irradiations no Au radioisotopes are generated. Co-produced Ir-194 (T1/2=19.2 h) activities are 6 GBq and 9 GBq at 15 MeV or 18 MeV incident 139 S. Takacs et al., Appl Radiat Isotop 136 (2018) 133 140 F. Tarkanyi et al., Nucl Instrum Meth B 247 (2006) 210. F. Tarkanyi et al., Nucl Instrum Meth B 458 (2019) 105 Deliverable D8.3 74 deuteron energy, respectively. If required these could be reduced by additional decay time before transport or processing. The scenario with 18 MeV deuterons could be realized at ARRONAX or SPIRAL2, the scenario with 15 MeV deuterons in addition also at CERAD. The calculated high yields render this reaction more realistic for the production of activities sufficient for clinical Auger-electron-TRT than all other charged-particle induced reactions. 9.2.12 Ir-193(n, g )Ir-194(n, g )Ir-195m(β-)Pt-195m Enriched Ir-193 targets can also serve for Pt-195m production via a multi-step process involving two successive neutron captures and a beta decay. Due to the short half-life of the intermediate radionuclide Ir-194 (T1/2=19.2 h) very high thermal or epithermal neutron fluxes should be favourable to increase the chances of the second neutron capture taking place before Ir-194 decays. This reaction had first been proposed and tested by Mirzadeh and Knapp et al141. They predicted yields of »0.1 GBq Pt-195m/mg Ir-193 target and specific activities of »3 GBq/mg Pt after 10 days of irradiation at thermal neutron fluxes up to 2.5 × 1015 cm-2 s-1 in HFIR (ORNL, USA) followed by radiochemical separation. A patent by Mirzadeh et al142. reports much higher yields of »3 GBq and 10 GBq Pt-195m/mg Ir-193 target after 1 day of irradiation and calculated specific activities of »3 GBq/mg Pt after 1 day of irradiation, dropping to 0.5 »3 GBq/mg Pt after 10 days of irradiation. Madumarov et al143. used the same reaction for a 17-day irradiation in the IBR-2 reactor at an average thermal neutron flux of 2.3 × 1012 cm-2 s-1 and an epithermal flux of 2×1011 cm-2 s-1 and obtained experimental yields of 0.005 to 0.019 MBq Pt-195m/mg Ir-193 and a calculated specific activity of 39 MBq/mg. A patent by De Groot et al144. describes irradiation of Ir-193 targets for two weeks in a medium flux position of the HFR reactor (flux of the order of 2 × 1014 cm-2 s-1) reaching specific activities after Ir/Pt separation of »15 GBq Pt-195m/mg Pt. We summarize that a high-flux neutron irradiation of Ir-193 targets could produce Pt-195m by a multi-step reaction, followed by a radiochemical Ir/Pt separation. This leads to relatively high specific activities, but the yields and specific activities reported in literature show extremely large scatter indicating possibly unreliable literature cross-sections for the Ir-194(n,g) reactions to Ir-195g and Ir-195m, respectively. Consequently, it is challenging to predict yields or specific activities that could be obtained by high-flux irradiations at ILL, BR2, JHR, ESS or MYRRHA, respectively. More test irradiations in well-known neutron spectra are required to solve this problem and improve the predictability of Pt-195m yields and specific activities achievable in this production path. 9.3 Production at existing and emerging PRISMAP facilities 9.3.1 Production at JHR Cadarache The production in the JHR Cadarache has been calculated considering Pt targets with natural isotopic composition. Two configurations were compared: unfiltered thermal plus epithermal neutron flux and Cdfiltered neutron flux that cuts thermal neutrons below about 0.5 eV and just leaves the epithermal component. In the unfiltered neutron spectrum the Pt-195(n,n’) production dominates (75% of the total Pt-195m production) over the Pt-194(n,g) production (24% contribution) and the Pt-196(n,2n) is negligible (0.4% contribution) since only a small fraction of the fission neutron spectrum has sufficient energy (>8.2 MeV). 141 F.F. Knapp Jr et al., J Radioanal Nucl Chem 263 (2005) 503. 142 S. Mirzadeh et al., US patent 2004/0032923 A1. 143 A.S. Madumarov et al., Nucl Med Biol 134-135 (2024) 108928 144 S. de Groot et al., EP2023072728W Deliverable D8.3 75 In the Cd-filtered neutron spectrum the Pt-195(n,n’) production dominates even more (89%), the Pt-194(n,g) contribution drops to 10% and Pt-196(n,2n) represents 0.5%. Specific activities (=yields) and radionuclidic purity have been calculated for an irradiation of 4 days followed by 1 day decay with the cross-sections of the EAF2010 library. However, this library does not contain the previously discussed 13,000 b thermal destruction cross-section of Pt-195m, but has instead some strong resonances in the epithermal range (>10 eV). We note that the location and magnitude of these resonances are not based on experimental information (because there are no Pt-195m targets available for energyresolved neutron-capture experiments), but have simply been assumed in the library. To account for the possible effect of a very high thermal cross-section, a second calculation has been performed with a manually renormalized Pt-195m(n,g) cross-section where the low-energy part was enhanced to reach 13,000 b at the thermal energy (25.3 meV), see Figure 34. Figure 34. Renormalization of the EAF2010 Pt-195m(n, g ) cross-section to account for the high thermal neutron capture probability. Table 16. Calculated specific activity and radionuclidic purity (RNP) for the irradiation of natural or enriched Pt targets for 4 days in the JHR neutron spectrum, followed by 1 d decay and assumed radiochemical removal of other elements. Hence, the RNP only considers other Pt radioisotopes. EAF2010 XS library + Renormalized Pt-195m(n, g ) c.s. Target Specific Activity Pt-195m MBq/mg Pt RNP Bq Pt-195m/Bq Pt Specific Activity Pt-195m MBq/mg Pt RNP Bq Pt-195m/Bq Pt Nat Pt 65 21.1% 16 6.2% Pure Pt-194 48 97.3% 12 89.9% Pure Pt-195 144 96.5% 35 87.2% Pure Pt-196 1.1 0.14% 0,28 0.03% Nat Pt + Cd filter 52 24.2% 26 13.9% Pure Pt-194 + Cd filter 16 92.5% 8.1 86.2% Pure Pt-195 + Cd filter 137 98.2% 69 96.5% Pure Pt-196 + Cd filter 1.1 0.20% 0.55 0.10% Deliverable D8.3 76 The highest specific activity is calculated with enriched Pt-195 targets exploiting the Pt-195(n,n’) reaction. With the unmodified EAF-2010 cross-sections, Cd-filtering slightly reduces the specific activity but improves the radionuclidic purity because other Pt radioisotopes are less produced. With the renormalized cross-section, the Pt-195(n,n’) route still provides the best specific activity but at a factor two reduction compared to unmodified EAF-2010 cross-sections. Now the Cd-filtering becomes beneficial as it doubles the achievable specific activity. We note that the main uncertainty stemming from the not precisely known Pt-195m destruction crosssection is inherent to these calculations and a final conclusion on the optimum neutron spectrum and irradiation conditions would require more reliable cross-section data and/or experimental verification in an actual neutron spectrum. 9.3.2 Production at MYRRHA The production in MYRRHA has been calculated considering Pt targets with natural isotopic composition. The fast neutron spectrum at MYRRHA has less thermal contribution compared to JHR, but the main contribution to Pt-195m production is again the Pt-195(n,n’) reaction with 71.1%, followed by Pt-194(n,g) with 28.8%. An additional calculation has been performed with the BR2 reactor spectrum (mainly thermal + epithermal) to study the influence of a 13,000 b Pt-195m destruction cross-section. Introducing the latter “manually” by rescaling the JEFF4-T4 cross-section to 13,000 b at 25.3 meV leads to a reduction of the Pt-195m activity by one third. Figure 35. Pt-196 excitation function from Pt-195m neutron irradiation 9.3.3 Production at IFMIF-DONES or SPIRAL2 The facilities IFMIF-DONES and SPIRAL2 provide neutron spectra peaked around 14 MeV. Therefore, among the different production routes based on Pt targets, the Pt-196(n,2n) route is expected to provide the highest yields. No detailed Monte Carlo simulations of Pt-196 have been performed, but the competitiveness of this reaction can be verified with a simple scaling exercise. A prior detailed calculation of Mo-99 production from Mo-100(n,2n) reactions had been published for IFMIF-DONES145. With 6 days irradiation of a 95% enriched 145 E. López-Melero et al., Nuclear Materials and Energy 38 (2024) 101575. Deliverable D8.3 83 pure samples. However, scenario 1 has the major drawback that the full source term of the target needs to be chemically processed. This is not required in scenario 2 as actinium is extracted online from the target. Scenario’s 3 and 4 explore the Th-232(p,x)Ra-225 route followed by three milking periods to extract Ac-225 from the Ra-225 generator. These two scenarios produce comparable amounts of around 130 MBq of Ac225. This is more than one order of magnitude lower than the Th-232(p,x)Ac-225 route. Scenario 3 goes over the difficult radiochemical purification of the full target which brings its own challenges. Moreover, only scenario 3 does not have a mass separation step and hence small amounts of Th-228 can be found in these samples (via Ra-228 (6y) -> Ac-228 (6h) -> Th-228 (2y)). A contamination level of 2.7 × 10-5 Bq Th-228/Bq Ac225 was found. Today it is unclear if this could hinder its usefulness as a radiopharmaceutical because of the long half-life of Th-228. The main conclusion of this study is that the preferred production route to produce Ac-225 with a 100 MeV beam is the online extraction of Ac-225 via the ISOL method. It is estimated that up to 2500 MBq per 10-day irradiation-cycle can be produced. This requires high-power targets (10 kW) and a proton-beam current up to 200 µA. 10.4 Production of Ac-225 with epithermal and fast neutrons on Ra-226 at ESS 10.4.1 Production To determine the activation calculations, ENDF70 libraries were used together with the fluxes shown in Figure 24. For consistency with the calculations performed for the other radionuclides of interest, a mass of 10 g of Ra-226 was assumed, even though such a large mass would not be possible in the specific case, due to the large amount of Rn produced (typically, masses in the range of mg of Ra-226 are used); results are however scalable per unit of mass. An irradiation time of 10 days, followed by 10 days of decay was considered. The main reactions for the production of 225Ac are: ¡ Ra-226(n,2n)Ra-225, followed by b-decay to Ac-225. ¡ Ra-226(g, n)Ra-225 followed by b-decay to Ac-225. The dominant reaction for the co-production of unwanted isotopes is Ra-226(n,g)Ra-227 followed by b-decay to Ac-227. These three reactions are responsible for the production of Ac-225 and of the unwanted Ac-227. Additionally, (n, np), (n,3n), (n,4n), (n, a) reactions were calculated. After 10 days of irradiation, followed by 10 days of decay, the following activities are calculated. Table 19. Activities from a Ra-226 sample of 10 g irradiated for 10 days, followed by 10 days of cooling. The Rn-222 activity from the a decay of Ra-226 is of GBq. Radionuclide Activity (GBq) Reaction Ac-225 550 (n,2n), (g,n) Ac-227 0.62 (n, g) Rn-219 1.68 (n,4n) Rn-220 78.7 (n,3n) Rn-223 0 (n,a) Deliverable D8.3 84 10.4.2 Discussion For both ESS models it was assumed that the ESS linear accelerator will provide a proton beam of 2 GeV, 62.5 mA peak current, a repetition rate of 14 Hz and 2.86 ms pulse length, corresponding to a time average current of 2.5 mA and a beam power of 5 MW. In the considered geometries, a peak of fast neutrons is present in the 1-MeV range, with a high-energy tail extending to the energy of the incoming proton beam. In addition, in the two-moderators model, a second peak of low-energy neutrons is present below 1 eV, due to thermalization in the large liquid deuterium moderator. The radionuclidic purity of Ac-225, i.e. the fraction of activity of the wanted nuclide to all activity, where only the activities of Ac-225 and Ac-227 after separation are considered, is of 99.9% after 10 days of decay. 10.5 Production of Ac-225 with epithermal and fast neutrons on Ra-226 at JHR (CEA) 10.5.1 Production This study considers the irradiation of a 10 g radium metal target composed of 100% Ra-226 with a density of 5 g cm-3. If this quantity of Ra-226 were available, the target’s activity prior to irradiation would be 366 GBq. When irradiated with fast neutrons, Ra-226 undergoes a transmutation to Ra-225 via the (n,2n) reaction in the high-energy region. The (n,2n) reaction, which has a threshold energy of 6.4 MeV, is followed by the subsequent beta-decays of Ra-225 into Ac-225. (In terms of fast neutron flux above 6.4 MeV, the JHR irradiation characteristics are 7.4 × 1012 n·cm-2·s-1 and 6.3 × 1011 n·cm-2·s-1 in the core and reflector irradiation positions, respectively. The irradiation of Ra-226 also results in the generation of Ra-227 through the process of radiative capture (n,g). This phenomenon is particularly evident in the thermal energy range of neutrons. The Ra-227 radionuclide then undergoes rapid decay (T₁/₂ = 42 min) to the Ac-227, which is a beta emitter characterised by a relatively long half-life of 21.8 years. For the purpose of targeted alpha therapy applications, it is crucial to minimise the production of Ac-227 to ensure the required radioisotopic purity of Ac-225. In order to facilitate effective comparison, a common irradiation scenario was adopted. The baseline scenario posits a 10-day irradiation period, during which half of the saturation activity of Ac-225 is formed, followed by a 10-day decay period. However, this scenario is not realistic regarding the expected operation of the JHR, which assumes that the irradiation positions within the core are only accessible during the outage period of refuelling. Therefore, in order to provide a more comprehensive analysis, a more realistic scenario was investigated, which assumes a 36-day irradiation period, corresponding to the expected cycle length of the reactor. The calculation results obtained for these two scenarios are presented in Table 20. Deliverable D8.3 85 Table 20. Production of Ac-225 by fast neutron irradiation of a 10 g Ra-226 target in the JHR. Neutron flux characteristics Cd-filtered, core Cd-filtered, reflector Cd-filtered, core Unfiltered, core Irradiation time at EOI (days) Ac-225 Activity (GBq) 10 24 10 2 36 124 36 114 Decay time at EOP #1 (days) Ac-225 Activity (GBq) Ac-227/Ac-225 (Bq/Bq) Ac-226/Ac-225 (Bq/Bq) Radioisotopic purity (%) Total activity (GBq) 1 25 3.91 0.04 4.8% 2960 1 2 15.92 0.01 3.1% 2080 1 114 2.86 0.05 6.3% 8790 1 108 9.05 0.23 0.6% 79510 Decay time at EOP #2 (days) Ac-225 Activity (GBq) Radioisotopic purity (%) Total activity (GBq) 9 23 100% 1880 9 2 100% 1670 9 50 100% 2650 9 59 100% 7490 10.5.2 Discussion The results of the in-reflector irradiation show that the activity of Ac-225 is approximately ten times lower than that of the in-core irradiation (25 GBq compared to 2 GBq). Using an irradiation device equipped with a cadmium filter significantly reduces the thermal neutron flux, thereby limiting the production of Ra-227 through radiative capture. This improves the radioisotopic purity of Ac-225, increasing it from 0.6% to 6.3%. However, these levels of purity remain incompatible with the requirements for medical applications. Highpurity Ac-225 can be produced by separating Ra and Ac through a series of chemical processes. Following a 10-day (or a 36-day) irradiation period, the target is extracted from the JHR core and permitted to decay for 24 hours. Thereafter, a preliminary chemical separation of radium and actinium species is conducted. For the sake of simplicity, it is assumed that this chemical process yields an efficiency of 90%. This hypothetical working assumption has the effect of penalising the quantities of radium and actinium that can be extracted from the target. In this preliminary separation, the actinium must be discarded due to the presence of a large amount of the undesirable Ac-227, which is co-produced during the irradiation. Meanwhile, 90% of the remaining radium isotopes are extracted and left to decay for a further period of 9 days. During this period of cooling, the Ac-225 is naturally produced and accumulates through the beta decay of Ra-225 (designated as the “cow” in the so-called “milking process”). Subsequent to this cooling period, a secondary separation can be performed in order to yield a highly pure Ac-225, practically free from any other isotope of actinium. The residual radium can be reused as a novel Ra-226 irradiation target, or alternatively as a new Ra-225 cow for repeated milking, as illustrated in Figure 39. A similar milking scenario was considered to assess the production rate of Ac-225 in the JOYO fast reactor156. 156 D. Iwahashi, K. Kawamoto, Y. Sasaki, N.Takaki, Neutronic study on Ac-225 production for cancer therapy by (n,2n) reaction of Ra226 or Th-230 using fast reactor JOYO, Processes 2022, 10, 1239. Deliverable D8.3 86 Figure 39. Process of milking to produce the Ac-225 in the JHR. 10.6 Production of Ac-225 via high-energy proton (1.4 GeV) irradiation of Th-232 at CERN-MEDICIS 10.6.1 Production Isotope (mass)-separation has a long history, both for the separation of stable or very long-lived radionuclides, and later for the separation of radionuclides from thick irradiated targets, either online as done at ISOLDE at CERN or offline in batch mode as done more recently at MEDICIS, specifically for the production of radionuclides for biomedical research. As opposed to the production of short-lived isotopes, the process related to the separation of Ac-225 and Ra-225 from irradiated thorium targets rely on the past experience accumulated over decades at ISOLDE and elsewhere. The separation of Ac-225 and Ra-225 proceeds with the release by ohmic heating of the target material under vacuum, and the ionisation of the released isotopes using specialised compact 1+ ion sources. The irradiation of the thick thorium target with high energy protons induces different nuclear reactions, notably spallation and fission reactions. Crosssections are in the mb region above 150 MeV, and are thus low compared to low energy nuclear reactions. The activity at EOI can therefore be compensated with thick targets of several tens of g. The preferred mode of operation proceeds with the release under vacuum of Ac-225 and Ra-225 from targets processed in the form of a refractory compounds. Thorium targets are made of refractory oxides, and of carbides eventually with excess graphite, with high porosity. Similar targets made of natural or depleted uranium have also been operated. The material must sustain high temperatures during the separation process, which is around 1600 C for radium release and 2200 C for actinium release, coinciding with the proton beam impact. The release rate and the total efficiency will depend on a number of parameters, notably of the diffusion and evaporation from the material, and the efficiency of the operated ion source. The duration of the process will for instance depend on the impurities and gas load from the target and structural materials brought at the high temperatures. The CERN-MEDICIS production steps are reported in Figure 40. They fit in the more general supply chain with the different institutes in PRISMAP. Mass separation fits well with the specific supply chain for Ra-225 and Ac225, due to the long half-life of more than a week, allowing thus to fit different steps with moderate decay. Deliverable D8.3 87 Figure 40. Mass separation process at MEDICIS157. Some operational parameters have recently been reported for MEDICIS, and serve as the basis for the estimates provided in this report158,159. For all scenarios, an irradiation of 10 days and a separation of 3 days have been considered. Separation from oxide or carbide targets of 50 and 100 g, respectively, and a separation of Ra-225 or Ac-225 ion beam at 72 or 12% efficiency, respectively, has been considered (see Table 21). Table 21. Ac-225 and Ra-225 activities reached at end of separation process. A subsequent radiochemical “milking” is required to extract Ac-225 from Ra-225 separated sources. Scenario Target Ac-225 or Ra-225 activity (MBq) Ac-225 / Ac-225 (Bq/Bq) Ra-225 separation ThCx 886 100.00% Ac-225 separation ThO2 395 100.00% Ra-225 separation ThO2 372 100.00% Ac-225 separation ThCx 942 100.00% 10.6.2 Discussion The production of large activities of Ac-225 with spallation reactions with high energy proton beams on thorium targets has been developed already from the past few years, mainly in the DOE program managed in a “Tri-Lab” effort by ORNL, BNL and ANL. In parallel, mass separation of radionuclides produced in a thorium target has been developed over the past years at MEDICIS. The related target and ion source technologies require specialists and appropriate accelerator environment. For industrial scale production, access to thorium targets at level of kg/year must be guaranteed, as well as the related synthetical and characterization infrastructures for actinide materials. The operation of a related infrastructure must comply with handling open radiological sources, with appropriate monitoring and skilled staff, notably dealing with radioprotection matters. Radon containment and monitoring must be managed, at levels much lower than those foreseen for the handling of gram quantities of Ra-226 targets. The upscaling of this process can be foreseen, with careful checks of the possible limiting factors. At MEDICIS itself, a foreseen increase of beam intensity and energy is planned, going to 6 µA and 2 GeV proton beam on average. In addition, parallel isotope collection is possible, allowing collection of Ra-223,224,225 or Ac-225,226 sources. The handling and disposal of irradiated thorium targets in different chemical forms is a final aspect that needs careful attention in the implementation of Ac-225 production with such a technique. 157 T Stora et al 2024 J. Phys.: Conf. Ser. 2687 082039 158 C. Bernerd at al, JaCOW IPAC’25 proceedings (2025) https://doi.org/10.18429/JACoW-IPAC25-TUPB047 159 J.D. Johnson et al. Sci Rep 13, 1347 (2023). https://doi.org/10.1038 Deliverable D8.3 88 11. Conclusions 11.1 Overview The goal of the study reported in this white paper is to evaluate various reaction channels and production routes for a selected set of radionuclides, with the aim of identifying the most efficient and promising methodologies and infrastructures for each of the investigated radionuclides. The paper consolidates findings on the production of innovative radionuclides for nuclear medicine, specifically Sc-47, Cu-67, Tb-152, Tb-155, Pt-195m, and Ac-225. It explores the capabilities of emerging facilities in this domain, building on existing expertise and ongoing R&D efforts at both established and newly developed centres. Particular emphasis is placed on production routes enabled by emerging infrastructures, with comparative analyses to existing methods at established institutions wherever possible. Particular attention was given to key parameters at each stage of the production chain, including total and specific activity, radionuclidic purity, and sample composition (natural vs enriched). Due to the stringent high purity requirements, detailed considerations were made regarding mass and chemical separation processes. The efficiencies of these processes were estimated and taken into account for several of these studies. The production methods discussed, which concern different types of irradiations, different purification processes, and the scale in power of the facilities, are complementary. For preclinical research, a large-scale production is not required; instead, a coexistence of facilities with different capacities is needed to support flexible and efficient development. European infrastructures will capitalize on the existing experience developed in PRISMAP, and will notably reinforce the position of Europe, with new mass separation facilities, new neutron sources covering a wide energy spectrum and different high-power ion beam facilities. The following section summarizes the main outcomes and challenges associated with each of the individual radionuclides that have been studied. 11.2 Conclusion from scandium studies ¡ Radiochemical separation of scandium from irradiated calcium or titanium targets is well established. ¡ Separation of scandium isotopes from ISOL targets was demonstrated, which removes co-produced impurities. Further margins exist on separation efficiencies. ¡ The use of enriched Ti-50 targets would be favourable in proton irradiations to enhance yield and radionuclidic purity, however, the contamination with Sc-46 and Sc-48 is not completely avoidable and needs experimental verification ¡ The chemical form of the target plays a crucial role. Metallic Ti outperforms TiO2 in the irradiation yield. TiC could be used optionally. In each case, the dissolution may require aggressive conditions. The identification of the proper collection substrate and optimal chemical separation route is challenging. ¡ A high enrichment of Ti-47 is mandatory for the (n,p) reaction, to minimise the co-production of longlived radionuclide impurities such as Sc-46 and Sc-48. ¡ The Ca-44(a,p)Sc-47 route requires enriched Ca-44 and has the advantage of a low co-production of other Sc radionuclides. Enriched Ca-44 is readily available in the chemical form of carbonate, but this has poor thermal stability. ¡ In each of the above, recycling of enriched target material is essential to make the process economically feasible and sustainable. 11.3 Conclusions from copper studies ¡ The use of light charged particles can produce Cu-67 at activity levels for preclinical studies or phase I clinical trials. ¡ Charged particle induced reactions on Zn-70 allow the production of Cu-67 without Cu-64 admixture. Deliverable D8.3 89 ¡ Production by fast neutrons at emerging facilities may offer higher batch activities, compatible with clinical trials and clinical use. The chemical processing of such targets is still to be investigated. ¡ Production by photonuclear reactions (Bremsstrahlung facilities) allows for the production of high activity and high radionuclidic purity. This route is being pursued by industrial producers. 11.4 Conclusions from terbium studies ¡ The need for chemical separation, after mass separation for elemental Tb separation, is typically required to reach high radionuclidic purity. ¡ Mass separation of Tb isotopes is still challenging as shown by the large scattering of extracted efficiencies. The most recent developments point to efficiencies of 1 to 5% that could be reliably achieved. The experimental scattering is due to the need to manage the targets at very high temperature due to the low volatility of elemental Tb. Alternatives show that. extraction of parent dysprosium or of Tb as a volatile molecule (i.e. TbFx, to obtain pure Tb-155) is a competitive process. ¡ For low energy facilities, the main challenges are the availability of sufficiently enriched targets to minimize the co-production of Tb contaminants as well as the challenges of high beam power-density in the targets. High activities could be produced if these challenges can be solved. ¡ Both, emerging and established facilities within PRISMAP, have complementary on-going R&D programs to improve the production of Tb. 11.5 Conclusions from platinum studies ¡ Carrier-added Pt-195m with specific activity not exceeding »0.1 GBq/mg is obtained by irradiation of enriched Pt-194 with thermal neutrons or of enriched Pt-195 with epithermal/fast neutrons. ¡ Resonant laser ionisation could provide isomeric selectivity between Pt-195m and Pt-195g. ¡ Certain enriched Pt isotopes are not commercially available at present. Research towards Pt isotope enrichment by AVLIS (Atomic Vapour Laser Isotope Separation) is recommended. ¡ Non-carrier-added Pt-195m and Pt-193m with much higher specific activities can be obtained by irradiation of Os-192 with a beams, but at very limited yield. High specific activities and much higher yields could be obtained by irradiating Ir-193 targets, either in a high thermal neutron flux by double neutron capture, or with deuteron or a beam irradiations. Efficient and rapid dissolution of irradiated Ir targets and radiochemical Ir/Pt separations should be investigated in a joint effort of accelerator and reactor labs. 11.6 Conclusions from actinium studies ¡ Existing experience on irradiation with high-energy particles of a thorium target and associated mass separation is a strong asset to prepare for future facilities. ¡ Combination of processes, such as Ra/Ac chemical separation and subsequent milking of the produced sources, allows to practically achieve theoretical radionuclidic purities. The exact achievable yields will depend to a large extend on the exact process followed. ¡ Production with low energy particles, neutrons and photonuclear reactions will require large targets of Ra-226. Their procurement, the handling of large activities and associated radon release and containment, the licensing and the recycling constitute the major points of focus for these production pathways. ¡ The access and procurement, handling and disposal of irradiated thorium compound targets will constitute a major point of focus for the production pathways based on thorium. ¡ In all cases, upscaling these processes that so far has been tested at lower scales, will be associated with major points of validation of the performance of the processes, and of their associated infrastructures and monitoring. ¡ While different Ac-225 production routes have been demonstrated, upscaling to production of larger activities is still a challenge.