SOLID-PHASE ISOTOPE HARVESTING: HARVESTING RADIOISOTOPES FROM BEAM IRRADIATED TUNGSTEN, BORON, AND LUTETIUM By Samridhi Satija A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Chemistry – Doctor of Philosophy 2024 ABSTRACT During the operation of radioactive ion-beam facilities, such as the recently decommissioned National Superconducting Cyclotron Laboratory (NSCL) and the newly operational Facility for Rare Isotope Beams (FRIB), many radioisotopes get deposited along beamlines in components such as in mass-separator slits, fragment catchers, and beam dumps. The techniques to recover the by-product radioisotopes from these components is called ‘solid-phase isotope harvesting’ or ‘isotope mining’ and can help to meet the increasing demand for radioisotopes in a variety of fields. This technique offers a unique approach to access rare isotopes that are often challenging to produce using conventional methods. A comparative analysis with aqueous-phase (harvesting from water) and gaseous-phase (harvesting from gases) methods also shows the distinct advantages of solid-phase harvesting, including simplified chemistry and enhanced flexibility in collection location and material choice. This work comprises four primary aims, each aiming at advancing the understanding and implementation of solid-phase isotope harvesting. The first is to establish the of proof-of-concept for solid-phase isotope harvesting, with a specific focus on the extraction of 88Zr from irradiated tungsten foils. This aim seeks to demonstrate the feasibility and efficacy of solid-phase harvesting, and enables comparison with aqueous-phase harvesting. The second aim involves the harvesting of 172Hf from a tungsten ‘heavy-met’ alloy beam-blocker/beam-dump that was used to stop unused beams at NSCL. The beam blocker was retrieved after the NSCL facility was decommissioned and was chemically processed to extract 172Hf and to develop a 172Hf/172Lu generator. This work utilizes the proof of concept study from the first aim and exhibits harvesting from a decommissioned component. The third aim involves the production of 172Hf from the conventional method of irradiating natural lutetium foil targets at the Brookhaven Linear Isotope Producer (BLIP), coupled with the development of methodologies for extracting 172Hf followed by creation of several 172Hf/172Lu generators. This aim helps to compare radionuclides produced by a conventional technique to those obtained from the isotope harvesting efforts shown in the second aim. The fourth aim demonstrates production and separation of 7Be from natural and enriched boron targets, followed by chemical separation, 7Be source preparation and 7Be beam delivery to users at NSCL. This aim demonstrates the practical application of solid-phase isotope harvesting by producing a specific radionuclide for use in scientific experiments. Collectively, this work describes the radioanalytical techniques and chemical methodologies used in the extraction and purification of embedded radioisotopes from beam- irradiated solid materials of tungsten, boron and lutetium. These four aims have together laid the groundwork for developing robust methodologies to extract and separate long-lived radioisotopes that will accumulate along the FRIB beamline. Ultimately, this work aims to enable solid-phase isotope harvesting on a larger scale at FRIB, when FRIB is operational at full beam power, and at other similar ion-beam facilities. The methods described herein allow harvesting of difficult-to- produce byproduct radionuclides without the need for dedicated beamtime at FRIB. Copyright by SAMRIDHI SATIJA 2024 Dedicated to my loving parents Kanti Raman and Sunita Satija, and younger sister Navani Satija for their unconditional love and support v ACKNOWLEDGEMENTS Reflecting on my PhD journey fills me with profound gratitude for the many remarkable individuals who have contributed to my success. First and foremost, I want to extend my deepest gratitude towards my parents and my first teachers in life, Kanti Raman and Sunita Satija and for their unwavering support throughout. The sacrifices they have made to provide the best of everything for me and my sister is still beyond my comprehension and conscious realm. I’m immensely grateful for all the difficult times they stood by me, checked on me every day and every night on video calls for the past five years and made me feel like I was never away from home. I’m beyond blessed to have them as my parents. To my sister, Navani, thank you for always acting like the little ‘elder sister’ taking care of everything while I was here doing my PhD, I’ll be forever indebted. You’re simply the best sister and friend I could ask for. Next, my heartfelt thanks to my PhD advisor, Dr. Greg Severin. Thank you for being infinitely patient and considerate throughout these five years, every time I faltered. You’re a great scientist and I’ve learnt a lot from you in that regard, but you’re an even better person, and I’ve been fortunate to learn from your example in that aspect too. My grad school experience has been exceptional and your support along with the work culture you’ve ensured in our lab is a huge reason behind it. Looking back, joining your lab was undoubtedly one of the best decisions of my life. I also want to extend a huge thank you to my committee members, Dr. Greg Swain, Dr. Sean Liddick and Dr. Kyle Brown for being so accommodating and always encouraging. I must express my gratitude to my labmates in this wonderful journey. Dr. Katharina Domnanich, who’s been by my side since Day 1. You’re the best mentor and a really great friend who I’ve learnt a great deal from. Thank you for your encouragement, it meant the world to me. Our ‘harry potter nights’, coffee and food celebrations are memories which will always hold a vi very special place in my heart. Thank you, Dr. Chirag Vyas for all your guidance, both professionally and personally all these years. You’ve always been like an elder brother to me and I knew I could always count on you to solve anything. A big thank you to Chloe, Vlad, and Scott for all the help with so many different and incredibly time consuming experiments, especially when I had gone to BNL. I am really going to miss all the fun we had these years, inside and outside the lab. Thank you to Jose, Tracy, Wes, and Xyla as well for all the assistance I received for all the experiments, you have all been the best labmates ever. A special thanks to the team at BNL - Dr. Cathy Cutler, Dr. Dmitri Medvedev, Vanessa, Michael, Leah, Jasmine and Courtney. I am grateful for the opportunity to work alongside you for 8 months. It was great to see the National Lab set up for the first time, and you all made it such a special experience for me. Thank you to the BMIS team at NSCL, especially Dr. Chandana Sumithrarachchi, for providing the data I needed even in a short time. I’d also like to thank my advisor, Dr. Sameer Sapra and Dr. Arkajyoti Chakrabarty while I was in IIT, for helping me gain research experience that made me want to pursue PhD at MSU. To my friends - Debashree, Aamna, Atif, Nidhi, Eagle, Ankita, Haritha, Sunanda, Souvik, Avirup, Himanshu, Benji, Apurva, and Paige - a big thank you for being so kind, for being my family away from home and for all the fun times we had, cooking and celebrating everything. All of you have played a big role in keeping me sane, motivated, and going. Finally, I am thankful to God, who is omnipotent and omnipresent, and from whom I derive the energy and guidance to navigate life's challenges. vii TABLE OF CONTENTS CHAPTER 1: INTRODUCTION..…………………………………………………………......1 1.1 Radioactive-ion beam facilities – NSCL and FRIB ......................................................... 1 Isotope harvesting ............................................................................................................ 2 1.2 1.3 The four aims for this thesis ............................................................................................. 7 1.4 The objective .................................................................................................................. 13 REFERENCES ......................................................................................................................... 15 CHAPTER 2: 88Zr HARVESTING: PROOF OF CONCEPT STUDY1 ............................... 19 2.1 Introduction .................................................................................................................... 19 2.2 Experimental .................................................................................................................. 20 2.3 Results and discussion .................................................................................................... 30 2.4 Conclusion ...................................................................................................................... 37 REFERENCES ......................................................................................................................... 38 CHAPTER 3: 172Hf HARVESTING FROM TUNGSTEN ‘HEAVY-MET’ ALLOY BEAM-BLOCKER…………………………………………………………………………... .. 40 Introduction .................................................................................................................... 40 3.1 3.2 Experimental .................................................................................................................. 42 3.3 Results and discussion .................................................................................................... 50 3.4 Conclusion ...................................................................................................................... 59 REFERENCES ......................................................................................................................... 60 APPENDIX ............................................................................................................................... 61 CHAPTER 4: 172Hf HARVESTING FROM LUTETIUM FOIL TARGET AT THE BROOKHAVEN LINEAR ISOTOPE PRODUCER……………………………………..…..62 4.1 Introduction .................................................................................................................... 62 4.2 Experimental .................................................................................................................. 64 4.3 Results and discussion .................................................................................................... 75 4.4 Conclusion ...................................................................................................................... 95 REFERENCES ......................................................................................................................... 96 APPENDIX ............................................................................................................................... 99 CHAPTER 5: 7Be HARVESTING: PRODUCTION AND SEPARATION FROM BORON AND BEAM DELIVERY …………………………………………………………..102 5.1 Introduction .................................................................................................................. 102 5.2 Experimental ................................................................................................................ 105 5.3 Results and discussion .................................................................................................. 114 5.4 Conclusion .................................................................................................................... 126 REFERENCES ....................................................................................................................... 127 APPENDIX ............................................................................................................................. 131 CHAPTER 6: CONCLUSION AND RECOMMENDATIONS ........................................... 132 6.1 Chemical methodologies applicability ......................................................................... 133 Future work .................................................................................................................. 134 6.2 Final remarks ................................................................................................................ 135 6.3 viii REFERENCES ....................................................................................................................... 137 ix CHAPTER 1: INTRODUCTION 1.1 Radioactive ion-beam facilities – NSCL and FRIB At radioactive ion-beam research facilities, such as the recently decommissioned National Superconducting Cyclotron Laboratory (NSCL) and the recently commissioned Facility of Rare Isotope Beams (FRIB), primary beams of heavy ions are accelerated with the aim to produce secondary radioactive beams for nuclear physics experiments. In this process, a vast array of exotic radionuclides gets produced that finds applicability in multiple fields enabling basic and applied research [1]. Some of the fields where these rare isotopes prove indispensable are medicine, biochemistry, plant biology, astrophysics, stockpile stewardship science, and materials science [1]. The operation of facilities like FRIB and NSCL will result in a significant advancement in rare isotope research capabilities resulting from the production of a broad distribution of radioactive isotopes. FRIB will exhibit intensities from the fragmentation of 400-kW primary beams with energies of about 200 MeV/nucleon [2], [3], [4]. This results in the production of 80% of all known isotopes of elements, and very recently five new isotopes were discovered that had never been seen on the surface of the earth [5]. This is indicative of the promise of FRIB and rare isotope beam science. During the operation of these radioactive ion-beam facilities, and generally for other facilities as well, byproduct radioisotopes get deposited at multiple places along the beamline, such as mass-separator slits, fragment catchers, beam-dumps and other locations. Over time, this process leads to a considerable accumulation of exotic radionuclides at these locations. These components accumulate long-lived radionuclides throughout their service period and typically have replacement time in the order of years. By the time these components are removed from 1 operation, potentially useful activities of radionuclides have been built up. In most cases, these accumulated radionuclides are difficult to produce through other conventional means and recovering and extracting them from decommissioned components might be the best option for obtaining them. 1.2 Isotope harvesting The accumulated rare isotopes can be chemically extracted and purified using radiochemical techniques in a process known as ‘isotope harvesting’ or ‘isotope mining’. Specifically for this thesis, ‘solid-phase isotope harvesting’ is the focus as it allows access to rare, usually difficult-to-produce radionuclides that find application in a variety of fields, mentioned previously [1], [6], [7]. Other forms of isotope harvesting include the ‘aqueous-phase harvesting’ which is harvesting from irradiated water, and ‘gaseous-phase harvesting’ which is harvesting from activated gases [8], [9]. Figure 1.1: A simplified schematic showing the process of solid-phase isotope harvesting at the NSCL. Over the last 20 years, primary beams at NSCL have been stopped in a tungsten beam-blocker during the delivery of secondary beams to experimental users. The activation products resulting from the dumped beam are isolated in the process of solid-phase isotope harvesting. 2 In order to understand how to best implement these novel approaches at FRIB, experiments have been performed at the NSCL, which delivered beams with properties similar to those that are delivered at FRIB, though at lower intensity by several orders of magnitude. A simplified schematic of the process of solid-phase isotope harvesting at the decommissioned NSCL, is shown in Figure 1.1. Briefly, before the NSCL was decommissioned, many of the unused nuclides and unused primary beams were embedded into a tungsten ‘heavy-met’ alloy beam-dump/beam-blocker/beam-stop (referred to as the ‘tungsten beam-blocker’ herein). The decades-long accumulation of embedded and activation-product radionuclides in the tungsten beam blocker provided a unique opportunity to study isotope harvesting from solid materials. Similarly, it is expected that materials in the FRIB beamlines will become available for harvesting as they are replaced over the course of many years of operation. All of the solid components shown in Figure 1.2 are potential locations in the FRIB beamline that will accumulate radionuclides continuously without the need for a dedicated beamtime [10]. Figure 1.2: Schematic of the potential isotope harvesting locations at FRIB that includes beam-dump, fragment catchers, and mass slits. Reprinted from ref [10] with permission from Elsevier. 3 While there are many advantages of building an ‘Isotope harvesting program’ for FRIB, some notable ones are as follows - this is a synergistic process and can keep happening without interrupting the beam delivery for physics experiments. Secondly, traditional facilities for isotope production, such as nuclear reactors and medical cyclotrons, face limitations in terms of the resources they can allocate to producing specific radionuclides. This restricts both the quality and quantity of the produced isotopes [1]. It is also a good approach to access hard-to-produce radionuclides from conventional methods. 1.2.1 Isotope harvesting from aqueous and gaseous phases So far, aqueous isotope harvesting approaches have been investigated extensively at the NSCL, in a series of experiments carried out by delivering NSCL beams into a dedicated aqueous endstation. An experiment facilitating the collection from primary beam of 48Ca and directing these unused primary beams to an aqueous target and replacing the solid tungsten beam-blocker already present there has been demonstrated [11]. Mastren et al. studied harvesting 67Cu using an aqueous beam stop built by Pen et al. at the NSCL demonstrating the recovery of the radiochemically pure 67Cu [4], [12], [13]. This isotope finds great relevance for radiopharmaceutical applications. Loveless et al. investigated harvesting experiments utilizing the same aqueous beam stop for recovery of 48V [14]. These proof-of-concept studies have demonstrated the importance of aqueous harvesting for providing rare isotopes which are difficult to produce by conventional methods. In parallel, the harvesting of gaseous radionuclides has been another main focus. Noble gas radionuclides that were created in the water targets have been transferred into the gas phase and trapped on selective traps. Recently, the collection of krypton radioisotopes to generate 76Br and 77Br depicted successful demonstration of the feasibility of gas-phase isotope harvesting from irradiated accelerator cooling-water [15]. 4 Another promising avenue under investigation is using Metal-Organic Frameworks (MOFs) that have the potential to increase the selectivity of krypton or other noble gas capture to increase collection efficiencies, especially at FRIB. The research suggests that when FRIB reaches full power, substantial quantities of these above-mentioned isotopes could be harvested daily, highlighting the need for these proof-of-concept experiments. 1.2.2 Solid-phase isotope harvesting Just like the aqueous and gaseous phase harvesting efforts, solid-phase isotope harvesting at NSCL/FRIB has a great potential to provide access to rare isotopes. Similar efforts in solid phase harvesting at other facilities worldwide have been highly successful. Among one of the first such efforts, the ERAWAST (Exotic Radionuclides from Accelerator Waste for Science and Technology) initiative at the Paul Scherrer Institut (PSI) successfully extracted the long-lived isotopes 44Ti (t1/2 = 59.1 y), 26Al (t1/2 = 7.17 x 105 y), 53Mn (t1/2 =3.74 x106 y), 60Fe (t1/2 = 2.62 x106 y) and 59Ni (t1/2 = 7.6 x104 y) that are of importance to astrophysics experiments from a copper beam dump [16], [17], [18], [19], [20], [21], [22]. This copper beam dump was a part of accelerator waste generated at PSI. The extracted 60Fe (t1/2 = 2.62 x106 y) was used for studying the 60Fe(n, γ)61Fe cross-section and for half-life measurement experiments [17], [23], [24]. In another study, 10Be (t1/2 = 1.51 x106 y) was extracted from irradiated graphite wheels, and 207Bi (t1/2 = 31.55 y) from an irradiated lead target at PSI [25], [26]. Because of the success of the ERAWAST project, other facilities are also developing programs for the collection of these by- product radionuclides. For the majority of radioisotopes that will not emerge as gases, aqueous-phase or solid- phase harvesting will be the only feasible way to collect them. Although, in principle, separating radionuclides from the solid materials will be complicated, there are a few instances where solid- 5 phase harvesting provides a distinct advantage over the aqueous-phase harvesting procedures. In the case of aqueous-phase harvesting, the chemistry gets complicated because, at least at FRIB, there is an enormous stock of water that must be processed for efficient harvesting. In this situation, the radionuclide of interest, present at sub-nanogram scales, would need to be isolated from up to 8000 L water [27], [28]. This renders the extraction of radionuclides that are prone to hydrolysis very difficult. Additionally, there is also the problem of formation of highly oxidative species, such as hydrogen peroxide, from the stopping of high intensity heavy-ion beams in water. This would complicate the collection of radionuclides with sensitive redox chemistry [29], [30]. Additionally, with this kind of volume, the beam products must be concentrated by reducing the volume of the water by evaporation, co-precipitation, or adsorption to a solid phase such as an ion-exchange resin. Whereas in solid-phase harvesting, smaller volumes are required for the dissolution of collectors/targets which considerably simplifies the post-irradiation chemistry. This offers more flexibility in the development of specific methodologies, such as dissolving the collector/target in the appropriate acid matrix that restricts hydrolysis. Solid-phase harvesting is a general method that should work for nearly all chemical species. Advantages of solid-phase harvesting also includes flexibility in terms of collection location and the collector material. As shown in Figure 1.2, the solid collectors can be deliberately placed in various locations along the beamline, allowing the collectors to intercept the beam for a short time and potentially providing access to shorter-lived species. Additionally, beam products will accumulate in solid beam-dumps whenever the facility is in operation and will therefore offer a wide variety of radionuclides. This will potentially require more 6 sophisticated chemical separation methods to be developed to harvest these short- and long-lived radionuclides. Many radionuclides that have been identified as being important for astrophysics and stockpile stewardship have half-lives longer than 30 days and will build up over months or years of facility operation. Therefore, in preparation for the large-scale isotope harvesting at FRIB, it was important to investigate solid-phase harvesting approaches at NSCL and establish proof-of- concept procedures. With a focus on solid-phase, this thesis will involve the harvesting of radionuclides from beam irradiated tungsten, boron, and lutetium. Any of these materials can be used to either accumulate radionuclides from FRIB’s rare isotope beams, or can be activated themselves by interactions with beams. Depending on the application, materials and irradiation conditions can be selected and matched with appropriate radiochemical processes to provide useful quantities of rare radionuclides for science. 1.3 The four aims for this thesis 1.3.1 AIM I - 88Zr harvesting: Proof of concept study for establishing solid-phase isotope harvesting approaches Previously, a study by Shusterman et. al. investigated the extraction and collection of 88Zr and its daughter 88Y, when a high-energy secondary 88Zr beam from a 140-MeV/u 92Mo primary beam was stopped in an aqueous endstation at the NSCL [31]. This was the first study to access the recovery of a Group IV element that is prone to hydrolysis form an aqueous matrix. As a result, recovery efficiency for 88Zr post chemical processing from aqueous harvesting was only (26±2)%. Another study was aimed at the collection of 48V, and showed similar low recovery yields of 47% [14]. It was inferred that for radionuclides that exhibit hydrolysis in water such as Zr, Hf, and V, aqueous harvesting presents considerable challenges due to irreversible formation 7 of hydrolyzed species and their adsorption on container walls. When micromolar concentrations are exceeded for metals, these group IV (Zr and Hf) and V (V) metals tend to exist in tetravalent and pentavalent oxidation states, and form oxide and hydroxide complexes in aqueous conditions that are not highly acidic [32]. Thus, solid-phase isotope harvesting is a viable option here. Additionally, several Zr isotopes are valuable for nuclear science, such as 95Zr for astrophysics, 86Zr and 89Zr for nuclear medicine, and 88Zr for stockpile-stewardship applications. For example, 88Zr (t1/2 = 83.4 d) with its characteristic 392.87-keV γ ray (emitted with a γ-ray intensity Iγ = 97.29%) is of interest for interpreting data from radiochemical detectors used in underground nuclear tests [33], [34]. Since the United States no longer conducts nuclear tests, the National Nuclear Security Administration (NNSA) uses a science-based assessment called the Stockpile Stewardship Program for maintaining and enhancing the safety and security of the nuclear weapons stockpile [35]. The recently determined large thermal-neutron-capture cross section of 88Zr, amounting to (8.61 ± 0.69) x105 barn, represents an important addition to the cross section data of radioactive nuclei required for the Stockpile Stewardship Program [36]. The simplified decay scheme for 88Zr is shown in Figure 1.3. Furthermore, the daughter of 88Zr, 88Y (t1/2 = 106.6 d) is of interest to the astrophysics community for p-process investigations and for application as a surrogate for the medically relevant 90Y in research. Its prolonged half-life and the emittance of easily detectable 898.04 and 1836.06 keV gamma rays (Iγ = 93.7% and 99.2%, respectively) enables the performance of tracer studies, instead of directly employing the diagnostic and therapeutic counterpart [37], [38]. Considering the many uses of zirconium and yttrium, it is worthwhile to explore its chemistry from solid-phase perspective. In this first proof-of-concept experiment, the beamtime at the NSCL was devoted to the fragmentation of a 92Mo primary beam and the implantation of the formed secondary 88Zr beam 8 in diverse collector materials, such as copper, aluminum, gold and tungsten. These materials were deliberately chosen because they are expected to be used as components in FRIB or have already accumulated beam at NSCL. Thereby, the radionuclides of 88,89Zr, 85-88Y, and 83-86Sr were produced along with other species such as 90Nb and 83,84Rb. This allowed for the investigation of extraction and collection of 88Zr and its daughter 88Y from the solid collector of tungsten by the method of solid-phase harvesting. Many isotopes of Y which is a lanthanide analog have applications in nuclear medicine and batteries, stockpile stewardship, and materials science. To sum up the AIM I, the main tasks for solid harvesting of 88Zr involve the development of chemistries needed to purify Zr, Y, and Sr from the collector material of tungsten, development of a solid harvesting endstation for the proof-of-concept experiments at the NSCL, conducting the heavy-ion irradiation of the solid collector materials using NSCL beams followed by the separation of the collected products by implementing the newly-developed techniques. Figure 1.3: Simplified scheme for the decay of 88Zr to produce 88Y which decays into stable 88Sr [41]. 9 This proof of concept experiment would provide insight into tungsten and zirconium chemistry, and is particularly important because it will allow comparison to the aqueous collection methods developed for this isotope. This is described in further detail in Chapter 2. 1.3.2 AIM II - 172Hf harvesting from tungsten ‘heavy-met’ beam-blocker AIM I served as proof-of-concept for AIM II. The interest in harvesting from the tungsten beam-blocker recovered from the NSCL is due to the presence of radionuclides 172Hf (t1/2 = 1.87 y), and 173Lu (t1/2 = 1.37 y) which are the most abundant after a one-year decay period [39], [40]. Establishing proof of concept methodologies are necessary before attempting harvesting and processing at a higher scale. After achieving AIM I, and having understood the tungsten, zirconium, and yttrium chemistry from the irradiation of the tungsten target in AIM-I, the next step was to process the ‘tungsten beam-blocker’, and develop separation techniques to primarily separate out 172Hf from medium Z and high Z radionuclides embedded in the beam blocker. The chemical homologs, 88Zr and 88Y, were suitable surrogates for developing separation chemistries (in AIM I) before attempting hafnium and lutetium chemistry from beam- blocker. This tungsten beam blocker (shown in Figure 1.4) was used as a beam stop since 2000. It was water-cooled and located at the exit of the first dipole magnet to block the primary beam (or a primary beam charge state); in the A1900 fragment separator along the beamline. It is an 89- mm (3.5 inches) in length ‘heavy-met’ alloy (W 95%, Cu 1.5% and Ni 3.5%), 51-mm (2 inches) wide and 102-mm (4 inches) deep. The activity in the beam-blocker (described in more detail in Chapter 3) is due to implantation of radioactive ions and activation from fast secondary light ions and neutrons [41]. Decommissioning of the NSCL had made it possible to harvest isotopes from the beam-blocker which have applications in a number of fields. 10 Figure 1.4: Different sides of the tungsten ‘heavy-met’ beam-blocker is shown to provide the reader with dimension estimation. After the extraction, a 172Hf/172Lu generator was created for application of 172Lu in Perturbed Angular Correlation Spectroscopy (PAC) described in further detail in Chapter 3. Since the chemistry of zirconium is similar to hafnium and chemistry of yttrium is similar to lutetium, separation techniques developed from the proof-of-concept tungsten irradiation experiment had proved useful. The chemical processing of this beam-blocker at NSCL and potentially the ones when FRIB is in operation in the future could yield substantial quantities of 172Hf and consequently its daughter, 172Lu, which finds applicability in several fields. This is described in further detail in Chapter 3. 1.3.3 AIM III - 172Hf harvesting from lutetium foil target at the Brookhaven Linear Isotope Producer The third AIM of solid-phase isotope harvesting is to produce 172Hf from a conventional method of irradiating a natural lutetium foil with protons. This irradiation happened at the Brookhaven Linear Isotope Producer (BLIP) involving the 175Lu(p,4n)172Hf nuclear reaction and was followed by developing chemical methodologies for setting up the 172Hf/172Lu generator from the extracted parent, 172Hf. A simplified decay scheme is also shown in Figure 1.5. 11 One prime motive of using the solid-phase isotope harvesting approach is to be able to compare separation yields from decommissioned activated material with those that are processed after production from conventional methods, such as after being produced from a cyclotron or reactor. Comparison of the 172Hf/172Lu generator obtained from AIM II and AIM III was also made and is provided in Chapter 4. Figure 1.5: A simplified decay chain of 172Hf, 172Lu, and 172Yb is shown [43]. 1.3.4 AIM IV - 7Be harvesting: Production and separation from boron and beam delivery Having understood chemistry of the extraction of radionuclides from solid material of tungsten and lutetium in AIM I, II, and III, an opportunity arose for the production and extraction of 7Be (t1/2 = 53.2 d) from a solid natural (10B and 11B) and (10B) enriched boron pellet target. This was required to provide the 7Be beam to experimenters. 12 This experiment involved investigating the feasibility of the 10B(p,α)7Be nuclear reaction (Figure 1.6) for the production of 7Be, carrying out the production at the University of Alabama at Birmingham Cyclotron facility, followed by performing the developed separation methodologies for extracting 7Be from the irradiated boron with high purity. This was followed by 7Be source preparation for use in the Batch-Mode Ion Source (BMIS) to facilitate the re- accelerator operation at NSCL. The entire pipeline was demonstrated here for creating offline radioactive beams, and the 100% pure 7Be4+ ion beam was delivered to users. Figure 1.6: The relevant portion of the chart of the nuclides is shown to depict the possible production pathways for various shorter-lived radioisotopes in the vicinity of 7Be, which can be co-produced by proton irradiation of a boron target. This is discussed in greater detail in Chapter 5. 1.4 The objective In preparation for the large-scale isotope harvesting program at FRIB, it is important to establish proof-of-concept procedures and investigate solid-phase harvesting approaches using decommissioned components from the NSCL [41][42]. Collectively, achieving the above- mentioned four AIMs have established guidelines and approaches needed to be able to extract and chemically separate the enormous quantities of long-lived radioisotopes which will 13 accumulate in several locations along the FRIB beamline during beam delivery to end users (Figure 1.7). Figure 1.7: The schematic of solid-phase isotope harvesting under the broad umbrella of isotope harvesting is shown here, comprising of four AIM’s. The first aim involves the harvesting of 88Zr from a tungsten foil target as a proof of concept experiment, the second aim is to harvest 172Hf from the tungsten ‘heavy-met’ beam-blocker, the third aim is to harvest 172Hf from lutetium foil target, and the fourth aim is to harvest 7Be from boron pellet target. 14 REFERENCES [1] E. P. Abel et al., ‘Isotope harvesting at FRIB: Additional opportunities for scientific discovery’, Journal of Physics G: Nuclear and Particle Physics, vol. 46, no. 10, 2019, doi: 10.1088/1361-6471/ab26cc. A. Gade and B. M. Sherrill, ‘NSCL and FRIB at Michigan State University: Nuclear [2] science at the limits of stability’, Physica Scripta, vol. 91, no. 5. Institute of Physics Publishing, Apr. 13, 2016. doi: 10.1088/0031-8949/91/5/053003. [3] A. B. Balantekin et al., ‘Nuclear theory and science of the facility for rare isotope beams’, Modern Physics Letters A, vol. 29, no. 11. World Scientific Publishing Co. Pte Ltd, Apr. 10, 2014. doi: 10.1142/S0217732314300109. [4] A. Pen et al., ‘Design and construction of a water target system for harvesting radioisotopes at the National Superconducting Cyclotron Laboratory’, Nucl Instrum Methods Phys Res A, vol. 747, pp. 62–68, 2014, doi: 10.1016/j.nima.2014.02.010. ‘FRIB creates five new isotopes’. Accessed: Mar. 22, 2024. [Online]. Available: [5] https://frib.msu.edu/news/2024/new-isotopes D. Schumann, E. Maugeri, and R. Dressler, ‘Exotic radionuclides - What are they good [6] for?’, Chimia (Aarau), vol. 74, no. 12, pp. 932–938, 2020, doi: 10.2533/CHIMIA.2020.932. D. Schumann and J. Neuhausen, ‘Accelerator waste as a source for exotic radionuclides’, [7] Journal of Physics G: Nuclear and Particle Physics, vol. 35, no. 1, 2008, doi: 10.1088/0954- 3899/35/1/014046. E. P. Abel and G. W. Severin, ‘Isotope Harvesting of Aqueous Phase Ions from Heavy- [8] Ion Fragmentation Facilities for the Production of a 47Ca/47Sc Generator’, United States -- Michigan, 2020. [Online]. Available: https://ezproxy.msu.edu/login?url=https://www.proquest.com/dissertations-theses/isotope- harvesting-aqueous-phase-ions-heavy-ion/docview/2476547502/se-2?accountid=12598 H. K. Clause and G. W. Severin, ‘Demonstrating Gas-Phase Harvesting Capabilities at [9] the NSCL Through the Production and Collection of 76Kr and 77Kr’, United States -- Michigan, 2021. [Online]. Available: https://ezproxy.msu.edu/login?url=https://www.proquest.com/dissertations- theses/demonstrating-gas-phase-harvesting-capabilities/docview/2572564689/se- 2?accountid=12598 [10] W. Mittig, ‘Possible evolutions of rare isotope production methods and the associated instrumentation’, Nucl Instrum Methods Phys Res B, vol. 317, no. PART B, pp. 186–193, 2013, doi: 10.1016/j.nimb.2013.07.026. [11] E. P. Abel et al., ‘Production, collection, and purification of 47Ca for the generation of 47Sc through isotope harvesting at the national superconducting cyclotron laboratory’, ACS Omega, vol. 5, no. 43, pp. 27864–27872, 2020, doi: 10.1021/acsomega.0c03020. 15 [12] T. Mastren et al., ‘Feasibility of isotope harvesting at a projectile fragmentation facility: 67Cu’, Sci Rep, vol. 4, pp. 1–6, 2014, doi: 10.1038/srep06706. [13] T. Mastren et al., ‘Harvesting 67Cu from the Collection of a Secondary Beam Cocktail at the National Superconducting Cyclotron Laboratory’, Anal Chem, vol. 87, no. 20, pp. 10323– 10329, 2015, doi: 10.1021/acs.analchem.5b02322. [14] C. S. Loveless et al., ‘Harvesting 48V at the National Superconducting Cyclotron Laboratory’, Applied Radiation and Isotopes, vol. 157, no. November 2019, p. 109023, 2020, doi: 10.1016/j.apradiso.2019.109023. [15] H. K. Clause et al., ‘Harvesting krypton isotopes from the off-gas of an irradiated water target to generate 76Br and 77Br’, Sci Rep, vol. 12, no. 1, Dec. 2022, doi: 10.1038/s41598-022- 05500-8. [16] R. Dressler et al., ‘44Ti, 26Al and 53Mn samples for nuclear astrophysics: The needs, the possibilities and the sources’, Journal of Physics G: Nuclear and Particle Physics, vol. 39, no. 10, 2012, doi: 10.1088/0954-3899/39/10/105201. [17] D. Schumann et al., ‘Preparation of a 60Fe target for nuclear astrophysics experiments’, Nucl Instrum Methods Phys Res A, vol. 613, no. 3, pp. 347–350, 2010, doi: 10.1016/j.nima.2009.09.072. [18] M. Ayranov and D. Schumann, ‘Preparation of 26Al, 59Ni, 44Ti, 53Mn and 60Fe from a proton irradiated copper beam dump’, J Radioanal Nucl Chem, vol. 286, no. 3, pp. 649–654, 2010, doi: 10.1007/s10967-010-0732-0. J. Chen, B. Singh, and J. A. Cameron, ‘Nuclear Data Sheets for A = 44’, Nuclear Data [19] Sheets, vol. 112, no. 9, pp. 2357–2495, 2011, doi: 10.1016/j.nds.2011.08.005. [20] M. S. Basunia, ‘Nuclear Data Sheets for A=59’, Nuclear Data Sheets, vol. 151, pp. 1– 333, 2018, doi: 10.1016/j.nds.2018.08.001. [21] M. S. Basunia and A. M. Hurst, ‘Nuclear Data Sheets for A = 26’, Nuclear Data Sheets, vol. 134, pp. 1–148, 2016, doi: 10.1016/j.nds.2016.04.001. [22] E. Browne and J. K. Tuli, ‘Nuclear data sheets for A = 60’, Nuclear Data Sheets, vol. 114, no. 12, pp. 1849–2022, 2013, doi: 10.1016/j.nds.2013.11.002. [23] G. Rugel et al., ‘New measurement of the Fe60 half-life’, Phys Rev Lett, vol. 103, no. 7, pp. 14–17, 2009, doi: 10.1103/PhysRevLett.103.072502. [24] A. Wallner et al., ‘Settling the half-life of Fe 60: Fundamental for a versatile astrophysical chronometer’, Phys Rev Lett, vol. 114, no. 4, pp. 1–6, 2015, doi: 10.1103/PhysRevLett.114.041101. [25] D. Schumann, T. Stowasser, R. Dressler, and M. Ayranov, ‘Possibilities of preparation of exotic radionuclide samples at PSI for scientific investigations’, vol. 101, no. 8, pp. 501–508, 2013, doi: doi:10.1524/ract.2013.2058. 16 [26] S. Heinitz, D. Kiselev, N. Kivel, and D. Schumann, ‘Separation of weighable amounts of 10Be from proton irradiated graphite’, Applied Radiation and Isotopes, vol. 130, pp. 260–263, 2017, doi: https://doi.org/10.1016/j.apradiso.2017.10.012. [27] A. Visser and N. Scielzo, ‘Feasibility of Harvesting Radon from FRIB Beam Dump’, Llnl-Tr-709438, 2016. [28] M. Avilov et al., ‘Thermal, mechanical and fluid flow aspects of the high power beam dump for FRIB’, Nucl Instrum Methods Phys Res B, vol. 376, pp. 24–27, 2016, doi: 10.1016/j.nimb.2016.02.068. [29] K. A. Domnanich, E. P. Abel, H. K. Clause, C. Kalman, W. Walker, and G. W. Severin, ‘An isotope harvesting beam blocker for the National Superconducting Cyclotron Laboratory’, Nucl Instrum Methods Phys Res A, vol. 959, no. January, p. 163526, 2020, doi: 10.1016/j.nima.2020.163526. [30] E. P. Abel et al., ‘Durability test of a flowing-water target for isotope harvesting’, Nucl Instrum Methods Phys Res B, vol. 478, no. May, pp. 34–45, 2020, doi: 10.1016/j.nimb.2020.05.011. J. A. Shusterman et al., ‘Aqueous harvesting of Zr 88 at a radioactive-ion-beam facility [31] for cross-section measurements’, Phys Rev C, vol. 103, no. 2, pp. 1–12, 2021, doi: 10.1103/PhysRevC.103.024614. [32] T. Kobayashi, T. Sasaki, I. Takagi, and H. Moriyama, ‘Solubility of zirconium(IV) hydrous oxides’, J Nucl Sci Technol, vol. 44, no. 1, pp. 90–94, 2007, doi: 10.1080/18811248.2007.9711260. [33] E. A. McCutchan and A. A. Sonzogni, ‘Nuclear Data Sheets for A = 88’, Nuclear Data Sheets, vol. 115, no. 1, pp. 135–304, 2014, doi: 10.1016/j.nds.2013.12.002. [34] R. D. Hoffman, K. Kelley, F. S. Dietrich, R. Bauer, and M. Mustafa, ‘Modeled Neutron and Charged-Particle Induced Nuclear Reaction Cross Sections for Radiochemistry in the Region of Yttrium, Zirconium, Niobium, and Molybdenum’, Ucrl-Tr-222275, p. UCRL-TR-222275, 2006. [35] V. Reis, R. Hanrahan, and K. Levedahl, ‘The big science of stockpile stewardship’, AIP Conf Proc, vol. 1898, no. November 2017, 2017, doi: 10.1063/1.5009218. J. A. Shusterman et al., ‘The surprisingly large neutron capture cross-section of 88Zr’, [36] Nature, vol. 565, no. 7739, pp. 328–330, 2019, doi: 10.1038/s41586-018-0838-z. [37] A. C. Larsen et al., ‘Experimentally constrained (p,γ) y 89 and (n,γ) y 89 reaction rates relevant to p -process nucleosynthesis’, Phys Rev C, vol. 93, no. 4, pp. 1–13, 2016, doi: 10.1103/PhysRevC.93.045810. [38] G. F. Steyn et al., ‘Large-scale production of 88Y and 88Zr/88Y generators: A proof of concept study for a 70 MeV H− cyclotron’, Applied Radiation and Isotopes, vol. 168, no. June 2020, 2021, doi: 10.1016/j.apradiso.2020.109469. 17 [39] B. Singh, ‘Nuclear Data Sheets for A=172’, Nuclear Data Sheets, vol. 75, no. 2, pp. 199– 376, 1995, doi: https://doi.org/10.1006/ndsh.1995.1025. [40] V. S. Shirley, ‘Nuclear Data Sheets for A = 173’, Nuclear Data Sheets, vol. 75, no. 2, pp. 377–488, 1995, doi: https://doi.org/10.1006/ndsh.1995.1026. [41] S. Satija et al., ‘Harvesting 88Zr from heavy-ion beam irradiated tungsten at the National Superconducting Cyclotron Laboratory’, Applied Radiation and Isotopes, vol. 197, p. 110831, 2023, doi: https://doi.org/10.1016/j.apradiso.2023.110831. J. A. Bence et al., ‘Solid-phase isotope harvesting of 88Zr from a radioactive ion beam [42] facility’, Applied Radiation and Isotopes, vol. 189, no. August, p. 110414, 2022, doi: 10.1016/j.apradiso.2022.110414. ‘National Nuclear Data Center’. Accessed: Mar. 22, 2024. [Online]. Available: [43] https://www.nndc.bnl.gov/nudat3/ 18 CHAPTER 2: 88Zr HARVESTING: PROOF OF CONCEPT STUDY1 This work was the first in the series to demonstrate proof-of-concept for establishing solid-phase isotope harvesting where extraction of radioisotopes from solid tungsten is explored. Tungsten is a widely used material in numerous radioactive ion-beam facilities, where it often undergoes activation through interactions with the beam. The resulting activated products have many potential applications in both basic and applied sciences, provided they can be efficiently recovered with high extraction efficiency. To explore the radiochemistry and harvest Group (IV) elements, an experiment was conducted at NSCL, embedding a heavy-ion 88Zr beam into a stack of tungsten foils. A method was then developed to separate and recover 88Zr. This involved dissolving the tungsten foils in 30% H2O2 and chemically purifying the 88Zr from the tungsten matrix and other co-implanted radionuclides like 85Sr and 88Y. The purification process utilized strong cation-exchange chromatographic resin (AG MP-50) in H2SO4 media. The resulting 88Zr was obtained in approximately 60 mL of 0.5 M H2SO4 with no detectable radio-impurities, achieving an overall recovery yield of (92.3 ± 1.2)%. This is then compared to the recovery of (26±2)% from aqueous methods, making solid-phase harvesting a favorable approach for hydrolyzable elements. This experiment contributes to the development of methods for harvesting elements from tungsten and tungsten-alloy parts commonly irradiated in heavy-ion beam facilities like NSCL/FRIB. 2.1 Introduction The 92Mo primary beam was fragmented, with the resulting secondary beam, specifically tuned to 88Zr, implanted into various collector materials including copper, aluminum, gold, and 1 This chapter draws heavily from the published paper S. Satija et al., “Harvesting 88Zr from heavy-ion beam irradiated tungsten at the National Superconducting Cyclotron Laboratory,” Applied Radiation and Isotopes, vol. 197, p. 110831, 2023, doi: https://doi.org/10.1016/j.apradiso.2023.110831. 19 tungsten. While the studies focusing on copper, aluminum, and gold collectors have been described in Bence et al., this chapter explains the extraction and collection process of 88Zr and its daughter 88Y from the solid tungsten collector [1]. 2.2 Experimental 2.2.1 Materials Before the irradiation, the tungsten foils (99.95%, 0.05- mm thick, Alfa Aesar) and tantalum foil backing (99.95%, 0.127-mm thick, Alfa Aesar) were cleaned with ethanol (ACS Reagent, 99.5%, Sigma Aldrich). Post-irradiation chemical processing and dissolution of the tungsten foils was performed with the following reagents: sulfuric acid (ACS reagent, 95.0– 98.0%, Sigma Aldrich and OmniTrace, 93–98%, EMD Millipore), hydrogen peroxide (29–31%, Sigma Aldrich), and MilliQ water (obtained from the Thermo Scientific MicroPure Ultrapure Water System, 18.2 MΩ cm). TraceCERT® Zirconium ICP Standard (998 ± 3 ppm Zr, in 2% HNO3 and 0.2% HF, Sigma Aldrich), Yttrium ICP Standard (10,000 ppm, in 5% HNO3, GFS Chemicals), and nitric acid (ACS Plus, 15.8 M, Fisher Chemical) were used in methodology development experiments with stable elements. Column construction using cation-exchange resin - Cation-exchange resin AG MP-50 (100–200 mesh, H+ form, BioRad) was slurry packed in columns and employed for the separations. The column was constructed from rigid polycarbonate tubing (3/8′′ OD, 1/4′′ ID, McMaster-Carr, PN:9176T1) and chemical-resistant push-to-connect fittings (Straight Reducer, for 3/8′′ x 1/4′′ Tube OD). Two pieces of glass wool were inserted on each end of the column to hold the resin in place. Instruments - All the separations were performed with a peristaltic pump (Ismatec® Reglo Peristaltic Pumps with independent channel control operated with pump tubing of 1.65- 20 mm internal diameter). For the development of the separation method, the identification and quantification of stable tungsten, zirconium, and yttrium was performed with a 5900 Agilent inductively coupled plasma-optical emission spectrometer (ICP-OES) together with the Agilent ICP Expert software. An HPGe Canberra BEGe γ-ray Detector (BE2020) was used to detect and quantify the radionuclides. A second HPGe detector from PHDS Inc. (Fulcrum Compact HPGe Gamma-ray Spectrometer) was used for detection of radionuclides immediately after the irradiation in NSCL. Phosphor imaging of the tungsten foils was done with an Amersham Typhoon™ Biomolecular Imager (GE Healthcare). 2.2.2 Experimental setup 2.2.2.1 Solid-phase harvesting apparatus and irradiation of the tungsten target The solid-phase collection setup comprised a series of tungsten foils arranged in a stack, depicted in Figure 2.1. The outer body of this solid-phase collection setup was made from an aluminum-based KF-40 flange, as shown in Figures 2.2 (a) and 2.2 (b). This KF-40 flange underwent machining to accommodate metal foils within a cavity which was secured by an outer ring and screws as shown in Figures 2.2 (a) and 2.2 (b). Ten tungsten foils, each measuring 0.05 mm in thickness and weighing between 540 and 570 mg, along with a 0.127-mm tantalum backing foil, were utilized, resulting in a combined foil stack thickness of 0.627 mm. Before assembling together, the tungsten foils and tantalum backing were wiped with ethanol, then stacked sequentially and securely clamped into the collector assembly between the ring clamp and the KF-40 flange. The foils were labeled from 1 to 10, with the lowest numbered foil positioned closest to the ring clamp. The tantalum backing, situated furthest downstream, served to stop any beam that was not stopped by the foils. This collector assembly was subsequently plugged at the end of the beamline, as depicted in Figure 2.2 (c). 21 At the NSCL, 88Zr39+ secondary beam was generated through the fragmentation of a 140- MeV/u primary beam of 92Mo, directed at a beryllium production target with a thickness of 446 mg/cm2 [2]. A mostly unpurified 88Zr39+ secondary beam was requested to showcase the collection of a diverse array of beam products. Throughout the 8-hour irradiation period targeting the tungsten material, the beam current was monitored continuously. This was accomplished by inserting a Faraday cup positioned immediately upstream of the endstation, resulting in an integrated beam current of 3.90 × 10-8 eA on the tungsten target. Figure 2.1: A schematic of the solid-phase harvesting apparatus used for the irradiation experiment at the NSCL is shown. A thin foil stack of 10 tungsten foils, each 0.05-mm thick, and a 0.127-mm thick tantalum backing foil is housed in the solid collector made of aluminum. Foil 7 and 8 marked here are the most radioactive ones [22]. 22 Figure 2.2: The disassembled aluminum solid collector apparatus with the ring and screws is shown in (a) and the assembled aluminum solid collector set up containing the clamped tungsten foils and the tantalum backing held by ring and screws is shown in (b). This assembled solid collector body plugged at the end of the S1/S2 user beamline at NSCL is shown in (c) [22]. 2.2.2.2 Quantification of radionuclides and comparison with estimated production Before the irradiation, the LISE++ simulation software was employed to model nuclear reactions at the NSCL and optimize beam parameters with the A1900 fragment separator. This software estimated the radionuclide production rates when fragments interact and get stopped with a solid tungsten matrix. This is illustrated in Figure 2.3. These estimated production rates were then used to predict radionuclide activities [3]. Furthermore, LISE++ enabled the determination of range distributions for implanted radionuclides within the tungsten target, which is shown in Figure 2.3, where 100% of fragmentation products were expected to stop within the first 440 µm of tungsten. Through this analysis, the main radionuclides identified were 88Zr, 88Y, and 85Sr isotopes. 23 Figure 2.3: This range distribution generated from LISE++ for the tungsten catcher illustrates that 100 % of the beam product mixture is expected to stop within the first 0.44 mm of tungsten. Image provided by Dr. Thomas Baumann. The experimental activities of 88Zr and 88Y were determined through γ-ray spectroscopy. At the end of the bombardment (EOB), the solid collector was removed from the NSCL S1/S2 beamline and positioned 50 cm in front of the PHDS HPGe detector, shown in Figure 2.4, where it was counted as a complete assembly over varying durations, ranging from approximately 19 minutes to 2.8 hours. To maintain consistent positioning, a ring-stand was employed to elevate the sample as shown in Figure 2.4, ensuring the holder remained centered in front of the detector. Immediate counting post-bombardment allowed for the identification of short-lived radionuclides. 24 Figure 2.4: The solid collector housing irradiated tungsten foils counted as a whole in front of the PHDS HPGe detector outside the S1/S2 vault in NSCL to assess the short-lived species. Following transportation to the MSU Chemistry laboratory, the solid collector underwent disassembly after being counted on the HPGe γ-ray detector as a whole. Each foil was then individually counted to evaluate the radionuclide inventory using the HPGe γ-ray detector. All the calibrations of energy and efficiency for fractions obtained from resin-based separations described in section 2.2.4.2 was conducted using a 152Eu source (1 μCi, Eckert & Ziegler Analytics), evenly distributed in 10 mL of epoxy within a sealed 15 mL falcon tube positioned at distances of 15 cm and 25 cm from the detector face. To ensure uniform geometry with the calibration setup made for the HPGe detector, all resulting fractions from separations in this study were collected and measured in 15 mL falcon tubes, filled up to 10 mL, and placed in a fixed acrylic holder at the specified distances of 15 and 25 cm. The analysis of energy spectra was performed using the Genie 2000 software (Mirion Technologies), which identified the characteristic γ-ray emissions of the radionuclides, conducted the background subtraction, and computed the activities based on the net photopeak areas. Throughout all the measurements, the average dead time remained below 3%. Additionally, other long-lived, co-implanted radionuclides, such as 85Sr, were identified and quantified when observable above the limit of detection. The details of the characteristic γ-ray 25 emissions and intensities can be found in Table 2.1. A comparison was then drawn between the determined activities from gamma spectroscopic measurements and the numbers obtained from the LISE++ models previously mentioned. 2.2.3 Post-irradiation techniques and chemical processing of irradiated tungsten foils Autoradiography was carried out for the examination of the distribution profile of the radioactive species within the tungsten foils. The assembled tungsten collector along with individual active foils were exposed to a film measuring 20 × 40 cm2 (BAS-IP SR, 2040; GE Healthcare) for durations ranging from 10 to 60 seconds. The Image acquisition was performed using the Amersham Typhoon™ software (version 1.1.0.7, GE Healthcare), while ImageQuant TL (version 8.2.0.0, GE Healthcare) was used for image analysis. The autoradiograph of the Figure 2.5: Autoradiograph generated from the individual tungsten foils numbered 1 through 10 and the tantalum backing where Foil 7 and 8 are the most radioactive. 26 individual foils numbered 1 through 10 and the tantalum backing foil is shown in Figure 2.5. By utilizing this technique, the most radioactive foils i.e., Foil 7 and 8 were identified, and were required to be chemically processed. Table 2.1: This shows the short- and the long-lived radionuclides present in the secondary beam and their associated gamma-ray emissions used for identification [4]. Radionuclide Half-life Gamma-ray Energy [keV] (Intensity) 81Sr 82Sr 83Sr 85Sr 83Rb 84Rb 83Y 84Y 85Y 86mY 86Y 87mY 87Y 88Y 86Zr 87Zr 88Zr 89Zr 88Nb 89Nb 90Nb 91Nb 90Mo 93Mo 22.3 min 25.34 days 32.41 hours 64.85 days 86.2 days 32.82 days 7.08 min 39.5 min 2.68 hours 47.4 min 14.74 hours 13.37 hours 79.8 hours 106.63 d 16.5 hours 1.68 hours 83.4 d 78.41 h 14.55 min 2.03 hours 14.6 h 680 years 5.56 hours 4000 years 147.76 (30), 153.54 (34), 188.27 (15.4), 443.34 (17.5) None 762.65 (26.7) 514.0048 (96) 520.3991 (45), 529.5945 (29.3), 552.5512 (16) 881.6041 (68.9) 489.9 (5.4), 882.1 (6.2) 793.1 (98.3), 974.3 (78), 1039.7 (56) 231.65 (84), 504.44 (60) None 627.72 (32.6), 777.37 (22.4), 1076.63 (82.5), 1153.05 (30.5), 1920.72 (20.8) None 388.5276 (82.2), 484.805 (89.8) 898.042 (93.7), 1836.063 (99.2) 242.8 (95.84) 1227 (2.8) 392.87 (97.3) 909.15 (99.04) 271.8 (30.1), 399.4 (31.8), 502.9 (60), 671.2 (64), 1057.1 (100) 1627.2 (3.5) 141.178 (66.8), 1129.224 (92.7), 2318.959 (82.0) None 257.34 (77) None 27 2.2.4 Separation chemistry development 2.2.4.1 Separation chemistry development with stable elements Establishing chemistry techniques with stable tracers is an important prerequisite before attempting radiochemical separations. In order to establish a procedure for separating 88Zr and 88Y from the tungsten matrix, solutions of stable Zr and Y in a H2SO4 medium were prepared by using the ICP-OES standards. Subsequently, these solutions were separated from 18-500 mg of stable dissolved tungsten foil in 30% H2O2. The separation process involved employing the cation exchange chromatographic resins, specifically DOWEX 50W-X8 and AG MP-50. The separation was carried out using a peristaltic pump, following the calibration of the flow rate, and, fractions of 10-14 mL were collected. To assess the efficacy of the separation, aliquots were withdrawn from each fraction and appropriate dilutions in 0.6% tartaric acid + 2% HNO3 were made for subsequent analysis via ICP-OES. Calibration curves for sample analysis were established by diluting standards to 0.01 - 50 ppm. The separation efficiency was quantified by expressing the total eluted amount in all fractions as a percentage of the loaded amount for each element. 2.2.4.2 Separation with active foils 7 and 8 Post autoradiography, a waiting period of 10 weeks was observed to allow for the decay of shorter-lived isotopes such as 90Nb (t1/2 = 14.6 hours), 83Sr (t1/2 = 32.4 hours), 85Y (t1/2 = 2.7 hours), 86Y (t1/2 = 14.7 hours), 87Y (t1/2 = 79.8 hours), and 89Zr (t1/2 = 78.4 hours) [4]. A schematic is shown in Figure 2.6 which is a snippet of the chart of the nuclides of the radionuclides embedded. Before chemically processing, approximately 1 mm of the inactive circular edges of both foil 7 and foil 8 were cut to reduce the excess tungsten mass. 28 A total of three separations were carried out, one with foil 7 and two with foil 8 and the results were compared. The active tungsten foils 7 and 8 were dissolved in 8.3 mL and 8.5 mL of 30% hydrogen peroxide, respectively. To ensure complete dissolution, these solutions were heated at approximately 60°C for a duration of 8 hours. Upon dissolution, volumes of 1.9 mL and 2.0 mL were withdrawn from the solutions corresponding to a mass of approximately 100 mg tungsten each. These aliquots were subsequently adjusted to achieve a concentration of ≤0.2 M H2O2 and 0.1 M H2SO4 through the addition of dilute sulfuric acid. The resulting loading solutions, with volumes of 93 mL and 98 mL, respectively, were used for the subsequent radiochemical separations. Figure 2.6: Schematic of the radionuclides generated when the secondary 88Zr beam is embedded in the tungsten collector. Three columns slurry packed with AG MP-50 resin (9 mm outer diameter, 6 mm inner diameter, and 15–18 cm in length) were prepared for use. Rinsing was done sequentially with 30 mL of MilliQ water, 24 mL of 2 M H2SO4, and finally conditioned with 24 mL of a mixture comprising 0.2 M H2O2 + 0.1 M H2SO4. Subsequently, the prepared loading solutions were passed over these pre-conditioned columns. The majority of tungsten present in the loading 29 solution passed through the columns and was collected in 10 mL fractions. To ensure a thorough removal of any remaining tungsten, a rinse step was performed using 28–30 mL of 0.1 M H2SO4. In the next step, 88Zr was eluted by passing 60–70 mL of 0.5 M H2SO4 through the columns. This was followed by the elution of 88Y, 85Sr, and any residual 88Zr using 30 mL of 6 M H2SO4. The flow rate was maintained at a constant 0.8 mL/min throughout the process of conditioning and separation. 2.2.5. Stable element analysis The load and rinse fractions obtained from the radiochemical separations were analyzed using ICP-OES to determine the concentration of tungsten. Each fraction was used for preparing two different dilutions of the same concentration, by extracting an aliquot and diluting in a solution containing 0.6% tartaric acid + 2% HNO3. The quantification and analysis were performed using Agilent ICP software. For the analysis, the emission lines with the highest intensity for tungsten (207.912, 209.475, 220.449, 224.876, 203.000 nm), zirconium (343.823, 327.307, 349.619, 327.927, 339.198 nm), and yttrium (371.029, 360.074, 377.433, 361.104, 378.869 nm) were selected. The ICP-OES analysis was conducted under conditions of axial plasma view, with a plasma gas flow rate of 12 L/min, auxiliary gas flow of 1.0 L/min, nebulizer gas flow of 0.7 L/min, and make-up flow of 0.00 L/min. 2.3 Results and discussion 2.3.1. Radionuclide identification and quantification As observed from Figure 2.3, The LISE++ simulation indicated that the majority of beam particles were expected to penetrate and stop in a depth ranging from 0.34 to 0.44 mm within the tungsten foil [3]. This is supported by the observation that foil 7 (0.30–0.35 mm) and 30 foil 8 (0.35–0.40 mm) showed the highest levels of activation following irradiation, as confirmed by subsequent post-irradiation γ-spectroscopy (Figure 2.8) and autoradiography measurements (Figure 2.5). The beam current measurements showed that, on average, (1.59 ± 0.01) × 107 particles per second per particle nano-Ampere of 92Mo primary beam (pps/pnA) of 88Zr39+ impinged on the tungsten foils. Further details regarding beam tuning can be found in Bence et al. [1]. Consequently, a total of 4.42 × 1011 atoms of 88Zr were accumulated on the tungsten foils by the end of bombardment, which spanned a duration of 7.70 hours. Through gamma spectroscopic measurements post irradiation, no activity was detected on the collector body, tantalum backing, or the screws, while the ring exhibited minimal activity, indicating the beam was well centered on the target. The evaluation of all the foils started 92 hours post-irradiation, and an HPGe spectrum of the non-active foil 3 and active foil 7 is depicted in Figure 2.7 and Figure 2.8, respectively. The radionuclides 88Y and 85Sr, as the radioactive daughters of 88Zr and 85Y, respectively, were detected in foil 8, given the short half-life of 85Y (t1/2 = 2.68 hours) at the time of measurement [5]. By the time of radiochemical separations, conducted 10 weeks after EOB, the short-lived 87Zr (t1/2 = 1.68 hours) and 89Zr (t1/2 = 78.4 hours) had already decayed, leaving 88Zr as the sole radioactive zirconium isotope [6], [7]. 31 Figure 2.7: Gamma spectrum for the non-active foil 3 indicating the absence of 88Zr and peaks for other radionuclides. Figure 2.8: Gamma spectrum of the active foil 7. HPGe spectrum of foil 7 recorded 92 h after EOB. The co-implanted radionuclides include 88Zr, 89Zr, 86Y, 87Y, and 90Nb. The remaining peaks are those from the background [22]. Based on gamma spectroscopic measurements, the cumulative activity of 88Zr measured in tungsten foils 7 and 8 (decay corrected to EOB) was determined to be 42.55 ± 0.37 kBq, 32 aligning within 1σ agreement with the theoretically predicted yield of 42.51 kBq for 88Zr. 2.3.2. Post-irradiation techniques and chemical processing of irradiated foils 2.3.2.1 Separation chemistry Tungsten foils are typically dissolved using either hydrofluoric acid or concentrated peroxide, as noted in prior studies [8], [9], [10], [11]. In this study, hydrogen peroxide was selected over hydrofluoric acid to prevent fluoride ion-related effects and the creation of zirconium fluoride complexes. The irradiated tungsten completely dissolved in the hydrogen peroxide solution without any visible precipitation. Additionally, previous literature describes several methods utilizing acidic hydrogen peroxide solutions for cation exchange-based separations to elute tungsten as an anionic peroxy complex, thereby preventing W(VI) hydrolysis [8], [12], [13]. Hence, this separation is based on the inability of the prevalent anionic peroxy complex of tungsten to retain on cation exchange resin under dilute sulfuric acid conditions. Sulfuric acid was selected due to the enhanced stability of W(VI) in H2SO4 compared to HCl or HNO3. Given that the foils underwent heating in 30% H2O2 for approximately 8 hours, peroxide decomposition and consumption were inevitable, thereby establishing 0.2 M H2O2 as the ‘upper limit’ for achievable peroxide concentration. Employing the previously outlined elution conditions, tungsten was quantitatively eluted with high radiopurity. Table 2.2: Kd values in H2SO4 with a sulfonated polystyrene resin for Zr(IV) and Y(III) [14]. 0.05 M H2SO4 546 4 >10 Zr (IV) Y(III) 0.1 M 0.25 M 0.5 M 1.0 M 1.5 M 474 4 >10 98 1380 4.6 253 1.4 49.9 1.2 18.0 At low acidity, such as 0.1 M H2SO4, the cationic Zr(IV)SO4 2+complex is adsorbed to cationic resin, as shown from distribution coefficient (Kd) values in Table 2.2. [15]. The elution 33 of Zr becomes possible when the molarity is increased to 0.5 M H2SO4, which promotes the formation of anionic Zr species Zr(SO4)3 2−and Zr(SO4)4 4−with poor affinity towards the cationic resin [15]. Additionally, with 0.5 M H2SO4, the distribution coefficients of Y(III) and Sr(II) are still high compared to Zr(IV), thus allowing the elution of pure zirconium [14]. The gamma spectrum analysis of the 88Zr fraction, conducted immediately after the separation of foil 8 (shown in Figure 2.10), showed no gamma-ray peaks attributable to 88Y or 85Sr. The retrieval of 88Zr amounted to (94.2 ± 1.4)% of the loading solution for foil 7 and (91.3 ± 1.1)% for foil 8, across a volume range of 55–58 mL, without any detectable 88Y or 85Sr peaks. This represents a significant increase in 88Zr recovery compared to the aqueous harvesting results of (26 ± 2)% discussed earlier [16]. In subsequent fractions eluted using 6 M H2SO4, the recovery of 88Y averaged (95.8 ± 5.7)% of the loaded solution for both foil 7 and foil 8, within a volume range of 33–37 mL, accompanied by the co-elution of the remaining 88Zr and the entirety of 85Sr. It's worth noting that while the yttrium fraction wasn't radiopure, there exists several methods for separating Y from Sr and Zr [17], [18], [19]. One of the methods that was tried but wasn’t successful is detailed below. The separation process utilizing AG MP-50 described in this study is illustrated in Figure 2.9, and Figure 2.11 serves as a representative elution profile for the separation involving foil 7. To further purify the obtained 88Y fractions from the first separation step, the 85Sr-, 88Zr- and 88Y-containing mixture in 6 M H2SO4 was complexed with Oxalic acid and adjusted to yield a final concentration of 0.3 M Oxalic acid + 3 M H2SO4. This solution was loaded on the DGA- resin containing column. In this case, an average of (107.0 ± 5.6)% 88Zr eluted for Foil 7 and Foil 8 separation, and an average of (86.6 ± 5.5)% 85Sr eluted for Foil 8 separation for the load 34 solution and the 0.3 M Oxalic acid + 3 M H2SO4 rinse leaving only 88Y adsorbed to the DGA column. However, only (32.6 ± 3.7)% 88Y could be recovered with 0.01 M H2SO4 for Foil 7, while no elution was observed in the case of Foil 8 separations. This irreversible adsorption was observed despite attempts of increasing the H2SO4 concentration and thereafter changing the medium to HCl. This adsorption behavior of 88Y onto the DGA is in contrast to the cold separations, where >80 - 90 % recovery of Y was observed with 0.01 M H2SO4 and the largest losses in the entire process were due to the irreversible adsorption of 88Y to the DGA column. One of the reasons could be the adsorption of 88Y to the resin backbone [20]. An average quantitative recovery of (73.0 ± 5.9)% 88Y for Foil 7 and 8 was only possible with the addition of pure ethanol, which also eluted the TODGA functional groups from the solid support material. A Figure 2.9: A schematic of the separation scheme (from left to right) employing the use of AG MP-50 and DGA resin for the recovery of 88Zr and 88Y. 35 possible suggested extraction of the 88Y from the eluted TODGA-ethanol mixture could be done by the addition of dodecane, followed by a back-extraction with 0.1 M HNO3 [20], [21]. Figure 2.10: Gamma spectrum for the purified 88Zr fraction, directly after elution showing peaks from 88Zr with no accompanying 88Y peaks, overlaid with the background spectrum. Figure 2.11: The elution profile obtained for tungsten foil 7 separation using AG MP-50 resin for the extraction of 88Zr [22]. 36 2.4 Conclusion This first proof-of-concept study aimed to establish and evaluate radiochemical techniques for separating Zr from a tungsten matrix using radioactive samples. The implanted 88Zr was successfully isolated from the tungsten matrix utilizing cation-exchange resin. The methodologies developed in this work yielded a recovery rate of (92.3 ± 1.2)% for 88Zr with high radionuclidic purity, establishing that solid-phase harvesting is a better approach over aqueous- phase methods, which only achieved a recovery rate of (26 ± 2)% of the 88Zr content [16]. Similar conclusions were drawn from the study of Al, Cu, and Au collectors for solid-phase harvesting [1]. Given the analogous chemistry of Hf to Zr and Lu to Y, it is expected that the solid-phase harvesting techniques developed in this study could be applied to recover Hf and Lu from tungsten-alloy components at NSCL, one of which is the heavy-met alloy beam blocker discussed in the next chapter. This will provide a viable route to obtaining 172Hf, which can be challenging to produce in high purity through alternative means. Additionally, the solid-phase harvesting system constructed as part of this endeavor can be used for future solid-phase isotope harvesting endeavors at FRIB. 37 REFERENCES J. A. Bence et al., “Solid-phase isotope harvesting of 88Zr from a radioactive ion beam [1] facility,” Applied Radiation and Isotopes, vol. 189, no. August, p. 110414, 2022, doi: 10.1016/j.apradiso.2022.110414. D. J. Morrissey, B. M. Sherrill, M. Steiner, A. Stolz, and I. Wiedenhoever, [2] “Commissioning the A1900 projectile fragment separator,” Nucl Instrum Methods Phys Res B, vol. 204, pp. 90–96, 2003, doi: 10.1016/S0168-583X(02)01895-5. O. B. Tarasov and D. Bazin, “LISE++: Radioactive beam production with in-flight [3] separators,” Nucl Instrum Methods Phys Res B, vol. 266, no. 19–20, pp. 4657–4664, 2008, doi: 10.1016/j.nimb.2008.05.110. “National Nuclear Data Center.” Accessed: Mar. 22, 2024. [Online]. Available: [4] https://www.nndc.bnl.gov/nudat3/ B. Singh and J. Chen, “Nuclear data sheets for A = 85,” Nuclear Data Sheets, vol. 116, [5] no. 1, pp. 1–162, 2014, doi: 10.1016/j.nds.2014.01.001. T. D. Johnson and W. D. Kulp, “Nuclear Data Sheets for A = 87,” Nuclear Data Sheets, [6] vol. 129, pp. 1–190, 2015, doi: 10.1016/j.nds.2015.09.001. B. Singh, “Nuclear Data Sheets for A = 89,” Nuclear Data Sheets, vol. 114, no. 1, pp. 1– [7] 208, 2013, doi: 10.1016/j.nds.2013.01.001. [8] J. S. Fritz and J. J. Topping, “Chromatographic separation of vanadium, tungsten and molybdenum with a liquid anion-exchanger,” Talanta, vol. 18, no. 9, pp. 865–872, 1971, doi: 10.1016/0039-9140(71)80143-1. [9] S. U. Ganiev, T. Artykbae, and G. A. Tsyganov, “Kinetics and Dissolution Products of Molybdenum and Tungsten in Hydrogen-Peroxide,” Zhurnal Neorganicheskoi Khimii, vol. 18, no. 3, pp. 709–711, 1973. [10] K. Smith and J. R. Dean, “Investigation into the Spontaneous Dissolution of Tungsten Using Acid Media,” vol. 118, no. November, pp. 1445–1447, 1993, doi: 10.1039/AN9931801445. [11] K. N. Kmak, J. D. Despotopulos, and N. D. Scielzo, “Development of Chemical Procedures for Isotope Harvesting: Separation of Trace Hafnium from Tungsten,” Solvent Extraction and Ion Exchange, vol. 40, no. 7, pp. 718–734, 2022, doi: 10.1080/07366299.2022.2079502. J. S. Fritz and L. H. Dahmer, “Cation Exchange Separation of Molybdenum, Tungsten, [12] Niobium, and Tantalum from Other Metal Ions,” Anal Chem, vol. 37, no. 10, pp. 1272–1274, 1965, doi: 10.1021/ac60229a026. [13] L. H. Dahmer, “Chromatographic separations of niobium , tantalum , molybdenum , and tungsten,” 1966. 38 [14] F. W. E. Strelow, R. Rethemeyer, and C. J. C. Bothma, “Ion Exchange Selectivity Scales for Cations in Nitric Acid and Sulfuric Acid Media with a Sulfonated Polystyrene Resin,” Anal Chem, vol. 37, no. 1, pp. 106–111, 1965, doi: 10.1021/ac60220a027. J. Korkisch, Handbook of Ion Exchange Resins: Their Application to Inorganic [15] Analytical Chemistry Volume VI. Boca Raton / London / Tokyo: CRC Press, 2017. doi: 10.1201/9781315140520. J. A. Shusterman et al., “Aqueous harvesting of Zr 88 at a radioactive-ion-beam facility [16] for cross-section measurements,” Phys Rev C, vol. 103, no. 2, pp. 1–12, 2021, doi: 10.1103/PhysRevC.103.024614. [17] S. A. Kandil, B. Scholten, Z. A. Saleh, A. M. Youssef, S. M. Qaim, and H. H. Coenen, “A comparative study on the separation of radiozirconium via ion-exchange and solvent extraction techniques, with particular reference to the production of 88Zr and 89Zr in proton induced reactions on yttrium,” J Radioanal Nucl Chem, vol. 274, no. 1, pp. 45–52, Oct. 2007, doi: 10.1007/s10967-006-6892-2. [18] M. J. O’Hara, N. J. Murray, J. C. Carter, and S. S. Morrison, “Optimized anion exchange column isolation of zirconium-89 (89Zr) from yttrium cyclotron target: Method development and implementation on an automated fluidic platform,” J Chromatogr A, vol. 1545, pp. 48–58, 2018, doi: 10.1016/j.chroma.2018.02.053. [19] S. Lahiri, B. Mukhopadhyay, and N. R. Das, “Simultaneous production of 89Zr and 90,91m,92mNb in α- particle activated yttrium and their subsequent separation by HDEHP,” Applied Radiation and Isotopes, vol. 48, no. 7, pp. 883–886, 1997, doi: 10.1016/S0969- 8043(96)00338-7. [20] Y. Sasaki, Y. Sugo, Y. Kitatsuji, A. Kirishima, T. Kimura, and G. R. Choppin, “Complexation and back extraction of various metals by water-soluble diglycolamide,” Analytical Sciences, vol. 23, no. 6, pp. 727–731, 2007, doi: 10.2116/analsci.23.727. [21] Y. Sasaki, Z. X. Zhu, Y. Sugo, and T. Kimura, “Extraction of various metal ions from nitric acid to n-dodecanen by diglycolamide (DGA) compounds,” J Nucl Sci Technol, vol. 44, no. 3, pp. 405–409, 2007, doi: 10.1080/18811248.2007.9711301. [22] S. Satija et al., “Harvesting 88Zr from heavy-ion beam irradiated tungsten at the National Superconducting Cyclotron Laboratory,” Applied Radiation and Isotopes, vol. 197, p. 110831, 2023, doi: https://doi.org/10.1016/j.apradiso.2023.110831. 39 CHAPTER 3: 172Hf HARVESTING FROM TUNGSTEN ‘HEAVY-MET’ ALLOY BEAM- BLOCKER Despite their relevance for many applications, the radioisotope 172Hf and its daughter 172Lu are not readily available. One way to obtain this is from the decommissioned beam-blocker (described briefly in the Introduction chapter) obtained from NSCL. Generally, amongst all the solid components of an accelerator, a beam dump will accumulate a wide variety of long-lived isotopes and can be potentially challenging to chemically separate as will be seen in this Chapter. 3.1 Introduction Figure 3.1: A schematic depicting the location of the tungsten ‘heavy-met’ alloy beam- blocker at the decommissioned facility, NSCL, is shown. It was located after the first large dipole magnet of the A1900 separator and was used to intercept the unused primary beam. Here, the beam-blocker is still bound to its fixture, and the beam-spot is also visible on the front side [9]. 40 The beam-blocker had been irradiated over the course of about twenty years and now has substantial quantities of useful radionuclides such as 7Be, 22Na, 42Ar, 46Sc, 54Mn, 56,57,60Co, 68Ga, 68Ge, 88Y, 88Zr, 101,102,102mRh, 110mAg, 109Cd, 125Sb, 133Ba, 143,144Pm, 145Sm, 148,151Gd, 172,173Lu, 172,175Hf, 182Ta, 184,184mRe, and 181W embedded in it. This list is not exhaustive and reasons are provided in section 3.3. This beam-blocker was later replaced with a water-filled beam blocker for harvesting byproduct radioisotopes in some experiments and its position before being decommissioned is shown in Figure 3.1. Of particular interest in the beam-blocker is 172Hf (t1/2 = 1.87 y, Еγ = 125.8 keV (Iγ = 11.3%), 23.9 keV (Iγ = 20.3%)) which undergoes electron capture and generates 172Lu (t1/2 = 6.7 d, Еγ = 1093.6 keV (Iγ = 63.0%), 181.5 keV (Iγ = 20.6%), 900.7 keV (Iγ = 29.8%)) [1]. The half- life of the daughter 172Lu here is also adequate, lasting weeks of investigation making it suitable for a variety of purposes. Apart from being a radiotracer which could be used for compound labelling and biodistribution studies, 172Lu also has suitable properties for being a PAC probe due to the 91-1094 keV -cascade in the 172Yb daughter [2], [3]. This PAC probe finds application in wide Band-Gap Semiconductors [4]. It is also acknowledged that there is a major limitation in availability and production of radioisotopes with appropriate properties for PAC which makes purifying of 172Lu a worthwhile endeavor [5], [6]. Efforts are also underway to harvest the radiolanthanides and entrapped radioactive gases from the beam-blocker. In this chapter, a general chemical procedure for recovering 172Hf from the NSCL/FRIB beam-blocker is described. A significant amount (approximately 500 g) of tungsten alloy was dissolved in hydrogen peroxide in batches (notably without the use of hydrofluoric acid), followed by precipitation-based debulking of tungsten and then purification on LN resin to isolate hafnium. In order to obtain the pure 172Lu samples necessary for applications, a 41 172Hf/172Lu generator system was developed. The generated 172Lu was used for PAC measurements, and after several milkings, the generator as remade on another LN resin. Decommissioning of the components such as this from NSCL has made it possible to harvest isotopes and opens up numerous possibilities for research and applications. The methods developed here will also have applicability to any future efforts to extract byproducts from the tungsten targets at other facilities (described in detail in Chapter 6). 3.2 Experimental 3.2.1 Materials 3.2.1.1 Reagents Before processing the beam-blocker, a tungsten foil (99.95%, 0.05- mm thick, Alfa Aesar), and a scrap piece of heavy-met alloy obtained from NSCL, were used for methodology development. The dissolution and chemical processing tests for the beam-blocker was performed with the following reagents: hydrogen peroxide (29-31%, Sigma Aldrich), hydrochloric acid (HCl, Optima Grade, 32.0-35.0%, Fisher Chemical and 99.99% Trace metals basis, Sigma-Aldrich), nitric acid (HNO3, 65%, Suprapur, Sigma-Aldrich), oxalic acid (C2H2O4, ACS Reagent, J.T. Baker), sulfuric acid (H2SO4, ACS reagent, 95.0–98.0%, Sigma Aldrich and OmniTrace, 93– 98%, EMD Millipore), acetonitrile (MeCN, Sigma-Aldrich, ACS Reagent >= 99.5%), tartaric acid (C4H6O6, Sigma Aldrich, ACS Reagent >= 99.5%), sodium hydroxide (NaOH, Sigma Aldrich, ACS reagent >= 97%), and MilliQ water (obtained from the Thermo Scientific MicroPure Ultrapure Water System, 18.2 MΩ cm). Transition metal mix 1 comprising of V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Ag and Cd (2%, w/w HNO3, 100 mg/L each, Sigma-Aldrich), transition metal mix 2 comprising of Ti, Zr, Hf, Nb, Ta, Mo, W and Re (2%, w/w HNO3, 1% w/w HF, 100 42 mg/L each, Sigma-Aldrich), Lu ICP standard (1000 μg/mL, in 2% HNO3, Assurance), Y ICP standard (10,000 ppm, in 5% HNO3, GFS Chemicals), Na ICP standard (1000 μg/mL, in 2% HNO3, Assurance), Rh ICP standard (1000 μg/mL, in 2% HNO3, Assurance), Be ICP Standard (1000 ppm, <5% HNO3, GFS Chemicals) were used in methodology development experiments with stable elements. 3.2.1.2 Column construction with ion-exchange and extraction chromatography resins The experiments were conducted using the following resins: LN-extraction chromatographic resin (2 mL Cartridges, particle size 50-100 μm, Eichrom LN-R50-S), QMA strong anion exchanger resin (Sep-Pak Accell Plus QMA Plus Light Cartridge, Part no. WAT023525, Lot no. 034639256A). 3.2.1.3 Instruments The drill bits used for drilling the tungsten beam-blocker were obtained from McMaster- Carr (Part number 8871A28). The separations were performed with a peristaltic pump (Ismatec® Reglo Peristaltic Pump with independent channel control) operated with pump tubing of 1.65- mm internal diameter, and with a syringe pump (New Era Pump Systems, NE-1000 SyringeONE one channel programmable syringe pump). During the separation method development, the identification and quantification of stable elements was performed with a 5900 Agilent ICP-OES together with the Agilent ICP Expert software. The radionuclides were detected and quantified with the HPGe Canberra Broad Energy BEGe γ-ray Detector (BE2020). Phosphor imaging of the tungsten foils was conducted with an Amersham Typhoon™ Biomolecular Imager (GE Healthcare). 43 3.2.2 Experimental setup Figure 3.2: (a) The drilled holes in the most radioactive face of the tungsten beam-blocker using a carbide tipped steel bit and drill press is shown (b) Tungsten shavings resulting from the drilling process is shown. Once NSCL was decommissioned, the beam-blocker was transferred to chemistry laboratory, and autoradiography was performed to identify the most radioactive face. The most radioactive face was then drilled out using a drill press with carbide tipped steel bits. The drill press was located inside of a fume hood with filtered ventilation, and enclosed in a large plastic bag during machining (to prevent dust migration). Drilling was done on the beam-spot and adjacent areas in batches and ~500 g tungsten shavings were collected for the extraction of radioisotopes. Figure 3.2 (a) shows the drilled beam-blocker where the drilling was done in batches. Figure 3.2 (b) shows the tungsten shavings collected after the drilling process. 3.2.3 Experimental quantification of radioisotopes through  spectroscopy Initially, the beam-blocker was measured as a whole using the HPGe Detector after using autoradiography to identify the most radioactive face and the spectra analysis was done using the 44 Canberra Genie 2000 software (Mirion Technologies). All energy spectra were analyzed by identifying the characteristic -ray emissions of the radionuclides embedded in the beam-blocker, performing background subtraction, and calculating activities based on the net photopeak areas. A list of radionuclides detected along with their half-lives and characteristic gamma lines are given in section 3.3. The average dead time was less than 5% for all the measurements. The energy and efficiency calibrations were performed with a 152Eu source (1 μCi, Eckert & Ziegler) with the activity evenly distributed in 10 mL of epoxy in a sealed 15 mL falcon tube at 15, 25, 50, 100, 150 and 180 cm from the face of the HPGe detector. To maintain a consistent geometry with the calibration, all resulting fractions were positioned in a fixed acrylic holder at the above- mentioned distances from the HPGe detector and collected and measured in 15 mL falcon tubes. Counting of material that did not fit the geometry specifications is described in further sections. 3.2.4 Separation chemistry development: Dissolution of the tungsten beam-blocker and radiochemical separation with resins 3.2.4.1 Dissolution of the tungsten beam-blocker Initial dissolution testing was done with a piece of scrap heavy-met alloy obtained from NSCL. That same alloy was used to fabricate the beam blocker. The dissolution of the beam- blocker shavings was attempted in batches, and the procedure described here is representative of all dissolutions attempts. Approximately 0.25 g of the shavings were weighed into a 100 mL beaker and 5.0 mL of 10% H2O2 (v/v) was added. The reaction began in about a minute and the reaction vessel was placed into an ice-bath to curtail the vigorous reaction and bubbling. A total of 35 mL of 30% H2O2 was then added to the beaker 2.5 mL at the time. At this point, the addition of more peroxide did not result in vigorous bubbling; however, some grey dust size particles were still observed at the bottom of the beaker. Using peroxide strips, the concentration 45 of peroxide in the solution was determined and the beaker was left on heating at 65 °C, and then left overnight at room temperature. H2O2 concentration was once again determined using peroxide strips. The solution was again heated at 70 °C and was left for several hours to decrease the peroxide concentration. The dissolved solution is shown in Figure 3.3 (a). At this stage, an aliquot was withdrawn from this dissolved solution and radionuclides were identified and quantified by gamma spectroscopy. The procedure was repeated with varying amount of shavings. While most of the shavings dissolved in the first attempt, some undissolved dust (shown in Figure 3.3 (b)) was left behind. Dissolution was attempted in a controlled manner similarly in an ice-bath with approximately 290 mg of this dust by first adding 5 mL of 10% H2O2 followed by a total of 16 mL of 30% H2O2 (1 mL at the time with 10-15 min breaks in between additions). At this point, the beaker was removed from the ice bath and placed on the hot plate at 60 °C for 1.5 hours. Another 24 mL 30% H2O2 (2 mL at a time with breaks in between addition) was added and left overnight. Most of the dust was dissolved and addition of another 9 mL 30% H2O2 the following day and heating up the solution to 50 °C (2 mL at a time with breaks in between addition) dissolved the dust entirely. The peroxide test was also performed using strips and temperature was increased to allow the peroxide concentration to reduce. 46 Figure 3.3: (a) The beam-blocker dissolved in 30% H2O2 is shown (b) the left-over tungsten beam-blocker shavings after the first round of dissolution is shown and it was dissolved separately (c) the tungstic acid precipitate after the de-bulking of tungsten with 7 M HCl is shown. 3.2.4.2 Precipitation reactions with the dissolved solution In one instance, 5.2 mL of the dissolved solution was taken at a time and 10.1 mL of concentrated HCl was added. This was set up in the shaking incubator for 16 hours at 300 rpm at 60 °C. 50 mL tube with the solution was taken and centrifuged for 5 minutes at 4000 rpm following which the supernatant was decanted, passed through millex filters. The remaining precipitate was washed with 1 mL 7 M HCl in two rounds and combined with the supernatant from previous decantation. This process was repeated several times with varying amounts of dissolved solutions, and the supernatants were combined. In instances when the dissolved solution volume was high, the solution volume was reduced in a rotary evaporator in batches prior to the tungsten precipitation step. Next, high molarity HCl (10.6 M HCl) was added to bring up the final molarity to 7 M HCl to induce precipitation. It was set up in the shaking incubator in the same manner described above. After settling and centrifugation, the supernatant was filtered with a 500 mL vacuum filtration flask 47 (0.22 µm PES membrane), as shown in Figure 3.3 (c) and the filtrand was washed with approximately 10 mL 7 M HCl. 3.2.4.3 Separation with LN resin and 172Hf/172Lu generator Following the process outlined in section 3.2.4.2, several batches of 7 M HCl supernatant were obtained and assayed with gamma spectrometry. These supernatant volumes were combined, and load solutions were prepared in 7 M HCl for loading on LN resin column. Separations were carried out using both the peristaltic pump and the syringe pump and the peristaltic pump was calibrated before each separation to ensure a constant flow rate. As an example, a load solution (that was previously supernatant) of 28 mL was prepared for loading on the 2 mL LN cartridge preconditioned with 10 mL water followed by 10 mL 7 M HCl at a flow rate of 0.25 mL/min. Using the syringe pump, the prepared load solution was then loaded on the resin at a flow rate of 0.25 mL/min. 0.5 mL and 1 mL fraction volumes were collected of the load solution and analyzed on the HPGe for radionuclides that eluted with the load solution. 10 mL of 7 M HCl rinse solution was passed through the column to ensure a complete removal of radionuclides.172Hf was left adsorbed onto the resin for the 172Lu daughter to grow in. Elution cycles: The 172Lu daughter was eluted with 7 M HCl when required and 0.5 mL elute fractions were collected at a flow rate of 0.25 mL/min and analyzed on HPGe for breakthrough. The column was stored in 1 M HCl until the next elution cycle was performed. 3.2.4.4 Regeneration of 172Hf/172Lu generator The first step towards the regenerating of the above LN resin generator included eluting the adsorbed 172Hf with 12 mL 0.5 M Oxalic acid. This was done at a flow rate of 0.25 mL/min and 2 fractions were collected for HPGe analysis. Once the 172Hf was eluted from the column, 48 QMA resin was chosen for the hafnium-oxalic acid cleanup. The QMA resin was conditioned with 10 mL acetonitrile (flow rate of 1 mL/min), followed by 10 mL water (flow rate of 1 mL/min) and 10 mL 0.5 M Oxalic acid (flow rate of 0.25 mL/min) sequentially. This was followed by passing the 12 mL load solution at 0.25 mL/min. 10 mL MilliQ water was then passed through the cartridge to ensure complete washing of oxalic acid at 0.25 mL/min. 172Hf was then eluted from the cartridge with 10 mL of 1 M HCl (2 fractions total) followed by rinsing with 5 mL 2 M HCl at 0.25 mL/min. The fractions were monitored with HPGe throughout. 3.2.4.5 Establishment of new LN-resin generator Once 172Hf was eluted from QMA resin, it was acidified to 7 M HCl for further loading and LN resin was chosen again for the new generator setup. It was also conditioned in a similar manner described in section 2.4.3. The load solution of 172Hf in 7 M HCl was then passed through the LN resin at 0.25 mL/min. Once 172Lu was allowed to grow in, the generator was eluted with 7 M HCl, 0.5 mL per fraction for 172Lu when required and the column was stored in 1 M HCl until the next elution was required. A schematic of the entire process is shown in Figure 3.4. 3.2.5 Quantification of stable elements with ICP-OES Samples were measured using ICP-OES at different stages of the chemical processing by extracting an aliquot, diluting in 2% HNO3, quantifying and analyzing with Agilent ICP software. Single and multi-elements standards in various dilutions (0.01 ppm - 1 ppm) were used to establish the calibration curves for the sample analysis. The emission lines which had the highest intensity for the different elements were chosen for analysis and the separation efficiency was determined as a percentage of the loaded amount for different elements. 49 Figure 3.4: A schematic of the entire process of processing the tungsten beam-blocker from debulking, precipitation, separation on resin, and generator set-up is shown. 3.3 Results and discussion 3.3.1 Identification of radionuclides in the tungsten beam-blocker The available 172Hf was contained throughout the beam blocker with highest concentrations near the beamstrike that was approximately 5 cm2 aerially, and a few mm deep. While all faces of the beam-blocker were surveyed initially through  spectroscopy and autoradiography, only the most active face, with the beam-spot was found to have enough 172Hf to warrant chemical processing. Examples of isotopes embedded in the other faces not displaying the beam-spot include the cobalt isotopes. Not much can be commented on how these radioisotopes were produced since there is limited information on which beams were stopped in the beam blocker over its many years of service. Several radioisotopes were identified that were likely produced due to interactions between fast secondary protons and neutrons with W, Cu, and Ni (e.g. 172Hf), and others were 50 found that were likely implanted during heavy-ion irradiation (e.g. 110mAg). The complete list of radioisotopes that could be identified is given in Table 3.1. The pure alpha emitter 148Gd could only be detected with liquid scintillation counting after cursory separations were performed to isolate 151Gd (method not presented here). Similarly, radionuclides with small gamma branches were only identifiable after the bulk of the gamma- emitting background was removed in chemical separations. As more separations are performed in the future, additional radionuclides may be found. Table 3.1: List of radionuclides found in the tungsten beam-blocker measured through repeated gamma spectroscopy at different stages of chemical processing [1]. Radionuclide T1/2 Gamma rays (keV) Radionuclide T1/2 1.87 y 23.9 (20.3 %), 125.8 (11.3 %) 184Re 35.4 d Gamma rays (keV) 903.2 (38.1 %), 792 (37.7 %) Radionu- clide T1/2 Gamma rays (keV) 145Sm 340 d 61.2 (12.1 %) 172Hf 172Lu 173Lu 6.7 d 1.7 y 1093.6 (63 %), 900.7 (29.8 %), 181.5 (20.6 %) 272.1 (21.2 %), 78.6 (11.9 %) 343.4 (84 %) 175Hf 70 d 22Na 56Co 2.6 y 1274.5 (99.9 %) 77.2 d 846.77 (99.9 %), 1238.28 (66.46 %) 184mRe 169 d 104.7 (13.6 %) 181W 121. 2 d 6.24 (1.0 %) 54Mn 312.2 d 182Ta 114.7 d 143Pm 265 d 741.9 (38.5 %) 101Rh 3.3 y 198.0 (73 %), 127.2 (68 %) 109Cd 461. 9 d 88.0 (3.6 %) 834.8 (99.9 %) 67.7 (42.9 %), 1121.2 (35.2 %) 356 (62.0 %), 80.9 (32.9 %) 1077 (3.22 %) 68Ga 67.7 m 125Sb 2.8 y 68Ge 270.9 d - 144Pm 363 d 51 427.8 (29.6 %), 600.5 (17.65 %) 696.4 (99.5 %), 618.01 (98 %) 7Be 53.2 d 477.6 (10.44 %) 133Ba 10.5 y 42K 12.3 h 1524.6 (18.08 %) Table 3.1 (cont’d) 57Co 110mAg 60Co 88Zr 88Y 271. 7 d 249. 8 d 192 5.3 d 83.4 d 106. 6 d 122.1 (85.6 %), 136.5 (10.7 %) 657.7 (95.6 %), 884.6 (75 %) 1173.2 (99.85 %), 1332.5 (99.9 %) 392.8 (97.3 %) 898 (93.7 %), 1836 (99.2 %) 46Sc 83.7 d 102Rh 207.3 d 102mRh 3.7 y 889.27 (99.98 %), 1120.54 (99.98 %) 475 (46 %), 628 (4.5 %) 475 (95 %), 631.3 (56 %) 148Gd 71.1 y - 151Gd 123.9 d 243.2 (5.6 %), 153.6 (6.2 %) 3.3.2 Dissolution of the tungsten beam-blocker As the activity was contained in the face of the beam blocker identified by autoradiography, the chemical processing of 172Hf from the entire beam blocker mass was not necessary. However, even with partial dissolution of the beam blocker, a considerable amount of the tungsten alloy mass had to be chemically processed. The dissolution of the shavings was done in batches using an optimized volume of 125.5 mL 30% H2O2 per gram of the tungsten beam-blocker shavings, and the prime advantage of the dissolution described in section 3.2.4.1 is the non-requirement of HF. Using peroxide strips, the concentration of peroxide during the dissolution stage in the solution was determined to be >50 ppm. The rate of bubbling was a good indication of how the reaction was proceeding during the dissolution of shavings. Moderate to little bubbling was a good indication to add more peroxide. Since the reaction was vigorous, caution was observed throughout, and dissolution was performed in batches. H2O2 concentration was determined again using peroxide strips towards 52 the near end of the dissolution process and determined to be around 10 ppm. Finally, at the end of dissolution, peroxide concentration dropped to 2 ppm and later to <1 ppm. The solution once dissolved was determined to be 0.9% Cu, 3.9% Ni and 97.2% W which was verified through ICP-OES. Approximately 77.1% of the tungsten shavings were dissolved in the first attempt (Figure 3.3 (a)) and 22.9% stayed as the black residue (Figure 3.3 (b)) which was dissolved separately. This was also determined using ICP-OES. 3.3.3 De-bulking of tungsten (precipitation) and separation process A precipitation of the bulk tungsten matrix while keeping the radioactive species in solution represents a viable alternative to the ion exchange based separation. In several pre- experiments, a precipitation of tungsten as yellow tungsten trioxide (its hydrated form is tungstic acid) in solutions of >3M HCl has been observed [7]. By performing such a precipitation step, the desired radionuclides are transferred into an HCl matrix and there is a considerably better availability of distribution coefficient data for HCl media. Additionally, large amounts of tungsten could cause interferences with the elution of other species from the ion exchange column, hence precipitation was chosen. The precipitation of bulk tungsten with 7 M HCl results in tungstic acid H2WO4 and tungsten dichloride dioxide, or tungstyl chloride WO2Cl2 (speciation was not determined in this work), shown in Figure 3.3 (c). The HCl molarity was optimized to be 7 M HCl for the precipitation process and 99.7% tungsten was successfully de-bulked with 22(1)% Cu and 11(1)% Ni also being co-precipitated. De-bulking of tungsten was also necessary because it was observed that the separation efficiency of the proceeding step drops as the mass of tungsten is increased. As a note, the yellow precipitate was readily dissolvable in NaOH solutions, so it should be possible to recover co-precipitated radioisotopes, if needed in the future. 53 For the separation of 172Hf from the remainder tungsten and other radionuclides after precipitation, LN resin was used. Di(2-ethylhexyl)phosphoric acid (HDEHP) based resin is widely used for adsorbing hafnium and allowing trivalent lutetium to elute out [8]. With 7 M HCl 172Hf, Ge, Zr was held up in the column, while Lu, Gd, Pm, Co isotopes and stable W, Cu, Ni were washed down the column. The radiolanthanides eluted from the load fraction can be further processed and chemically separated. 172Hf stayed on the column and the gamma lines of 125.8 and 23.9 KeV of 172Hf are not present in the eluted fractions from the LN resin (see Figure 3.6), thus indicating a successful separation from other radionuclides. Approximately (80 ± 2)% 172Hf could be eluted with 0.5 M Oxalic acid. The 172Hf/172Lu generator was eluted with 7 M HCl and used for PAC experiments and (104 ± 2)% 172Lu could be eluted from the generator within 4 mL. The entire separation scheme is shown in Figure 3.5 with the aim of tracking hafnium. Figure 3.6 shows the gamma spectrum of the load fractions to show absence of hafnium and Figure 3.7 shows the presence of hafnium on the column. An elution profile of the eluted 172Lu from the regenerated generator is shown in Figure 3.9 and gamma spectra is shown in Figure 3.8. 54 Figure 3.5: A schematic is shown with the radioisotopes separating at different stages. The primary aim was to track 172Hf. 55 Figure 3.6: A gamma spectrum of the load solution passing through the column indicating the absence of the 172Hf peak. Figure 3.7: A gamma spectrum of the generator column indicating the presence of the 172Hf peak is shown. 56 Figure 3.8: Gamma spectra, 100-350 keV (top) and 900-1100 keV (bottom) of a 0.5 mL fraction eluted from the regenerated 172Hf/172Lu generator showing the presence of 172Lu peaks, and absence of 172Hf peak. 57 Figure 3.9: An elution profile of the eluted daughter 172Lu with 7 M HCl from the generator. Figure 3.10: A schematic summarizing the process followed in this chapter. 58 3.4 Conclusion Since tungsten and tungsten alloys (e.g. ‘heavymet’ W 95%, Cu 1.5%, and Ni 3.5%) will be routinely used to stop beams at FRIB and other facilities, performing this experiment was important to establish methods for recovering implanted and induced radionuclides, such as 172Hf from those materials. A schematic is given in Figure 3.10 of the process that was detailed in this chapter. The generator that was set up in this chapter will be compared to those set up in the next chapter and comparison will be drawn between harvesting through isotope harvesting and production through a conventional technique. 59 REFERENCES “National Nuclear Data Center.” Accessed: Mar. 22, 2024. [Online]. Available: [1] https://www.nndc.bnl.gov/nudat3/ [2] R. L. Rasera and A. Li-Scholz, “Perturbed-Angular-Correlation Study of Electric- Quadrupole Interactions in 172Yb in Thulium Metal and Oxide Lattices,” Phys Rev B, vol. 1, no. 5, pp. 1995–2006, Mar. 1970, doi: 10.1103/PhysRevB.1.1995. A. Lerf and T. Butz, “Nuclear quadrupole interactions in compounds studied by time [3] differential perturbed angular correlations/distributions,” Hyperfine Interact, vol. 36, no. 3, pp. 275–370, 1987, doi: 10.1007/BF02395636. [4] R. Valentini and R. Vianden, “Angular correlation studies on 172Lu(172Yb) in GaN and measurements at low temperatures,” Hyperfine Interact, vol. 197, no. 1, pp. 149–153, 2010, doi: 10.1007/s10751-010-0229-9. [5] S. Chakraborty, S. Pallada, J. T. Pedersen, A. Jancso, J. G. Correia, and L. Hemmingsen, “Nanosecond Dynamics at Protein Metal Sites: An Application of Perturbed Angular Correlation (PAC) of γ-Rays Spectroscopy,” Acc Chem Res, vol. 50, no. 9, pp. 2225–2232, Sep. 2017, doi: 10.1021/acs.accounts.7b00219. [6] L. Hemmingsen, K. N. Sas, and E. Danielsen, “Biological applications of perturbed angular correlations of γ-ray spectroscopy,” Chem Rev, vol. 104, no. 9, pp. 4027–4061, Sep. 2004, doi: 10.1021/cr030030v. [7] F. W. E. Strelow, “Distribution coefficients and cation-exchange behaviour of 45 elements with a macroporous resin in hydrochloric acid/methanol mixtures,” Anal Chim Acta, vol. 160, no. C, pp. 31–45, 1984, doi: 10.1016/S0003-2670(00)84506-7. P. Horwitz and D. Mcalister, “Eichrom’s LN Series of Resins: Characterization and [8] Novel Applications,” 2008. “Virtual tour.” Accessed: Mar. 22, 2024. [Online]. Available: [9] https://nscl.msu.edu/public/virtual-tour.html 60 THEORY OF TECHNIQUES UTILIZED IN CHAPTER 3 Perturbed Angular Correlation Spectroscopy (PAC) APPENDIX PAC is a technique providing insights into the chemical environment surrounding a target metal ion by probing its nuclear spin interaction with an electric field gradient [5], [6]. For this technique to be effective, it necessitates the presence of cascade gamma rays emitted by the probe nucleus and a lifetime of the intermediate level exceeding 1 nanosecond, a criterion that 172Lu fulfills. The nuclear level diagram illustrating the characteristics of 172Lu is depicted in Figure 1.5 in Chapter 1. The 172Lu eluted samples were transferred to FRIB and placed between 14 LaBr3 scintillation detectors that use 1.5 x 1.5 inch LaBr3(Ce) crystals, arranged in a circular array, operating at room temperature and the purpose was to measure the correlation of γ-rays of 172Lu. 61 CHAPTER 4: 172Hf HARVESTING FROM LUTETIUM FOIL TARGET AT THE BROOKHAVEN LINEAR ISOTOPE PRODUCER This chapter details the conventional method of producing 172Hf at the BLIP by the proton irradiation of natural lutetium metal foil through the 175Lu(p,4n)172Hf nuclear reaction pathway. The production was carried out to obtain pure 172Lu (PAC probe) generated from the extracted parent, 172Hf. 172Lu can also be used a surrogate for the more popular 177Lu theragnostic agent to test chemistry and the PAC technique has the potential to provide insights into radiolabeling, pharmacokinetics, in new Lu-based pharmaceuticals. Following irradiation, the lutetium foil was dissolved in HCl and subsequent separation of 172Hf radioisotope from bulk lutetium and other lutetium isotopes was achieved using various columns loaded with LN resin, ZR resin, as well as in-house synthesized hydroxamate and methyl-substituted hydroxamate resins. The efficacy of these different 172Hf/172Lu resin-based generator systems was assessed to monitor 172Hf breakthrough. Subsequently, all four generators were consolidated into a single higher activity generator (130 µCi) using ZR resin, which exhibited similar performance characteristics to the original ZR resin generator. This is detailed in this chapter and a comparison is also drawn with the 172Hf/172Lu generator set up from Chapter 3. The experimental methodologies described in Chapter 3 and 4 have advanced the development of methodologies and our understanding for producing high-purity 172Lu, which is essential for PAC and radiolabeling studies. 4.1 Introduction The pioneering FDA-approved radiopharmaceutical, Lutathera® [177Lu]Lu-DOTATATE, for radioligand therapy (RLT), has gathered significant attention in cancer treatment, indicating a 62 growing interest in 177Lu-based radiopharmaceuticals [1]. Research facilities around the world aiming to innovate new 177Lu-based therapeutics find value in the ability to conduct lutetium radiolabeling tests independently of reactor irradiation schedules. An effective substitute for exploring lutetium chemistry is 172Lu (t1/2 = 6.7 days), which can be generated in high purity from its long-lived parent, 172Hf (t1/2 = 1.87 years) [2]. Amongst the possible routes to obtain 172Lu, some options for its production are: (I) irradiation of Yb with protons or deuterons such as natYb(p,xn)172Lu, 172Yb(p,n)172Lu, natYb(d,xn)172Lu and 172Yb(d,2n)172Lu [3], [4], (II) 169Tm(α,n)172Lu reaction [5] (III), 172Hf/172Lu radionuclide generators [6], [7], and (IV) 170Er(7Li,5n)172Lu reaction [8]. For the purpose of comparison with the 172Hf/172Lu generator set up detailed in Chapter 3, proton irradiation of natural lutetium foil was chosen as shown in Figure 4.1. Figure 4.1: Schematic of a section of the chart of the nuclides demonstrating the 175Lu(p,4n)172Hf reaction when a natural lutetium foil is irradiated for the production of 172Hf. While extensive literature exists on the radiopharmaceutical applications of 177Lu, there are not studies focusing on the development of 172Hf/172Lu generator setups, which is the primary focus of this chapter [7], [9], [10], [11]. Specifically, there is a lack of column-based generators, despite the investigation of numerous promising substrates for analogous separations like 44Ti/44Sc and 89Zr/natY, which are widely employed in radiopharmaceutical production [11]. 63 The four resins used in this study include ZR resin, in-house synthesized hydroxamate resin, methyl-substituted hydroxamate resin, and LN resin [12], [13], [14], [15]. Dirks et al. shows that ZR resin, having the hydroxamate functionality, exhibits less selectivity for lutetium in the presence of HCl, thus enabling the separation of hafnium from lutetium [12]. The hafnium can subsequently be eluted from the columns by complexing it with oxalic acid, behaving similarly to its chemical analogue zirconium [16]. Similarly for LN resin, which is based on di- (2-ethylhexyl)phosphoric acid (HDEHP), also shows selectivity for hafnium over lutetium in high molarity hydrochloric acid solutions [17]. The methodologies developed for chemical separation of hafnium and lutetium in this chapter also hold significant relevance in isotope geochemistry separations, given that Hf-Lu serves as an important alternative chronometer to Sm-Nd [18], [19], [20], [21], [22]. Once the generators were set up, a comparative analysis was conducted with the four resins, considering various parameters such as elution volumes, elution patterns, and the occurrence of 172Hf breakthrough. Following this, the generators were regenerated into a single higher activity generator to enable the cyclic elution of higher activity of 172Lu. The eluted high purity 172Lu was used for PAC spectroscopy measuring the lifetimes of excited states and the angular correlations of  rays emitted in cascade using the LaBr3(Ce) detector array at FRIB and radiolabeling small peptides (results not presented here). 4.2 Experimental 4.2.1 Materials 4.2.1.1 Reagents The lutetium foils (99%, 0.25 mm thick, Goodfellow) were cleaned with ethanol (ACS Reagent, 99.5%, Sigma Aldrich) in preparation for the irradiation. The dissolution of the 64 lutetium foils and the chemical processing post-irradiation was performed with the following reagents: hydrochloric acid (HCl, Optima Grade, 32.0-35.0%, Fisher Chemical and 99.99% Trace metals basis, Sigma-Aldrich), nitric acid (HNO3, 65%, Suprapur, Sigma-Aldrich), oxalic acid (C2H2O4, Analytical Reagent, Mallinckrodt), oxalic acid dihydrate (ACS Reagent, J.T.Baker) acetonitrile (MeCN, Sigma-Aldrich, ACS Reagent >= 99.5%) and MilliQ water (Deionized Milli-Q water, 18.20 MΩ, Millipore and Thermo Scientific MicroPure Ultrapure Water System, 18.20 MΩ cm). Hafnium ICP Standard (1000 μg/mL, in 2% HCl, Assurance), lutetium ICP Standard (1000 μg/mL, in 2% HNO3, Assurance), and lutetium (III) chloride hexahydrate (>99.99% trace metal basis, Sigma Aldrich) were used in methodology development experiments with stable elements. Synthesis of the covalently bonded hydroxamate and methyl-substituted hydroxamate resins was performed with the following reagents: AccellTM PLUS CM resin (Waters, Lot: 006501071), 2,3,5,6-Tetrafluorophenol (97%, Alfa Aesar, Lot: 10233607), 1-Ethyl-3-(3ʹ- dimethylaminopropyl)carbodiimide (EDAC) hydrochloride (98%, Millipore Sigma, Lot: 3996375), hydroxylamine hydrochloride (ReagentPlusR, 99%, Sigma-Aldrich, SourceMKCS8067), N Methylhydroxylamine hydrochloride (98%, Sigma-Aldrich, Source:STBJ7422), Acetonitrile (ACS Reagent, ≥99.5%, Sigma-Aldrich, source:SHBN8541), methyl alcohol (ACS Reagent, ≥99.8%, Sigma-Aldrich, source:MKCPO477). 4.2.1.2 Column construction with Extraction Chromatographic Resins The experiments conducted at Brookhaven lab included using the following resins and columns - LN-extraction chromatographic resin (Di-(2-ethylhexyl)phosphoric acid, particle size 100-150 μm, Eichrom) was slurry packed in columns and employed for the separations. The columns used for them were Glass chromatography columns (Bio-Rad, 1.0 x 5 cm, Econo-Column 65 Chromatography Columns). Parts from Low Pressure Fittings Kit obtained from Bio-Rad were used to connect tubing (Silicone tubing, 0.8 mm ID, Bio-Rad) to the column. A piece of glass wool was inserted on the top end of the columns to hold the resin in place. The QMA cartridge used was obtained from Waters Corp (Sep-Pak Accell Plus QMA Plus Light Cartridge, 130 mg Sorbent per Cartridge, 37 - 55 µm). The experiments conducted at MSU included the use of ZR-extraction chromatographic resin (2 mL Cartridges, particle size 50-100 μm, Eichrom ZR-R10-S), LN-extraction chromatographic resin (2 mL Cartridges, particle size 50-100 μm, Eichrom LN-R50-S), QMA strong anion exchanger resin (Sep-Pak Accell Plus QMA Plus Light Cartridge, Part no. WAT023525, Lot no. 034639256A). In-house synthesized hydroxamate and methyl-substituted hydroxamate resin was slurry packed in columns and employed for the separations. The column was constructed from rigid polycarbonate tubing (3/8′′ OD, 1/4′′ ID, McMaster-Carr, PN:9176T1) and chemical-resistant push-to-connect fittings (Straight Reducer, for 3/8′′ x 1/4′′ Tube OD). Two pieces of glass wool were inserted on each end of the column to hold the resin in place. 4.2.1.3 Instruments At Brookhaven Lab, the stable and radiochemical separations were performed with a peristaltic pump (Bio-Rad) operated with tubing of 0.8 mm internal diameter. During the separation method development, the identification and quantification of stable lutetium and hafnium was performed with a Perkin Elmer Optima 7300 Inductively coupled plasma-optical emission spectrometer (ICP-OES) together with the Syngistix software. The radionuclides were detected and quantified with an Ortec HPGe γ-ray Detector. At MSU, the separations were performed with a syringe pump (New Era Pump Systems, NE-1000 SyringeONE one channel 66 programmable syringe pump) and the radionuclides were detected and quantified with an HPGe Canberra Broad Energy BEGe γ-ray Detector (BE2020). 4.2.2 Experimental setup 4.2.2.1 Target design and irradiation Figure 4.2: Natural lutetium foil enclosed between aluminum target holder and held in place by screws is shown. Figure 4.3: A schematic of the target assembly at BLIP is shown. Lutetium foil labeled 5, sandwiched between several components of beryllium window labeled 1, AlBeMet window labeled 2, water gaps in blue, stainless steel window labeled 3, and aluminum window labeled 4, in the beamline is shown. The natural lutetium foil (97.4% 175Lu and 2.6% 176Lu), with weights of 1.5836 g (target 1) and 1.4707 g (target 2), approximately 0.254 mm thick and 28.5 mm in diameter, were inserted into a screw-bolted aluminum target holder crafted at BNL, following ethanol cleaning, as depicted in Figure 4.2 [23]. Two proton irradiations were conducted at the BLIP using a focused beam with incident energies of 66 MeV (21.05 hours, 0.48 µA) and 117 MeV (23.99 67 hours, 2.82 µA). The attenuation of the beam in the natural lutetium foil target and other beamline components was estimated using the SRIM (Stopping and Range of Ions in Matter) compilation [24]. For these two irradiations, the other components included the beam passing through the following components in that order: Be window (beryllium), AlBeMet window (AlBeMet), beamline window (stainless steel), water gap, BOX front window (stainless steel), water gap, aluminum degrader (Aluminum), water gap, bolted holder window (Aluminum), Lutetium foil (Lutetium), Bolted holder back (Aluminum), water pocket, water gap, aluminum degrader (Aluminum), water gap, aluminum degrader (Aluminum), and water gap. A schematic representation is provided in Figure 4.3. Al, Cu, and Ni foils served as monitor foils for the second target. 4.2.3 Theoretical production yield of radioisotopes with FLUKA The excitation functions and the stopping characteristics of the particle beam in the target material need to be considered when the beam energy and target thickness for an irradiation needs to be determined. For this work, the theoretical activities for each isotope were obtained by simulating the target array (above-mentioned components of beamline windows, water cell, degraders, and target foil) in the Monte Carlo particle transport code FLUKA (Figure B.1.). The calculations were performed using the average beam intensity and irradiation duration matching the two irradiation scenarios using the default FLUKA parameters for activation and recommended for secondary particle generation and transport. Comparison was made for the activities that were calculated and to those determined in this work. The target array and the energy propagation calculations for 117 and 66 MeV beam have also been shown in the section 4.3.1. 68 4.2.4 Experimental production yield of radioisotopes through γ spectroscopy Following the irradiation, the targets underwent measurement at MSU and Brookhaven lab using an HPGe Detector, with spectra analysis conducted using the Canberra Genie 2000 software (Mirion Technologies) at MSU and the GammaVision-32 software, version 6.01 at Brookhaven lab. Both these software packages facilitated the analysis of energy spectra, including identification of characteristic γ-ray emissions of the different radioisotopes, background subtraction, and activity calculations based on net photopeak areas. Table 4.1 and Table 1 in Appendix presents the gamma lines utilized for activity assessment across the separations. Energy and efficiency calibrations of HPGe detector were carried out using a mixed gamma solution prepared from NIST traceable calibration standard (Eckert and Ziegler Analytics, GA, USA), positioned 10-15 cm from the detector at Brookhaven Lab to ensure accurate quantification. For precise quantification, aliquots of each collected fraction from the radiochemical separations were extracted, diluted with MilliQ water in 3 mL Eppendorf screw- cap conical centrifuge tubes, and analyzed with γ-ray spectroscopy. During all the measurements, the average dead time remained below 3%. At MSU, the energy and efficiency calibrations were done with a 152Eu source (1 μCi, Eckert & Ziegler Analytics) distributed evenly in 10 mL of epoxy within a sealed 15 mL Falcon tube at distances of 15, 25, 50, 100, and 180 cm from the face of the HPGe detector. In this case, the average dead time during measurements was kept below 5%. To maintain consistency in geometry with calibration, all resulting fractions from separations were collected and measured in 15 mL Falcon tubes, positioned in a fixed acrylic holder at the specified distances from the HPGe detector, as with previous experiments in previous chapters. Both targets underwent a decay period of several weeks to allow for the decay of shorter-lived species before initiating chemical processing. 69 4.2.5 Separation chemistry development 4.2.5.1 Dissolution of irradiated targets and radiochemical separation with resins The initial disassembly of the first lutetium target involved removing the screws in a hot cell, after which the lutetium foil was dissolved in a fume hood incrementally in 20 mL 1 M HCl and 20 mL 2 M HCl, added in 5 mL batches. While most of the foil dissolved upon treatment with HCl, residual black residue was observed, needing filtration of the solution. Scaling up the solution to a total volume of 50 mL with 7 M HCl prepared it for loading onto resin. At this stage, an aliquot was extracted from the target solution to quantify the quantities of hafnium and lutetium isotopes via gamma spectroscopy. Concentrated HCl was introduced to the remaining black residue in an attempt to dissolve it, employing heating to elevate the temperature to around 250°C. For the second lutetium foil, dissolution was done with 12 mL 7 M HCl in 2 mL increments dropwise, followed by further dilution with an additional 21 mL 7 M HCl, with the solution placed in an ice bath. Subsequently, this solution was divided to create load solutions for the various resins. A total of five separations were conducted, one with the initial target and four with the subsequent one, enabling a comparative analysis of the separation processes. 4.2.5.2 In-house resin synthesis In-house covalently bonded hydroxamate resin was synthesized through functionalization of the commercially available AccellTM Plus CM resin. Unsubstituted hydroxamate resin was synthesized using procedure described by J. P. Holland et. al. and the methyl substituted hydroxamate resin was synthesized following procedures described by L. Gajecki et. al. [12], [14]. 70 4.2.5.3 Separation with LN resin and 172Hf/172Lu generator To process the first target with lower activity, approximately 900 mg of Eichrom LN resin was packed into a column as a slurry and conditioned with 2 BV of 7 M HCl after being rinsed with 30 mL of MilliQ water at a flow rate of 0.5 mL/min. Subsequently, 50 mL 7 M HCl, dissolved above, was loaded onto the column at the same flow rate using a peristaltic pump. The eluate, comprising approximately 10 mL fractions of the load solution containing stable lutetium and lutetium isotopes, was collected and quantified using HPGe. Following the load solution, the column was rinsed with 7 M HCl (approximately 30 mL) to recover any remaining stable lutetium and lutetium isotopes, with 10 mL fractions collected for analysis using HPGe and ICP- OES. The matrix on the column was then changed to 1 M HCl. The adsorption of 172Hf/175Hf onto the resin allowed for the growth of the 172Lu daughter and whenever the elution of the generator was necessary, 6-10 mL 7 M HCl was used for eluting 172Lu. For the processing of the second target with higher activity using LN resin, a load solution consisting of 20 mL 7 M HCl was prepared and loaded onto a 2 mL LN cartridge. The resin was previously conditioned with 10 mL water followed by 10 mL 7 M HCl at a flow rate of 0.25 mL/min. The prepared load solution was then loaded onto the resin at the same flow rate using a syringe pump. Four fractions, each containing 5 mL of the load solution, were collected and measured on HPGe for lutetium isotopes. Subsequently, the column was washed with a minimum of 20 mL 7 M HCl to recover any residual stable lutetium and lutetium isotopes. The matrix on the column was then changed to 1 M HCl. The adsorption of 172Hf/175Hf onto the resin allowed for the in-growth of the 172Lu daughter. Elution cycles for the LN resin involved the use of 10 mL 7 M HCl for nine cycles to analyze for 172Hf breakthrough, with 2 mL eluate fractions collected at a flow rate of 0.25 mL/min. 71 4.2.5.4 Separation with ZR resin and 172Hf/172Lu generator To process a fraction of the second target (of higher activity), a load solution comprising 20 mL 7 M HCl was prepared for loading onto a 2 mL ZR resin cartridge. Prior to loading, the cartridge underwent preconditioning with 10 mL of water followed by 10 mL 7 M HCl, both carried out at a flow rate of 0.25 mL/min. The prepared solution was then loaded onto the resin at the same flow rate using a syringe pump. Following this, four fractions, each containing 5 mL of the load solution, were collected and measured on HPGe for the presence of lutetium isotopes. The column was then washed with 25 mL of 7 M HCl to recover any residual stable lutetium and lutetium isotopes. As with the previous resins, the matrix of the resin was changed to 1 M HCl and the daughter 172Lu was allowed to grow in from the adsorption of 172Hf/175Hf onto the resin. Regarding the elution cycles, the 172Lu daughter was eluted cyclically using 10 mL of 0.1 M HCl to analyze for breakthrough of 172Hf. Within each cycle, 2 mL eluate fractions were collected at a flow rate of 0.25 mL/min. 4.2.5.5 Separation with in-house synthesized hydroxamate resin and 172Hf/172Lu generator Around 682 mg of hydroxamate resin, synthesized in-house, was packed into a column (9 mm outer diameter, 6 mm inner diameter, and a length of 15 cm). Following this, the resin was rinsed with 10 mL MilliQ water and conditioned with 10 mL of 7 M HCl. Subsequently, the 14 mL 7 M HCl load solution was passed over the pre-conditioned hydroxamate resin-packed column which is when the majority of lutetium and lutetium isotopes passed through the column and collected in 3.5 mL fractions. To ensure the removal of any remaining lutetium, a rinse step with 14 mL 7 M HCl, distributed in 3.5 mL fractions, was done. This was followed by changing the matrix to 1 M HCl. 172Hf/175Hf adsorbed onto the resin and was followed by the in-growth of 72 the 172Lu daughter. Throughout the conditioning and separation processes, the flow rate remained constant at 0.25 mL/min. Regarding elution cycles, the 172Lu daughter was eluted cyclically using 10 mL of 0.1 M HCl for nine cycles to check for breakthrough of 172Hf. Within each cycle, 2 mL eluate fractions were collected at a flow rate of 0.25 mL/min and subsequently analyzed on the HPGe detector. 4.2.5.6 Separation with in-house synthesized methyl-substituted hydroxamate resin and 172Hf/172Lu generator Approximately 680 mg of methyl-substituted hydroxamate resin, synthesized in-house, was utilized for slurry packing into a column with the same dimensions as the previous one. The rinsing and preconditioning was also done in the same way as the hydroxamate resin at the same constant flow rate 0.25 mL/min. The loading step was also the same as the hydroxamate and 172Hf/175Hf was left adsorbed onto the resin which allowed for the in-growth of the 172Lu daughter after changing the matrix of the column to 1 M HCl. Similar to the hydroxamate resin, the 172Lu daughter underwent cyclic elution with 10 mL of 0.1 M HCl for nine cycles to check for breakthrough of 172Hf. Within each cycle, 2 mL eluate fractions were collected at a flow rate of 0.25 mL/min and subsequently analyzed with the HPGe detector. 4.2.5.7 Regeneration of 172Hf/172Lu generators using QMA resin To initiate the regeneration of the generators, the initial step involved stripping hafnium from the various resins. In all established generator setups, hafnium was stripped by elution with 10 mL of 0.5 M Oxalic acid at a flow rate of 0.3 mL/min, with fractions being collected for subsequent analysis using the HPGe detector (result in Figure B.2 for target 1). 73 Following the elution of 172Hf/175Hf, fractions from the four generators were combined, resulting in a total volume of 40 mL. For the hafnium-oxalic acid cleanup process, QMA resin was selected. Two QMA resin cartridges underwent conditioning with 6 mL acetonitrile, followed by 10 mL water and 10 mL 0.5 M Oxalic acid at a flow rate of 1 mL/min. Following that, the 20 mL load solution was passed through each of the two QMA columns at a flow rate of 0.25 mL/min. To ensure the complete removal of oxalic acid, 25 mL of MilliQ water was then passed through the cartridge at 0.25 mL/min. The hafnium retained on the cartridge was subsequently eluted with 10 mL of 1 M HCl at a flow rate of 0.25 mL/min. 4.2.5.8 Establishment of new higher activity ZR-resin generator After the elution of hafnium from the QMA resin using 1 M HCl, ZR resin was utilized for the establishment of a new generator setup. This resin underwent conditioning with 10 mL of MilliQ water, followed by 10 mL of 1 M HCl, with a flow rate of 1 mL/min for the initial 5 mL and 0.25 mL/min for the subsequent 5 mL of 1 M HCl. The load solution containing 172Hf in 1 M HCl, previously eluted from the QMA resin, was then passed through the resin at a flow rate of 0.25 mL/min. When necessary, the generator was eluted with 0.1 M HCl for the extraction of 172Lu. 4.2.6 Quantification of stable elements with ICP-OES For the purpose of measuring using ICP-OES, aliquots were withdrawn from each fraction collected during the inactive separations and two dilutions of identical concentrations were prepared for each fraction by diluting them in 2% HNO3. These diluted samples were then quantified and analyzed using Agilent ICP software. The separation efficiency was determined as a percentage of the loaded amount for lutetium. Calibration curves for sample analysis were established using single-element lutetium 74 standards at various dilutions ranging from 0.01-1 ppm. The emission lines exhibiting the highest intensity for lutetium were selected for analysis; as shown in the Appendix for the specific wavelengths chosen. 4.3 Results and discussion 4.3.1 Radionuclide identification and quantification The SRIM simulation package was utilized to perform the energy propagation calculations as the proton beam traversed through various components including aluminum, beryllium, stainless steel, AlBeMet alloy, and a water gap within the target array. This is demonstrated in Figure 4.4 and Figure 4.5. Figure 4.4: Irradiation of Lu foil with an energy of 66 MeV at BLIP (SRIM package was used for the proton beam traversing through several components in the target array and energy propagation calculations are shown). Image provided by Dr. Dmitri Medvedev. 75 Figure 4.5: Irradiation of Lu foil with an energy of 117 MeV at BLIP (SRIM package was used for the proton beam traversing through several components in the target array and energy propagation calculations are shown). Image provided by Dr. Dmitri Medvedev. For the first target, the proton beam enters the lutetium foil target with an initial energy of 22.5 MeV after passing through the preceding components. Upon exiting the foil, the beam energy was measured at 18.7 MeV. The discrepancy with the SRIM simulation is due to the fact that one of the water layers was actually 4.5 mm larger than the design documentation states. This propagates into a large difference in beam energy. Monitor foils for the second irradiation were used to correct for this. This correction resulted in a better and closer energy impinging on the second target where the specific energy of the proton beam upon entering is 35.2 MeV and exiting the lutetium foil target is 32.5 MeV. To investigate the co-production of radioisotopes resulting from proton irradiation of natural lutetium foil at energies of 66 MeV and 117 MeV, γ-spectroscopy was employed which established the presence of various isotopes including 171, 172, 173,175Hf, as well as 171, 172, 173, 174, 76 177Lu, and 169Yb, identifiable by their characteristic gamma lines. Figure 4.6 shows a gamma spectrum of the second target prior to any attempted chemical separation. Additionally, no activity was detected on the aluminum target body or screws, suggesting that the proton beam was well centered onto the lutetium foil. The total activity of 172Hf in the first and second lutetium foil targets, decay corrected to the end of irradiation, was determined to be 0.39 ± 0.02 MBq and 230 ± 13 µCi (8.51 ± 0.48 MBq). This finding is compared with the theoretical yield of 13.2 MBq for the second target, as indicated in Table 4.1. Additionally, through the same γ-spectroscopic measurements, the activity of 172Lu for the second target, decay corrected to the end of irradiation, was measured at 731 ± 269 µCi (27.0 ± 9.9 MBq). Figure 4.6: A gamma spectrum of the second target before chemical separation was attempted is shown. 77 Table 4.1: The production of different radioisotopes and their theoretical and experimental activity (decay corrected to EOB) is shown resulting from the proton irradiation of natural lutetium foil. The shorter-lived radionuclides are not shown here and given in Table B.1 [2]. Radioisotope Half-life Gamma lines (Intensity) 172Hf 1.87 y 125.8 keV (11.3 %), 23.9 keV (20.3 %), 67.4 keV (5.3 %) 70 d 343.4 keV (84.0 %) 175Hf 171Lu 8.2 d 172Lu 6.7 d 173Lu 1.37 y 174gLu 3.31 y 169Yb 32.0 d 739.8 keV (48.7 %), 667.4 keV (11.2 %) 1093.6 keV (63.0 %), 181.5 keV (20.6 %), 900.7 keV (29.8 %) 272.1 keV (21.2 %), 78.6 keV (11.9 %) 1241.8 keV (5.1 %), 76.4 keV (5.9 %) 197.9 keV (35.9 %), 177.2 keV (22.3 %), 63.1 keV (43.6 %) Target 2 Experimental activity in µCi (MBq) Target 2 Theoretical activity in µCi (MBq) 230 ± 13 (8.5 ± 0.5) 357 ± 0.5 (13.2 ± 0.02) 49.5 ± 1.6 (1.8 ± 0.05) 143.5 ± 17.2 (5.3 ± 0.6) 88.6 ± 0.9 (3.3 ± 0.03) 162.3 ± 12.8 (6.0 ± 0.5) 731 ± 269 (27.0 ± 9.9) 285.6 ± 28.7 (10.6 ± 1.1) 198.8 ± 7.0 (7.4 ± 0.3) 16.3 ± 12.4 (0.6 ± 0.5) 93.4 ± 0.3 (3.5 ± 0.01) 27.1 ± 0.5 (1.0 ± 0.01) 17.8 ± 0.6 (0.7 ± 0.02) 29.3 ± 0.8 (1.1 ± 0.02) 4.3.2 Dissolution, radiochemical separation, and 172Hf/172Lu generator setups During the dissolution of the lutetium foil with HCl for both the first and second targets, an observation consistent with the work in Holland et al., when working with irradiated yttrium foils, was noted: the majority of the lutetium foil dissolved, yet a residual trace of black precipitate remained despite repeated attempts at dissolution [12]. This precipitate is presumed to be an insoluble form of lutetium oxide, and it was discovered to contain minimal 172Hf activity, approximately (~2 µCi). At the time of the radiochemical separations, which occurred several weeks after EOB for both the first and second targets, the short-lived radioisotopes 173Hf and 171Hf had already 78 decayed, leaving behind 172Hf, 175Hf, 172Lu, 173Lu, 174Lu, 177Lu, and 169Yb as the radioactive species at the stage of radiochemical separation. For the second target, due to the high activity levels of the collected lutetium fractions, ICP-OES analysis was omitted, and only HPGe was used to track lutetium radioisotopes (and consequently stable lutetium). 4.3.2.1 LN resin separation and 172Hf/172Lu setup The separation of hafnium from stable lutetium and lutetium isotopes in both the first and second targets utilized di(2-ethylhexyl)orthophosphoric acid (HDEHP) extractant-based LN resin, known for its efficacy with di, tri, and tetravalent metals. In the case of the first target, while stable lutetium and lutetium isotopes passed through the column with the 7 M HCl load solution, hafnium isotopes 172Hf and 175Hf were adsorbed onto the resin. Approximately (94.4 ± 2.0)% of stable lutetium was eluted with the load fractions and quantified using ICP-OES. Additional rinsing with 7 M HCl ensured complete removal, resulting in approximately (5.6 ± 0.3)% lutetium elution. Gamma spectrometry was employed to monitor the separation, confirming the absence of hafnium isotopes in the eluate. The 172Lu daughter was allowed to grow on the column, and the generator was eluted with 7 M HCl when necessary. Despite the low 172Hf activity, it remained valuable as a research tool, even with minimal total activity for the generator. The gamma spectra for separation with the first target is shown in Figure 4.7. 79 Figure 4.7: For the first target, the first spectrum is that obtained from a load solution (7 M HCl) fraction, which is representative of other load solution fractions. The next two spectra are the zoomed-in versions of this main spectrum indicating the presence of the 1093.6 keV peak for 172Lu and the absence of the 125.8 keV peak for 172Hf. This indicates a successful separation. 80 For the second target, which had higher activity, 173Lu and 172Lu were utilized to monitor stable lutetium on the HPGe. Approximately (87.3 ± 1.6)% 173Lu and (90.5 ± 1.7)% 172Lu were eluted with the load fractions. Further rinsing with 7 M HCl ensured complete removal, resulting in approximately (10.0 ± 0.5)% 173Lu and (11.6 ± 0.5)% 172Lu elution. Gamma spectrometry confirmed successful separation, with no hafnium isotopes detected in the eluate. Figure 4.8 (a) illustrates the gamma spectrum of the load fractions, demonstrating the absence of hafnium peaks. Figure 4.8 (b) depicts the remaining 172Hf/172Lu generator column after the collection of load and wash fractions. For the elution cycles, Figure 4.9 displays the gamma spectrum of the LN 172Hf/172Lu generator just before the start of elution for cycle 1, revealing no 173Lu peak, indicating the absence of stable lutetium remaining on the column. Approximately (85.0 ± 1.0)% of the 172Lu was captured within three fractions consisting of 2 mL 7 M HCl each, consistently across each cycle, with no observed hafnium breakthrough. Figure 4.10 illustrates the elution profile for the LN resin generator for cycle 2 and cycle 8 of elution. 81 Figure 4.8: (a) A gamma spectrum of the load fraction showing the absence of hafnium peaks after separation was attempted with the LN resin for the second target (b) A gamma spectrum of the LN resin column once the separation is carried out and lutetium is allowed to grow in from the hafnium. 82 Figure 4.9: A gamma spectrum of the 172Hf/172Lu generator just before starting elution for cycle 1 demonstrating no 173Lu peak (at 272 keV). Figure 4.10: Elution profile for different cycles of 172Lu for LN resin is shown. 83 4.3.2.2 ZR resin separation and 172Hf/172Lu setup The ZR resin, based on hydroxamate extractant, exhibits selectivity for hafnium over lutetium. As a result, approximately (98.0 ± 1.8)% of 173Lu and (89.3 ± 4.0)% of 172Lu were eluted with the 7 M HCl load fractions and quantified using HPGe detector. To ensure complete removal of stable lutetium and lutetium isotopes, additional rinsing with 7 M HCl was performed, resulting in approximately (7.6 ± 0.4)% 173Lu and (8.2 ± 0.4)% 172Lu elution. Gamma spectrometry was employed throughout to monitor the separation, confirming the absence of hafnium isotopes in both the load and rinse fractions. At the end of the separation, the matrix was changed to 1 M HCl, and the 172Lu daughter was allowed to grow in on the ZR resin column. The generator was eluted with 0.1 M HCl as needed. Figure 4.11 (a) illustrates the gamma spectrum of the load fractions from the ZR resin separation, demonstrating the absence of hafnium peaks and indicating successful separation. Figure 4.11 (b) depicts the remaining 172Hf/172Lu generator column after the collection of load and wash fractions. Regarding the elution cycles, Figure 4.12 displays the gamma spectrum of the ZR 172Hf/172Lu generator just before the start of elution for cycle 1, revealing no 173Lu peak at 272 keV, indicating the absence of stable lutetium remaining on the column. Approximately (95.0 ± 2.2)% of the 172Lu was captured within the first fraction consisting of 2 - 4 mL of 0.1 M HCl consistently across each cycle, with no observed hafnium breakthrough until cycle 6. Figure 4.13 illustrates the elution profile for the ZR resin generator for cycle 2 and cycle 6, which marked the breakthrough cycle. 84 Figure 4.11: (a) A gamma spectrum of the load fraction showing the absence of hafnium peaks after separation was attempted with the ZR resin for the second target (b) A gamma spectrum of the ZR resin column once the separation is carried out and lutetium is allowed to grow in from the hafnium. 85 Figure 4.12: A gamma spectrum of the ZR 172Hf/172Lu generator just before starting elution for cycle 1 demonstrating no 173Lu peak (at 272 keV). Figure 4.13: Elution profile for different cycles of 172Lu for ZR resin is shown. 86 4.3.2.3 In-house synthesized hydroxamate resin and methyl-substituted hydroxamate resin separation and 172Hf/172Lu setup The two in-house synthesized hydroxamates underwent identical processing procedures. Upon elution with 7 M HCl, (40.9 ± 1.3)% 173Lu and (49.8 ± 1.2)% 172Lu were eluted from the hydroxamate resin, while (41.0 ± 1.0)% 173Lu and (50.7 ± 0.8)% 172Lu were eluted from the methyl-substituted hydroxamate resin. Additional rinsing with 7 M HCl was carried out to ensure the complete removal of stable lutetium and its isotopes, resulting in (53.3 ± 1.7)% 173Lu and (52.4 ± 1.4)% 172Lu elution from the hydroxamate resin, and (31.5 ± 1.3)% 173Lu and (38.6 ± 1.0)% 172Lu elution from the methyl-substituted hydroxamate resin. Throughout the separation process, gamma spectrometry was employed to monitor the separation, indicating the absence of hafnium isotopes in both the load and rinse fractions. The elution matrix was switched to 1 M HCl towards the end of the separation, and 172Lu daughter was allowed to grow in on the hydroxamate resin columns. The generator was then eluted with 0.1 M HCl in successive elution cycles. Figure 4.14 (a) shows the gamma spectrum of the load fractions from the hydroxamate resin separation, the absence of hafnium peaks shows a successful separation. Figure 4.17 (a) shows the same for the methyl-substituted hydroxamate resin. Figure 4.14 (b) shows the 172Hf/172Lu hydroxamate generator that was eluted with 0.1 M HCl. Figure 4.17 (b) shows the methyl-substituted 172Hf/172Lu hydroxamate generator that was eluted with 0.1 M HCl. Figure 4.15 shows the gamma spectrum of the hydroxamate 172Hf/172Lu generator just before starting elution for cycle 1 demonstrating no 173Lu peak (at 272 keV), indicating that no stable lutetium remains on the column. Figure 4.18 shows the same for methyl-substituted hydroxamate generator. Additionally, the elution cycles were done, with (88.0 ± 2.1)% 172Lu 87 elution within the first two fractions for the hydroxamate resin in cycle 2. The elution profiles for subsequent cycles were consistent, with no observed hafnium breakthrough. Similar results were obtained for the methyl-substituted hydroxamate resin with (93.2 ± 1.3)% 172Lu elution for cycle Figure 4.14: (a) A gamma spectrum of the load fraction showing the absence of hafnium peaks after separation was attempted with the hydroxamate resin for the second target (b) A gamma spectrum of the hydroxamate resin column once the separation is carried out and lutetium is allowed to grow in from the hafnium. 88 Figure 4.15: A gamma spectrum of the hydroxamate 172Hf/172Lu generator before beginning elution during cycle 1 demonstrating no 173Lu peak (at 272 keV). Figure 4.16: Elution profile for different cycles of 172Lu for hydroxamate resin is shown. Figure 4.16 shows the elution profile for the hydroxamate resin generator for cycles 2 and 7, and Figure 4.19 shows the elution profile for the methyl-substituted hydroxamate resin 89 generator for cycles 2 and 8. Figure 4.20 summarizes the radiochemical separation scheme for the four resins. Figure 4.17: (a) A gamma spectrum of the load fraction showing the absence of hafnium peaks after separation was attempted with the methyl-substituted hydroxamate resin for the second target (b) A gamma spectrum of the methyl-substituted hydroxamate resin column once the separation is carried out and lutetium is allowed to grow in from the hafnium. 90 Figure 4.18: A gamma spectrum of the methyl-substituted hydroxamate 172Hf/172Lu generator before beginning elution during cycle 1 demonstrating no 173Lu peak (at 272 keV). Figure 4.19: Elution profile for cycles 2 and 8 of 172Lu for methyl-substituted hydroxamate resin is shown. 91 Figure 4.20: A schematic showing the sequence (from left to right) of the radiochemical separation on LN resin, ZR resin, hydroxamate, and methyl-substituted resin packed column. 4.3.2.5 - Comparison, and regeneration of 172Hf/172Lu generators using ZR resin For all the generators set up above the 172Lu eluted activity is more than sufficient to carry out PAC experiments which typically need only a few hundred kBq for a study and to test radiolabeling parameters for development of new tracers. One disadvantage of LN resin in this regard is the elution of 172Lu in higher molarity HCl which requires an added step of dilution or evaporation and reconstitution in lower molarity HCl to use it for an application like radiolabeling. Some comparative features are listed in Table 4.2. 92 After passing the Hf-oxalic acid solution through QMA resin, and obtaining the Hf in HCl matrix, ZR resin was selected out of the four generators to be made into a short-term, high activity generator. Unlike LN resin having to use high molarity HCl to elute 172Lu, ZR resin eluted 172Lu in 0.1 M HCl, which is a more preferable matrix to perform radiolabeling studies without having to dilute to a lower molarity or evaporate to dryness to be reconstituted in another matrix. Compared to the two in-house hydroxamate based synthesized resins, the ZR resin was selected as a 2 mL cartridge that can be readily purchased making preparations of a generator routine. Following construction of the high activity generator, the elution volume was reduced to 2.1 mL 0.1 M HCl as a majority of the 172Lu is eluted in the first fraction as shown for the hydroxamate based resins. Specific activity was performed using a modified method from Holland et. al following the concentration of 172Lu using an Agilent Bond Elute 1 mL SCX column with a modified method from Domnanich et al. [12], [25]. The specific activity was determined to range from 19 to 51 μCi/nmol. Figure 4.21 shows the elution profile for cycle 1 for the higher activity ZR 172Hf/172Lu generator. 93 Table 4.2: A comparison is drawn between the different resins with respect to volume for elution and 172Hf breakthrough. Feature LN resin ZR resin methyl- hydroxamate substituted resin hydroxamate resin Volume for Broad elution eluting generators profile, 6 - 8 mL First 2 - 4 mL First 2-4 mL First 2-4 mL 172Hf breakthrough No breakthrough Cycle 6 of eluting No breakthrough No breakthrough Figure 4.21: An elution profile for different cycles of 172Lu for the higher activity ZR resin is shown. 4.3.2.6 Comparison of these generators with those in Chapter 3 The generators set up in this work exhibit high radiochemical purity as compared to the generator set up from the beam-blocker. The generators set up from the beam-blocker work are 94 good for PAC studies, but, the generators set up here are good for radiolabeling studies as well, and also provide higher 172Lu activity in one elution. Nevertheless, the generators from the isotope harvesting technique met the requirements where high purity was not needed. 4.4 Conclusion In this work, radiochemical methodologies were established for the extraction of 172Hf from stable lutetium and lutetium isotopes in the hydrochloric acid media and four generator systems with commercially available LN, ZR, and in-house synthesized hydroxamate and methyl-substituted hydroxamate were studied. In all cases, the ease of separation, elution, and the long shelf-life of the 172Hf/172Lu generator systems makes it ideal for experiments where pure lutetium isotope is needed and procuring 177Lu is not always feasible. Generators were compared on parameters like volume for elution and 172Hf breakthrough. This endeavor of obtaining pure 172Lu finds abundant applicability in PAC studies, radiolabeling and in Hf-Lu isotope geochemistry studies. 95 REFERENCES J. Strosberg et al., “Phase 3 Trial of 177Lu-Dotatate for Midgut Neuroendocrine [1] Tumors,” New England Journal of Medicine, vol. 376, no. 2, pp. 125–135, Jan. 2017, doi: 10.1056/nejmoa1607427. “National Nuclear Data Center.” Accessed: Mar. 22, 2024. [Online]. Available: [2] https://www.nndc.bnl.gov/nudat3/ [3] A. Hermanne, S. Takacs, M. B. Goldberg, E. Lavie, Y. N. Shubin, and S. Kovalev, “Deuteron-induced reactions on Yb: Measured cross sections and rationale for production pathways of carrier-free, medically relevant radionuclides,” Nucl Instrum Methods Phys Res B, vol. 247, no. 2, pp. 223–231, Jun. 2006, doi: 10.1016/j.nimb.2006.03.008. F. Tárkányi et al., “Activation cross sections of proton induced nuclear reactions on [4] ytterbium up to 70 MeV,” Nucl Instrum Methods Phys Res B, vol. 267, no. 17, pp. 2789–2801, Sep. 2009, doi: 10.1016/j.nimb.2009.05.075. S. Santos, S. De Barros, and J. C. Suita, “Cross sections and thermonuclear reaction rates [5] for 181Ta(a,n)185Re, 169Tm(a,n)172Lu, 191Ir(a,n)194Au and 197Au(a,n)200Tl,” J. Phys. G: Nucl. Part. Phys, vol. 26, pp. 301–308, 2000. S. K. Das, A. G. C. Nair, R. K. Chatterjee, R. Guin, and S. K. Saha, “The performance of [6] a new 172Hf-172Lu generator,” Applied Radiation and Isotopes, vol. 47, no. 7, pp. 643–644, 1996, doi: https://doi.org/10.1016/0969-8043(96)00025-5. P. M. Grant, G. E. Montero, A. M. Newman, and H. A. O’Brien, “First use of millicurie [7] levels of172Hf−172Lu in the industrial sector,” J Radioanal Nucl Chem, vol. 96, no. 6, pp. 629– 633, 1985, doi: 10.1007/BF02165043. T. Venkova et al., “Pseudo-spin band in the odd-odd nucleus 172Lu,” Eur Phys J A [8] Hadron Nucl, vol. 18, no. 1, pp. 1–4, 2003, doi: 10.1140/epja/i2003-10024-4. [9] R. J. Daniels, P. M. Grant, and H. A. O’Brien, “The production, recovery, and purification of 172Hf for utilization in nuclear medicine as the generator of 172Lu,” Int J Nucl Med Biol, vol. 5, no. 1, pp. 11–17, 1978, doi: 10.1016/0047-0740(78)90085-2. [10] P. M. Grant, R. J. Daniels, W. J. Daniels, G. E. Bentley, and H. A. O’ Brien, “The generator production of 172Lu from 172Hf for limited nuclear medicine research,” 1983. J. A. Dadakhanov et al., “172Hf → 172Lu Radionuclide Generator Based on a Reverse- [11] Tandem Separation Scheme,” Radiochemistry, vol. 60, no. 4, pp. 415–426, Jul. 2018, doi: 10.1134/S1066362218040112. J. P. Holland, Y. Sheh, and J. S. Lewis, “Standardized methods for the production of high [12] specific-activity zirconium-89,” Nucl Med Biol, vol. 36, no. 7, pp. 729–739, Oct. 2009, doi: 10.1016/j.nucmedbio.2009.05.007. 96 [13] W. E. Meijs et al., “Production of highly pure no-carrier added 89Zr for the labelling of antibodies with a positron emitter,” Applied Radiation and Isotopes, vol. 45, no. 12, pp. 1143– 1147, 1994, doi: https://doi.org/10.1016/0969-8043(94)90029-9. [14] L. Gajecki, C. M. Marino, C. S. Cutler, and V. A. Sanders, “Evaluation of hydroxamate- based resins towards a more clinically viable 44Ti/44Sc radionuclide generator,” Applied Radiation and Isotopes, vol. 192, Feb. 2023, doi: 10.1016/j.apradiso.2022.110588. I. Verel, G. W. M. Visser, R. Boellaard, M. S. Walsum, G. B. Snow, and G. A. M. S. van [15] Dongen, “89Zr Immuno-PET: Comprehensive Procedures for the Production of 89Zr-Labeled Monoclonal Antibodies,” Journal of Nuclear Medicine, vol. 44, no. 8, p. 1271, Aug. 2003, [Online]. Available: http://jnm.snmjournals.org/content/44/8/1271.abstract [16] “ZR resin.” Accessed: Mar. 22, 2024. [Online]. Available: https://www.triskem- international.com/scripts/files/61eef0f5e9fa69.45400285/PS_ZR-Resin_EN_210908.pdf [17] P. Horwitz and D. Mcalister, “Eichrom’s LN Series of Resins: Characterization and Novel Applications,” 2008. [18] Y. heng Yang, H. fu Zhang, Z. yin Chu, L. wen Xie, and F. yuan Wu, “Combined chemical separation of Lu, Hf, Rb, Sr, Sm and Nd from a single rock digest and precise and accurate isotope determinations of Lu-Hf, Rb-Sr and Sm-Nd isotope systems using Multi- Collector ICP-MS and TIMS,” Int J Mass Spectrom, vol. 290, no. 2–3, pp. 120–126, Feb. 2010, doi: 10.1016/j.ijms.2009.12.011. [19] I. C. Kleinhanns, K. Kreissig, B. S. Kamber, T. Meisel, T. F. Nägler, and J. D. Kramerst, “Combined chemical separation of Lu, Hf, Sm, Nd, and REEs from a single rock digest: Precise and accurate isotope determinations of Lu-Hf and Sm-Nd using multicollector-ICPMS,” Anal Chem, vol. 74, no. 1, pp. 67–73, Jan. 2002, doi: 10.1021/ac010705z. [20] M. Bizzarro, J. A. Baker, and D. Ulfbeck, “A new digestion and chemical separation technique for rapid and highly reproducible determination of Lu/Hf and Hf isotope ratios in geological materials by mc-ICP-MS,” Geostandards Newsletter, vol. 27, no. 2, pp. 133–145, Jul. 2003, doi: 10.1111/j.1751-908X.2003.tb00641.x. J. N. Connelly, “Improved dissolution and chemical separation methods for Lu-Hf garnet [21] chronometry,” Geochemistry, Geophysics, Geosystems, vol. 7, no. 4, Apr. 2006, doi: 10.1029/2005GC001082. [22] B. Le Fèvre and C. Pin, “A straightforward separation scheme for concomitant Lu-Hf and Sm-Nd isotope ratio and isotope dilution analysis,” Anal Chim Acta, vol. 543, no. 1–2, pp. 209– 221, Jul. 2005, doi: 10.1016/j.aca.2005.04.044. [23] D. G. Medvedev, L. F. Mausner, G. A. Greene, and A. L. Hanson, “Activation of natural Hf and Ta in relation to the production of 177Lu,” Applied Radiation and Isotopes, vol. 66, no. 10, pp. 1300–1306, 2008, doi: https://doi.org/10.1016/j.apradiso.2008.02.090. 97 J. F. Ziegler, M. D. Ziegler, and J. P. Biersack, “SRIM - The stopping and range of ions [24] in matter (2010),” Nucl Instrum Methods Phys Res B, vol. 268, no. 11–12, pp. 1818–1823, Jun. 2010, doi: 10.1016/j.nimb.2010.02.091. [25] K. A. Domnanich et al., “Production and separation of 43Sc for radiopharmaceutical purposes,” EJNMMI Radiopharm Chem, vol. 2, no. 1, p. 14, 2017, doi: 10.1186/s41181-017- 0033-9. 98 APPENDIX 172Hf PRODUCTION AND 172Hf/172Lu CHEMISTRY DEVELOPMENT ICP-OES measurement wavelengths for lutetium and hafnium For the ICP-OES analysis, the emission lines which had the highest intensity for Lu (261.542, 291.139 nm), and Hf (232.247, 277.336, 264.141 nm) were chosen for analysis. Figure B.1: A comparison is shown here between data from TALYS-2019 from Tendl library and FLUKA code values for the production of 172Hf and 175Hf when lutetium foil is irradiated with protons. Image provided by Dr. Dmitri Medvedev. FLUKA data generated by Dr. Dohyun Kim. Separation chemistry development with stable elements To develop a procedure for the separation of hafnium from stable lutetium and other radioisotopes, hafnium and lutetium solutions in the HCl medium were prepared using commercially available Hf and Lu ICP-OES standards and from dissolution of Lutetium (III) chloride hexahydrate. The methodology was developed to separate hafnium tracer from 1.5 g 99 stable lutetium in the hydrochloric acid medium to mimic the radioactive conditions, and loading 50 - 60 mL of 7 M HCl on the LN resin. Separations were performed with the peristaltic pump after calibrating the flow rate and 10 - 13 mL fractions were collected. Aliquots from each fraction was withdrawn and appropriate dilutions in 2% HNO3 were prepared for measurement with ICP- OES. Standards in various dilutions (0.01 ppm - 10 ppm) were used to establish the calibration curves for the sample analysis. The separation efficiency was determined as expressing the total eluted amount in all fractions as a percentage of the loaded amount for each element. Table B.1: The gamma lines used for tracking the radionuclides not mentioned in the main text [2]. Radioisotope 173Hf 174mLu 177Lu Half-life 23.6 h 142 d 6.6 d Gamma lines (Intensity) 123.7 keV (83.0 %), 296.9 keV (33.9 %), 139.6 keV (12.7 %) 67.058 keV (7.25), 44.6 keV (12.4 %) 208.4 keV (10.4 %), 112.9 keV (6.23 %) 100 Figure B.2: For the first target, the first spectrum is obtained from the elute fraction of 0.5 M Oxalic acid. The next two spectra are the zoomed-in versions of this main spectrum indicating the presence of the 1093.63 keV peak for 172Lu generated from 172Hf and the presence of the 125.812 keV peak for the elution of 172Hf. 101 CHAPTER 5: 7Be HARVESTING: PRODUCTION AND SEPARATION FROM BORON AND BEAM DELIVERY The cosmogenic radionuclide 7Be was produced by the irradiation of natural boron and (10B) enriched boron pellet targets. Following this, the targets underwent dissolution in nitric acid, after which 7Be was separated from bulk boron via the macroporous cation-exchange chromatographic resin AG-MP 50. Six such separations were carried out from 7 irradiations and an average recovery rate of (99.4 ± 3.7)% was achieved. Source preparation was then carried out form the isolated 7Be before being introduced into an ion source at NSCL/FRIB called the Batch- Mode Ion Source. The resulting 7Be beam was then delivered to end users at the NSCL, maintaining an average delivery rate of 5 x 105 particles per second. 5.1 Introduction Numerous proposed experiments conducted in radioactive ion beam facilities necessitate the utilization of radioactive source materials, such as the cosmogenic radionuclide 7Be. This proton-rich isotope of beryllium possesses a half-life of 53.2 days and emits a distinct gamma- ray with 477.6035(20) keV and an intensity of 10.44(4)% [1]. One notable application of this involves researchers needing 7Be for investigating the impact of weakly-bound nuclei structures on nuclear fusion reaction outcomes. Additionally, 7Be beams are of interest in astrophysics to solve the ‘cosmological lithium problem’. The abundance of 7Be at the time of its production and its subsequent decay rate influence the amount of 7Li that is ultimately produced during Big Bang Nucleosynthesis (BBN). Therefore, understanding the processes related to 7Be and its interactions with other nuclei is important for accurately predicting the abundance of 7Li and addressing the cosmological lithium problem [2]. 102 In order to provide the 7Be beam to experimenters, a source of 7Be was required that could be inserted into an offline ion source, at the NSCL/FRIB called the Batch-Mode Ion Source (BMIS) [3]. The ion source facilitates the stand-alone operation of the ReAccelerator (ReA) as well as stopped-beam instrumentation at the NSCL/FRIB [4]. The BMIS consists of a sample oven connected to a VD5 ion source which is the Versatile Arc Discharge Ion Source (VADIS) based on electron impact ionization [5]. In the oven, samples of stable isotopes and radioisotopes with relatively long half-lives can be placed [3], [6]. The samples are then heated and atomized in the oven and effuse into the plasma ion- source, where they get ionized, accelerated and delivered as beams to various user end-stations [3], [6]. The temperature of the oven is adjusted in accordance with the volatility of the introduced compounds. In the BMIS operation, nitrogen trifluoride (NF3) is often used as a reaction gas to create fluorides for lower temperature release of non-volatile substances and to decrease their surface adsorption like for beryllium. Figure 5.1 shows the location of BMIS in the NSCL building and the BMIS module. The production of the 7Be radioisotope commonly involves several methods, including proton irradiation of a lithium target (7Li(p,n)7Be), followed by chemical separation techniques to isolate 7Be, as outlined by [7], [8], [9], [10]. Additionally, alternative methods such as photonuclear reactions and sequential charged particle reactions induced by neutron bombardment have been employed for 7Be production [11], [12]. However, for the specific purpose of generating source material for extracting a radioactive beam, the preferred approach was the production of 7Be via proton bombardment of a boron target through the 10B(p,α)7Be 103 nuclear reaction [10], [13]. This method is favored as it eliminates the introduction of 7Li, which could serve as an isobaric impurity contaminating the delivered beam [13]. Figure 5.1: (Left) A schematic of the NSCL building is shown with the BMIS area highlighted [36] (Right) The BMIS is shown where the oven is the cylinder in the middle [6]. Both the 7Be isotope and the 10B(p,α)7Be reaction pathway have found application in diverse fields, spanning environmental radiochemistry, nuclear astrophysics encompassing cross- section analyses, plasma physics, boron-related reactions in fusion reactors, and materials science [8], [12], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27]. Despite the extensive investigation into nuclear reactions, there is a noticeable dearth of references addressing the chemical separation of 7Be from irradiated boron. One potential approach to purify 7Be from boron involves cation-exchange chromatography in nitric acid media, leveraging favorable distribution coefficients [28]. Consequently, a separation method based on cation-exchange resin was devised for the present study. This study details the proton irradiation of natural boron (consisting of both 11B and 10B isotopes) and 10B-enriched pellet targets, followed by a chemical separation technique to isolate 7Be from the boron targets using the macroporous cation-exchange AG-MP 50 resin. Conducted at the radiochemistry facility of the University of Alabama at Birmingham (UAB), this method 104 purified 7Be, which was subsequently transferred to MSU. At MSU, appropriate source samples were prepared and transferred into the BMIS. Successful extraction and delivery of 7Be beams to end users were achieved. Additionally, the study provides insights into the overall yields of production, separation, and BMIS source performance. 5.2 Experimental 5.2.1 Cyclotron production of 7Be at UAB Materials: For the preparation of the boron pellet targets, the materials obtained were natural amorphous boron powder (19.9% 10B and 80.1% 11B) from Sigma Aldrich (15580, ≥ 95% boron amorphous powder) and 10B enriched boron powder from Cambridge Isotopes Laboratories, Inc. (10B, 98% chemical purity). The “coin” target holder and proton energy degrader were made of tantalum (tantalum sheet, 3 mm thickness, 3N8 purity, KND6098, ESPI metals) and were cleaned with acetone (OptimaTM, A929SK-4, Fisher Chemical). The chemical processing was performed with nitric acid (HNO3, 65%, Suprapur) and hydrochloric acid (HCl, 37 wt. % in H2O, 99.99% Trace metals basis, Sigma-Aldrich). MilliQ water (Thermo Scientific MicroPure Ultrapure Water System, 18.2 MΩ cm) was used to prepare all solutions. Other reagents used during the method development stage include Beryllium ICP Standard (1000 ppm, <5% HNO3, GFS Chemicals), Boron ICP Standard (1000 ppm, <5% HNO3, GFS Chemicals), metallic boron foil (3 mm thick, hot pressed, 99.6%, GoodFellow Cambridge Limited), and nitric acid (HNO3, ACS Plus, 15.8 M, Fisher Chemical). Instruments: The 7Be was detected and quantified with a Canberra S5000 High-Purity Germanium Detector and the analysis of spectra was performed with Canberra Genie 2000 software. At UAB, an inductively coupled plasma mass spectrometer (ICP-MS) (Agilent Technologies 7800 ICP-MS) and Agilent software, ICP-MS MassHunter v4, was used for 105 quantifying the amount of stable boron in the collected fractions from the radiochemical separations. For source preparation efforts at MSU Chemistry laboratory, 7Be was detected and quantified with an HPGe Canberra BEGe γ-ray Detector (BE2020). The determination of boron and beryllium during the separation methodology development at MSU was performed with an Agilent inductively coupled plasma-optical emission spectrometer (Agilent 5900 SVDV ICP- OES) and analyzed with Agilent ICP software. 5.2.1.1 Target Preparation Figure 5.2: Closed view of the tantalum coin with enclosed boron pellet having the tantalum degrader and retaining ring at the top is shown in (a) and (b) shows the open view of the coin containing the pressed boron pellet (c) schematic of Type 2 targets in shown. Two categories of targets, referred to as Type 1 and Type 2, were designed for the irradiation process. Type 1 targets comprised approximately 100-102 mg of natural amorphous boron powder, that has an average particle size of less than 1.0 micron. This boron powder was compacted into 1-mm-thick circular pellets with a 10-mm diameter using a hydraulic press (model 3664, Carver, Inc.) and a 10-mm die (FTIR Evacuable Pellet Die, GS03100, Specac). The press applied incremental force, ranging from 1 to 5 tons, with each ton maintained for one minute before reaching 5 tons, which was then sustained for 5 minutes. These pellets were then 106 placed within the 1.5-mm-deep indentation of a 2-mm-deep tantalum coin (as illustrated in Figure 5.2 (b)). A tantalum foil, 0.25-mm thick and cut to a 10-mm diameter using a punch and die set (Precision Brand Products), was placed atop the boron pellet. Functioning as a degrader, the tantalum foil was secured in position by a single-turn spiral internal retaining ring (3/8” OD in 18-8 Stainless steel, McMaster-Carr) inserted around its circumference (as depicted in Figure 5.2 (a)). The tantalum coin itself, with a thickness of 0.5 mm, served as the backing material for the target after subtracting the divot's depth from the coin's total thickness. In total, three Type 1 targets were prepared. Type 2 targets were prepared using 199-204 mg of 10B-enriched powder, which was compacted into a 1.8-mm-thick circular pellet with a 10-mm diameter. This pellet was then inserted into the 2.5-mm-deep indentation of a 3-mm-deep tantalum coin. The tantalum degrader and the thickness of the tantalum backing remained consistent with those of the Type 1 targets. A total of four Type 2 targets were fabricated and a schematic is shown in Figure 5.2 (c). The tantalum coins utilized for both Type 1 and Type 2 targets underwent machining processes at UAB. 5.2.1.2 Irradiation at UAB The pressed Type 1 and 2 targets underwent irradiation at the TR-24 Cyclotron, located within the UAB Cyclotron Facility. The proton beam extracted for irradiation had an energy of 17.5 MeV. In total, seven irradiation sessions were conducted, with three involving Type 1 targets and four involving Type 2 targets. These targets were inserted into the coin holder of the solid cyclotron target station at UAB [29]. The front side of the target underwent helium cooling, while the back side was water- cooled. Irradiation durations ranged from 30 minutes to 7 hours, with beam currents varying 107 between 5 and 10 µA. Subsequently, the bombarded targets were left within the target station for a minimum of 4 hours to allow for the decay of the co-produced 11C, which has a half-life of 20.36 minutes. 5.2.1.3 Experimental production yield of 7Be Following the bombardment process, the dissolved irradiated targets were analyzed using an HPGe Detector and assessed using the Canberra Genie 2000 software to quantify 7Be. This involved detecting and integrating peaks based on the distinctive γ-ray emission, conducting background subtraction, and calculating activity based on the net peak area. Energy and efficiency calibrations were made utilizing a 1 mL mixed nuclide source (manufactured by Eckert & Ziegler) enclosed in a sealed 1.5 mL micro centrifuge vial, positioned at distances of 5 and 25 mm away from the detector face. To ensure precise quantification, samples from each collected fraction of the radiochemical separations were pipetted into 1.5 mL Eppendorf screw- cap conical centrifuge tubes, diluted with MilliQ water to reach a volume of 1 mL, and subjected to γ-ray spectroscopy. The distance from the detector face was selected based on the sample's activity to maintain a dead time below 5% for all measurements. 108 5.2.2 Separation of 7Be from boron pellet targets 5.2.2.2 Dissolution of target Figure 5.3: The stepwise dissolution process of the boron pellet targets is shown leading to pure boric acid powder in the end. All target dissolutions were conducted within a fume hood. The disassembly of the target involved the removal of the retaining ring and the tantalum degrader. Subsequently, the boron pellets were carefully extracted from the divot using a spatula and transferred into a beaker containing 40-60 mL of concentrated HNO3 (Suprapur). If a significant amount of boron powder remained adhered to the tantalum coin, the coin itself was immersed in the nitric acid solution and removed once the powder had completely dissolved. The solution was then heated to 100°C for a duration of 10 to 12 hours until the boron pellet dissolved entirely. Upon dissolution, the nitric acid solution was allowed to evaporate at 120°C, resulting in the formation of white boric acid powder residue (Figure 5.3). This residue was subsequently reconstituted in 0.1 M HNO3 to achieve a concentration of 1 mg of boron per mL, serving as the loading solution for the separation process. An aliquot was withdrawn from this target solution, and the activity of 7Be was determined through gamma spectroscopy. 5.2.2.1 Development of the separation methodology with stable elements To refine the separation procedure, various conditions were tested with stable elements, involving the dissolution of boron powder ranging from 10.8 - 538 mg in 10-75 mL of 109 concentrated HNO3, heated to 100°C. Complete evaporation at 120°C resulted in boric acid formation, which was then dissolved in 0.1 M HNO3. Following the addition of 1 mg of Be tracer, the separation of boron and beryllium was carried out in 0.1 M HNO3 using an AG MP-50 column, slurry packed with 1.5 - 2.5 grams of resin. Before separation, the column underwent pre-conditioning with 20 mL MilliQ water, followed by sequential washes with 20 mL 2 M HNO3 and 20 mL 0.1 M HNO3, all at a flow rate of 1 mL/min. Once the 0.1 M HNO3 load solution passed through the column at 0.8 mL/min, the column was rinsed with 1 M HNO3, and Be was eluted using 4 M HNO3 at a constant flow rate of 0.8 mL/min. To evaluate separation efficiency, aliquots of each eluted fraction were withdrawn and diluted in 3% (v/v) HNO3 to volumes of 5-10 mL. The quantities of stable boron and beryllium were determined by ICP-OES using calibration curves generated from standard solutions containing Be and B in the concentration range of 0.1 - 50 ppm. The percentage of eluted boron and beryllium was calculated by dividing the amount of eluted element by the initial amount present in the loading solution for all cold separations. This optimized separation methodology effectively removed bulk boron and was employed to process the irradiated targets. 5.2.2.3 Radiochemical separation The radiochemical separations were conducted employing a peristaltic pump (Ismatec® Reglo Peristaltic Pump with independent channel control, utilizing pump tubing with a 1.65- millimeter internal diameter). Before each separation, the pump underwent calibration to ensure a consistent flow rate. The columns were assembled by slurry-packing the cation-exchange resin AG MP-50 (100−200 mesh, in H+ form, BioRad) into transparent polycarbonate tubes measuring approximately 15 cm in length (with a 3/8″ outer diameter and 1/4″ inner diameter, obtained 110 from McMaster-Carr, PN:9176T1). Glass wool inserts were placed at both ends of the columns. Push-to-connect fittings consisting a poly(butylene terephthalate) (PBT) body (Pneumatic NITRA Union Reducer, 3/8″ to 1/4″) were used to link one end of the column to the loading solution and the other end to a peristaltic pump using clear Tygon tubing. For each separation, a fresh column was prepared, resulting in a total of six columns (with an outer diameter of 9 mm, inner diameter of 6 mm, and length ranging from 15-18 cm), each containing 2.5 g AG MP-50 resin. Pre-conditioning of the columns involved passing 25 mL MilliQ water, followed by 20 mL 2 M HNO3, and then 20 mL 0.1 M HNO3 sequentially, at a flow rate of 1 mL/min. The 0.1 M HNO3 load solutions containing boric acid and 7Be were passed through the columns at a flow rate of 0.8 mL/min, with fractions collected. Residual boron was washed from the column using 10 mL 0.1 M HNO3 and collected in subsequent fractions. Further rinsing was achieved through sequential use of 6 mL 1 M HNO3. Finally, 7Be was eluted in 3-4 fractions, each containing 8-9 mL 4 M HNO3. A final rinse with 10-13 mL MilliQ water was performed to reduce acidity and ensure suitable storage conditions for the equipment. This entire procedure was repeated six times for seven irradiated boron targets. In one instance, two targets containing 203.5 mg and 200.5 mg of boron powder, respectively, were 111 combined to assess the reproducibility of the separation method by scaling up the boron quantity. The separation methodology utilizing AG MP-50 resin in this study is illustrated in Figure 5.4. Figure 5.4: A schematic of the entire separation process for boron and beryllium on AG-MP 50 is shown. 5.2.2.4 Quantification with ICP-MS At the end of the separation process for the combined target, 5 and 7, ICP-MS samples were prepared for measurement at UAB for quantification of 10B and 11B. Due to the high levels of radioactivity of 7Be, the elute fractions were not analysed. Multi-element standards in 2% HNO3 was used to create a calibration curve of 50, 100, 200, 500, 800 and 1000 ppb for both 10B and 11B. All ICP-MS samples were made in triplicates. For the load fractions, two different dilutions for ICP-MS sample were prepared in 2 % HNO3: 1:1200 and 1:1400 dilution. For the rinse fractions and water fraction, a 1:2 and 1:4 dilutions were made in 2 % HNO3. 112 5.2.3 7Be in BMIS 5.2.3.1 Source preparation A total of four source samples containing 7Be activities were prepared. From the separations conducted using Type 1 targets, the fraction of 4 M HNO3 displaying the highest 7Be activity was identified and utilized for two separate source sample preparation endeavors. From the separations using Type 2 targets, the fractions containing 7Be within the 4 M HNO3 were combined, subjected to evaporation until dryness at temperatures ranging from 150-200 °C, and then reconstituted in 0.1 M HCl before being divided between two micro-centrifuge vials. Upon transportation to the MSU radiochemistry laboratory, gamma spectroscopic measurements were conducted with each Eppendorf vial, after which this activity was further concentrated through evaporation. The concentrated solution was then pipetted onto a 1x1 cm2 tantalum foil, which measured 6 µm in thickness (99.9% purity, Goodfellow). This process involved repeatedly pipetting 10-12 µL droplets onto the foil while heating it to temperatures of 150-200 °C using a hot plate. Subsequently, these tantalum foil source samples, with the 7Be deposited, were then transported to NSCL. 5.2.3.2 7Be Beam delivery The tantalum foils having 7Be deposits underwent a tight folding process before their insertion into BMIS oven. Typically, one foil sample was introduced into the BMIS oven during each operation. The oven was resistively heated to temperatures ranging from 700 - 1800 °C. NF3 gas served as a reactive carrier gas, introduced into the oven to facilitate the conversion of 7BeO, a refractory compound, into 7Be-Fx adduct molecules possessing higher vapor pressure. Subsequently, the evaporated sample materials effused through a tantalum transfer tube into the 113 VD5 ion source, where the molecules underwent ionization, primarily being extracted as [7Be- Fx]+ molecular ions. Control over the 7Be ion beam rate was achieved by regulating the oven temperature and NF3 gas flow (partial pressure). The extracted ion beam, comprising the 7Be molecular ions alongside other stable contaminants, was accelerated to 30 keV and sent to the magnetic mass separator, where the desired 7Be molecular ions were selectively isolated. Eventually, the selected ion beam was transported to the stopped-beam area (LEBIT) or the ReA accelerator post-charge breeding to the 4+ atomic charge state. Four 7Be sources were employed to provide beams for five experiments. 5.3 Results and discussion 5.3.1 Theoretical production yield determination of 7Be For the calculation of the theoretical production yields before the irradiations were carried out, the Stopping and Range of Ions in Matter (SRIM) compilation was used to determine the entrance and exit proton energies in a boron target, using the measured density of (1.40 ± 0.3) g/cm3 (Table C.1 and C.2). According to the SRIM compilation, for Type 1 targets, the proton beam entered the boron target after passing through the tantalum degrader with 11.5 MeV energy and left the boron pellet with 5.8 MeV energy (Figure 5.5). The proton beam was finally stopped in the first 0.1 mm of the 0.5 mm Ta backing of the coin. The experimental (p,α) reaction cross- section data for 10B target nuclei were retrieved from the Experimental Nuclear Reaction Data database over the 0.5 - 12 MeV proton energy range, while the theoretical reaction cross-section data was computed with the TALYS-2019 code and retrieved from the TENDL library (Figure 5.6). The production calculations were made using Equation 5.1 below and the integral was approximated by numerical integration over the energy range of 11.5 - 5.8 MeV. 114 𝑬𝒇 𝑷 = ∫ 𝑬𝒊 𝝈(𝑬)𝑵𝑨𝝆 𝑺𝑷(𝑬) ∗ 𝑴 𝒅𝑬 (5.1) P is the production rate of 7Be (atoms/proton), σ is the energy-dependent reaction cross section (cm2), NA is the Avogadro constant (6.02 x 1023 atoms/mol), ρ is the density of 10B, dE is the difference in energy (MeV), SP is the stopping power of the beam in the target material (MeV/cm) and M is the molar mass of the target material (g/mol). Equation 5.2 was used to compute the produced activity of 7Be over the irradiation time where A is the activity of produced 7Be (Bq/μA*h), P is the production rate of 7Be(atoms/proton), obtained from Equation 5.1, I is the beam current (protons/s), λ is the decay constant (s-1) and t is the irradiation time (s). 𝑨 = 𝑷 ∗ 𝑰(𝟏 − 𝒆−𝝀𝒕) (5.2) Figure 5.5: A schematic of the 17.5 MeV proton beam traversing the tantalum degrader, boron pellet, and finally stopping in the tantalum coin backing is shown. 115 Figure 5.6: Cross-section versus energy for the 10B(p,α)7Be nuclear reaction is shown where the experimental data was chosen from Kalinin, 1957 and is largely comparable to the data Figure 5.5: A schematic of the 17.5 MeV proton beam traversing the tantalum degrader, set obtained from the TALYS-2019 code [37], [38]. shown. 5.3.2 Proton irradiation of boron targets Table 5.1: The parameters for the irradiations of boron target Type 1 and Type 2 and the experimental yields are given. *note the targets for irradiations 5 and 7 were combined for the measurement of activity. Irradiation Target type Target Irradiation Time (h) 1 2 3 4 5 6 7 Type 1 natB Type 1 Type 1 natB natB Type 2 10B Type 2 10B Type 2 10B Type 2 10B 5+7 (combined)* Type 2 10B 0.5 0.5 1 3 6 4 7 116 Proton Beam current (μA) μAh Total 7Be MBq (μCi) experimental 5 2.5 1.3 ± 0.03 (33.7) 10 5 2.3 ± 0.1 (62.5) 10 10 4.8 ± 0.1 (129.1) 5 6 6 6 15 44.5 ± 0.3 (1203.8) 36 24 72.9 ± 0.5 (1969.5) 42 78 217.4 ± 1.3 (5876.2) The boron targets underwent irradiation using proton beam currents ranging from 5-10 µA. During this process, noticeable scorching or burn marks were detected on the second and third targets when subjected to a beam current of 10 µA. As a result of this observation, adjustments were made to lower the beam currents for subsequent irradiations. Detailed parameters for all irradiations involving both Type 1 and Type 2 targets can be found in Table 5.1. For these experiments, another nuclear-reaction channel was 11B(p,n)11C, with a theoretical production rate of 7.74 GBq 11C/µAh, making it the only other radioisotope with a half-life greater than 1 minute. The resulting 11C was allowed to decay prior to target processing. Notably, the contribution of the 11B(p,αn)7Be nuclear reaction to the 7Be production was considered negligible, given its minimal cross-section below the 11.2 MeV threshold [30], [31]. The values obtained for the yields are roughly 5 x 10-1 MBq 7Be/μAh for type 1 targets, and 3 x 100 MBq 7Be/μAh for Type 2 targets. These irradiations were not designed for precise determination of nuclear data but rather practical 7Be production. In the above cases, it was not possible to determine if all of the delivered beam current actually interacted with the boron target because the entire target body, including the retaining ring, was conductive. This is why the experimental yield numbers are not meaningful to compare to theoretical yields. A gamma spectrum of the assembled target coin # 6 after irradiation is shown in Figure 5.7, which is representative of other irradiated targets. The presence of 56Co is due to the activation of the retaining ring, which can be seen in another gamma spectrum of only the retaining ring and tantalum foil, shown in the Figure 5.8. Through the γ-spectroscopic measurements, the total activity of 7Be measured in seven targets was 343.2 ± 1.4 MBq. 117 Figure 5.7: The gamma spectrum of the assembled target #6, counted for 30 minutes is shown where the 56Co peaks are due to the activation of the retaining ring. Figure 5.8: The gamma spectrum of the retaining ring and tantalum foil minus the boron pellet is shown. 118 5.3.3 Separation of 7Be from natB and enriched 10B targets Earlier research has shown that elemental boron can be dissolved through treatment with nitric acid, either with or without the addition of hydrogen peroxide [32]. In this study, the use of hot concentrated nitric acid alone led to the complete dissolution of both natural and enriched boron pellets. It's worth noting that the dissolution process involving boron and nitric acid produced brown NOx fumes, emphasizing the necessity for adequate ventilation during such procedures. Subsequent evaporation and reconstitution of the targets using 0.1 M nitric acid proved to be straightforward in all cases. Table 5.2: Distribution coefficient values obtained for Be(II) in the nitric acid medium is shown [28]. 0.1 M HNO3 0.2 M 0.5 M 1.0 M 2.0 M 3.0 M 4.0 M Be(II) 553 183 52 14.8 6.6 4.5 3.1 For the seven irradiations outlined in Table 5.1, a total of six separation attempts were undertaken. To evaluate the reliability of the separation procedure, targets 5 and 7 were combined prior to chemical processing. The separation relied on the fact that boron, as boric acid, does not absorb onto cation exchange resin AG MP-50, whereas beryllium is strongly adsorbed by cation exchange resin from low-molarity nitric acid, as shown by the distribution coefficient (Kd) values in Table 5.2 [28], [33]. During the passage of load solutions through the columns, boron was completely eluted in every separation, with no detectable breakthrough of 7Be, even when the boron content was increased (approximately 404 mg) by combining targets 5 and 7. Additionally, there was no discernible difference in elution efficiency between natural boron and enriched 10B. 119 For the combined targets (5 and 7), the ICP-MS analysis yielded a calculated mass of (359.8 ± 2.2) mg for 10B in the load fractions where initially, the total mass of targets 5 and 7 was 404 mg. To ensure the complete elution of boron, additional rinse fractions containing 0.1 M HNO3 were employed, followed by a subsequent increase in molarity to 1 M HNO3, guaranteeing the thorough removal of boron while preserving the adsorption of 7Be [34]. The quantity of boron, as determined by ICP-MS analysis of the combined targets (5+7) in the rinse fractions, amounted to (3.3 ± 0.2) mg. Subsequently, the elution of 7Be was achieved by raising the molarity to 4 M HNO3, resulting in an average recovery of (99.4 ± 3.7)% compared to the loading solution, within 8-9 mL eluate each [33]. The overall error in these measurements encompasses both the dilution and activity measurement errors. The gamma spectrum of the load solution and 7Be fraction, captured before and immediately after the separation of the foils, is depicted in Figure 5.9 and Figure 5.10, respectively. Figure 5.11 illustrates a representative elution profile of the separation process involving the target # 6. 120 Figure 5.9: The gamma spectrum of the load solution before attempting radiochemical processing is shown. Figure 5.10: The gamma spectrum of the purified 7Be fraction, obtained after the radiochemical separation, counted for an hour, is shown. 121 5.3.4 Quantification of boron with ICP-MS Figure 5.11: The elution profile obtained for separation with target # 6 using AG MP-50 resin for the extraction of 7Be is shown. The load fractions (L1 - L9) comprising of 0.1 M HNO3 are approximately 33 mL each. The rinse fraction R1 is 10 mL 0.1 M HNO3 followed by rinse fraction R2 which is 6 mL of 1 M HNO3, and the elute fractions (E1 - E3) are 9 mL each of 4 M HNO3 containing 7Be. This is followed by wash fraction W1 comprising water. Samples were extracted from each eluate and measured to ascertain the boron distribution in the fractions. The ICP-MS findings for the combined targets (5 and 7) are outlined in Figure 5.12 and in Table 5.3. The boron collection fraction exhibited an average recovery of 359.8 ± 2.2 mg, corresponding to approximately 89% recovery from the initial mass of the two targets, totaling 404 mg. Both elution 1 and the wash displayed minimal presence of 10B and 11B, with ≤0.03 mg detected, thereby confirming the efficacy of the purification method for isolating the 7Be radionuclide from the 10B target. Elution 2 and 3 were not directly measured due to elevated levels of radioactivity; however, their estimated masses were extrapolated using data from 122 elution 1. The estimated boron content for combined elutions 1-3 amounted to 0.05 mg of 10B and less than 0.01 mg of 11B. Figure 5.12: The elution profile for 10B and 11B for the combined targets 5 and 7 is shown. The mass of recovered 10B and the mass of the recovered 11B is reported on left and right y- axis, respectively. The concentrations of E2 and E3 were estimated using E1. The load fractions (L1 - L9) comprising of 0.1 M HNO3 are approximately 33 mL each. The rinse fraction R1 is 10 mL 0.1 M HNO3 followed by rinse fraction R2 which is 6 mL of 1 M HNO3. The elute fractions (E1 - E3) are 9 mL each of 4 M HNO3 containing 7Be. This is followed by wash fraction W1 comprising 13.5 mL of water. Image provided by Shelbie Cingoranelli. Table 5.3: The ICP-MS results of the 10B and 11B separation. Note that those marked *are estimated mass of B based on measured FR12 (E1). FRACTION L1 L2 L3 L4 L5 L6 L7 L8 L9 R1 R2 MATRIX 0.1 M HNO3 0.1 M HNO3 0.1 M HNO3 0.1 M HNO3 0.1 M HNO3 0.1 M HNO3 0.1 M HNO3 0.1 M HNO3 0.1 M HNO3 0.1 M HNO3 1 M HNO3 10B MG 28.31 39.99 40.94 45.37 35.59 44.54 38.00 44.31 42.78 3.27 <0.01 123 S.D 1.16 3.45 3.35 3.39 0.55 3.20 2.41 3.54 0.77 0.28 11B MG 0.67 0.99 0.87 0.94 0.70 1.01 0.72 0.90 0.88 0.08 <0.01 <0.01 S.D 0.06 0.12 0.13 0.05 0.08 0.08 0.04 0.06 0.12 <0.01 <0.01 L1L2L3L4L5L6L7L8L9R1R2E1 E2 E3 W1 02040600.00.51.01.5Mass of10B (mg)Mass of11B (mg)10B11B Table 5.3 (cont’d) E1 E2 E3 W1 4 M HNO3 Water 0.02 0.02 0.02 0.01 <0.01 * * <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 * * <0.01 5.3.5 Source preparation & 7Be beam delivery Figure 5.13 shows the 7Be pipetted on the tantalum foil for making the source and when it was folded tightly and inserted into the BMIS oven. The rates of 7Be4+ ion beam was assessed post-Re-accelerator, with the highest rate achieved when fine-tuning the pre-separator to direct the ([7Be]BeF2)+ molecular ion beam into the charge breeder across the entire temperature range of the oven (700-1800 °C). There was no notable alteration in molecular composition throughout the operating temperature range for the formation of BeF, BeF2, and BeF3 at various temperatures. Figure 5.14 illustrates that the purity of the delivered beam was 100%, independently confirmed via mass spectroscopy conducted by LEBIT at FRIB [35]. The overall efficiency of the ion source for 7Be was determined to be 3%, utilizing a 0.126 μCi (4.7 kBq) 7Be-loaded foil. A comprehensive account of the procedure and results pertaining to the efficiency measurement can be found in Sumithrarachchi et al. [3]. The remaining three sources, having activities of 1.2 mCi (44.4 MBq), 1.8 mCi (66.6 MBq), and 1 mCi (37 MBq), were allocated for three ReA experiments and one experiment in the LEBIT 124 Figure 5.13: (left) 7Be pipetted on the tantalum foil for the purpose of making the 7Be source (right) The tantalum foil is folded and placed inside the BMIS oven. Images provided by Dr. Katharina Domnanich. Figure 5.14: Silicon-detector spectrum of 7Be4+ beam measured after the ReA accelerator when the magnetic mass separator was selected at mass of 45 amu ([7Be]BeF2). The silicon detector measures the total kinetic energy of the incident particle which is a characteristic identification of 7Be before experiments [3]. Image provided by Dr. Chandana Sumithrarachchi. area. An average rate of 5x105 particles per second (pps) was supplied to all experiments over a total duration of 350 hours. 125 5.4 Conclusion The findings of this study includes the production of 7Be utilizing both natural and enriched boron targets. A highly efficient separation method was devised, employing a cation- exchange resin, resulting in pure 7Be fractions with an average extraction efficiency of (99.4 ± 3.7)% and 100% radionuclidic purity (within the limit of detection). Following this, preparation of 7Be source samples was conducted, and these samples were provided to the BMIS group to generate a 7Be beam. Subsequently, the beam was utilized Figure 5.15: A schematic of the entire process till obtaining pure 7Be. in four nuclear physics experiments at the NSCL. This study demonstrated the comprehensive process involved in establishing offline radioactive beams at the NSCL, encompassing 7Be isotope production, separation from boron target material, source sample preparation, and the delivery of 7Be4+ beams to scientific users. This was also the first source sample prep where the entire pipeline was demonstrated, from production to source sample prep and is demonstrated in the schematic in Figure 5.15. 126 REFERENCES “National Nuclear Data Center.” Accessed: Mar. 22, 2024. [Online]. Available: [1] https://www.nndc.bnl.gov/nudat3/ [2] C. Broggini, L. Canton, G. Fiorentini, and F. L. Villante, “The cosmological 7Li problem from a nuclear physics perspective,” Journal of Cosmology and Astroparticle Physics, vol. 2012, no. 6, Jun. 2012, doi: 10.1088/1475-7516/2012/06/030. [3] C. S. Sumithrarachchi et al., “The new Batch Mode Ion Source for stand-alone operation at the Facility for Rare Isotope Beams (FRIB),” Nucl Instrum Methods Phys Res B, vol. 541, pp. 301–304, Aug. 2023, doi: 10.1016/j.nimb.2023.05.061. A. Lapierre et al., “Commissioning results of the ReA EBIT charge breeder at the NSCL: [4] First reacceleration of stable-isotope beams,” Nucl Instrum Methods Phys Res B, vol. 317, no. PART B, pp. 399–401, Dec. 2013, doi: 10.1016/j.nimb.2013.06.034. L. Penescu, R. Catherall, J. Lettry, and T. Stora, “Development of high efficiency [5] Versatile Arc Discharge Ion Source at CERN ISOLDE,” in Review of Scientific Instruments, 2010. doi: 10.1063/1.3271245. K. A. Domnanich et al., “Preparation of stable and long-lived source samples for the [6] stand-alone beam program at the Facility for Rare Isotope Beams,” Applied Radiation and Isotopes, vol. 200, Oct. 2023, doi: 10.1016/j.apradiso.2023.110958. [7] M. Gaelens, M. Cogneau, M. Loiselet, and G. Ryckewaert, “Post-acceleration of 7Be at the Louvain-la-Neuve radioactive ion beam facility,” in Nuclear Instruments and Methods in Physics Research, Section B: Beam Interactions with Materials and Atoms, May 2003, pp. 48– 52. doi: 10.1016/S0168-583X(02)01889-X. N. Gharibyan et al., “Production and separation of carrier-free 7Be,” Applied Radiation [8] and Isotopes, vol. 107, pp. 199–202, Jan. 2016, doi: 10.1016/j.apradiso.2015.10.028. H. R. Haymond, W. M. Garrison, J. G. Hamilton, and E. Org, “Lawrence Berkeley [9] National Laboratory Recent Work Title CARRIER-FREE RADIOISOTOPES FROM CYCLOTRON TARGETS. XII. PREPARATION AND ISOLATION OF Be 7 FROM LITHIUM Permalink https://escholarship.org/uc/item/74j712fb Publication Date,” 1950. [Online]. Available: https://escholarship.org/uc/item/74j712fb [10] L. Gialanella et al., “Off-line production of a 7Be radioactive ion beam,” Nucl Instrum Methods Phys Res B, vol. 197, no. 1, pp. 150–154, 2002, doi: https://doi.org/10.1016/S0168- 583X(02)01386-1. [11] T. Ohtsuki, S. I. Fujikawa, and H. Yuki, “Preparation of carrier-free 7Be by ion-exchange following charged particle and photonuclear reactions,” Applied Radiation and Isotopes, vol. 59, no. 4, pp. 221–223, Oct. 2003, doi: 10.1016/S0969-8043(03)00200-8. 127 [12] F. Maekawa, Y. M. Verzilov, D. L. Smith, and Y. Ikeda, “Experimental study on beryllium-7 production via sequential reactions in lithium-containing compounds irradiated by 14 MeV neutrons,” Journal of Nuclear Materials, vol. 283–287, pp. 1448–1452, 2000, doi: https://doi.org/10.1016/S0022-3115(00)00182-3. [13] M. Loiselet, M. Gaelens, and G. Ryckewaert, “THE DEVELOPMENT OF A BE-7 RADIOACTIVE BEAM : FROM NUCLEAR PHYSICS TO APPLICATIONS FOR INDUSTRY,” 2004. [Online]. Available: https://api.semanticscholar.org/CorpusID:225101083 [14] M. Wiescher, R. J. Deboer, J. Görres, and R. E. Azuma, “Low energy measurements of the B 10 (p,α) Be 7 reaction,” Phys Rev C, vol. 95, no. 4, Apr. 2017, doi: 10.1103/PhysRevC.95.044617. [15] A. J. Hidy et al., “A new 7Be AMS capability established at CAMS and the potential for large datasets,” Nucl Instrum Methods Phys Res B, vol. 414, pp. 126–132, Jan. 2018, doi: 10.1016/j.nimb.2017.11.017. [16] A. Caciolli et al., “A new study of 10B(p, α)7Be reaction at low energies,” European Physical Journal A, vol. 52, no. 5, May 2016, doi: 10.1140/epja/i2016-16136-8. [17] A. Caciolli, “Nuclear Astrophysics at LNL: The 25Mg(α,n)28Si and 10B(p,α)7Be cases,” in Journal of Physics: Conference Series, Institute of Physics Publishing, Jan. 2018. doi: 10.1088/1742-6596/940/1/012035. [18] G. G. Rapisarda et al., “Study of the 10B(p, α1)7Be reaction by means of the Trojan Horse Method,” European Physical Journal A, vol. 54, no. 11, Nov. 2018, doi: 10.1140/epja/i2018-12622-3. [19] L. Lamia et al., “Boron depletion: indirect measurement of the 10B(p,α)7Be S(E)-factor,” Nucl Phys A, vol. 787, no. 1-4 SPEC. ISS., pp. 309–314, May 2007, doi: 10.1016/j.nuclphysa.2006.12.049. [20] A. N. N. M. BOESGAARD, “STELLAR ABUNDANCES OF LITHIUM, BERYLLIUM, AND BORON,” Publications of the Astronomical Society of the Pacific, vol. 88, no. 524, pp. 353–366, 1976, [Online]. Available: http://www.jstor.org/stable/40675939 [21] C. ~P. Deliyannis, “Lithium and Beryllium as diagnostics of stellar interior physical processes,” in Stellar Clusters and Associations: Convection, Rotation, and Dynamos, R. Pallavicini, G. Micela, and S. Sciortino, Eds., in Astronomical Society of the Pacific Conference Series, vol. 198. Jan. 2000, p. 235. [22] R. Depalo et al., “New direct measurement of the 10B(p, α)7Be reaction with the activation technique,” in EPJ Web of Conferences, EDP Sciences, Dec. 2017. doi: 10.1051/epjconf/201716501021. [23] L. Mabit and W. Blake, Assessing recent soil erosion rates through the use of beryllium-7 (Be-7). Springer International Publishing, 2019. doi: 10.1007/978-3-030-10982-0. 128 [24] H. Nagai, W. Tada, and T. Kobayashi, “Production rates of 7Be and 10Be in the atmosphere,” Nucl Instrum Methods Phys Res B, vol. 172, no. 1, pp. 796–801, 2000, doi: https://doi.org/10.1016/S0168-583X(00)00124-5. [25] E. A. Maugeri et al., “Preparation of 7Be targets for nuclear astrophysics research,” Journal of Instrumentation, vol. 12, no. 2, Feb. 2017, doi: 10.1088/1748-0221/12/02/P02016. [26] S. ~M. ~R. Puglia et al., “Study of the ^10B(p,alpha )^7Be reaction via the Trojan Horse Method .,” Memorie della Societa Astronomica Italiana Supplementi, vol. 14, p. 43, Jan. 2010. [27] T. Rauscher and G. Raimann, “Astrophysical reaction rates for 10 B"p,… 7 Be and 11 B"p,… 8 Be from a direct model,” 1996. [28] F. W. E. Strelow, R. Rethemeyer, and C. J. C. Bothma, “Ion Exchange Selectivity Scales for Cations in Nitric Acid and Sulfuric Acid Media with a Sulfonated Polystyrene Resin,” Anal Chem, vol. 37, no. 1, pp. 106–111, 1965, doi: 10.1021/ac60220a027. [29] S. L. Queern et al., “Production of Zr-89 using sputtered yttrium coin targets 89Zr using sputtered yttrium coin targets,” Nucl Med Biol, vol. 50, pp. 11–16, Jul. 2017, doi: 10.1016/j.nucmedbio.2017.03.004. [30] W. J. Huang, M. Wang, F. G. Kondev, G. Audi, and S. Naimi, “The AME 2020 atomic mass evaluation (I). Evaluation of input data, and adjustment procedures,” Chinese Physics C, vol. 45, no. 3, Mar. 2021, doi: 10.1088/1674-1137/abddb0. [31] M. Wang, W. J. Huang, F. G. Kondev, G. Audi, and S. Naimi, “The AME 2020 atomic mass evaluation (II). Tables, graphs and references,” Chinese Physics C, vol. 45, no. 3, Mar. 2021, doi: 10.1088/1674-1137/abddaf. J. M. Donaldson and F. Trowell, “Dissolution of Elemental Boron.,” Anal Chem, vol. 36, [32] no. 11, p. 2202, Oct. 1964, doi: 10.1021/ac60217a057. [33] J. Korkisch, Handbook of Ion Exchange Resins: Their Application to Inorganic Analytical Chemistry Volume VI. Boca Raton / London / Tokyo: CRC Press, 2017. doi: 10.1201/9781315140520. J. Korkisch, Handbook of Ion Exchange Resins: Their Application to Inorganic [34] Analytical Chemistry Volume 5. New York, 1989. [35] S. Schwarz et al., “The low-energy-beam and ion-trap facility at NSCL/MSU,” Nucl Instrum Methods Phys Res B, vol. 204, pp. 507–511, 2003, doi: https://doi.org/10.1016/S0168- 583X(02)02122-5. “Virtual tour.” Accessed: Mar. 22, 2024. [Online]. Available: [36] https://nscl.msu.edu/public/virtual-tour.html [37] S. P. Kalinin, A. A. Ogloblin, and Yu. M. Petrov, “Excitation functions for the reactions Li7(p. n)Be7, B10(p.α)Be7, and B11(p. n)C11, and energy levels in Be8, C11, and C12 nuclei,” 129 The Soviet Journal of Atomic Energy, vol. 2, no. 2, pp. 193–196, 1957, doi: 10.1007/BF01832091. [38] A. J. Koning, D. Rochman, J. C. Sublet, N. Dzysiuk, M. Fleming, and S. van der Marck, “TENDL: Complete Nuclear Data Library for Innovative Nuclear Science and Technology,” Nuclear Data Sheets, vol. 155, pp. 1–55, 2019, doi: 10.1016/j.nds.2019.01.002. 130 7Be PRODUCTION CALCULATIONS APPENDIX Table C.1: Theoretical 7Be activities are shown for the 10B(p,α)7Be reaction, for each Type II target. t irradiation [h] 3 6 4 7 # target 4 5 6 7 5+7 I [µA] µA*h A, predicted [µCi] 5 6 6 6 15.0 36.0 24.0 42.0 78.0 Sum 2542.6 6102.2 4068.1 7119.3 13221.5 19832.2 Table C.2: Theoretical 7Be activities are shown for the 10B(p,α)7Be reaction, for each Type I target. # target 1 2 3 t irradiation [h] 0.5 0.5 1 I [µA] µA*h A, predicted [µCi] 5 10 10 2.5 5.0 10.0 Sum 54.8 109.6 219.2 383.6 131 CHAPTER 6: CONCLUSIONS AND RECOMMENDATIONS To summarize, the Isotope harvesting program at NSCL/FRIB has been identified as a source of otherwise-difficult-to-obtain radioisotopes for use in multiple basic and applied research fields. The goal of this dissertation was to establish a solid-phase isotope harvesting program, complement the already existing aqueous-phase and gaseous-phase isotope harvesting program, and establish proof of concept methodologies which also find wide applicability in studies at different facilities aiming towards the extraction and chemical processing of different radionuclides from solid targets. As FRIB ramps up beam power for the higher production of radioisotopes, more accelerator parts like beam dumps, fragment catchers, mass slits, will become available for harvesting opportunities in the future. This necessitated testing on components such as the decommissioned beam blocker from NSCL and other materials that were part of this dissertation. Moreover, at FRIB, the complexity of extraction chemistry is anticipated to increase due to the considerable size of the aqueous beam dump, approximately 8000 L, and the prolonged residence time of several hours for radionuclides before encountering the extraction medium, which makes solid-phase isotope harvesting approach a viable alternative [1], [2], [3]. Establishing this solid-phase isotope harvesting approach in this dissertation composed of achieving four aims viz. harvesting 88Zr from a tungsten target, where a proof-of-concept irradiation was carried out at NSCL with tungsten foils (commonly used material at FRIB and other radioactive ion beam facilities), and (92.3 ± 1.2)% 88Zr was recovered. Moreover, the long- lived isotopes of zirconium hold significant importance in practical nuclear research. Hence, the exploration of solid-phase extraction methods for zirconium was a valuable pursuit, particularly 132 when considering the challenges associated with extracting it from aqueous solutions, as discussed in Chapter 1 and Chapter 2. Having understood chemistry of tungsten and zirconium from the experiment in Chapter 2, the knowledge was utilized in designing methodologies for processing the tungsten beam- blocker where 172Hf was extracted from the tungsten beam-blocker and other co-embedded radionuclides followed by establishing the 172Hf/172Lu generator and the 172Lu generated was used for PAC experiments. This was followed by comparison of generators set up with those from the conventional method to produce 172Hf (through the proton irradiation of a natural lutetium target) where the generators on different resins were eluted cyclically to carry out PAC and radiolabeling studies with the ZR resin set up as the higher activity generator. Additionally, the ease of separation, elution, and the long shelf-life of the 172Hf/172Lu generator system makes it suitable for experiments where pure lutetium isotope is needed and procuring 177Lu is not always feasible. Furthermore, beryllium and boron chemistry was also explored with the production of 7Be, its separation from boron target material, source preparation efforts, and 7Be beam delivery. In all cases, experiments were first conducted with stable elements, to understand the chemistry of the selected element and then replicated with radionuclides. 6.1 Chemical methodologies applicability The techniques and methodologies developed in this work will find applicability in chemical separation to extract Hf and Zr isotopes from irradiated tungsten parts in the accelerator beamline at heavy-ion beam facilities like FRIB. These extracted radioisotopes can then be used to establish generators that will provide pure daughter radioisotopes for an extended period of time. 133 Other facilities harboring tungsten parts are the European Spallation Source (ESS) in Sweden, Los Alamos Neutron Science Center (LANSCE) facility at Los Alamos National Laboratory (LANL), and the Spallation Neutron Source (SNS) at Oak Ridge National Laboratory (ORNL) amongst others. At the ESS, the heavy metal, tungsten target is a 2.6 m diameter stainless steel disk/wheel weighing almost five tons [4]. At the LANSCE facility, a water-cooled tungsten target assembly is used to produce spallation neutrons [5]. The SNS facility houses a rotating tungsten target, which is hit by a 700-kilowatt proton beam resulting in spallation [6]. When the tungsten parts at these facilities are decommissioned or have run their service period, many exotic radionuclides can be harvested due to years of operation. There can be many other components in the beamline that would accumulate radionuclides due to years of operation at these facilities as well. 6.2 Future work Conducting the experiments mentioned in this thesis also resulted in the identification of future directions to investigate to improve these methodologies. Further scope for improvement remains in the context of the purified fractions from the tungsten beam-blocker generators to achieve radiopurity. While it was not a major obstacle for PAC experiments, radiopurity needs to be achieved for radiolabeling study purposes, followed by exploration of more novel ligand systems to bind the eluted 172Lu. For the radiolanthanides eluted from the LN resin with the load solution fractions in Chapter 3, further separation can be achieved using a DGA resin column to harvest the radiolanthanides as well apart from 172Hf. Furthermore, entrapped gases can be harvested from the tungsten beam-blocker. More novel ligand systems can also be synthesized and explored for chelation with 172Lu eluted from higher activity ZR resin generator set up in Chapter 4. In context of 7Be, the BMIS extraction efficiency can be improved. 134 Further exploration can be conducted into additional materials suitable for online implantation and activation that are capable of regeneration between experiments. Among the promising candidates are vitreous carbon and alumina, known for their ability to withstand high temperatures and resistance against beam damage during irradiation. Notably, these materials have large surface to volume ratios, meaning that embedded radionuclides would be close to surfaces. This proximity to the surface streamlines the removal process via an acid wash post- irradiation, or could allow thermal distillation in an ISOL-like process. Furthermore, depending on the end use of the harvested radionuclides, the purity of the beam at the location of the collector must be considered. In instances where solid collections occur on the slits of mass separators, encountering less mass and a potentially smaller array of radionuclides compared to those found in the beam dump is expected due to the prior selectivity in the fragment separator at this point. 6.3 Final remarks Through this work, different target materials like tungsten, lutetium, and boron have been tested and chemical methodologies were developed, and the first true test of the solid-phase harvesting capabilities at the NSCL was demonstrated. The insight gained from the experiments performed here will lead to successful solid-phase harvesting experiments at FRIB to produce and collect useful radionuclides. The capability to harvest and study rare isotopes has significant implications in fields like nuclear medicine, where they are used for diagnosis and treatment, as well as in basic scientific research to understand the fundamental properties of matter. Additionally, considering the specific radionuclide under investigation and the desired balance between quantity and purity, certain modes may present advantages over others. Therefore, it is 135 imperative to thoroughly explore all avenues through which the nuclear science community can harness the wealth of radionuclides generated at FRIB. 136 REFERENCES A. Visser and N. Scielzo, “Feasibility of Harvesting Radon from FRIB Beam Dump,” [1] Llnl-Tr-709438, 2016. [2] M. Avilov et al., “Thermal, mechanical and fluid flow aspects of the high power beam dump for FRIB,” Nucl Instrum Methods Phys Res B, vol. 376, pp. 24–27, 2016, doi: 10.1016/j.nimb.2016.02.068. S. Satija et al., “Harvesting 88Zr from heavy-ion beam irradiated tungsten at the National [3] Superconducting Cyclotron Laboratory,” Applied Radiation and Isotopes, vol. 197, p. 110831, 2023, doi: https://doi.org/10.1016/j.apradiso.2023.110831. “How the ESS Target works,” 2024, Accessed: Mar. 22, 2024. [Online]. Available: [4] https://europeanspallationsource.se/target/how-it-works [5] A. T. Nelson, J. A. O’Toole, R. A. Valicenti, and S. A. Maloy, “Fabrication of a tantalum-clad tungsten target for LANSCE,” Journal of Nuclear Materials, vol. 431, no. 1–3, pp. 172–184, Dec. 2012, doi: 10.1016/j.jnucmat.2011.11.041. “Second Target Station Target Systems ,” 2024, Accessed: Mar. 22, 2024. [Online]. [6] Available: https://sns.gov/sts/sts-target-systems 137