Highly Enriched Uranium-Free Medical Radioisotope Production Methods: An Integrative Review

Abstract

The ever-growing need for radiopharmaceuticals, i.e., compounds containing pharmaceutical drugs and radioisotopes used for medical diagnostic imaging (SPECT/PET scan) and treating neoplasms, is significantly leading to an increased demand for such substances in hospitals and clinics worldwide. Currently, most large-scale productions of radioisotopes required for radiopharmaceuticals are carried out in research reactors, via the fission of highly enriched uranium. However, because large amounts of radioactive waste are produced as byproducts in this process, new greener methods are needed for radioisotope production. This work presents an integrative literature review and summarizes enriched uranium-free methods for radioisotope production, accomplished through the adoption of new reaction routes, distinct acceleration technologies, or by using other physical processes. This review considered forty-eight studies published from 2010 to 2021 on three established virtual databases. Among these selected works, a cyclotron is the most adopted HEU-free method for radioisotope production, and ${}^{44}$Sc, ${}^{68}$Ga, and ${}^{99\mathrm{m}}$Tc are the medical radioisotopes most often reported as produced by using the investigated HEU-free production methods.

1. Introduction

Hospitals, clinics, and health centers make daily use of a wide range of highly complex diagnoses and treatments resulting from technological advances achieved in the last decades, with nuclear medicine being an area that benefits from such advances. In nuclear medicine, radiopharmaceuticals, which are compounds containing radionuclides (or radioisotopes) bound to pharmaceutical drugs [1], are administered to patients via intravenous or oral injections. Depending on their properties, these organic molecules allow the radionuclides to be fixed in specific human body organs. The radioisotope corpuscular and/or the electromagnetic radiation it emits in this organ is used for imaging or treatment purposes [2]. Radionuclides emitting either ${\beta }^{+}$ particles and/or $\gamma$-rays are usually employed in diagnoses. On the other hand, Auger electrons and ${\beta }^{-}$-emitter radioisotopes are administered in therapy [3]. Medical radioisotopes must have half-lives long enough to allow image acquisition or therapeutic effects, but their half-lives must be short enough to minimize the patient’s exposure to ionizing radiation [4]. In this sense, not all radioisotopes are appropriate to be used in radiopharmaceuticals.

Some examples of medical radioisotopes are the ${}^{99\mathrm{m}}$Tc (half-life of approximately 6 h), ${}^{18}$F (half-life of 1.83 h), and ${}^{64}$Cu (half-life of 12.7 h). Moreover, ${}^{99\mathrm{m}}$Tc, the most used radioisotope in nuclear medicine, is a $\gamma$-ray-emitter radioisotope; its emitted radiation is detected through a gamma camera in a SPECT (single photon emission tomography) equipment, enabling internal imaging of the patient for diagnostic purposes [5]. Routine clinical applications of ${}^{99\mathrm{m}}$Tc include bone scintigraphy [6], myocardial perfusion [7], ventriculography [8], brain [9], sentinel node [10], immunoscintigraphy [11], blood-pool labeling [12], sulfur colloid [13], and Meckel’s diverticulum [14] scan procedures. On the other hand, ${}^{18}$F is a ${\beta }^{+}$-emitter radioisotope applied in positron emission tomography (PET) for the early detection and diagnosis of Hodgkin lymphoma [15], and colorectal, head and neck, lung, cervical, and ovarian cancers [16]. Finally, ${}^{64}$Cu is a ${\beta }^{-}$ (39 %) and ${\beta }^{+}$ (61 %) emitter used for both PET imaging [17,18,19] and cancer therapy [20].

Currently, most of the radioisotope demand is met by two production methods: the fission of highly enriched uranium (HEU) in research reactors, or the acceleration of electrically charged particles (usually protons) in cyclotrons, toward gaseous, liquid, or solid targets, depending on the radioisotope to be produced [21]. Cyclotrons are most often utilized to produce short-life, neutron-deficient, ${\beta }^{+}$-emitter radioisotopes, which are mainly used for PET scans. Because of their short half-lives, these radioisotopes need to be used right after production. Hence, they are mostly produced in medical cyclotrons and are immediately utilized in the same sites in which these facilities are installed [22]. On the other hand, long-lived ${\beta }^{-}$ and $\gamma$-ray-emitter radioisotopes, generally applied to internal radiotherapy and SPECT scans, respectively, are primarily produced in research reactors. In addition to enabling large-scale production, research reactors also offer logistic advantages. Because most of the radioisotopes produced in these facilities have longer half-lives, if compared to those typically produced in cyclotrons [22], they can be widely distributed before being used. However, as it will be soon discussed, this comes at the cost of transporting highly radioactive material over long distances.

In a research reactor, a flux of up to 10${}^{14}$ neutrons/$\mathrm{c}{\mathrm{m}}^2 \mathrm{s}$ hits an HEU target [23], starting the uranium fission. This process produces over forty distinct radioisotopes [24], including the ${}^{99}$Mo (half-life of 66 h), which later decays to ${}^{99\mathrm{m}}$Tc in a ${}^{99}$Mo/${}^{99\mathrm{m}}$Tc generator. Due to the molybdenum’s large half-life, this generator can be transported all over the world to support the worldwide medical demand of ${}^{99\mathrm{m}}$Tc. Despite the significant advantages of the radioisotope production scale, multiple substantial concerns are associated with the fission of HEU in research reactors. The first (and most dangerous) one is the generation of large amounts of environmentally hazardous radioactive waste, which can take from decades to hundreds of years to decay. Dealing with radioactive waste often requires relocating and storing it in a deep–underground repository [25], which leads to storage space/safety problems since the waste can be active for hundreds of years. Another issue caused by having most of the radioisotope production centralized in research reactors is the consequent need of transporting highly radioactive material over long distances [26]. Although this activity has been safely and efficiently conducted for several decades [27], there are issues associated with it, such as the higher production costs and logistics complexities, which are even more critical for developing countries [28]. Finally, as mentioned by (Van Noorden, R. 2013) [29], the shutdown of two research reactors in 2009, the Chalk River reactor (in Canada) and the Petten high flux nuclear reactor (in the Netherlands), made it clear that the world’s radioisotope supply chain was fragile, as it greatly relied on a few research reactors built in the 1950s and 1960s. In those times, these two reactors were responsible for producing most of the world’s supply of ${}^{99\mathrm{m}}$T; thus, the shutdowns created a shortage of this radioisotope. This crisis motivated the research into ${}^{99}$Mo production in cyclotrons via proton irradiation, and in linear accelerators via the bremsstrahlung $\gamma$-ray reaction route ${}^{100}$Mo($\gamma$,n)${}^{99}$Mo [29].

In addition to the aforementioned issues associated with using research reactors, international nuclear non-proliferation policies emphasize the need of eliminating the use of HEU for medical radioisotope production [30]. Hence, developing alternatives to the HEU-based production methods, i.e., HEU-free production routes through different acceleration technologies and/or physical processes, is a crucial step toward finding a greener solution to meet the ever-growing demand for medical radioisotopes. In order to contribute to this relevant topic, this integrative review [31] aims to identify and synthesize the recently reported HEU-free medical-radioisotope production methods, as well as the medical radioisotopes produced by using such methods.

This review is organized as follows. Section 2 presents an overview of the method adopted for the literature review and describes the process used to analyze the results. In Section 3, the results obtained are presented and discussed. Finally, in Section 4, the conclusions and future directions are presented.

2. Materials and Methods

This study aims to identify which radioisotopes used in medicine have recently received interest from the scientific community and the methodologies used in their production. Thus, this research has two guiding questions: “What alternative, HEU-free methods have been recently used to produce medical radioisotopes?” and “What medical radioisotopes can be produced by these alternative technologies?”.

The literature review was performed in January 2022 on the databases Virtual Health Library (VHL) [32], Science Direct [33], and Scopus [34], with the following search strings: “radiopharmaceutical production” OR “radioisotope production”. The search filters used were the following: peer-reviewed papers; English-written; available electronically; and published from 2010 to 2021. By using these filters, 91 papers were selected on VHL, 604 on Science Direct, and 294 on Scopus, resulting in 989 papers. After removing duplicate papers, this number was reduced to 927 unique papers. After a preliminary analysis, it was observed that, rather than reporting their own experimental radioisotope production, many of these publications only cited the productions from other works (eventually comparing them to their own simulation results) or estimated radioisotope reaction cross-sections, without informing the production itself. Since filtering this situation is non-trivial, once it involves context awareness, the 927 selected papers were manually analyzed. From this analysis, papers that did not report their own experimental radioisotope production, or did not present explicit radioisotope activity—in units of $\mathrm{Bq}$ or Ci—thick target yields, or saturation yields (which could be used to calculate the activity), were excluded. Using this methodology, from the 927 papers, 48 were selected [35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82], and their data are presented in

Table A1. Intended radioisotope production summary.

n°	Intended Radioisotope	Half-Life [h]	Decay	Daughter Isotope	Technology	Accelerated Particle	Production Route	Reference
1	${}^{64}$Cu	12.7	${\beta }^{-}$ (39%)	${}^{64}$Zn	Cyclotron	Proton	${}^{64}$Ni(p,n)${}^{64}$Cu	Thisgaard et al. [35]
			${\beta }^{+}$ (61%)
	${}^{120\mathrm{m}}$Sb	138.2	${\beta }^{+}$	${}^{120}$Sn			${}^{\mathrm{nat}}$Sn(p,x)${}^{120\mathrm{m}}$Sb
	${}^{122}$Sb	65.4	${\beta }^{-}$	${}^{122}$Te			${}^{\mathrm{nat}}$Sn(p,x)${}^{122}$Sb
2	${}^{18}$F	1.83	${\beta }^{+}$	${}^{18}$O	Cyclotron	Proton	${}^{18}$O(p,n)${}^{18}$F	Roeda et al. [36]
3	${}^{131}$I	192.5	${\beta }^{-}$	${}^{131}$Xe	Research reactor	Neutron	${}^{130}$Te(n,$\gamma$)${}^{131}$Te${\stackrel{}{\to }}^{131}$I	Achoribo et al. [37]
4	${}^{94\mathrm{m}}$Tc	0.867	${\beta }^{+}$	${}^{94}$Mo	Cyclotron	Proton	${}^{94}$Mo(p,n)${}^{94\mathrm{m}}$Tc	Hoehr et al. [38]
5	${}^{99}$Mo	66.0	${\beta }^{-}$	${}^{99\mathrm{m}}$Tc	Cyclotron	Proton	${}^{232}$Th(p,f)${}^{99}$Mo	Abbas et al. [39]
6	${}^{13}$N	0.166	${\beta }^{+}$	${}^{13}$C	Plasma focus device	Deuteron	${}^{13}$C(d,n)${}^{13}$N	Shirani et al. [40]
7	${}^{161}$Ho	2.48	Ec	${}^{161}$Dy	Cyclotron	Proton	${}^{\mathrm{nat}}$Dy(p,xn)${}^{161}$Ho	Tárkányi et al. [41]
8	${}^{67}$Ga	78.3	Ec	${}^{67}$Zn	Cyclotron	Proton	${}^{68}$Zn(p,2n)${}^{67}$Ga	Martins et al. [42]
9	${}^{169}$Er	225.4	${\beta }^{-}$	${}^{169}$Tm	Research reactor	Neutron	${}^{168}$Er(n,$\gamma$)${}^{169}$Er	Chakravarty et al. [43]
10	${}^{99\mathrm{m}}$Tc	6.01	$\gamma$	${}^{99}$Tc	Cyclotron	Proton	${}^{100}$Mo(p,2n)${}^{99\mathrm{m}}$Tc	Bénard et al. [44]
11	${}^{44}$Sc	4.04	${\beta }^{+}$	${}^{44}$Ca	Cyclotron	Proton	${}^{44}$Ca(p,n)${}^{44}$Sc	Hoehr et al. [45]
12	${}^{63}$Zn	0.641	${\beta }^{+}$	${}^{63}$Cu	Cyclotron	Proton	${}^{63}$Cu(p,n)${}^{63}$Zn	DeGrado et al. [46]
13	${}^{103}$Ag	1.10	${\beta }^{+}$	${}^{103}$Pd	Cyclotron	${}^3$He	${}^{\mathrm{nat}}$Pd(${}^3$He,pxn)${}^{103}$Ag	Al-Abyad et al. [47]
	${}^{104}$Ag	1.16	${\beta }^{+}$	${}^{104}$Pd			${}^{\mathrm{nat}}$Pd(${}^3$He,pxn)${}^{104}$Ag
	${}^{105}$Ag	991.0	${\beta }^{+}$	${}^{105}$Pd			${}^{\mathrm{nat}}$Pd(${}^3$He,pxn)${}^{105}$Ag
	${}^{106\mathrm{m}}$Ag	198.7	${\beta }^{+}$	${}^{106}$Cd			${}^{\mathrm{nat}}$Pd(${}^3$He,pxn)${}^{106\mathrm{m}}$Ag
	${}^{111}$Ag	178.8	${\beta }^{-}$	${}^{111}$Cd			${}^{\mathrm{nat}}$Pd(${}^3$He,pxn)${}^{111}$Ag
	${}^{112}$Ag	3.13	${\beta }^{-}$	${}^{112}$Cd			${}^{\mathrm{nat}}$Pd(${}^3$He,pxn)${}^{112}$Ag
	${}^{104}$Cd	0.962	${\beta }^{+}$	${}^{104}$Ag			${}^{\mathrm{nat}}$Pd(${}^3$He,xn)${}^{104}$Cd
	${}^{105}$Cd	0.925	${\beta }^{+}$	${}^{105}$Ag			${}^{\mathrm{nat}}$Pd(${}^3$He,xn)${}^{105}$Cd
	${}^{107}$Cd	6.50	${\beta }^{+}$	${}^{107\mathrm{m}}$Ag			${}^{\mathrm{nat}}$Pd(${}^3$He,xn)${}^{107}$Cd
	${}^{111\mathrm{m}}$Cd	0.808	${\beta }^{+}$	${}^{111}$Cd			${}^{\mathrm{nat}}$Pd(${}^3$He,xn)${}^{111\mathrm{m}}$Cd
14	${}^{211}$At	7.21	$\alpha$	${}^{207}$Bi	Cyclotron	Proton	${}^{209}$Bi($\alpha$,2n)${}^{211}$At	Martin et al. [48]
15	${}^{89}$Zr	78.4	${\beta }^{+}$	${}^{89}$Y	Cyclotron	Proton	${}^{89}$Y(p,n)${}^{89}$Zr	Siikanen et al. [49]
16	${}^{44}$Sc	4.04	${\beta }^{+}$	${}^{44}$Ca	Cyclotron	Proton	${}^{44}$Ca(p,n)${}^{44}$Sc	Valdovinos et al. [50]
17	${}^{99\mathrm{m}}$Tc	6.01	$\gamma$	${}^{99}$Tc	Cyclotron	Proton	${}^{100}$Mo(p,2n)${}^{99\mathrm{m}}$Tc	Schaffer et al. [51]
18	${}^{86}$Y	14.7	${\beta }^{+}$	${}^{86}$Sr	Cyclotron	Proton	${}^{\mathrm{nat}}$Sr(p,n)${}^{86}$Y	Oehlke et al. [52]
	${}^{89}$Zr	78.4	${\beta }^{+}$	${}^{89}$Y			${}^{89}$Y(p,n)${}^{89}$Zr
	${}^{68}$Ga	1.13	${\beta }^{+}$	${}^{68}$Zn			${}^{\mathrm{nat}}$Zn(p,x)${}^{68}$Ga
19	${}^{169}$Yb	1024	Ec	${}^{169}$Tm	Research reactor	Neutron	${}^{\mathrm{nat}}$Yb${}_2$O${}_3$(n,$\gamma$)${}^{169}$Yb	Saxena et al. [53]
20	${}^{11}$C	0.339	${\beta }^{+}$	${}^{11}$B	Cyclotron	Proton	${}^{14}$N(p,$\alpha$)${}^{11}$C	Moon et al. [54]
21	${}^{44}$Sc	4.04	${\beta }^{+}$	${}^{44}$Ca	Cyclotron	Proton	${}^{44}$Ca(p,n)${}^{44}$Sc	van der Meulen et al. [55]
22	${}^{44}$Sc	4.04	${\beta }^{+}$	${}^{44}$Ca	Cyclotron	Deuteron	${}^{44}$Ca(d,2n)${}^{44}$Sc	Duchemin et al. [56]
	${}^{44\mathrm{m}}$Sc	58.6	$\gamma$	${}^{44}$Sc			${}^{44}$Ca(d,2n)${}^{44\mathrm{m}}$Sc
23	${}^{61}$Cu	3.33	${\beta }^{+}$	${}^{61}$Ni	Cyclotron	Proton	${}^{\mathrm{nat}}$Zn(p,$\alpha$)${}^{61}$Cu	Asad et al. [57]
							${}^{64}$Zn(p,$\alpha$)${}^{61}$Cu
24	${}^{161}$Tb	165.6	${\beta }^{-}$	${}^{161}$Dy	Research reactor	Neutron	${}^{160}$Gd(n,$\gamma$)${}^{161}$Gd${\stackrel{}{\to }}^{161}$Tb	Aziz et al. [58]
25	${}^{186}$Re	89.0	${\beta }^{-}$	${}^{186}$Os	Cyclotron	Proton	${}^{\mathrm{nat}}$W(p,n)${}^{186}$Re	Kakavand et al. [59]
26	${}^{43}$Sc	3.89	${\beta }^{+}$	${}^{43}$Ca	Cyclotron	$\alpha$	${}^{\mathrm{nat}}$Ca($\alpha$,p)${}^{43}$Sc	Szkliniarz et al. [60]
							${}^{40}$Ca($\alpha$,p)${}^{43}$Sc
	${}^{44}$Sc	4.04	${\beta }^{+}$	${}^{44}$Ca			${}^{42}$Ca($\alpha$,np+pn)${}^{44}$Sc
	${}^{44\mathrm{m}}$Sc	58.6	$\gamma$	${}^{44}$Sc			${}^{42}$Ca($\alpha$,np+pn)${}^{44\mathrm{m}}$Sc
27	${}^{66}$Ga	9.49	${\beta }^{+}$	${}^{66}$Zn	Tandetron	Proton	${}^{\mathrm{nat}}$Zn(p,x)${}^{66}$Ga	Fraile et al. [61]
	${}^{68}$Ga	1.13	${\beta }^{+}$	${}^{68}$Zn			${}^{\mathrm{nat}}$Zn(p,x)${}^{68}$Ga
28	${}^{66}$Ga	9.49	${\beta }^{+}$	${}^{66}$Zn	Cyclotron	Proton	${}^{66}$Zn(p,n)${}^{66}$Ga	Cho et al. [62]
29	${}^{61}$Cu	3.33	${\beta }^{+}$	${}^{61}$Ni	Cyclotron	Proton	${}^{\mathrm{nat}}$Zn(p,$\alpha$)${}^{61}$Cu	do Carmo et al. [63]
	${}^{66}$Ga	9.49	${\beta }^{+}$	${}^{66}$Zn			${}^{\mathrm{nat}}$Zn(p,x)${}^{66}$Ga
	${}^{67}$Ga	78.2	Ec	${}^{67}$Zn			${}^{\mathrm{nat}}$Zn(p,x)${}^{67}$Ga
	${}^{68}$Ga	1.13	${\beta }^{+}$	${}^{68}$Zn			${}^{\mathrm{nat}}$Zn(p,n)${}^{68}$Ga
	${}^{65}$Zn	5857	${\beta }^{+}$	${}^{65}$Cu			${}^{\mathrm{nat}}$Zn(p,pn)${}^{65}$Zn
30	${}^{47}$Sc	80.4	${\beta }^{-}$	${}^{47}$Ti	Cyclotron	Proton	${}^{\mathrm{nat}}$Ca(p,2n)${}^{64}$Cu	Misiak et al. [64]
31	${}^{64}$Cu	12.7	${\beta }^{-}$ (39%)	${}^{64}$Zn	Cyclotron	Proton	${}^{64}$Ni(p,n)${}^{64}$Cu	Xie et al. [65]
			${\beta }^{+}$ (61%)
32	${}^{43}$Sc	3.89	${\beta }^{+}$	${}^{43}$Ca	Cyclotron	Deuteron	${}^{42}$Ca(d,n)${}^{43}$Sc	Sitarz et al. [66]
							${}^{43}$Ca(p,n)${}^{43}$Sc
	${}^{44}$Sc	4.04	${\beta }^{+}$	${}^{44}$Ca			${}^{\mathrm{nat}}$Ca(p,n)${}^{44}$Sc
						Proton	${}^{44}$Ca(p,n)${}^{44}$Sc
	${}^{47}$Sc	80.4	${\beta }^{-}$	${}^{47}$Ti			${}^{48}$Ca(p,2n)${}^{47}$Sc
							${}^{48}$Ti(p,2p)${}^{47}$Sc
33	${}^{100}$Rh	20.8	${\beta }^{+}$	${}^{100}$Ru	Cyclotron	${}^3$He	${}^{103}$Rh(${}^3$He,x)${}^{100}$Rh	Ali et al. [67]
	${}^{101\mathrm{m}}$Rh	104.2	Ec (94%)	${}^{101}$Ru			${}^{103}$Rh(${}^3$He,$\alpha$+n)${}^{101\mathrm{m}}$Rh
			$\gamma$ (6%)	${}^{101}$Rh
	${}^{103}$Ag	1.10	${\beta }^{+}$	${}^{103}$Pd			${}^{103}$Rh(${}^3$He,3n)${}^{103}$Ag
	${}^{104\mathrm{m}}$Ag	0.558	${\beta }^{+}$	${}^{104}$Pd			${}^{103}$Rh(${}^3$He,2n)${}^{104\mathrm{m}}$Ag
	${}^{104}$Ag	1.15	${\beta }^{+}$				${}^{103}$Rh(${}^3$He,2n)${}^{104}$Ag
	${}^{103}$Pd	407.8	Ec	${}^{103}$Rh			${}^{103}$Rh(${}^3$He,n)${}^{103}$Pd
34	${}^{65}$Zn	5857	Ec	${}^{65}$Cu	Research reactor	Neutron	${}^{64}$Zn(n,$\gamma$)${}^{65}$Zn	Karimi et al. [68]
35	${}^{106\mathrm{m}}$Ag	19.2	${\beta }^{+}$	${}^{106}$Pd	Cyclotron	$\alpha$	${}^{\mathrm{nat}}$Ag($\alpha$,x)${}^{106\mathrm{m}}$Ag	Ditrói et al. [69]
36	${}^{15}$O	0.034	${\beta }^{+}$	${}^{15}$N	Cyclotron	Deuteron	${}^{15}$N(d,n)${}^{15}$O	Iguchi et al. [70]
37	${}^{72}$Se	201.6	Ec	${}^{72}$As	Linear particle accelerator	Proton	${}^{75}$As(p,4n)${}^{72}$Se	DeGraffenreid et al. [71]
38	${}^{186}$Re	89.2	${\beta }^{-}$	${}^{186}$Os	Research reactor	Neutron	${}^{185}$Re(n,$\gamma$)${}^{186}$Re	Pourhabib et al. [72]
	${}^{188}$Re	17.0	${\beta }^{-}$	${}^{188}$Os			${}^{187}$Re(n,$\gamma$)${}^{188}$Re
39	${}^{67}$Cu	61.8	${\beta }^{+}$	${}^{67}$Zn	Cyclotron	${}^{70}$Zn${}^{15+}$	H${}_2$(${}^{70}$Zn${}^{15+}$,$\alpha$+p)${}^{67}$Cu	Souliotis et al. [73]
40	${}^{225}$Ac	240.0	$\alpha$	${}^{221}$Fr	Cyclotron	Proton	${}^{232}$Th(p,x)${}^{225}$Ac	Robertson et al. [74]
	${}^{225}$Ra	357.6	${\beta }^{-}$	${}^{225}$Ac			${}^{232}$Th(n,x)${}^{225}$Ra
41	${}^{94}$Tc	4.89	${\beta }^{+}$	${}^{94}$Mo	Cyclotron	Proton	${}^{\mathrm{nat}}$Mo(p,x)${}^{94}$Tc	Ahmed et al. [75]
	${}^{95}$Tc	20.0	${\beta }^{+}$	${}^{95}$Mo			${}^{\mathrm{nat}}$Mo(p,x)${}^{95}$Tc
	${}^{95\mathrm{m}}$Tc	1464	${\beta }^{+}$	${}^{95}$Mo			${}^{\mathrm{nat}}$Mo(p,x)${}^{95\mathrm{m}}$Tc
	${}^{96}$Tc	102.7	${\beta }^{+}$	${}^{96}$Mo			${}^{\mathrm{nat}}$Mo(p,x)${}^{96}$Tc
	${}^{99\mathrm{m}}$Tc	6.01	$\gamma$	${}^{99}$Tc			${}^{\mathrm{nat}}$Mo(p,x)${}^{99\mathrm{m}}$Tc
	${}^{99}$Mo	66.0	${\beta }^{-}$	${}^{99\mathrm{m}}$Tc			${}^{\mathrm{nat}}$Mo(p,x)${}^{99}$Mo
42	${}^{44}$Sc	4.04	${\beta }^{+}$	${}^{44}$Ca	Cyclotron	Proton	${}^{44}$Ca(p,n)${}^{44}$Sc	Alnahwi et al. [76]
43	${}^{99\mathrm{m}}$Tc	6.01	$\gamma$	${}^{99}$Tc	Cyclotron	Proton	${}^{\mathrm{nat}}$Mo(p,x)${}^{99\mathrm{m}}$Tc	Kambali et al. [77]
	${}^{99}$Mo	66.0	${\beta }^{-}$	${}^{99\mathrm{m}}$Tc			${}^{98}$Mo(p,x)${}^{99}$Mo
44	${}^{44}$Sc	4.04	${\beta }^{+}$	${}^{44}$Ca	Cyclotron	Proton	${}^{44}$Ca(p,n)${}^{44}$Sc	van der Meulen et al. [78]
45	${}^{13}$N	0.165	${\beta }^{+}$	${}^{13}$C	Cyclotron	Proton	${}^{16}$O(p,$\alpha$)${}^{13}$N	Abel et al. [79]
	${}^{18}$F	1.82	${\beta }^{+}$	${}^{18}$O			${}^{18}$O(p,n)${}^{18}$F
46	${}^{167}$Tm	222.0	Ec	${}^{167}$Er	Cyclotron	Proton	${}^{\mathrm{nat}}$Er(p,x)${}^{167}$Tm	Heinke et al. [80]
47	${}^{155}$Tb	127.7	Ec	${}^{155}$Gd	Cyclotron	Proton	${}^{155}$Gd(p,n)${}^{155}$Tb	Favaretto et al. [81]
							${}^{156}$Gd(p,2n)${}^{155}$Tb
48	${}^{89}$Zr	78.4	${\beta }^{+}$	${}^{89}$Y	Cyclotron	Proton	${}^{89}$Y(p,n)${}^{89}$Zt	Cisterino et al. [82]

,

Table A2. Cyclotron produced radioisotopes.

n°	Production Route	Kinetic Energy [MeV]	Current [$\mathsf{\mu }$A]	Irradiation Time [h]	${\mathbf{A}}_{\mathbf{EOB}}$ [kBq]	Reference
1	${}^{64}$Ni(p,n)${}^{64}$Cu	16.1	121	1.26	$8.2\times {10}^6$	Thisgaard et al. [35]
	${}^{\mathrm{nat}}$Sn(p,x)${}^{120\mathrm{m}}$Sb		150		$3.71\times {10}^4$
	${}^{\mathrm{nat}}$Sn(p,x)${}^{122}$Sb				$3.30\times {10}^4$
2	${}^{18}$O(p,n)${}^{18}$F	18	20	0.5	$2.96\times {10}^7$	Roeda et al. [36]
4	${}^{94}$Mo(p,n)${}^{94\mathrm{m}}$Tc	13	5	1	$1.1\times {10}^5$	Hoehr et al. [38]
5	${}^{232}$Th(p,f)${}^{99}$Mo	29.5	1	1	$3.7\times {10}^3$	Abbas et al. [39]
		26.5			$3.4\times {10}^3$
7	${}^{\mathrm{nat}}$Dy(p,xn)${}^{161}$Ho	36	61	1.18	$1.89\times {10}^6$	Tárkányi et al. [41]
8	${}^{68}$Zn(p,2n)${}^{67}$Ga	26	10	1	$3.9\times {10}^5$	Martins et al. [42]
10	${}^{100}$Mo(p,2n)${}^{99\mathrm{m}}$Tc	18	240	6.9	$3.48\times {10}^8$	Bénard et al. [44]
11	${}^{44}$Ca(p,n)${}^{44}$Sc	13	7.6	1	$5.5\times {10}^3$	Hoehr et al. [45]
12	${}^{63}$Cu(p,n)${}^{63}$Zn	14	20	1	$1.84\times {10}^6$	DeGrado et al. [46]
13	${}^{\mathrm{nat}}$Pd(${}^3$He,pxn)${}^{103}$Ag	27	0.150	1	$5.3\times {10}^2$	Al-Abyad et al. [47]
	${}^{\mathrm{nat}}$Pd(${}^3$He,pxn)${}^{104}$Ag				$6.83\times {10}^3$
	${}^{\mathrm{nat}}$Pd(${}^3$He,pxn)${}^{105}$Ag				30
	${}^{\mathrm{nat}}$Pd(${}^3$He,pxn)${}^{106\mathrm{m}}$Ag				35.9
	${}^{\mathrm{nat}}$Pd(${}^3$He,pxn)${}^{111}$Ag				26.9
	${}^{\mathrm{nat}}$Pd(${}^3$He,pxn)${}^{112}$Ag				102
	${}^{\mathrm{nat}}$Pd(${}^3$He,xn)${}^{104}$Cd				$1.48\times {10}^3$
	${}^{\mathrm{nat}}$Pd(${}^3$He,xn)${}^{105}$Cd				$5.95\times {10}^3$
	${}^{\mathrm{nat}}$Pd(${}^3$He,xn)${}^{107}$Cd				690
	${}^{\mathrm{nat}}$Pd(${}^3$He,xn)${}^{111\mathrm{m}}$Cd				466
14	${}^{209}$Bi($\alpha$,2n)${}^{211}$At	27.8	0.327	4	$3.91\times {10}^4$	Martin et al. [48]
		25.3	0.192		$7.91\times {10}^3$
15	${}^{89}$Y(p,n)${}^{89}$Zr	12.8	45	3	$6.3\times {10}^6$	Siikanen et al. [49]
16	${}^{44}$Ca(p,n)${}^{44}$Sc	15.6	25	1	$4.11\times {10}^6$	Valdovinos et al. [50]
17	${}^{100}$Mo(p,2n)${}^{99\mathrm{m}}$Tc	16.5	130	6	$1.74\times {10}^8$	Schaffer et al. [51]
18	${}^{\mathrm{nat}}$Sr(p,n)${}^{86}$Y	13	4.5	1	$7.4\times {10}^3$	Oehlke et al. [52]
	${}^{89}$Y(p,n)${}^{89}$Zr		7		$3.2\times {10}^4$
	${}^{\mathrm{nat}}$Zn(p,x)${}^{68}$Ga				$4.8\times {10}^5$
20	${}^{14}$N(p,$\alpha$)${}^{11}$C	13	60	0.5	$8.66\times {10}^7$	Moon et al. [54]
21	${}^{44}$Ca(p,n)${}^{44}$Sc	11	50	1.5	$1.9\times {10}^6$	van der Meulen et al. [55]
22	${}^{44}$Ca(d,2n)${}^{44}$Sc	30	0.054	0.5	$7.5\times {10}^4$	Duchemin et al. [56]
	${}^{44}$Ca(d,2n)${}^{44\mathrm{m}}$Sc				$1.9\times {10}^3$
23	${}^{\mathrm{nat}}$Zn(p,$\alpha$)${}^{61}$Cu	11.7	20	0.5	$1.4\times {10}^5$	Asad et al. [57]
	${}^{64}$Zn(p,$\alpha$)${}^{61}$Cu		40		$6.2\times {10}^5$
25	${}^{\mathrm{nat}}$W(p,n)${}^{186}$Re	15	20	5	$5.19\times {10}^4$	Kakavand et al. [59]
26	${}^{\mathrm{nat}}$Ca($\alpha$,p)${}^{43}$Sc	20	0.050	1.5	$5.5\times {10}^3$	Szkliniarz et al. [60]
	${}^{40}$Ca($\alpha$,p)${}^{43}$Sc				$5.8\times {10}^3$
	${}^{42}$Ca($\alpha$,np+pn)${}^{44}$Sc	29			$2.1\times {10}^3$
	${}^{42}$Ca($\alpha$,np+pn)${}^{44\mathrm{m}}$Sc				250
28	${}^{66}$Zn(p,n)${}^{66}$Ga	14.2	5	0.083	508	Cho et al. [62]
		8			195
29	${}^{\mathrm{nat}}$Zn(p,$\alpha$)${}^{61}$Cu	16.9	30	0.75	$2.72\times {10}^5$	do Carmo et al. [63]
	${}^{\mathrm{nat}}$Zn(p,x)${}^{66}$Ga				$3.45\times {10}^5$
	${}^{\mathrm{nat}}$Zn(p,x)${}^{67}$Ga				$2.24\times {10}^4$
	${}^{\mathrm{nat}}$Zn(p,n)${}^{68}$Ga				$2.21\times {10}^6$
	${}^{\mathrm{nat}}$Zn(p,pn)${}^{65}$Zn				225
30	${}^{\mathrm{nat}}$Ca(p,2n)${}^{47}$Sc	20.5	0.03	5	21.1	Misiak et al. [64]
31	${}^{64}$Ni(p,n)${}^{64}$Cu	12.5	20	5-7	$7.40\times {10}^6$	Xie et al. [65]
	${}^{42}$Ca(d,n)${}^{43}$Sc	6.8		4	$1.29\times {10}^5$
32	${}^{43}$Ca(p,n)${}^{43}$Sc	15.2	1	8	$9.1\times {10}^5$	Sitarz et al. [66]
	${}^{\mathrm{nat}}$Ca(p,n)${}^{44}$Sc				$5.0\times {10}^4$
	${}^{44}$Ca(p,n)${}^{44}$Sc				$2.24\times {10}^6$
	${}^{48}$Ca(p,2n)${}^{47}$Sc	22.8			$4.2\times {10}^5$
	${}^{48}$Ti(p,2p)${}^{47}$Sc	28			$2.0\times {10}^5$
33	${}^{103}$Rh(${}^3$He,x)${}^{100}$Rh	27	0.1	1	118	Ali et al. [67]
	${}^{103}$Rh(${}^3$He,$\alpha$+n)${}^{101\mathrm{m}}$Rh				30
	${}^{103}$Rh(${}^3$He,3n)${}^{103}$Ag				$2.88\times {10}^4$
	${}^{103}$Rh(${}^3$He,2n)${}^{104\mathrm{m}}$Ag				$1.13\times {10}^4$
	${}^{103}$Rh(${}^3$He,2n)${}^{104}$Ag				$8.87\times {10}^3$
	${}^{103}$Rh(${}^3$He,n)${}^{103}$Pd				270
35	${}^{\mathrm{nat}}$Ag($\alpha$,x)${}^{106\mathrm{m}}$Ag	37.6	2	1	238	Ditrói et al. [69]
36	${}^{14}$N(d,n)${}^{15}$O	3.5	40	0.017	$2.22\times {10}^6$	Iguchi et al. [70]
39	H${}_2$(${}^{70}$Zn${}^{15+}$,$\alpha$+p)${}^{67}$Cu	15	0.008	6.5	1.6	Souliotis et al. [73]
40	${}^{232}$Th(p,x)${}^{225}$Ac	480	72	36.5	$5.24\times {10}^5$	Robertson et al. [74]
	${}^{232}$Th(p,x)${}^{225}$Ra				$8.6\times {10}^4$
41	${}^{\mathrm{nat}}$Mo(p,x)${}^{94}$Tc	26	0.065	5	$2.00\times {10}^4$	Ahmed et al. [75]
	${}^{\mathrm{nat}}$Mo(p,x)${}^{95}$Tc				$1.55\times {10}^4$
	${}^{\mathrm{nat}}$Mo(p,x)${}^{95\mathrm{m}}$Tc				86.9
	${}^{\mathrm{nat}}$Mo(p,x)${}^{96}$Tc				$2.42\times {10}^3$
	${}^{\mathrm{nat}}$Mo(p,x)${}^{99\mathrm{m}}$Tc				$7.14\times {10}^4$
	${}^{\mathrm{nat}}$Mo(p,x)${}^{99}$Mo				$3.19\times {10}^3$
42	${}^{68}$Zn(p,n)${}^{68}$Ga	13	35	1.5	$1.44\times {10}^8$	Alnahwi et al. [76]
43	${}^{\mathrm{nat}}$Mo(n,x)${}^{99\mathrm{m}}$Tc	11	20	0.083	39	Kambali et al. [77]
	${}^{98}$Mo(n,x)${}^{99}$Mo				33
44	${}^{44}$Ca(p,n)${}^{44}$Sc	18	18.4	5.75	$3.21\times {10}^6$	van der Meulen et al. [78]
45	${}^{16}$O(p,$\alpha$)${}^{13}$N	16	5.6–34	3.3	$9.25\times {10}^6$	Abel et al. [79]
	${}^{18}$O(p,n)${}^{18}$F				$2.15\times {10}^5$
46	${}^{\mathrm{nat}}$Er(p,x)${}^{167}$Tm	22.8	50	8	$2.61\times {10}^5$	Heinke et al. [80]
47	${}^{155}$Gd(p,n)${}^{155}$Tb	10	50	8	$2.0\times {10}^5$	Favaretto et al. [81]
	${}^{156}$Gd(p,2n)${}^{155}$Tb	23			$1.7\times {10}^6$
48	${}^{89}$Y(p,x)${}^{89}$Zr	12	50	5	$3.45\times {10}^6$	Cisterino et al. [82]

,

Table A3. Research reactor-produced radioisotopes.

n°	Production Route	Target Mass [mg]	Neutron Flux [n/cm${}^2$s]	Irradiation Time [h]	${\mathbf{A}}_{\mathbf{EOB}}$ [kBq]	Reference
3	${}^{130}$Te(n,$\gamma$)${}^{131}$Te${\stackrel{}{\to }}^{131}$I	5000	$1.0\times {10}^{12}$	4	400	Achoribo et al. [37]
9	${}^{168}$Er(n,$\gamma$)${}^{169}$Er	10	$8.0\times {10}^{13}$	504	$3.88\times {10}^6$	Chakravarty et al. [43]
19	${}^{\mathrm{nat}}$Yb${}_2$O${}_3$(n,$\gamma$)${}^{169}$Yb	5	$5.0\times {10}^{13}$	168	$5.84\times {10}^5$	Saxena et al. [53]
24	${}^{160}$Gd(n,$\gamma$)${}^{161}$Gd${\stackrel{}{\to }}^{161}$Tb	100	$1.0\times {10}^{14}$	108	$3.92\times {10}^6$	Aziz et al. [58]
34	${}^{64}$Zn(n,$\gamma$)${}^{65}$Zn	1000	$4.5\times {10}^{13}$	0.5	$6.16\times {10}^3$	Karimi et al. [68]
38	${}^{185}$Re(n,$\gamma$)${}^{186}$Re	1	$3.0\times {10}^{13}$	96	$2.13\times {10}^6$	Pourhabib et al. [72]
	${}^{187}$Re(n,$\gamma$)${}^{188}$Re				$4.55\times {10}^6$

and

Table A4. Other technologies produced radioisotopes.

n°	Technology	Production Route	Kinetic Energy [MeV]	Current [$\mathsf{\mu }$A]	Irradiation Time [h]	${\mathbf{A}}_{\mathbf{EOB}}$ [kBq]	Reference
6	Plasma focus device	${}^{12}$C(d,n)${}^{13}$N	0.33	not applicable	0.097	0.014	Shirani et al. [40]
27	Tandetron	${}^{\mathrm{nat}}$Zn(p,n)${}^{66}$Ga	9	$6.3\times {10}^{-3}$	0.106	8.5	Fraile et al. [61]
		${}^{\mathrm{nat}}$Zn(p,n)${}^{68}$Ga				99
37	Linear particle accelerator	${}^{75}$As(p,4n)${}^{72}$Se	49.5	163.5	68.9	$1.38\times {10}^7$	DeGraffenreid et al. [71]

, in chronological order. In particular, in Table A2, Table A3 and Table A4, experimental results are reported with the same precision (number of decimal digits) informed by their respective authors. Moreover, quantities calculated from these results were rounded according to the number of significant digits.

With regard to the amount of radioisotope produced, all data collected from these papers were converted to activities at the end of the bombardment in order to present all results in the same physical unit. Each selected article presented the information differently, because the production technology used was not the same, or because different types of measurements were conducted.

Most of the selected studies utilizing cyclotron accelerators presented their production results in one of three ways: thick target yield, saturation yield, or activity at the end of the bombardment. In order to compare the production with studies that utilized other accelerator technologies or physical processes, all measurements or calculated results presented in these papers were converted to activities at the end of the bombardment, by using Equations (1) and (2) [83]:

$$\begin{array}{cc}\hfill {\mathrm{A}}_{\mathrm{EOB}}& =\frac{\mathrm{TTY}i}{\lambda }(1-e^{-\lambda t}),\hfill \end{array}$$

(1)

$$\begin{array}{cc}\hfill {\mathrm{A}}_{\mathrm{EOB}}& =\mathrm{SY}i(1-e^{-\lambda t}),\hfill \end{array}$$

(2)

where ${\mathrm{A}}_{\mathrm{EOB}}$ is the activity at the end of the bombardment [$\mathrm{M}\mathrm{Bq}$], t is the irradiation time [$\mathrm{h}$], TTY is the thick target yield [$\mathrm{M}\mathrm{Bq}/\mathrm{A}\mathrm{h}$], $\lambda$ is the decay constant of the produced radioisotope [${\mathrm{h}}^{-1}$], i is the beam current [$\mathrm{A}$], and SY is the saturation yield [$\mathrm{M}\mathrm{Bq}/\mathrm{A}$].

Some authors performed experimental activity measurements hours after the end of the bombardment. In those cases, the ${\mathrm{A}}_{\mathrm{EOB}}$ was estimated with Equation (3) using the measurement activity ${\mathrm{A}}_{\mathrm{T}}$, the time after irradiation T [$\mathrm{h}$], and the decay constant $\lambda$.

$${\mathrm{A}}_{\mathrm{EOB}}=\frac{{\mathrm{A}}_{\mathrm{T}}}{e^{-\lambda \mathrm{T}}}.$$

(3)

3. Results and Discussion

Table A1 and Figure 1a show that 81.3% of the selected papers describe the use of cyclotrons for radioisotope production. The reason behind this large number is that cyclotron is a well-established technology. The use of hospital cyclotrons to produce radioisotopes began in the 1950s [84], over 70 years ago, and the largest medical centers in the world already have cyclotrons. Globally, there are more than 1500 cyclotrons [85]. Protons are usually the types of particles accelerated in cyclotrons, justifying their large appearance numbers (34), as indicated in Table A1 and Figure 1b. In addition to cyclotrons, research reactors (without the presence of enriched uranium), linear particle accelerators (LINACS), plasma focus devices, and tandetrons have also been used for radioisotope production in the selected papers. These four production technologies add up to only 18.7% of the papers, as shown in Figure 1a.

Figure 2 shows that 53 unique radioisotopes were produced in the selected papers; the most cited medical radioisotopes were ${}^{44}$Sc, ${}^{66}$Ga, ${}^{99\mathrm{m}}$Tc, ${}^{68}$Ga, ${}^{99}$Mo, and ${}^{89}$Zr. Together, they appear in 25% of the selected papers.

As shown in Figure 2, the most cited (7 distant citations) radioisotope is the ${\beta }^{+}$-emitter radioisotope ${}^{44}$Sc. Its half-life is often described to be 3.97 h, but a recent study reported a half-life of 4.04 h [86]. This radioisotope was investigated for PET imaging [87] and might be a suitable alternative to ${}^{68}$Ga (with a half-life of 1.13 h) due to its half-life, almost four times greater than ${}^{68}$Ga, which allows the synthesis of a variety of radiotracers with longer pharmacokinetic profiles [88]. Additionally, ${}^{44}$Sc has been proposed as a matching pair of ${}^{47}$Sc (with a half-life of 80.4 h, a ${\beta }^{-}$-emitter that has been proposed for radioimmunotherapy [89]) for PET imaging, dosimetry estimation, and the assessment of radionuclide therapeutic responses [89,90,91]. Currently, there are two main routes for ${}^{44}$Sc production, i.e., via ${}^{44}$Ti/${}^{44}$Sc generators, but due to difficulties in the production of the parent isotope (${}^{44}$Ti), in the chemical separation, and the purification process, the applicability of this generator is very limited, making its clinical implementation difficult [92]. Cyclotron production is the other main source of the ${}^{44}$Sc radioisotope. In the selected papers, the production occurred via irradiation on ${}^{44}$Ca, ${}^{42}$Ca, or ${}^{\mathrm{nat}}$Ca targets through ${}^{44}$Ca(p,n)${}^{44}$Sc, ${}^{44}$Ca(d,2n)${}^{44}$Sc, ${}^{42}$Ca($\alpha$, np+pn)${}^{44}$Sc, and ${}^{\mathrm{nat}}$Ca(p,n)${}^{44}$Sc reaction routes, as indicated in Table A1.

With three distinct citations, ${}^{99}$Mo is a ${\beta }^{-}$-emitter radioisotope without direct application in medicine. However, ${}^{99}$Mo decays into ${}^{99\mathrm{m}}$Tc, the most used radioisotope for SPECT imaging, as mentioned in Section 1. It makes ${}^{99}$Mo and ${}^{99\mathrm{m}}$Tc the most relevant pair of radioisotopes in nuclear medicine, justifying a total of seven distinct paper citations (four papers cited ${}^{99\mathrm{m}}$Tc and three cited ${}^{99}$Mo).

Figure 2 also shows that two radioisotopes of gallium, namely ${}^{68}$Ga and ${}^{66}$Ga, had seven citations combined. Moreover, ${}^{66}$Ga (with a half-life of 9.49 h) is a ${\beta }^{+}$-emitter used in PET. Its long half-life makes it possible to perform next-day PET imaging, improving the image contrast [93]. ${}^{66}$Ga has been tested for studies of slow dynamic processes, such as lymphatic transport [94], and imaging of tumor angiogenesis using monoclonal antibodies [95]. Moreover, there have been reports of the use of ${}^{66}$Ga in a ${}^{66}$Ga-labeled somatostatin as an imaging agent for receptor-positive tumors [96] and in preclinical imaging of the GRPR expression in prostate cancer [97]. Table A1 shows that, in the papers found in our review, ${}^{66}$Ga was produced in cyclotrons via irradiation on zinc targets through the ${}^{\mathrm{nat}}$Zn(p,x)${}^{66}$Ga and ${}^{66}$Zn(p,n)${}^{66}$Ga reaction routes. Moreover, as used in positron emission tomography, ${}^{68}$Ga is a ${\beta }^{+}$-emitter radioisotope applied in the diagnostics of prostate cancer [98] and neuroendocrine neoplasms [99]. It is also produced in cyclotrons via irradiation on zinc targets through the ${}^{\mathrm{nat}}$Zn(p,x)${}^{68}$Ga and ${}^{68}$Zn(p,n)${}^{68}$Ga reaction routes, as shown in Table A1. Another possibility to produce this radioisotope is by a ${}^{68}$Ge/${}^{68}$Ga generator [100,101], which can consistently supply ${}^{68}$Ga for more than a year due to the ${}^{68}$Ge half-life of 271 days.

The last radioisotope with three citations in distinct papers is ${}^{89}$Zr (half-life of 78.4 h), a ${\beta }^{+}$-emitter, which is an effective imaging tool for antibody or immune-based PET, referred to as “immuno-PET” [102], which represents by far the widest field of the ${}^{89}$Zr medical application. However, there are studies on ${}^{89}$Zr-labeled nanoparticles (NPs) with promising results in tumor detection, drug monitoring, inflammation imaging, tumor-associated macrophages and sentinel lymph mapping [103]. Table A2 shows that the three papers produced ${}^{89}$Zr in a cyclotron via irradiation on yttrium targets through the ${}^{89}$Y(p,n)${}^{89}$Zr reaction route.

As shown in Table A2, Table A3 and Table A4, there is a wide range of ${\mathrm{A}}_{\mathrm{EOB}}$ values, spanning multiple orders of magnitudes (from a few $\mathrm{Bq}$ to ${10}^{14}$$\mathrm{Bq}$). This extreme variation is a consequence of the distinct production routes, technologies, irradiation times, and electric currents or neutron fluxes used for producing each radioisotope reported in these tables. As indicated in Table A2 n° 32, for the ${}^{44}$Sc production, the activity obtained with a natural target is two orders of magnitude lower than the one obtained with an enriched target. If compared to an equivalent enriched target, a natural target will contain only a fraction of the isotopes available for a given radioisotope production route. Hence, a much lower yield is expected for the natural material, if compared to the enriched target.

4. Conclusions and Future Directions

An integrative review of alternative methods for radioisotope production was carried out. Papers containing experimental production results, published from 2010 to 2021, were considered. A total of 48 papers were selected and analyzed. Among these works, cyclotrons were the most adopted particle accelerator technologies for radioisotope production, followed by HEU-free research reactors. Combined, these two methods were adopted in nearly 94% of the papers evaluated in this review.

Regarding the radioisotopes, the most-cited one was ${}^{44}$Sc, with seven distinct citations, followed by ${}^{68}$Ga and ${}^{99\mathrm{m}}$Tc with four citations, and ${}^{66}$Ga, ${}^{99}$Mo, and ${}^{89}$Zr with three citations each. All of these radioisotopes are applied in PET (${}^{44}$Sc, ${}^{68}$Ga, ${}^{66}$Ga, ${}^{89}$Zr), or SPECT (${}^{99}$Mo, ${}^{99\mathrm{m}}$Tc) diagnoses.

It is clear that new compact, environmentally safe, and cost-effective technologies for producing radioisotopes are needed, and that there is interest from the scientific community in developing new, HEU-free production routes that mitigate the generation of radioactive waste. Moreover, new, decentralized production methods could minimize possible future problems similar to the shortage crisis of ${}^{99}$Mo/${}^{99\mathrm{m}}$Tc generators in 2009. As shown in this review, HEU-free medical-radioisotope production methods have been adopted and reported in the last decade. However, more research in this area is crucial for attaining greener and high-yield radioisotope production methods, capable of competing with the current uranium-based standard production techniques.

A possible alternative would be producing radioisotopes via reaction routes triggered by bremsstrahlung $\gamma$-rays, obtained, for example, from the deceleration of high-energy electron beams. Depending on the target material, it is possible to estimate the energy range required for the photons to trigger photonuclear reactions capable of producing the radioisotope of interest. Moreover, as in this process, the radioisotope of interest is a product of the reaction, rather than a byproduct, as in the case of nuclear fission, the adoption of this process would mitigate the radioactive waste. Despite being reported in the literature as a possible scheme for producing radioisotopes [104,105], such as, for example, the ${}^{99}$Mo (which decays to ${}^{99\mathrm{m}}$Tc) [105,106], no experimental works were found among the papers selected for this review according to the aforementioned criteria.

The high-energy electron beams required to generate bremsstrahlung photons capable of producing radioisotopes via photonuclear reactions might be soon attainable by using laser wakefield accelerators (LWFAs) [107,108]. In this scheme, under the proper conditions, an intense laser pulse propagating in plasma drives a high-amplitude wakefield on its trail, which can be used as a compact accelerating structure for the plasma electrons (self-injection), or for an externally injected electron beam. LWFAs are capable of producing eight GeV electron beams at the exit of a 20 cm-long plasma capillary [109]. Moreover, stable long-term kHz operations of LWFAs have recently been reported [110]. In addition to the low amount of radioactive waste produced via bremsstrahlung $\gamma$-rays reaction routes, the compactness of this technology could allow for local radioisotope production, mitigating the aforementioned transportation problems associated with other technologies that require centralized production.

Despite the LWFA being a promising technology [111], recent studies [106,112,113] indicate that, even if high-energy, high-repetition-rate laser systems are adopted, due to the low yield provided by the photon activation process, very long irradiation times would still be required to produce a given radioisotope in quantities that could supply typical daily doses for medical applications. Hence, laser technology development, as well as further studies on LWFA optimization for radioisotope production—by using, for example, artificial intelligence algorithms [114,115]—may enable the development of greener, non-centralized radioisotope production processes.

Another possible laser-based radioisotope production method is the target normal sheath acceleration (TNSA) [112,116,117], in which an ultra-intense–ultra-short laser pulse interacts with a solid target. A fraction of the laser’s electromagnetic field is transferred to the target, creating plasma and heating electrons to relativistic temperatures. The electrons expand and exit the target, forming a charge separation localized next to its surface. This process generates a strong electrostatic field, perpendicular to the target surface, capable of accelerating ions from the target to several MeVs [118,119]. This ion irradiation can be used to produce radioisotopes via (p,x) and (d,x) reaction routes. The first experimental TNSA production of ${}^{99\mathrm{m}}$Tc through the ${}^{100}$Mo(p,2n)${}^{99\mathrm{m}}$Tc reaction route occurred in 2013 [120]. In the experiment, a 1 mm thick aluminum target was illuminated by a 50 J, 1 ps, $5\times {10}^{20}$ W/cm${}^2$ laser pulse. For a single shot, a production of 8.25 kBq of ${}^{99\mathrm{m}}$Tc was estimated. Moreover, simulation results with the 3D PIC simulation code Mandor [121] show that a 10 J, 100 fs laser might produce 1.93 kBq of ${}^{99\mathrm{m}}$Tc in a single shot through the same reaction route, ${}^{100}$Mo(p,2n)${}^{99\mathrm{m}}$Tc. Hence, in principle, it should be possible to produce 300 GBq of ${}^{99\mathrm{m}}$Tc after 6 h of continuous 10 kHz laser irradiation on a highly enriched ${}^{100}$Mo target [122].

Despite the encouraging progress in both experimental and theoretical research studies on laser-based radioisotope production, there are still many improvements to be made before laser-driven radioisotope production facilities become viable for medical applications.