- Email: [email protected]
Cross-section measurement of 44mSc,47Sc, 48Sc and 47Ca for an optimized 47Sc production with an 18 MeV medical PET cyclotron
Applied Radiation and Isotopes 143 (2019) 18–23
Contents lists available at ScienceDirect
Applied Radiation and Isotopes journal homepage: www.elsevier.com/locate/apradiso
Cross-section measurement of 44mSc,47Sc, 48Sc and 47Ca for an optimized 47 Sc production with an 18 MeV medical PET cyclotron
T
⁎
Tommaso Stefano Carzaniga , Saverio Braccini Albert Einstein Center for Fundamental Physics (AEC), Laboratory for High Energy Physics (LHEP), University of Bern, Sidlerstrasse 5, CH-3012 Bern, Switzerland
H I GH L IG H T S
cross-sections Ca(p,n) Sc, Ca(p,n) Sc, Ca(p,2n) Sc and Ca(p,pn) Ca were measured. • The 18 MeV proton cyclotron is suitable to produce Sc/ Sc theranostic pair. • An • 20–30 MeV cyclotrons are valuable options for further Sc production op- timization. 44
44m
48
48
48
47
43
48
47
47
47
A R T I C LE I N FO
A B S T R A C T
Keywords: Therapeutic radioisotopes Medical cyclotrons Scandium Cross-section Theranostics
Novel medical radioisotopes for both diagnostic and therapy are essential for the future development of personalized nuclear medicine. Among them, radiometals can be used to label both proteins and peptides and encompass promising theranostic pairs. The optimized supply of radiometals in quantity and quality for clinical applications represents a scientific and technological challenge. 47 Sc is a β − emitter that forms a theranostic pair together with one of the β + emitters 43Sc or 44Sc. It can be produced at a medical cyclotron by proton bombardment of an enriched calcium oxide target. The parasite production of 48Sc undermines the 47Sc purity, which strongly depends on the energy of protons impinging the target and on the thickness of the target material. For this purpose, an accurate knowledge of the production cross-sections is mandatory. In this paper, we report on the measurement of the cross-section of the reactions 44Ca(p,n)44 mSc,48Ca (p,n)48Sc, 48Ca(p,2n)47Sc and 48Ca(p,pn)47Ca using natCaCO3 targets performed at the Bern University Hospital cyclotron laboratory. On the basis of the obtained results and of the isotopic composition of commercially available enriched target materials, the thick target yields and the purity were calculated to assess the optimal irradiation conditions.
1. Introduction The production of novel radioisotopes is essential for the development of personalized nuclear medicine. The theranostic approach is based on a pair of isotopes, one for diagnostic and one for therapy, used to label the same compound. Several pairs of nucleides for theranostics are under study or in early adoption in clinical applications, as reported in Table 1. Radionuclide therapy is promising to be successful in treating persistent diseases with low toxic side-effects thanks to the localized dose deposition. The absorbed dose in target tissues in therapeutic procedures ranges between 20 and 60 Gy. Three families of isotopes are nowadays under study for adoption in therapy: β − , α and Auger electrons emitters. This paper will focus on β − emitters.
⁎
While for imaging hundreds of MBq are enough for one examination (typically 350 MBq of 18F-FDG and less than 200 MBq of 99mTc), more than one GBq of a β − emitter must be injected in each patient for therapy. An ideal therapeutic β − emitter decays with a low energy β − and low (100–300 keV) energy gamma suitable for SPECT imaging. A good example is 177Lu, that, together with 90Y, is becoming the main Radionuclide Therapy (RNT) agent. The PET isotope 68Ga is used as its diagnostic counterpart, to demonstrate the uptake in the tumor of the labeled molecule for the treatment of Neuroendocrine Tumors (NET). Activities of about 7 GBq of 177Lu per patient are needed. 47 Sc is very similar to 177Lu for its chemical and physical characteristics. 47Sc is a β − emitter and is foreseen as the therapeutic partner of the β + emitters 43Sc and 44Sc, both under study for PET imaging
Corresponding author. E-mail address: [email protected] (T.S. Carzaniga).
https://doi.org/10.1016/j.apradiso.2018.10.015 Received 11 July 2018; Received in revised form 13 September 2018; Accepted 10 October 2018 Available online 15 October 2018 0969-8043/ © 2018 Elsevier Ltd. All rights reserved.
Applied Radiation and Isotopes 143 (2019) 18–23
T.S. Carzaniga, S. Braccini
Table 1 Theranostic pairs under study or in early adoption. Imaging (PET/SPECT)
Therapy
123
131
124
I, I Y 61 64 Cu, Cu (Anderson and Ferdani, 2009) 111 In, 68Ga 43 Sc, 44Sc (Singh et al., 2015; van der Meulen et al., 2015; Rösch and Baum, 2011) 86
Table 3 Decay properties of 44mSc (Chen et al., 2011), 46Sc (Wu, 2000), 2007), 48Sc (Burrows, 2006) and 47Ca (Burrows, 2007). t1/2 [d]
I Y 67 Cu 90 177 Y, Lu 47 Sc (Domnanich et al., 2017) 90
44m 44m
Sc → Sc →
Target
Route
p
46
48
Ca Ca
48
Ti Ti Ca 49 Ti 44 Ca 46 Ca 48 Ca 47 Ti 50
d
α n
γ
46
48 48
Ca Ti
Ca(p, γ )47Sc Ca(p,2n)47Sc (Misiak et al., 2017a) 48 Ca(p,pn)47Ca → 47Sc 48 Ti(p,2p)47Sc (Kolsky et al., 1998) 50 Ti(p, α )47Sc 46 Ca(d,n)47Sc 49 Ti(d, α )47Sc 44 Ca(α , p)47Sc (Minegishi et al., 2016) 46 Ca(n, γ )47Ca → 47Sc (Domnanich et al., 2017) 48 Ca(n,2n)47Ca → 47Sc 47 Ti(n,p)47Sc (Hosseini et al., 2017; DeilamiNezhad et al., 2016; Domnanich et al., 2017) 48 Ca(γ , n)47Ca → 47Sc 48 Ti(γ , p)47Sc (Mamtimin et al., 2015) 48
2.442(4)
Ti
47
47
Ti
3.3492(6)
48
48
Ti
1.819(4)
142.6(7) - 68.4(6) 203.9(8) - 31.6(6) 227.3(25) - 80.98(18)
47
47
4.536(3)
242.66(49) - 73(15)
Sc →
46
Sc Ca
46
Sc →
Impinging particle
44
46
Sc →
Table 2 47 Sc production routes.
44
E β− [keV] - BR β− [%]
Ca →
Sc
83.79(4)
111.8(3) - 99.9964(7)
47
Sc (Burrows,
E γ [keV] – BRγ [%] 271.241(10) - 86.7(3) 1126.06(4) - 1.20(7) 1001.83(3) - 1.20(7) 1157.002(3) - 1.20(7) 1120.545(4) - 99.987(1) 889.277(3) - 99.984(1) 159.381(15) - 68.3(4) 1312.120(12) - 100.1(6) 983.526(12) - 100.1(6) 1037.522(12) - 97.6(7) 175.361(5) - 7.48(10) 1212.880(12) - 2.38(4) 1312.120(12) - 100.1(6) 1297.09(10) - 67(13) 489.23(10) - 5.9(12) 807.86(10) - 5.9(12)
radioisotope production and multi-disciplinary research activities (Braccini and Scampoli, 2016). The beam is brought to the second bunker by means of a 6 m long external Beam Transport Line (BTL), which was used for the measurements presented in this paper. The beam extracted to the BTL has a mean energy of (18.3 ± 0.3) MeV with an RMS of 0.40 MeV (Braccini, 2013; Nesteruk et al., 2017). The experimental method used in this work was very similar to the one described in our earlier publication on cross-section measurements for the production of the diagnostic partners 43Sc and 44Sc (Carzaniga et al., 2017). It is based on the irradiation of the full mass of a thin target by a proton beam with a flat profile. UniBEaM detectors (Auger et al., 2016) were used to monitor the beam on-line. These beam profilers were developed by our group and are based on silica doped fibers passing through the beam. Their commercialization was licensed by the University of Bern to the company D-Pace (Potkins et al., 2017). This method has the advantage that the target has not to be necessarily uniform in thickness provided that the energy of the protons can be considered constant. The target holder geometry was improved to assure the optimal centering of the target material with respect to the collimator, as shown in Fig. 1. The support material, i.e. the so called backing, for the target material was a 2 mm thick aluminum disc with a 0.6 mm deep pocket of 4.2 mm in diameter (Fig. 1 left). The target material was deposited in the pocket by sedimentation from a suspension of a few milligrams of nat CaCO3 in ultra-pure water. The isotopic composition of natural and commercially available CaCO3 is reported in Table 4. Ca-48 enriched material can be purchased in different forms (carbonate, chloride, oxide, nitrate, metal, gluconate, iodide, fluoride) and enrichment levels (64%-97 +%). The water was then evaporated at ∼ 70 °C by means of a small heating plate (Fig. 1 right). The mass of the deposited material was assessed with an analytical balance (METTLER TOLEDO AX26 DeltaRange) with a readability of 2 μ g and a repeatability of 4 μ g. After cooling, the targets were covered with a 10 μm thin aluminum foil for protection. Since the thickness of the target was estimated to be of the order of few tens of μm, the beam energy was considered constant within the uncertainties within the full irradiated mass.
(Carzaniga et al., 2017). Furthermore, 47Sc is characterised by a 159 keV gamma line (∼ 68% branching ratio), allowing for SPECT imaging during treatment. There are many routes to yield 47Sc (Table 2). Promising ones for a wide diffusion of this radioiotope could be based on medical cyclotrons (Braccini, 2016) commonly used to produce 18F, the main radioisotope for PET imaging. These accelerators provide proton (sometimes also deuteron) beams in the energy range of 15–25 MeV with intensities of 50–300 μ A. 47Sc can be produced using solid target stations. To optimize the yield together with the radio-nuclide purity, the precise knowledge of the cross-section of 47Sc and eventual impurities as a function of the beam energy is of paramount importance. Data reported in literature and accessible via the EXFOR database (iaea) do not fully cover the energy range of medical cyclotrons and are sometimes lacking or inconsistent. In the framework of a research program aimed at the production of scandium for theranostics ongoing at the Bern medical cyclotron laboratory (Braccini, 2013), 47Sc production via proton bombardment of 48 Ca enriched calcium oxide was considered. The main impurity of this production route is 48Sc that is unsuitable for therapy due to its high energy gammas. On the other hand, the concomitant production of 47 Ca, which then decays into 47Sc, enhances the produced activity during bombardment. Furthermore, 47Ca can be used as a generator if promptly separated after EoB. According to the enrichment level, other scandium isotopes can be produced, the long lived 44mSc in particular. The main properties of 47Sc,48Sc and 47Ca are summarized in Table. 3. In this paper we report on the measurement of the cross-sections of the reactions 44Ca(p,n)44mSc,48Ca(p,n)48Sc, 48Ca(p,2n)47Sc and 48Ca (p,pn)47Ca. The obtained results were used to assess the radio-nuclide purity when bombarding enriched calcium targets.
3. Cross-section measurements 2. Materials and methods The aim of this paper is to study the optimal production of 47Sc by maximising the activity while minimizing the impurities. To measure the cross-sections, irradiations were performed at several energies on target with currents of ∼10 nA for ∼30 mins. The energy was modulated by using aluminum absorbers (Carzaniga et al., 2017).
The laboratory at the Bern University Hospital (Inselspital) (Braccini, 2013) features an IBA Cyclone 18/18 high current cyclotron and two bunkers with independent access. This solution is unusual for a hospital-based facility and was conceived to perform both medical 19
Applied Radiation and Isotopes 143 (2019) 18–23
T.S. Carzaniga, S. Braccini
Fig. 1. Target holder geometry aimed at a optimal centering of the material with respect to the collimator (left). Targets drying on a heating plate kept at 70 °C (right). Table 4 Calcium isotopic percentages in calcium carbonate powder enriched in quoted from ISOFLEX (isoflex). 40
Natural (%) 48 Ca enrich. (%)
Ca
96.941 27.9
42
Ca
0.647 0.3
43
44
Ca
0.135 0.1
Ca
2.086 2.2
46
Ca
0.004 <0.1
48
48
Ca
Ca
0.187 69.2
47
Sc is obtained from 48Ca directly via the reaction 48Ca(p,2n)47Sc and indirectly via the reaction 48Ca(p,pn)47Ca that first produces 47Ca, which then decays in 47Sc with an half-life of ∼4.5 days. After each irradiation, the characteristic gamma-ray at 159 keV emitted by 47Sc and the 1297 keV gamma-ray emitted by 47Ca were measured with an HPGe detector for several hours, in order to reduce the statistical uncertainty on the integral area to a negligible level. The results of the 48Ca(p,2n)47Sc cross-section measurements are presented in Fig. 2; for completeness, the numerical values are reported in the Appendix (Table 5). Reasonable agreement is found with the TENDL (Koning and Rochman, 2012) code, especially below 14 MeV. No comparison with previous experimental measurements is possible. The results of the 48Ca(p,pn)47Ca cross-section measurements are presented in Fig. 3; for completeness the numerical values are reported in the Appendix (Table 6). Compared to TENDL (Koning and Rochman, 2012) predictions, the data measured in this study present a faster rise of the cross-section. A maximum is found at around 17 MeV, while TENDL suggests a value higher than 25 MeV. Data available in the literature (Michel et al., 1997) are scarse and are in reasonable agreement with both TENDL and our measurements. 48 Sc is the main impurity that would be produced by irradiating an
Fig. 3. Cross section of the reaction
48
Fig. 4. Cross section of the reaction
Fig. 2. Cross section of the reaction
48
Ca(p,pn)47Ca.
48
Ca(p,n)48Sc.
enriched 48CaO target. To study the 48Ca(p,n)48Sc reaction, the characteristic gamma-ray at 1312 keV emitted by 48Sc was measured as for 47 Sc and 47Ca. The results of the 48Ca(p,n)48Sc cross-section measurements are presented in Fig. 4; for completeness the numerical values are reported in the Appendix (Table 7). Good agreement is found with the TENDL (Koning and Rochman, 2012) code. The few experimental data below 18 MeV available in the literature (de Waal et al., 1971; Singh et al., 1982; Michel et al., 1997) concentrates at low energies (<5 MeV) and are found in reasonable agreement with our measurements. 44m Sc is another long lived impurity that would be produced by irradiating an enriched 48CaO target, since the fraction of 44Ca might not be negligible. To study the 44Ca(p,n)44mSc reaction, the
Ca(p,2n)47Sc. 20
Applied Radiation and Isotopes 143 (2019) 18–23
T.S. Carzaniga, S. Braccini
Fig. 5. Cross section of the reaction
44
Ca(p,n)
44m
Fig. 6. Thick target saturation yields for a 69% enriched
Fig. 8. 43,44,44m,48Sc impurities at saturation for a 69% enriched 48CaO target as a function of the target thickness. The entry energy in the pellet is 17.75 MeV.
Sc.
48
CaO target.
Fig. 9. 47Sc yield and purity for a 69% enriched 48CaO at saturation with an entry energy in the pellet of 17.75 MeV as a function of the target thickness.
(∼20%) of the 1297 keV gamma. The gamma branching ratio was a negligible source of uncertainties for the other reactions. A further evaluated impurity was 46Sc. According to TENDL the threshold for the 48Ca(p,3n)46Sc reaction is at ∼ 20 MeV. Since even in the enriched material a tiny amount of 46Ca is always present, 46Sc could also be produced via the reaction 46Ca(p,n)46Sc. Nevertheless, 46 Sc was searched for in all the measurements and no signal was found. It is known from other experiments (Misiak et al., 2017b) that 46Sc is created in the considered energy range but the produced activities were below our MDA (Minimum Detectable Activity). We are considering performing dedicated experiments to measure 46Sc production crosssections in future. Fig. 7. 43,44,44m,48Sc impurities for a thick 69% enriched turation.
48
CaO target at sa-
4. Study of production yield and purity A study of the thick target saturation yield has to be performed in order to optimize 47Sc production by 48Ca irradiation. Fig. 6 reports the thick target saturation yields for 47Sc, 48Sc and 47Ca using the enriched 48 Ca target material by ISOFLEX of Table 4 in the form of calcium oxide as a function of the impinging proton energy. CaO is our form of choice for Sc production since it gives a larger yield and the carbonate will dissociate carbon dioxide when heated up by proton bombardment, causing degassing issues. Since the purity of material enriched in 48Ca nowadays available on the market ranges from 64% to 97+%, 43Sc, 44 Sc and the longer lived 44mSc have also to be taken into account as impurities. 43Sc and 44Sc yields were studied in Carzaniga et al. (2017). The threshold for the 48Ca(p,n)48Sc reaction is found to be below 3 MeV and the maximal cross-section at around 10 MeV (Fig. 4). On the other hand, the threshold for the 48Ca(p,2n)47Sc reaction is found around 10 MeV with a probable maximum of its cross-section at around
characteristic gamma-ray at 271 keV emitted by 44mSc was measured as for 47Sc,48Sc and 47Ca. The results of the 44Ca(p,n)44mSc cross-section measurements are presented in Fig. 5; for completeness the numerical values are reported in the Appendix (Table 8). The experimental data available in the literature (Levkowskij, 1991) are found in reasonable agreement with our measurements. Up to 12 MeV good agreement is found with the TENDL (Koning and Rochman, 2012) code while at higher energies the prediction underestimates our results. The main experimental uncertainty for the reactions 48Ca(p,2n)47Sc, 48 Ca(p,n)48Sc and 44Ca(p,n)44mCa was due to the flatness of the beam (up to 5%). Other sources of uncertainties are the beam current (∼1%), the activity of the source used to calibrate the HPGe detector (∼1%) and the HPGe calibration procedure (∼1%). For the reaction 48Ca(p,pn)47Ca the main experimental uncertainty was due to the branching ratio 21
Applied Radiation and Isotopes 143 (2019) 18–23
T.S. Carzaniga, S. Braccini
Table 5 Cross-section data of the reaction
48
Ca(p,2n)47Sc.
E [MeV]
σ [mb]
E [MeV]
σ [mb]
E [MeV]
σ [mb]
18.4 ± 0.5 17.2 ± 0.5 16.0 ± 0.5
1107 ± 56 1105 ± 57 1058 ± 55
14.6 ± 0.5 13.2 ± 0.6 11.6 ± 0.6
967 ± 49 755 ± 39 342 ± 18
9.9 ± 0.6 8.9 ± 0.7
16 ± 3 0
Table 6 Cross-section data of the reaction
48
Ca(p,d)47Ca.
E [MeV]
σ [mb]
E [MeV]
σ [mb]
E [MeV]
σ [mb]
18.4 ± 0.5 17.2 ± 0.5
209 ± 42 233 ± 48
16.0 ± 0.5 14.6 ± 0.5
199 ± 43 120 ± 27
13.2 ± 0.6 11.6 ± 0.6
63 ± 20 0
Table 7 Cross-section data of the reaction
48
Ca(p,n)48Sc.
E [MeV]
σ [mb]
E [MeV]
σ [mb]
E [MeV]
σ [mb]
18.4 ± 17.2 ± 16.0 ± 14.6 ± 13.2 ±
199 ± 220 ± 243 ± 321 ± 425 ±
11.6 ± 0.6 9.9 ± 0.6 8.9 ± 0.7 6.7 ± 0.8
711 ± 890 ± 954 ± 822 ±
5.3 ± 4.4 ± 3.4 ± 3.0 ±
570 ± 448 ± 245 ± 122 ±
0.5 0.5 0.5 0.5 0.6
Table 8 Cross-section data of the reaction
44
28 19 33 32 41
44 51 49 49
0.9 1.0 1.1 1.2
39 30 14 10
Ca(p,n)44mSc.
E [MeV]
σ [mb]
E [MeV]
σ [mb]
E [MeV]
σ [mb]
18.4 ± 17.2 ± 16.0 ± 14.6 ±
37 ± 45 ± 60 ± 66 ±
13.2 ± 0.6 11.6 ± 0.6 10.9 ± 0.6 9.9 ± 0.6
58 ± 42 ± 24 ± 24 ±
8.9 ± 6.7 ± 5.3 ± 4.4 ±
18 ± 2 5±2 1±1 0
0.5 0.5 0.5 0.5
4 4 6 6
6 4 2 6
0.7 0.8 0.9 1.0
TENDL predictions, to minimize the 46Sc fraction in the final product, the energy window 20/14 MeV seems optimal. A maximum impinging energy of 17.75 MeV is achievable with an 18 MeV cyclotron equipped with a solid target station by employing commercial ∼10 μm Havar window foils. In this condition, the 47Sc target yield and the purity were studied as a function of the target thickness (Fig. 8). At saturation, for a 500 μm thick target a yield of 12 GBq/ μ A of 47Sc with a ∼85% purity (Fig. 9) should be achievable. 12 GBq is comparable with what is used for therapeutic interventions with similar theranostic isotopes. It has to be remarked that the purity will improve with time since the half-life of 47Sc is longer than the ones of 48,43,44,44mSc. Furthermore, 47Ca can be quickly separated from Sc and used as a 47Sc generator at a later stage.
18 MeV or above (Fig. 2). These data show that both the amount of produced 47Sc and the presence of the main impurity 48Sc strongly depend on the energy of the proton beam entering and exiting the target. Therefore, the impinging proton energy and the target thickness are of paramount importance. Using an 18 MeV proton cyclotron, the highest 47Sc isotopic purity can be reached with the maximum achievable impinging energy on target (Fig. 7). Moreover, the energy of the protons exiting the target should be higher than 12 MeV, requiring a target thickness in the order of 500 μm . To achieve better isotopic purity and/or larger 47Sc activities, higher energies (e.g. 24 MeV) or higher enrichment levels can be considered. While higher enrichment will surely be beneficial, the use of higher energy requires a careful evaluation. A recent study (Misiak et al., 2017b) using a higher energy proton accelerator showed a maximum 47Sc yield of 143.95 ± 9.36 kBq/ μ Ah, a 48Sc and 46Sc impurity at one hour of irradiation of 13% and 0.2%, respectively. For getting these results, a natCaCO3 target ∼1 mm thick was used with an impinging and exit energy of 24 and 17 MeV, respectively. This study shows also that the irradiation of natCaCO3 with protons of higher energies increases the 47Sc purity at EoB while also increasing production of 46Sc. With an half-life of 84 days, this represents an impurity to be taken in serious consideration. According to TENDL and using the isotopic ratio of 46Ca and 48Ca in the enriched material from Table 4, a 46Sc thick target yield of 5 kBq/ μ Ah will be produced with an impinging energy of 18 MeV. If the entry energy is increased to 24 MeV and 30 MeV, the 46Sc thick target yield will rise to ∼ 225 kBq/ μ Ah and ∼ 1.75 MBq/ μ Ah respectively, due to the contribution of the 48Ca(p,3n)46Sc reaction. This contribution cannot be reduced by higher enriched material. Therefore, following
5. Conclusions and outlook The production cross sections of the reactions 44Ca(p,n)44mSc, 48Ca (p,2n)47Sc, 48Ca(p,n)48Sc and 48Ca(p,pn)47Ca were measured at the Bern University Hospital cyclotron. The obtained results are instrumental for an optimized production of 47Sc in quantity and quality suitable for therapeutic interventions by irradiating highly enriched 48 CaCO3, reducible to 48CaO, with an 18 MeV medical cyclotron equipped with a solid target station. In particular, using an impinging energy of 17.75 MeV and a 500 μm thick target, a saturation yield of 12 GBq/ μ A of 47Sc with a 85% of purity at EoB can be achieved. The main impurity will be 48Sc at 14%, while the amount of 43Sc,44Sc and 44m Sc is predicted to be below 1% altogether. Since the purity will improve with time when the target is not irradiated, a fractionated 22
Applied Radiation and Isotopes 143 (2019) 18–23
T.S. Carzaniga, S. Braccini
2017), the results reported in this paper contribute to pave the way towards the use of medical cyclotrons for the production of 43,44Sc/47Sc for theranostics.
bombardment can be envisaged. A saturation yield of 1.3 GBq/ μ A of 47Ca can be achieved. This isotope could be separated from Sc after EoB and used as a high purity 47 Sc generator. The 47Sc production cross section was found to be rising at 18 MeV, indicating that an higher energy (20–30 MeV) cyclotron could be an option for a further optimization. This nevertheless requires particular attention on the 46Sc impurity coproduction. Since 43Sc and 44Sc can also be produced with 18 MeV proton beams in quality and quantity suitable for PET imaging (Carzaniga et al.,
Acknowledgements We acknowledge contributions from LHEP engineering and technical staff. This research project was financially supported by the Swiss National Science Foundation (grant CR23I2 _ 156852).
Appendix The numerical results shown in Figs. 2-5 are reported in the following tables.
Mamtimin, M., Harmon, F., Starovoitova, V.N., 2015. Sc-47 production from titanium targets using electron linacs. Appl. Radiat. Isot. 102, 1–4. https://doi.org/10.1016/j. apradiso.2015.04.012. van der Meulen, N.P., Bunka, M., Domnanich, K., Müller, C., Haller, S., Vermeulen, C., Türler, A., Schibli, R., 2015. Cyclotron production of 44Sc: from bench to bedside. Nucl. Med. Biol. 42, 745–751. Michel, R., Bodemann, R., Busemann, H., Daunke, R., Gloris, M., Lange, H.J., Klug, B., Krins, A., Leya, I., Lüpke, M., Neumann, S., Reinhardt, H., Schnatz-Büttgen, M., Herpers, U., Schiekel, T., Sudbrock, F., Holmqvist, B., Condé, H., Malmborg, P., Suter, M., Dittrich-Hannen, B., Kubik, P.W., Synal, H.A., Filges, D., 1997. Cross sections for the production of residual nuclides by low- and medium-energy protons from the target elements C, N, O, Mg, Al, Si, Ca, Ti, V, Mn, Fe, Co, Ni, Cu, Sr, Y, Zr, Nb, Ba and Au. Nucl. Instrum. Methods Phys. Res. Sect. B: Beam Interact. Mater. At. 129, 153–193. Minegishi, K., Nagatsu, K., Fukada, M., Suziki, H., Ohya, T., Zhang, M.R., 2016. Production of scandium-43 and -47 from a powdery calcium oxide target via the (nat/ 44)Ca(α,x)-channel. Appl. Radiat. Isot. 116, 8–12. https://doi.org/10.1016/j. apradiso.2016.07.017. Misiak, R., Walczak, R., Was, B., Bartyzel, M., Mietelski, J.W., Bile, A., 2017a. 47Sc production development by cyclotron irradiation of 48Ca. J. Radioanal. Nucl. Chem. 313, 429–434. https://doi.org/10.1007/s10967-017-5321-z. Misiak, R., Walczak, R., Wa¸s, B., Bartyzel, M., Mietelski, J.W., Bilewicz, A., 2017b. 47Sc production development by cyclotron irradiation of 48Ca. J. Radioanal. Nucl. Chem. 313, 429–434. https://doi.org/10.1007/s10967-017-5321-z.. URL 〈http://link. springer.com/10.1007/s10967-017-5321-z〉. Nesteruk, K.P., Auger, M., Braccini, S., Carzaniga, T.S., Ereditato, A., Scampoli, P., 2017. A system for online beam emittance measurements and proton beam characterization. Potkins, D.E., Braccini, S., Nesteruk, K.P., Carzaniga, T.S., Vedda, A., Chiodini, N., Timmermans, J., Melanson, S., Dehnel, M.P., 2017. A Low-Cost Beam Profiler Based On Cerium-Doped Silica Fibers. In: Proceedings of the CAARI-16, Physics Procedia. Rösch, F., Baum, R., 2011. Generator-based PET radiopharmaceuticals for molecular imaging of tumours: on the way to THERANOSTICS. Dalton Trans. 40, 6104–6111. Singh, A., Baum, R., Klette, I., van der Meulen, N., Mueller, C., Tuerler, A., Schibli, R., 2015. Scandium-44 DOTATOC PET/CT: first in-human molecular imaging of neuroendocrine tumors and possible perspectives for Theranostics. J. Nucl. Med. 56, 267. Singh, G., Kailas, S., Saini, S., Chatterjee, A., Balakrishnan, M., Mehta, M.K., 1982. Reaction 48Ca(p,n)48Sc from Ep =1.885 to 5.1 MeV. Pramana 19, 565–577. Carzaniga, Tommaso Stefano, Auger, Martin, Braccini, Saverio, Bunka, Maruta, Ereditato, Antonio, Nesteruk, Konrad Pawel, Scampoli, Paola, Türler, Andreas, van der Meulen, Nicholas, 2017. Measurement of 43Sc and 44Sc production cross-section with an 18 MeV medical PET cyclotron. Appl. Radiat. Isot. 129, 96–102. Wu, S.C., 2000. Nuclear data sheets for a = 46. Nucl. Data Sheets 91, 1–116. https://doi. org/10.1006/ndsh.2000.0014.. URL 〈http://www.sciencedirect.com/science/ article/pii/S0090375200900140〉.
References Anderson, C.J., Ferdani, R., 2009. Copper-64 radiopharmaceuticals for PET imaging of cancer: advances in preclinical and clinical research. Cancer Biother. Radiopharm. 24, 379–393. Auger, M., Braccini, S., Carzaniga, T.S., Ereditato, A., Nesteruk, K.P., Scampoli, P., 2016. A detector based on silica fibers for ion beam monitoring in a wide current range. J. Instrum. 11, P03027. Braccini, S., 2013. The new Bern PET cyclotron, its research beam line, and the development of an innovative beam monitor detector. In: Proceedings of the AIP Conference Proceedings, 1525, pp. 144–150. Braccini, S., 2016. Compact medical cyclotrons and their use for radioisotope production and multi-disciplinary research. In: Proceedings of the Cyclotrons 2016, pp. 229–234. 〈http://accelconf.web.cern.ch/AccelConf/cyclotrons2016/papers/tud01.pdf〉. Braccini, S., Scampoli, P., 2016. Science with a medical cyclotron. CERN Cour. 21–22. Burrows, T., 2006. Nuclear data sheets for a = 48. Nucl. Data Sheets 107, 1747–1922. https://doi.org/10.1016/j.nds.2006.05.005.. URL 〈http://www.sciencedirect.com/ science/article/pii/S0090375206000482〉. Burrows, T., 2007. Nuclear data sheets for a = 47. Nucl. Data Sheets 108, 923–1056. https://doi.org/10.1016/j.nds.2007.04.002.. URL 〈http://www.sciencedirect.com/ science/article/pii/S0090375207000403〉. Chen, J., Singh, B., Cameron, J.A., 2011. Nuclear data sheets for a = 44. Nucl. Data Sheets 112, 2357–2495. https://doi.org/10.1016/j.nds.2011.08.005.. URL 〈http:// www.sciencedirect.com/science/article/pii/S0090375211000779〉. de Waal, T.J., Peisach, M., Pretorius, R., 1971. Activation cross sections for proton-induced reactions on calcium isotopes up to 5.6 MeV. J. Inorg. Nucl. Chem. 33, 2783. Deilami-Nezhad, L., Moghaddam-Banaem, L., Sadeghi, M., Asgari, M., 2016. Production and purification of Scandium-47: a potential radioisotope for cancer theranostics. Appl. Radiat. Isot. 118, 120–130. Domnanich, K.A., Müller, C., Benesova, M., Dressler, R., Haller, S., Köster, U., Ponsard, B., Schibli, R., Tüerler, A., van der Meulen, N.P., 2017. 47Sc as useful β − -emitter for the radiotheragnostic paradigm: a comparative study of feasible production routes. EJNMMI Radiopharm. Chem. Hosseini, S.F., Sadeghi, M., Aboudzadeh, M.R., 2017. Theoretical assessment and targeted modeling of TiO2 in reactor towards the scandium radioisotopes estimation. Appl. Radiat. Isot. 127, 116–121. 〈https://www-nds.iaea.org/exfor/〉. 〈http://www.isoflex.com/〉. Kolsky, K., Joshi, V., Mausner, L., Srivastava, S., 1998. Radiochemical purification of nocarrier-added scandium-47 for radioimmunotherapy. Appl. Radiat. Isot. 49, 1541–1549. Koning, A., Rochman, D., 2012. Modern nuclear data evaluation with the TALYS code system. Nucl. Data Sheets 113 (2841–1934). Levkowskij, V.N., 1991. Activation Cross Sections for the Nuclides of Medium Mass Region (A = 40-100) with Protons and α-particles at Medium (E = 10–50 MeV) Energies. Inter-Vesti, Moscow.
23
Contents lists available at ScienceDirect
Applied Radiation and Isotopes journal homepage: www.elsevier.com/locate/apradiso
Cross-section measurement of 44mSc,47Sc, 48Sc and 47Ca for an optimized 47 Sc production with an 18 MeV medical PET cyclotron
T
⁎
Tommaso Stefano Carzaniga , Saverio Braccini Albert Einstein Center for Fundamental Physics (AEC), Laboratory for High Energy Physics (LHEP), University of Bern, Sidlerstrasse 5, CH-3012 Bern, Switzerland
H I GH L IG H T S
cross-sections Ca(p,n) Sc, Ca(p,n) Sc, Ca(p,2n) Sc and Ca(p,pn) Ca were measured. • The 18 MeV proton cyclotron is suitable to produce Sc/ Sc theranostic pair. • An • 20–30 MeV cyclotrons are valuable options for further Sc production op- timization. 44
44m
48
48
48
47
43
48
47
47
47
A R T I C LE I N FO
A B S T R A C T
Keywords: Therapeutic radioisotopes Medical cyclotrons Scandium Cross-section Theranostics
Novel medical radioisotopes for both diagnostic and therapy are essential for the future development of personalized nuclear medicine. Among them, radiometals can be used to label both proteins and peptides and encompass promising theranostic pairs. The optimized supply of radiometals in quantity and quality for clinical applications represents a scientific and technological challenge. 47 Sc is a β − emitter that forms a theranostic pair together with one of the β + emitters 43Sc or 44Sc. It can be produced at a medical cyclotron by proton bombardment of an enriched calcium oxide target. The parasite production of 48Sc undermines the 47Sc purity, which strongly depends on the energy of protons impinging the target and on the thickness of the target material. For this purpose, an accurate knowledge of the production cross-sections is mandatory. In this paper, we report on the measurement of the cross-section of the reactions 44Ca(p,n)44 mSc,48Ca (p,n)48Sc, 48Ca(p,2n)47Sc and 48Ca(p,pn)47Ca using natCaCO3 targets performed at the Bern University Hospital cyclotron laboratory. On the basis of the obtained results and of the isotopic composition of commercially available enriched target materials, the thick target yields and the purity were calculated to assess the optimal irradiation conditions.
1. Introduction The production of novel radioisotopes is essential for the development of personalized nuclear medicine. The theranostic approach is based on a pair of isotopes, one for diagnostic and one for therapy, used to label the same compound. Several pairs of nucleides for theranostics are under study or in early adoption in clinical applications, as reported in Table 1. Radionuclide therapy is promising to be successful in treating persistent diseases with low toxic side-effects thanks to the localized dose deposition. The absorbed dose in target tissues in therapeutic procedures ranges between 20 and 60 Gy. Three families of isotopes are nowadays under study for adoption in therapy: β − , α and Auger electrons emitters. This paper will focus on β − emitters.
⁎
While for imaging hundreds of MBq are enough for one examination (typically 350 MBq of 18F-FDG and less than 200 MBq of 99mTc), more than one GBq of a β − emitter must be injected in each patient for therapy. An ideal therapeutic β − emitter decays with a low energy β − and low (100–300 keV) energy gamma suitable for SPECT imaging. A good example is 177Lu, that, together with 90Y, is becoming the main Radionuclide Therapy (RNT) agent. The PET isotope 68Ga is used as its diagnostic counterpart, to demonstrate the uptake in the tumor of the labeled molecule for the treatment of Neuroendocrine Tumors (NET). Activities of about 7 GBq of 177Lu per patient are needed. 47 Sc is very similar to 177Lu for its chemical and physical characteristics. 47Sc is a β − emitter and is foreseen as the therapeutic partner of the β + emitters 43Sc and 44Sc, both under study for PET imaging
Corresponding author. E-mail address: [email protected] (T.S. Carzaniga).
https://doi.org/10.1016/j.apradiso.2018.10.015 Received 11 July 2018; Received in revised form 13 September 2018; Accepted 10 October 2018 Available online 15 October 2018 0969-8043/ © 2018 Elsevier Ltd. All rights reserved.
Applied Radiation and Isotopes 143 (2019) 18–23
T.S. Carzaniga, S. Braccini
Table 1 Theranostic pairs under study or in early adoption. Imaging (PET/SPECT)
Therapy
123
131
124
I, I Y 61 64 Cu, Cu (Anderson and Ferdani, 2009) 111 In, 68Ga 43 Sc, 44Sc (Singh et al., 2015; van der Meulen et al., 2015; Rösch and Baum, 2011) 86
Table 3 Decay properties of 44mSc (Chen et al., 2011), 46Sc (Wu, 2000), 2007), 48Sc (Burrows, 2006) and 47Ca (Burrows, 2007). t1/2 [d]
I Y 67 Cu 90 177 Y, Lu 47 Sc (Domnanich et al., 2017) 90
44m 44m
Sc → Sc →
Target
Route
p
46
48
Ca Ca
48
Ti Ti Ca 49 Ti 44 Ca 46 Ca 48 Ca 47 Ti 50
d
α n
γ
46
48 48
Ca Ti
Ca(p, γ )47Sc Ca(p,2n)47Sc (Misiak et al., 2017a) 48 Ca(p,pn)47Ca → 47Sc 48 Ti(p,2p)47Sc (Kolsky et al., 1998) 50 Ti(p, α )47Sc 46 Ca(d,n)47Sc 49 Ti(d, α )47Sc 44 Ca(α , p)47Sc (Minegishi et al., 2016) 46 Ca(n, γ )47Ca → 47Sc (Domnanich et al., 2017) 48 Ca(n,2n)47Ca → 47Sc 47 Ti(n,p)47Sc (Hosseini et al., 2017; DeilamiNezhad et al., 2016; Domnanich et al., 2017) 48 Ca(γ , n)47Ca → 47Sc 48 Ti(γ , p)47Sc (Mamtimin et al., 2015) 48
2.442(4)
Ti
47
47
Ti
3.3492(6)
48
48
Ti
1.819(4)
142.6(7) - 68.4(6) 203.9(8) - 31.6(6) 227.3(25) - 80.98(18)
47
47
4.536(3)
242.66(49) - 73(15)
Sc →
46
Sc Ca
46
Sc →
Impinging particle
44
46
Sc →
Table 2 47 Sc production routes.
44
E β− [keV] - BR β− [%]
Ca →
Sc
83.79(4)
111.8(3) - 99.9964(7)
47
Sc (Burrows,
E γ [keV] – BRγ [%] 271.241(10) - 86.7(3) 1126.06(4) - 1.20(7) 1001.83(3) - 1.20(7) 1157.002(3) - 1.20(7) 1120.545(4) - 99.987(1) 889.277(3) - 99.984(1) 159.381(15) - 68.3(4) 1312.120(12) - 100.1(6) 983.526(12) - 100.1(6) 1037.522(12) - 97.6(7) 175.361(5) - 7.48(10) 1212.880(12) - 2.38(4) 1312.120(12) - 100.1(6) 1297.09(10) - 67(13) 489.23(10) - 5.9(12) 807.86(10) - 5.9(12)
radioisotope production and multi-disciplinary research activities (Braccini and Scampoli, 2016). The beam is brought to the second bunker by means of a 6 m long external Beam Transport Line (BTL), which was used for the measurements presented in this paper. The beam extracted to the BTL has a mean energy of (18.3 ± 0.3) MeV with an RMS of 0.40 MeV (Braccini, 2013; Nesteruk et al., 2017). The experimental method used in this work was very similar to the one described in our earlier publication on cross-section measurements for the production of the diagnostic partners 43Sc and 44Sc (Carzaniga et al., 2017). It is based on the irradiation of the full mass of a thin target by a proton beam with a flat profile. UniBEaM detectors (Auger et al., 2016) were used to monitor the beam on-line. These beam profilers were developed by our group and are based on silica doped fibers passing through the beam. Their commercialization was licensed by the University of Bern to the company D-Pace (Potkins et al., 2017). This method has the advantage that the target has not to be necessarily uniform in thickness provided that the energy of the protons can be considered constant. The target holder geometry was improved to assure the optimal centering of the target material with respect to the collimator, as shown in Fig. 1. The support material, i.e. the so called backing, for the target material was a 2 mm thick aluminum disc with a 0.6 mm deep pocket of 4.2 mm in diameter (Fig. 1 left). The target material was deposited in the pocket by sedimentation from a suspension of a few milligrams of nat CaCO3 in ultra-pure water. The isotopic composition of natural and commercially available CaCO3 is reported in Table 4. Ca-48 enriched material can be purchased in different forms (carbonate, chloride, oxide, nitrate, metal, gluconate, iodide, fluoride) and enrichment levels (64%-97 +%). The water was then evaporated at ∼ 70 °C by means of a small heating plate (Fig. 1 right). The mass of the deposited material was assessed with an analytical balance (METTLER TOLEDO AX26 DeltaRange) with a readability of 2 μ g and a repeatability of 4 μ g. After cooling, the targets were covered with a 10 μm thin aluminum foil for protection. Since the thickness of the target was estimated to be of the order of few tens of μm, the beam energy was considered constant within the uncertainties within the full irradiated mass.
(Carzaniga et al., 2017). Furthermore, 47Sc is characterised by a 159 keV gamma line (∼ 68% branching ratio), allowing for SPECT imaging during treatment. There are many routes to yield 47Sc (Table 2). Promising ones for a wide diffusion of this radioiotope could be based on medical cyclotrons (Braccini, 2016) commonly used to produce 18F, the main radioisotope for PET imaging. These accelerators provide proton (sometimes also deuteron) beams in the energy range of 15–25 MeV with intensities of 50–300 μ A. 47Sc can be produced using solid target stations. To optimize the yield together with the radio-nuclide purity, the precise knowledge of the cross-section of 47Sc and eventual impurities as a function of the beam energy is of paramount importance. Data reported in literature and accessible via the EXFOR database (iaea) do not fully cover the energy range of medical cyclotrons and are sometimes lacking or inconsistent. In the framework of a research program aimed at the production of scandium for theranostics ongoing at the Bern medical cyclotron laboratory (Braccini, 2013), 47Sc production via proton bombardment of 48 Ca enriched calcium oxide was considered. The main impurity of this production route is 48Sc that is unsuitable for therapy due to its high energy gammas. On the other hand, the concomitant production of 47 Ca, which then decays into 47Sc, enhances the produced activity during bombardment. Furthermore, 47Ca can be used as a generator if promptly separated after EoB. According to the enrichment level, other scandium isotopes can be produced, the long lived 44mSc in particular. The main properties of 47Sc,48Sc and 47Ca are summarized in Table. 3. In this paper we report on the measurement of the cross-sections of the reactions 44Ca(p,n)44mSc,48Ca(p,n)48Sc, 48Ca(p,2n)47Sc and 48Ca (p,pn)47Ca. The obtained results were used to assess the radio-nuclide purity when bombarding enriched calcium targets.
3. Cross-section measurements 2. Materials and methods The aim of this paper is to study the optimal production of 47Sc by maximising the activity while minimizing the impurities. To measure the cross-sections, irradiations were performed at several energies on target with currents of ∼10 nA for ∼30 mins. The energy was modulated by using aluminum absorbers (Carzaniga et al., 2017).
The laboratory at the Bern University Hospital (Inselspital) (Braccini, 2013) features an IBA Cyclone 18/18 high current cyclotron and two bunkers with independent access. This solution is unusual for a hospital-based facility and was conceived to perform both medical 19
Applied Radiation and Isotopes 143 (2019) 18–23
T.S. Carzaniga, S. Braccini
Fig. 1. Target holder geometry aimed at a optimal centering of the material with respect to the collimator (left). Targets drying on a heating plate kept at 70 °C (right). Table 4 Calcium isotopic percentages in calcium carbonate powder enriched in quoted from ISOFLEX (isoflex). 40
Natural (%) 48 Ca enrich. (%)
Ca
96.941 27.9
42
Ca
0.647 0.3
43
44
Ca
0.135 0.1
Ca
2.086 2.2
46
Ca
0.004 <0.1
48
48
Ca
Ca
0.187 69.2
47
Sc is obtained from 48Ca directly via the reaction 48Ca(p,2n)47Sc and indirectly via the reaction 48Ca(p,pn)47Ca that first produces 47Ca, which then decays in 47Sc with an half-life of ∼4.5 days. After each irradiation, the characteristic gamma-ray at 159 keV emitted by 47Sc and the 1297 keV gamma-ray emitted by 47Ca were measured with an HPGe detector for several hours, in order to reduce the statistical uncertainty on the integral area to a negligible level. The results of the 48Ca(p,2n)47Sc cross-section measurements are presented in Fig. 2; for completeness, the numerical values are reported in the Appendix (Table 5). Reasonable agreement is found with the TENDL (Koning and Rochman, 2012) code, especially below 14 MeV. No comparison with previous experimental measurements is possible. The results of the 48Ca(p,pn)47Ca cross-section measurements are presented in Fig. 3; for completeness the numerical values are reported in the Appendix (Table 6). Compared to TENDL (Koning and Rochman, 2012) predictions, the data measured in this study present a faster rise of the cross-section. A maximum is found at around 17 MeV, while TENDL suggests a value higher than 25 MeV. Data available in the literature (Michel et al., 1997) are scarse and are in reasonable agreement with both TENDL and our measurements. 48 Sc is the main impurity that would be produced by irradiating an
Fig. 3. Cross section of the reaction
48
Fig. 4. Cross section of the reaction
Fig. 2. Cross section of the reaction
48
Ca(p,pn)47Ca.
48
Ca(p,n)48Sc.
enriched 48CaO target. To study the 48Ca(p,n)48Sc reaction, the characteristic gamma-ray at 1312 keV emitted by 48Sc was measured as for 47 Sc and 47Ca. The results of the 48Ca(p,n)48Sc cross-section measurements are presented in Fig. 4; for completeness the numerical values are reported in the Appendix (Table 7). Good agreement is found with the TENDL (Koning and Rochman, 2012) code. The few experimental data below 18 MeV available in the literature (de Waal et al., 1971; Singh et al., 1982; Michel et al., 1997) concentrates at low energies (<5 MeV) and are found in reasonable agreement with our measurements. 44m Sc is another long lived impurity that would be produced by irradiating an enriched 48CaO target, since the fraction of 44Ca might not be negligible. To study the 44Ca(p,n)44mSc reaction, the
Ca(p,2n)47Sc. 20
Applied Radiation and Isotopes 143 (2019) 18–23
T.S. Carzaniga, S. Braccini
Fig. 5. Cross section of the reaction
44
Ca(p,n)
44m
Fig. 6. Thick target saturation yields for a 69% enriched
Fig. 8. 43,44,44m,48Sc impurities at saturation for a 69% enriched 48CaO target as a function of the target thickness. The entry energy in the pellet is 17.75 MeV.
Sc.
48
CaO target.
Fig. 9. 47Sc yield and purity for a 69% enriched 48CaO at saturation with an entry energy in the pellet of 17.75 MeV as a function of the target thickness.
(∼20%) of the 1297 keV gamma. The gamma branching ratio was a negligible source of uncertainties for the other reactions. A further evaluated impurity was 46Sc. According to TENDL the threshold for the 48Ca(p,3n)46Sc reaction is at ∼ 20 MeV. Since even in the enriched material a tiny amount of 46Ca is always present, 46Sc could also be produced via the reaction 46Ca(p,n)46Sc. Nevertheless, 46 Sc was searched for in all the measurements and no signal was found. It is known from other experiments (Misiak et al., 2017b) that 46Sc is created in the considered energy range but the produced activities were below our MDA (Minimum Detectable Activity). We are considering performing dedicated experiments to measure 46Sc production crosssections in future. Fig. 7. 43,44,44m,48Sc impurities for a thick 69% enriched turation.
48
CaO target at sa-
4. Study of production yield and purity A study of the thick target saturation yield has to be performed in order to optimize 47Sc production by 48Ca irradiation. Fig. 6 reports the thick target saturation yields for 47Sc, 48Sc and 47Ca using the enriched 48 Ca target material by ISOFLEX of Table 4 in the form of calcium oxide as a function of the impinging proton energy. CaO is our form of choice for Sc production since it gives a larger yield and the carbonate will dissociate carbon dioxide when heated up by proton bombardment, causing degassing issues. Since the purity of material enriched in 48Ca nowadays available on the market ranges from 64% to 97+%, 43Sc, 44 Sc and the longer lived 44mSc have also to be taken into account as impurities. 43Sc and 44Sc yields were studied in Carzaniga et al. (2017). The threshold for the 48Ca(p,n)48Sc reaction is found to be below 3 MeV and the maximal cross-section at around 10 MeV (Fig. 4). On the other hand, the threshold for the 48Ca(p,2n)47Sc reaction is found around 10 MeV with a probable maximum of its cross-section at around
characteristic gamma-ray at 271 keV emitted by 44mSc was measured as for 47Sc,48Sc and 47Ca. The results of the 44Ca(p,n)44mSc cross-section measurements are presented in Fig. 5; for completeness the numerical values are reported in the Appendix (Table 8). The experimental data available in the literature (Levkowskij, 1991) are found in reasonable agreement with our measurements. Up to 12 MeV good agreement is found with the TENDL (Koning and Rochman, 2012) code while at higher energies the prediction underestimates our results. The main experimental uncertainty for the reactions 48Ca(p,2n)47Sc, 48 Ca(p,n)48Sc and 44Ca(p,n)44mCa was due to the flatness of the beam (up to 5%). Other sources of uncertainties are the beam current (∼1%), the activity of the source used to calibrate the HPGe detector (∼1%) and the HPGe calibration procedure (∼1%). For the reaction 48Ca(p,pn)47Ca the main experimental uncertainty was due to the branching ratio 21
Applied Radiation and Isotopes 143 (2019) 18–23
T.S. Carzaniga, S. Braccini
Table 5 Cross-section data of the reaction
48
Ca(p,2n)47Sc.
E [MeV]
σ [mb]
E [MeV]
σ [mb]
E [MeV]
σ [mb]
18.4 ± 0.5 17.2 ± 0.5 16.0 ± 0.5
1107 ± 56 1105 ± 57 1058 ± 55
14.6 ± 0.5 13.2 ± 0.6 11.6 ± 0.6
967 ± 49 755 ± 39 342 ± 18
9.9 ± 0.6 8.9 ± 0.7
16 ± 3 0
Table 6 Cross-section data of the reaction
48
Ca(p,d)47Ca.
E [MeV]
σ [mb]
E [MeV]
σ [mb]
E [MeV]
σ [mb]
18.4 ± 0.5 17.2 ± 0.5
209 ± 42 233 ± 48
16.0 ± 0.5 14.6 ± 0.5
199 ± 43 120 ± 27
13.2 ± 0.6 11.6 ± 0.6
63 ± 20 0
Table 7 Cross-section data of the reaction
48
Ca(p,n)48Sc.
E [MeV]
σ [mb]
E [MeV]
σ [mb]
E [MeV]
σ [mb]
18.4 ± 17.2 ± 16.0 ± 14.6 ± 13.2 ±
199 ± 220 ± 243 ± 321 ± 425 ±
11.6 ± 0.6 9.9 ± 0.6 8.9 ± 0.7 6.7 ± 0.8
711 ± 890 ± 954 ± 822 ±
5.3 ± 4.4 ± 3.4 ± 3.0 ±
570 ± 448 ± 245 ± 122 ±
0.5 0.5 0.5 0.5 0.6
Table 8 Cross-section data of the reaction
44
28 19 33 32 41
44 51 49 49
0.9 1.0 1.1 1.2
39 30 14 10
Ca(p,n)44mSc.
E [MeV]
σ [mb]
E [MeV]
σ [mb]
E [MeV]
σ [mb]
18.4 ± 17.2 ± 16.0 ± 14.6 ±
37 ± 45 ± 60 ± 66 ±
13.2 ± 0.6 11.6 ± 0.6 10.9 ± 0.6 9.9 ± 0.6
58 ± 42 ± 24 ± 24 ±
8.9 ± 6.7 ± 5.3 ± 4.4 ±
18 ± 2 5±2 1±1 0
0.5 0.5 0.5 0.5
4 4 6 6
6 4 2 6
0.7 0.8 0.9 1.0
TENDL predictions, to minimize the 46Sc fraction in the final product, the energy window 20/14 MeV seems optimal. A maximum impinging energy of 17.75 MeV is achievable with an 18 MeV cyclotron equipped with a solid target station by employing commercial ∼10 μm Havar window foils. In this condition, the 47Sc target yield and the purity were studied as a function of the target thickness (Fig. 8). At saturation, for a 500 μm thick target a yield of 12 GBq/ μ A of 47Sc with a ∼85% purity (Fig. 9) should be achievable. 12 GBq is comparable with what is used for therapeutic interventions with similar theranostic isotopes. It has to be remarked that the purity will improve with time since the half-life of 47Sc is longer than the ones of 48,43,44,44mSc. Furthermore, 47Ca can be quickly separated from Sc and used as a 47Sc generator at a later stage.
18 MeV or above (Fig. 2). These data show that both the amount of produced 47Sc and the presence of the main impurity 48Sc strongly depend on the energy of the proton beam entering and exiting the target. Therefore, the impinging proton energy and the target thickness are of paramount importance. Using an 18 MeV proton cyclotron, the highest 47Sc isotopic purity can be reached with the maximum achievable impinging energy on target (Fig. 7). Moreover, the energy of the protons exiting the target should be higher than 12 MeV, requiring a target thickness in the order of 500 μm . To achieve better isotopic purity and/or larger 47Sc activities, higher energies (e.g. 24 MeV) or higher enrichment levels can be considered. While higher enrichment will surely be beneficial, the use of higher energy requires a careful evaluation. A recent study (Misiak et al., 2017b) using a higher energy proton accelerator showed a maximum 47Sc yield of 143.95 ± 9.36 kBq/ μ Ah, a 48Sc and 46Sc impurity at one hour of irradiation of 13% and 0.2%, respectively. For getting these results, a natCaCO3 target ∼1 mm thick was used with an impinging and exit energy of 24 and 17 MeV, respectively. This study shows also that the irradiation of natCaCO3 with protons of higher energies increases the 47Sc purity at EoB while also increasing production of 46Sc. With an half-life of 84 days, this represents an impurity to be taken in serious consideration. According to TENDL and using the isotopic ratio of 46Ca and 48Ca in the enriched material from Table 4, a 46Sc thick target yield of 5 kBq/ μ Ah will be produced with an impinging energy of 18 MeV. If the entry energy is increased to 24 MeV and 30 MeV, the 46Sc thick target yield will rise to ∼ 225 kBq/ μ Ah and ∼ 1.75 MBq/ μ Ah respectively, due to the contribution of the 48Ca(p,3n)46Sc reaction. This contribution cannot be reduced by higher enriched material. Therefore, following
5. Conclusions and outlook The production cross sections of the reactions 44Ca(p,n)44mSc, 48Ca (p,2n)47Sc, 48Ca(p,n)48Sc and 48Ca(p,pn)47Ca were measured at the Bern University Hospital cyclotron. The obtained results are instrumental for an optimized production of 47Sc in quantity and quality suitable for therapeutic interventions by irradiating highly enriched 48 CaCO3, reducible to 48CaO, with an 18 MeV medical cyclotron equipped with a solid target station. In particular, using an impinging energy of 17.75 MeV and a 500 μm thick target, a saturation yield of 12 GBq/ μ A of 47Sc with a 85% of purity at EoB can be achieved. The main impurity will be 48Sc at 14%, while the amount of 43Sc,44Sc and 44m Sc is predicted to be below 1% altogether. Since the purity will improve with time when the target is not irradiated, a fractionated 22
Applied Radiation and Isotopes 143 (2019) 18–23
T.S. Carzaniga, S. Braccini
2017), the results reported in this paper contribute to pave the way towards the use of medical cyclotrons for the production of 43,44Sc/47Sc for theranostics.
bombardment can be envisaged. A saturation yield of 1.3 GBq/ μ A of 47Ca can be achieved. This isotope could be separated from Sc after EoB and used as a high purity 47 Sc generator. The 47Sc production cross section was found to be rising at 18 MeV, indicating that an higher energy (20–30 MeV) cyclotron could be an option for a further optimization. This nevertheless requires particular attention on the 46Sc impurity coproduction. Since 43Sc and 44Sc can also be produced with 18 MeV proton beams in quality and quantity suitable for PET imaging (Carzaniga et al.,
Acknowledgements We acknowledge contributions from LHEP engineering and technical staff. This research project was financially supported by the Swiss National Science Foundation (grant CR23I2 _ 156852).
Appendix The numerical results shown in Figs. 2-5 are reported in the following tables.
Mamtimin, M., Harmon, F., Starovoitova, V.N., 2015. Sc-47 production from titanium targets using electron linacs. Appl. Radiat. Isot. 102, 1–4. https://doi.org/10.1016/j. apradiso.2015.04.012. van der Meulen, N.P., Bunka, M., Domnanich, K., Müller, C., Haller, S., Vermeulen, C., Türler, A., Schibli, R., 2015. Cyclotron production of 44Sc: from bench to bedside. Nucl. Med. Biol. 42, 745–751. Michel, R., Bodemann, R., Busemann, H., Daunke, R., Gloris, M., Lange, H.J., Klug, B., Krins, A., Leya, I., Lüpke, M., Neumann, S., Reinhardt, H., Schnatz-Büttgen, M., Herpers, U., Schiekel, T., Sudbrock, F., Holmqvist, B., Condé, H., Malmborg, P., Suter, M., Dittrich-Hannen, B., Kubik, P.W., Synal, H.A., Filges, D., 1997. Cross sections for the production of residual nuclides by low- and medium-energy protons from the target elements C, N, O, Mg, Al, Si, Ca, Ti, V, Mn, Fe, Co, Ni, Cu, Sr, Y, Zr, Nb, Ba and Au. Nucl. Instrum. Methods Phys. Res. Sect. B: Beam Interact. Mater. At. 129, 153–193. Minegishi, K., Nagatsu, K., Fukada, M., Suziki, H., Ohya, T., Zhang, M.R., 2016. Production of scandium-43 and -47 from a powdery calcium oxide target via the (nat/ 44)Ca(α,x)-channel. Appl. Radiat. Isot. 116, 8–12. https://doi.org/10.1016/j. apradiso.2016.07.017. Misiak, R., Walczak, R., Was, B., Bartyzel, M., Mietelski, J.W., Bile, A., 2017a. 47Sc production development by cyclotron irradiation of 48Ca. J. Radioanal. Nucl. Chem. 313, 429–434. https://doi.org/10.1007/s10967-017-5321-z. Misiak, R., Walczak, R., Wa¸s, B., Bartyzel, M., Mietelski, J.W., Bilewicz, A., 2017b. 47Sc production development by cyclotron irradiation of 48Ca. J. Radioanal. Nucl. Chem. 313, 429–434. https://doi.org/10.1007/s10967-017-5321-z.. URL 〈http://link. springer.com/10.1007/s10967-017-5321-z〉. Nesteruk, K.P., Auger, M., Braccini, S., Carzaniga, T.S., Ereditato, A., Scampoli, P., 2017. A system for online beam emittance measurements and proton beam characterization. Potkins, D.E., Braccini, S., Nesteruk, K.P., Carzaniga, T.S., Vedda, A., Chiodini, N., Timmermans, J., Melanson, S., Dehnel, M.P., 2017. A Low-Cost Beam Profiler Based On Cerium-Doped Silica Fibers. In: Proceedings of the CAARI-16, Physics Procedia. Rösch, F., Baum, R., 2011. Generator-based PET radiopharmaceuticals for molecular imaging of tumours: on the way to THERANOSTICS. Dalton Trans. 40, 6104–6111. Singh, A., Baum, R., Klette, I., van der Meulen, N., Mueller, C., Tuerler, A., Schibli, R., 2015. Scandium-44 DOTATOC PET/CT: first in-human molecular imaging of neuroendocrine tumors and possible perspectives for Theranostics. J. Nucl. Med. 56, 267. Singh, G., Kailas, S., Saini, S., Chatterjee, A., Balakrishnan, M., Mehta, M.K., 1982. Reaction 48Ca(p,n)48Sc from Ep =1.885 to 5.1 MeV. Pramana 19, 565–577. Carzaniga, Tommaso Stefano, Auger, Martin, Braccini, Saverio, Bunka, Maruta, Ereditato, Antonio, Nesteruk, Konrad Pawel, Scampoli, Paola, Türler, Andreas, van der Meulen, Nicholas, 2017. Measurement of 43Sc and 44Sc production cross-section with an 18 MeV medical PET cyclotron. Appl. Radiat. Isot. 129, 96–102. Wu, S.C., 2000. Nuclear data sheets for a = 46. Nucl. Data Sheets 91, 1–116. https://doi. org/10.1006/ndsh.2000.0014.. URL 〈http://www.sciencedirect.com/science/ article/pii/S0090375200900140〉.
References Anderson, C.J., Ferdani, R., 2009. Copper-64 radiopharmaceuticals for PET imaging of cancer: advances in preclinical and clinical research. Cancer Biother. Radiopharm. 24, 379–393. Auger, M., Braccini, S., Carzaniga, T.S., Ereditato, A., Nesteruk, K.P., Scampoli, P., 2016. A detector based on silica fibers for ion beam monitoring in a wide current range. J. Instrum. 11, P03027. Braccini, S., 2013. The new Bern PET cyclotron, its research beam line, and the development of an innovative beam monitor detector. In: Proceedings of the AIP Conference Proceedings, 1525, pp. 144–150. Braccini, S., 2016. Compact medical cyclotrons and their use for radioisotope production and multi-disciplinary research. In: Proceedings of the Cyclotrons 2016, pp. 229–234. 〈http://accelconf.web.cern.ch/AccelConf/cyclotrons2016/papers/tud01.pdf〉. Braccini, S., Scampoli, P., 2016. Science with a medical cyclotron. CERN Cour. 21–22. Burrows, T., 2006. Nuclear data sheets for a = 48. Nucl. Data Sheets 107, 1747–1922. https://doi.org/10.1016/j.nds.2006.05.005.. URL 〈http://www.sciencedirect.com/ science/article/pii/S0090375206000482〉. Burrows, T., 2007. Nuclear data sheets for a = 47. Nucl. Data Sheets 108, 923–1056. https://doi.org/10.1016/j.nds.2007.04.002.. URL 〈http://www.sciencedirect.com/ science/article/pii/S0090375207000403〉. Chen, J., Singh, B., Cameron, J.A., 2011. Nuclear data sheets for a = 44. Nucl. Data Sheets 112, 2357–2495. https://doi.org/10.1016/j.nds.2011.08.005.. URL 〈http:// www.sciencedirect.com/science/article/pii/S0090375211000779〉. de Waal, T.J., Peisach, M., Pretorius, R., 1971. Activation cross sections for proton-induced reactions on calcium isotopes up to 5.6 MeV. J. Inorg. Nucl. Chem. 33, 2783. Deilami-Nezhad, L., Moghaddam-Banaem, L., Sadeghi, M., Asgari, M., 2016. Production and purification of Scandium-47: a potential radioisotope for cancer theranostics. Appl. Radiat. Isot. 118, 120–130. Domnanich, K.A., Müller, C., Benesova, M., Dressler, R., Haller, S., Köster, U., Ponsard, B., Schibli, R., Tüerler, A., van der Meulen, N.P., 2017. 47Sc as useful β − -emitter for the radiotheragnostic paradigm: a comparative study of feasible production routes. EJNMMI Radiopharm. Chem. Hosseini, S.F., Sadeghi, M., Aboudzadeh, M.R., 2017. Theoretical assessment and targeted modeling of TiO2 in reactor towards the scandium radioisotopes estimation. Appl. Radiat. Isot. 127, 116–121. 〈https://www-nds.iaea.org/exfor/〉. 〈http://www.isoflex.com/〉. Kolsky, K., Joshi, V., Mausner, L., Srivastava, S., 1998. Radiochemical purification of nocarrier-added scandium-47 for radioimmunotherapy. Appl. Radiat. Isot. 49, 1541–1549. Koning, A., Rochman, D., 2012. Modern nuclear data evaluation with the TALYS code system. Nucl. Data Sheets 113 (2841–1934). Levkowskij, V.N., 1991. Activation Cross Sections for the Nuclides of Medium Mass Region (A = 40-100) with Protons and α-particles at Medium (E = 10–50 MeV) Energies. Inter-Vesti, Moscow.
23