Radiochemical separation and quality assessment for the 68zn target based 64cu radioisotope production

Radiochemical separation and quality assessment for the 68Zn target based 64Cu radioisotope production Le Van So,* P.. The overlapped gamma-ray spectrum analysis method was developed to

Radiochemical separation and quality assessment for the 68Zn target

based 64Cu radioisotope production

Le Van So,* P Pellegrini, A Katsifis, J Howse, I Greguric

Radiopharmaceutical Research Institute, Australian Nuclear Science and Technology Organisation,

New Illawarra Road, Lucas Heights , P.M.B 1, Menai NSW 2234, Australia

(Received July 31, 2007)

The radiochemical separation of the different radionuclides ( 64 Cu, 67 Cu, 67 Ga, 66 Ga, 56 Ni, 57 Ni, 55 Co, 56 Co, 57 Co, 65 Zn, 196 Au ) induced in the Ni supported Cu substrate – 68 Zn target system, which was bombarded with the 29.0 MeV proton beam, was performed by ion-exchange chromatography using successive isocratic and/or concentration gradient elution techniques The overlapped gamma-ray spectrum analysis method was developed to assess the 67 Ga and 67 Cu content in the 64 Cu product and even in the post- 67 Ga production 68 Zn target solution without the support of radiochemical separation This method was used for the assessment of 64+67 Cu radioisotope separation from 67 Ga , the quality control of

64 Cu product and the determination of the 68 Zn (p,2p) 67 Cu reaction yield The improvement in the targetry and the optimization of proton beam energy for the 68 Zn target based 64 Cu and 67 Ga production were proposed based on the stopping power and range of the incident proton and on the excitation functions, reaction yields and different radionuclides induced in the target system

Introduction

64Cu which emits both beta and positron in high

abundance (39% β– and 17.4% β+) with a half-life of

12.7 hours can play a role of bi-functional radioisotope

for both positron emission tomography (PET) imaging

and endoradiotherapy (ERT) The better physical

characteristics of low intensity of the high energy

gamma-ray (1345.7 keV, 0.47%) and low energy β+

-emission (653 keV) of 64Cu compared to 61Cu and 62Cu

are particularly useful for high resolution PET imaging

Its 578 keV energy β–-emission, nearly identical to that

of the 67Cu beta-emitter, is suitable for the ERT of small

tumours Besides, the electron capture decay with its

associated Auger emissions can yield more efficient cell

killing when the 64Cu nuclide is deposited in the cell

nucleus The combined effect of three decay modes of

the 64Cu radioisotope gives a high cell killing efficacy.1

64Cu’s preferable properties involved in

physiological pathways1–4 – being almost invariably

bound to protein or peptide for transport purposes and

for functional use, being sequestered in mettalothionein

as Cu(I)-thiolate clusters, the redox and electron/oxygen

transfer properties based enzymatic uses – make

radioactive copper complex molecules useful (in their

own right) as targeting radiopharmaceuticals (the

bis-[thiosemicarbazone] chelating group based PTSM and

ATSM complexes, porphyrins, bleomycin) These

compounds displaying the metal-essential biochemical

properties of Cu are being used for brain and myocardial

imaging studies (PTSM) and for tumour treatment

(PTSM and ATSM).1,3

Another use of the Cu radioisotopes is based on the

high capability of forming stable coordinative

complexes with bifunctional chelators1,2 – acyclic polyaminocarboxylate (DTPA, EDTA), cyclic polyamines (Cyclam) and macrocyclic polyamino-carboxylates (TETA, DOTA, NOTA) The last two groups of chelator–Cu complexes with good in-vivo stability make the Cu radioisotopes amenable to coupling with the targetted molecules such as peptides and antibodies for the in-vivo applications This type of Cu-bifunctional chelator-biomolecule-conjugate further offers more versatile applications in PET imaging and in targeting radiotherapy The 64,67Cu-chelator-antibody conjugates (MAb35-against carcinoembryonic antigen, SEN7 and SWA20-against lung cancer antigen, VG76e, B72.3 antibodies) are the best-known targeting radiopharmaceuticals to this date The 64Cu labeled ligands targeting receptors (64Cu-DOTA-[Pro1, Tyr4]-bombesin[1-14] for targeting GRP receptors, 64 Cu-TETA-somatostatin analogs, 64Cu-DOTA-Annexin V.,) are conjugates being widely investigated.1–5

Currently two cyclotron based methods are used for

64Cu production The first is based on the 68Zn target and second on 64Ni The 64Ni target based production process is recently developed and based on the

64Ni(p,n)64Cu reaction.3,5,8,11–14 This production route, which uses a proton beam energy lower than 13 MeV for activation, has the advantages of no side nuclear reactions which could induce radionuclidic impurities in the 64Cu product However, the disadvantage of this process is the use of very expensive enriched 64Ni target The larger cyclotron with proton beam energy not adjustable to lower than 13 MeV is difficult to be used for the 64Ni target based 64Cu production due to the need for a larger amount of 64Ni target and a special target design

In contrast, the 68Zn target based production route

was developed more than 10 years ago.15–18 This

process was based on the 68Zn(p,αn)64Cu reaction and

performed by separating 64Cu from the “waste” solution

of the 68Zn(p,2n)67Ga reaction based 67Ga production

process This production route seems more economic in

the target utilization, because both 67Ga and 64Cu can be

produced from the same low cost target However, the

different side nuclear reactions inducing several

longer-lived radionuclides in the 68Zn target and its substrate

are the main disadvantages Besides, the potential of

67Ga contamination in 64Cu is high due to the much

higher reaction yield and much longer half-life of the

67Ga radionuclide The 68Zn(p,2p)67Cu reaction induced

67Cu and the target substrate contaminated “cold” Cu

elemental impurity in the 64Cu product are crucial

problems in performing the 68Zn target based production

route So, the effective target design, optimal proton

beam energy utilization and good radiochemical

separation should be the main issues concerned about

regarding the 68Zn target based 64Cu production As for

the quality assessment the analysis of the overlapping

gamma-ray spectra should be developed to identify 67Ga

and 67Cu radionuclide contaminations in the 68Zn target

based 64Cu product The lack of an analysis method for

the evaluation of 67Ga and 67Cu radionuclide

contamination may create a pitfall that fails to draw

one’s attention on the performance of the 67Ga-64,67Cu

separation

Above are the reasons why the 68Zn target based

64Cu production and its quality assessment are worthy of

further investigation In this paper some important

aspects regarding the 68Zn target based 64Cu +67Ga

production technology and the quality of the 64Cu

product are reported

Experimental

Reagent and material

The commercially available anion-exchange resin

AG1-X4 and cation-exchange resin AG50W-X4

(Bio-Rad) with average particle size of 200–400 mesh were

used for the radiochemical separations Analytical grade

hydrochloric acid and Milli-Q purified water were used

for the whole experimental process Isotopically

enriched 68Zn target was purchased from Trace-Sciences

International Inc USA.27 The target isotopic

compositions were 68Zn (>99.4%), 67Zn (0.43%), 66Zn

(0.08%), 64Zn (<0.01%), 70Zn (0.09%) and other

chemical impurities (<10 ppm)

Apparatus, radioactivity and radionuclide calibration

The radioactivity of the different radioisotopes was

Radioactivity measurement and radionuclide identification were carried out using an Ortec gamma-ray spectrometer coupled with a high purity Ge detector The gamma-ray energy and radioactivity calibration of this analyser system was performed using a 152Eu radioisotope solution standard The chromatographic elution profile was recorded on-line during the radionuclide separation process using a HPLC radioisotope detector coupled with the LAURA software supported computer-radioactive counter system

The targetry and the cyclotron proton bombardment of the 68 Zn target

The Ni–Cu (substrate)–68Zn target system, which is routinely used for the 67Ga radioisotope production, was applied in co-operation with the radioisotope production unit of the National Medical Cyclotron (NMC) Australia The target system was prepared by plating the metallic 68Zn onto a Ni coated Cu substrate The surface area of 68Zn target was 9.2854 cm2 and the Cu substrate surface (33.5 cm2) was coated with approximate 780 mg

Ni Approximately 800 mg 68Zn was used for each proton bombardment with the 29 MeV protons

Radiochemical separation of radionuclides from the cyclotron irradiated 68 Zn target and the 64 Cu radioisotope product

A glass column of 1.0 cm in diameter × 15 cm length loaded with anion exchange resin AG1-X4 was used for the chromatographic separation of different radionuclides (64Cu, 67Cu, 67Ga, 56Ni, 57Ni, 55Co, 56Co,

57Co, 65Zn, 68Zn) from the 68Zn target solution from which the majority of 67Ga radionuclide was removed during the 67Ga production process The 67Ga separation from the 68Zn target solution was carried out by retaining 67Ga in a cation-exchange resin AG50W-X4 and eluting all other radionuclide and 68Zn with 8M HCl solution.18

The anion-exchange resin column loading was performed as follows: The required amount of AG1-X4 resin was soaked in 0.1M HCl solution in a glass vial Then the degassing of resin was performed by shaking the vial in an ultra-sonic water bath for 5 minutes with vacuum application The degassed resin was then packed into the above mentioned glass column A peristaltic pump applied on the out-let of the column was used to draw eluent through the column After flushing with 6M HCl solution and reconditioning with HCl solution of relevant concentration, the column was ready to use for chromatographic separation

The 67Ga-pre-eliminated 68Zn target solution, as mentioned above, was added with some drops of H2O2 (30%), boiled for 5 minutes and cooled to room

resin column The separation process was performed

with successive isocratic elution and/or concentration

gradient elution Samples were also taken from each

elution fraction for the gamma spectrometric evaluation

The overlapped gamma-ray spectrum analysis

and its application for the assessment

of 67 Cu and 67 Ga radionuclide content

The overlapped gamma-ray spectrum of the mixture

of two radioisotopes A and B is depicted in Fig 1

(a) Gamma-ray peak area:

0

,γ

A

S = partial area of photo-peak E0 belonging to

radionuclide A;

0

,γ

B

S = partial area of photo-peak E0 belonging to

radionuclide B;

0

γ

S =S A,γ0+S B,γ0: total area of photo-peak E0

1

,γ

A

S = partial area of photo-peak E1 belonging to

radionuclide A;

1

,γ

B

S = partial area of photo-peak E1 belonging to

radionuclide B;

1

γ

S =S A,γ1+S B,γ1: total area of photo-peak E1

1

, n

A

S γ = partial area of photo-peak En belonging to

radionuclide A;

1

, n

B

S γ = partial area of photo-peak En belonging to

radionuclide B;

n

Sγ =S A,γn1+S B,γn1: total area of photo-peak En

(b) Relative intensity ratio of gamma-rays

For radionuclide A:

The corrected count numbers calculated from the gamma-peak area and its counting efficiencies,

n f f f

fγ0, γ1, γ2, , γ (%), are the following:

0

0 100

, 0

γ

γ

f

S

1

1 100

, 1 γ

γ

f

S

n

n f

S

A n A

γ

γ 100

, ⋅

was set up as follows:

From the definition of gamma-ray absolute intensity

of the nuclide A, I A,γ (%), it is deducible that:

0

1

,

, 0

1

γ

γ

A

A I

I

A A =

By giving

0

1

,

, 0 / 1

γ

γ

A

A I

I

the following is deduced: a1/0.A0=A1

The relative intensity ratio a n/0, intensity of the

photo-peak E n versus intensity of the photo-peak E0,

was set up in the same manner as above:

0

,

,

γ

A

A n I

I A

=

By giving

0

,

, 0 /

γ

γ

A

A

I

the following is deduced: a n/0 A0=A n

For radionuclide B:

The corrected count numbers calculated from the

gamma peak area and its counting efficiencies,

n f f

f

fγ0, γ1, γ2, , γ (%), are the following:

0

0 100

,

0

γ

γ

f

S

1

1 100

,

1

γ

γ

f

S

n

n

f

S

γ

γ 100

, ⋅

In the same manner as for radionuclide A the

following ratios are deduced from the gamma-ray

absolute intensity of the nuclide B, I B,γ (%):

photo-peak E1 versus intensity of the photo-peak E0, is:

0 / 1 ,

, 0

1

0

I

I B

B

B

B

=

=

γ

γ

photo-peak E n versus intensity of the photo-peak E0, is:

0 / ,

,

0 B 0 n

B

I

I B

=

=

γ

γ

or b n/0 B0=B n

As experienced in our previous work, the gamma-ray

measurement of a rather weakly radioactivity sample is

increases with the growing-up of the background counts beneath the photo-peak To reduce the calculation error and achieve the lower detection limits, the gamma-ray

of the input data for the calculation process is shown in Table 1

(c) Formulation of the overlapped gamma-ray spectrum analysis for the mixture of two radionuclides A and B

At the photo-peak E0:

photo-peak area Sγ0 as follows (see Sections a and d):

0 0

0 0 0

0 0

0

γ

γ γ

γ γ

γ

f

S f

S f

S

+

⋅

=

⋅

=

At the photo-peak E1:

Total corrected count numbers N1 is calculated from photo-peak area S as follows (see the Sections a and γ1

d):

1 1

1 1 1

1 1

1

γ

γ γ

γ γ

γ

f

S f

S f

S

+

⋅

=

⋅

=

By introducing the relative intensity ratios a1/0 and

b1/0into this equation (see Section d), it leads to:

N1 = a1/0.A0+b1/0.B0 (2)

Table 1 Input database for the overlapped gamma-ray spectrum analysis

Gamma-ray and energy E0, keV E1, keV E2, keV E3, keV … E n, keV

Absolute gamma-ray intensity, I A,γn for nuclide A, % I A,γ0 I A,γ1 I A,γ2 I A,γ3 I A,γn

Absolute gamma-ray intensity,

n B

I ,γ for nuclide B, %

0

, γ

B

I

1

, γ

B

I

2

, γ

B

I

3

, γ

B

I

n B

I ,γ

Gamma-intensity ratio, a n/0 for nuclide A a1/0 a2/0 a3/0 a n/0

Gamma-intensity ratio, b n/0 for nuclide B b1/0 b2/0 b3/0 b n/0

Gamma-counting efficiency,

n

fγ , %

0 γ

Gamma-peak area,

n

Sγ for the counting time t counting, counts

0 γ

S

1 γ

S

2 γ

S

3 γ

S

n

Sγ

By putting B0 from Eq (1) into Eq (2) we have the

following:

N1 = a1/0.A0+b1/0.N0–b1/0.A0

0

1 b N (a b ) A

N

⎠

⎞

⎜⎜

⎝

⎛

−

By dividing both sides of this equation with b1/0 and

rearranging, we have the corrected count numbers A0

belonging to the radionuclide A as follows:

⎟⎟

⎠

⎞

⎜⎜

⎝

⎛

−

⋅

⎟⎟

⎠

⎞

⎜⎜

⎝

⎛

−

⋅

=

1 1

0 / 1

0 / 1

0 0

/ 1 0 1

0

b a

N b

N

N

By introducing the photo-peak area to the A0

calculation, we have the following:

→

⎟⎟

⎞

⎜⎜

⎛

−

⎟

⎟

⎠

⎞

⎜

⎜

⎝

⋅

⎟

⎟

⎠

⎞

⎜

⎜

⎝

⎛

−

⋅

⋅

⋅

=

=

⎟⎟

⎠

⎞

⎜⎜

⎝

⎛

−

⋅

⎟⎟

⎞

⎜⎜

⎛

−

⋅

=

⋅

=

1

100 1

] / 100

[

] / 100

[

1

1 100

0 / 1

0 / 1

0 / 1

0 / 1

0 / 1

0 0

/ 1 0

1 ,

0

0 0 0

0

1 1

0

0

b a

f

S f

b

S

f S

b a

N b

N N f

S

γ

γ γ

γ

γ γ

γ

γ

⎟⎟

⎠

⎞

⎜⎜

⎝

⎛

−

⎟

⎟

⎠

⎞

⎜

⎜

⎝

⋅

⎟

⎟

⎠

⎞

⎜

⎜

⎝

⎛

−

⋅

⋅

⋅

=

1

100 1

] [

] [

0 / 1

0 / 1

0 / 1

0 1

0

0 1

b a

f

S b

f

S

f S

γ γ

γ

γ γ

(counts) (3b)

The radioactivity of radionuclide A calculated from

follows:

(Bq) 1

100

100 1

] [

] [

100

0 0 0 1

0

0 1

0

, 0

/ 1

0 / 1

0 / 1

,

0 1 0 ,

t I b

a

f

S b

f

S

f S

t I

A R

A

A A

⋅

⋅

⎟⎟

⎞

⎜⎜

⎛

−

⋅

⎟

⎟

⎠

⎞

⎜

⎜

⎝

⋅

⎟

⎟

⎠

⎞

⎜

⎜

⎝

⎛

−

⋅

⋅

⋅

=

=

⋅

⋅

=

↔

γ γ

γ γ

γ

γ γ

γ

(3c)

where t is the counting time in seconds

radioactivity of radionuclide A in the mixture:

(Bq) 1

100

100 1

] [

] [

100

0 0 0 0

0

0

, 0

/

0 /

0 /

,

0 0

,

t I b

a

f

S b

f S

f S

t I

A R

A n

n n

A n A

n n

⋅

⋅

⎟⎟

⎠

⎞

⎜⎜

⎝

⎛

−

⋅

⎟

⎟

⎠

⎞

⎜

⎜

⎝

⋅

⎟

⎟

⎠

⎞

⎜

⎜

⎝

⎛

−

⋅

⋅

⋅

=

=

⋅

⋅

=

↔

γ γ

γ γ

γ

γ γ

γ

(3d)

value [Eq (1)] as follows:

) (

1 1

0 / 1 0 / 1

1 0 / 1 0

0 / 1

0 / 1

0 0

/ 1 0 1 0

0 0 0

b a

N a N

b a

N b

N

N N

A N B

−

−

⋅

=

=

⎟⎟

⎠

⎞

⎜⎜

⎝

⎛

−

⋅

⎟⎟

⎞

⎜⎜

⎛

−

⋅

−

=

−

=

value as follows:

⎟⎟

⎞

⎜⎜

⎛

−

⋅

⎟⎟

⎞

⎜⎜

⎛

−

⋅

=

1 1

0 / 1

0 / 1

0 0

/ 1 0 1 0

a b

N a

N

N

calculation, we have the following:

→

⎟⎟

⎠

⎞

⎜⎜

⎝

⎛

−

⎟

⎟

⎠

⎞

⎜

⎜

⎝

⋅

⎟

⎟

⎠

⎞

⎜

⎜

⎝

⎛

−

⋅

⋅

⋅

=

=

⎟⎟

⎞

⎜⎜

⎛

−

⋅

⎟⎟

⎠

⎞

⎜⎜

⎝

⎛

−

⋅

=

⋅

=

1

100 1

] / 100 [

] / 100 [

1

1 100

0 / 1

0 / 1

0 / 1

0 / 1

0 / 1

0 0

/ 1 0

1 ,

0

0 0 0

0

1 1

0 0

a b

f

S f

a S

f S

a b

N a

N N f

S

γ

γ γ

γ

γ γ

γ γ

⎟⎟

⎠

⎞

⎜⎜

⎝

⎛

−

⎟

⎟

⎠

⎞

⎜

⎜

⎝

⋅

⎟

⎟

⎠

⎞

⎜

⎜

⎝

⎛

−

⋅

⋅

⋅

=

1

100 1

] [

] [

0 / 1

0 / 1

0 / 1

0 1

0

0 1

a b

f

S a

f S

f S

γ γ

γ

γ γ

(counts) (4b)

The radioactivity of radionuclide B calculated from

follows:

(Bq) 1

100

100 1

] [

] [

100

0 0 0 1

0

0

1

0

, 0

/ 1

0 / 1

0 / 1

,

0 1 0 ,

t I a

b

f

S a

f

S

f

S

t I

B R

B

B B

⋅

⋅

⎟⎟

⎠

⎞

⎜⎜

⎝

⎛

−

⋅

⎟

⎟

⎠

⎞

⎜

⎜

⎝

⋅

⎟

⎟

⎠

⎞

⎜

⎜

⎝

⎛

−

⋅

⋅

⋅

=

=

⋅

⋅

=

↔

γ γ

γ γ

γ

γ

γ

γ

(4c)

where t is the counting time in seconds

radioactivity of radionuclide B in the mixture:

(Bq) 1

100

100 1

] [

] [

100

0 0 0 0

0

0

, 0

/

0 /

0 /

,

0 0

,

t I a

b

f

S a

f

S

f

S

t I

B R

B n

n n

B n B

n

n

⋅

⋅

⎟⎟

⎠

⎞

⎜⎜

⎝

⎛

−

⋅

⎟

⎟

⎠

⎞

⎜

⎜

⎝

⋅

⎟

⎟

⎠

⎞

⎜

⎜

⎝

⎛

−

⋅

⋅

⋅

=

=

⋅

⋅

=

↔

γ γ

γ γ

γ

γ

γ

γ

(4d)

The mean values R A and R B of all these calculation

results are the actual radioactivity values of

radionuclides A an B in the mixture, respectively

Radioactivity of radionuclide A:

(Bq) )

1 (

100 )

100 (

) 1 ] [

] [

(

1

0 /

0 /

0 /

1 0 ,

0 0 0 0

0

∑

∑

⋅

⋅

−

⋅

⋅

⋅

−

⋅

⋅

⋅

=

=

=

↔

n

A n

n n

n

n A A

t I b

a

f

S b

f

S

f S

n

n

R R

n n

γ γ

γ γ

γ

γ

Radioactivity of radionuclide B:

(Bq) 1

100

100 1

] [

] [

1

1

, 0

/

0 /

0 /

1 0 ,

0 0 0 0

0

∑

∑

⋅

⋅

⎟⎟

⎞

⎜⎜

⎛

−

⋅

⎟

⎟

⎠

⎞

⎜

⎜

⎝

⋅

⎟

⎟

⎠

⎞

⎜

⎜

⎝

⎛

−

⋅

⋅

⋅

=

=

=

↔

n

B n

n n

n

n B B

t I a

b

f

S a

f S

f S

n

n

R R

n n

γ γ

γ γ

γ

γ

Gamma spectrometric measurement and the overlapping

gamma-ray spectrum analysis for the assessment of

67 Cu and 67 Ga in the 64 Cu solution

The above equations will be applied with the

following harmonizations:

nuclide

For 67Ga radioisotope:

gamma-ray photo-peak [Eq (3d)]:

(Bq) 1

100

100 1

] [

] [

0 0 0 0

0

, 67 0

/

0 /

0 /

0 , 67

t I

b a

f

S b

f S

f S

R

Ga n

n n

n Ga

n n

⋅

⋅

⎟⎟

⎞

⎜⎜

⎛

−

⋅

⎟

⎟

⎠

⎞

⎜

⎜

⎝

⋅

⎟

⎟

⎠

⎞

⎜

⎜

⎝

⎛

−

⋅

⋅

⋅

=

=

−

↔

−

γ γ

γ γ

γ

γ γ

(7)

393 keV [Eq (5)]:

(Bq) 1

100

100 1

] [

] [

3 1

3

3 1

, 67 -Ga 0

/

0 /

0 /

3 1 0 , 67 -Ga 67

-Ga

0 0 0 0

0

∑

∑

=

=

=

⋅

⋅

⎟⎟

⎞

⎜⎜

⎛

−

⋅

⎟

⎟

⎠

⎞

⎜

⎜

⎝

⋅

⎟

⎟

⎠

⎞

⎜

⎜

⎝

⎛

−

⋅

⋅

⋅

=

=

=

n

n

n n n

n

n

n

t I

b a

f

S b

f S

f S

R R

n n

γ γ

γ γ

γ

γ

For 67Cu radioisotope:

gamma-ray photo-peak [Eq (4d)]:

(Bq) 1

100

100 1

] [

] [

0 0 0 0

0

, 67 -Cu 0

/

0 /

0 /

0 , 67 -Cu

t I

a b

f

S a

f S

f S

R

n n n

n

n n

⋅

⋅

⎟⎟

⎠

⎞

⎜⎜

⎝

⎛

−

⋅

⎟

⎟

⎠

⎞

⎜

⎜

⎝

⋅

⎟

⎟

⎠

⎞

⎜

⎜

⎝

⎛

−

⋅

⋅

⋅

=

=

↔

γ γ

γ γ

γ

γ γ

(9)

393.0 keV, [Eq (6)]:

(Bq) 1

100

100 1

] [

] [

3 1

3

3 1

, 67 -Cu 0

/ 0 /

0 /

3 1 0 , 67 -Cu 67

-Cu

0 0 0 0

0

∑

∑

=

=

=

⋅

⋅

⎟⎟

⎞

⎜⎜

⎛

−

⋅

⎟

⎟

⎠

⎞

⎜

⎜

⎝

⎛ ⋅

⋅

⎟

⎟

⎠

⎞

⎜

⎜

⎝

⎛

−

⋅

⋅

⋅

=

=

=

n n

n n n

n n

n

t I

a b

f

S a

f S

f S

R R

n n

γ γ

γ γ

γ γ

Results and discussion

Radiochemical separation

For the purpose of identify the radioisotopes present

in the 68Zn target solution, the solution sample (67

Ga-pre-eliminated 68Zn target solution), which comes from

the 67Ga production process, was left to decay for 3 days

and then chromatographic separation was performed

The typical separation profiles and the radioisotopes

(64Cu, 67Cu, 66Ga, 67Ga, 56Ni, 57Ni, 55Co, 56Co, 57Co,

65+68 Zn) identified for the relevant chromatographic

fractions were shown in Figs 2 and 3 The good

separations can be achieved using both successive

isocratic elution and a concentration gradient Although

67Ga has been removed from the 68Zn target solution in

the 67Ga production process, a significant amount of

67Ga can be found and further separated from this

post-67Ga-production solution Due to the identical

gamma-ray spectrum of 67Ga and 67Cu, the result of gamma-ray

spectrometric analysis of 67Cu content in the 68Zn target

and/or 64Cu solution could be wrong without the support

of the chromatographic separation Obviously the

chromatographic separation combined with the

overlapped gamma-ray spectrum analysis described in

the next section offers a secure way to evaluate the 67Ga

and 67Cu content in both the target and final 64Cu

product

As for the separation performance, the advantage of

this anion-exchange chromatographic technique is that

the column loading can be performed with post 67 Ga-production target solution of high HCl concentration (8M HCl) and the following elution of the different nuclides can be performed with HCl solutions of lower concentration This avoided the time consumption and corrosive evaporation step involved in the separation process using the organic solvent eluents.18 Besides, the HCl concentration gradient elution, as shown in Fig 3, offers a better chance of automating of the high radioactivity separation which is definitely important for large scale radioisotope production The gamma-ray spectrometric analysis (Fig 4) of the Cu radioisotope fraction collected from the chromatographic separation present in Figs 2 and 3 has shown no contamination of 66+67Ga, 56+57Ni, 55+56+57Co and 65+68 Zn The 67Cu contamination in the 64Cu product, as shown in Fig 4, cannot be definitely removed by any chemical separation method and the issue is discussed in the following section of the proton bombardment optimization

The 66+67Ga fraction clearly separated from the 64+67Cu fraction strongly supported the overlapped spectrum analysis for accurate identification of the 67Cu and 67Ga contents in the less 67Ga contaminated 68Zn target solution used for the 64Cu production and that in the final 64Cu product The 65Zn fraction separation was useful for monitoring the elemental Cu leakage from the

Cu substrate, so that the intactness of the protective Ni interlayer between the 68Zn layer and Cu substrate (Fig 6) could be justified

Fig 2 Chromatographic diagram of the 68 Zn target solution separation using a successive isocratic elution technique

Column: A glass column of 1.0 cm in diameter × 15 cm length loaded with anion-exchange resin AG1-X4

Successive elutions: 5M HCl (0–65 ml) → 3M HCl (65–85 ml) → 2.5M HCl (85–110 ml) → H2O (110–150 ml)

Fig 3 Chromatographic diagram of the 68 Zn target solution separation using a concentration gradient elution technique (a) and HCl concentration gradient (b) Chromatographic column: A glass column of 1.0 cm in diameter × 15 cm length loaded

with anion-exchange resin AG1-X4

Fig 4 Gamma-ray spectrum of 64 Cu and 67 Cu radioisotopes in the 64 Cu product ( 64+67 Cu radioisotope elution fraction shown in Fig 2)

Fig 5 Gamma-ray spectrum of the different nuclides in the 68 Zn target solution

Table 2 Radionuclides and radioactivity induced in the Ni–Cu (substrate)–68 Zn target system used at NMC

The reactions of threshold energy E Threshold <22.8 MeV were selected for the natural Ni, E Threshold<16 MeV for the

natural Cu and E Threshold<29 MeV for 68 Zn Only radionuclides with the half-life longer than 3 hours were listed

Target RI Half-life Reaction Q-value,MeV E ThresholdMeV , EMeV Max, σMax,

mb

T.T.Y,*

µCi/µAh

68 Zn 67 Ga 78.3 h 68 Zn(p,2n) 67 Ga –11.981 12.158 21 700 8108.0

68 Zn 66 Ga 9.4 h 68 Zn(p,3n) 66 Ga –23.207 23.552 34 235 1800.0

68 Zn 64 Cu 12.7 h 68 Zn(p,αn) 64 Cu –7.790 7.905 30 20 946.0

68 Zn 67 Cu 58.6 h 68 Zn(p,2p) 67 Cu –9.977 10.125 45 3.5 6.7 Nat Ni 56 Ni 6.0 d 58 Ni(p,p2n) 56 Ni –22.466 22.857 42 21 –

Nat Ni 57 Ni 37 h 58 Ni(p,pn) 57 Ni –12.217 12.429 26 260 96

Nat Ni 55 Co 18 h 58 Ni(p,α) 55 Co

60 Ni(p,α2n) 55 Co

–1.336 –21.723

1.359 22.088

16

45

40

14

440 – Nat Ni 56 Co 77 d 58 Ni(p,2pn) 56 Co

60 Ni(p,αn) 56 Co

–19.548 –11.639

19.888 11.835

40

28

350

72

0.0176 – Nat Ni 57 Co 270 d 58 Ni(p,2p) 57 Co

60 Ni(p,α) 57 Co

–8.172 –0.263

8.314 0.268

22

51

540

200

8.8 – Nat Ni 58 Co 71 d 60 Ni(p,2pn) 58 Co

61 Ni(p,α) 58 Co

–19.986 –0.489

20.332 0.000

40

15

180

100

0.22 – Nat Ni 61 Cu 3.3 h 61 Ni(p,n) 61 Cu –3.019 3.069 9.5 480 200

Nat Ni 64 Cu 12.7 h 64 Ni(p,n) 64 Cu –2.457 2.496 11 670 200

Nat Cu 65 Zn 243.8 d 65 Cu(p,n) 65 Zn –2.134 2.167 10.5 230 23

Nat Cu 62 Zn 9.3 h 63 Cu(p,2n) 62 Zn –13.261 13.474 23.4 66 27

Nat Cu 64 Cu 12.7 h 65 Cu(p,pn) 64 Cu –9.910 10.064 26 93 10

Nat Cu 61 Cu 3.3 h 63 Cu(p,p2n) 61 Cu –19.738 20.054 35 149 40

* T.T.Y: Thick target yield for the 29 MeV proton bombardment

Quality assessment, targetry development and

proton-bombardment optimization for the improvement

of the 64 Cu and 67 Ga simultaneous production process

The gamma-ray spectrum analysis results of the 68Zn

target solution (Fig 5) and that of the

chromato-graphically separated elution fractions (Figs 2 and 3)

showed the presence of different radionuclides which

were induced by several nuclear reactions between the

through-penetrated protons and the different components

of the target assembly The radionuclides induced in the

Ni–Cu (substrate)–68Zn target system shown in Fig 6

and Table 2 have a good agreement with the results

obtained from gamma-ray spectrum analysis and

radiochemical separation

The 68Zn targetry was based on the solid target

assembly currently used at NMC The target surface

tilted at 6° with respect to the beam axis resulting in an

enlargement of the axial beam by a factor of 9.6 and a

corresponding reduction in material thickness

This target contained approximate 800 mg 68Zn The

thickness of 68Zn layer was 120.66 µm This offers the

1154.5 µm effective thickness of the 6° angle tilted

target This effective thickness is relevant to the

stopping power of the 29 MeV→17 MeV proton energy

range in the metallic 68Zn target

The thickness of the Ni layer underneath the 68Zn

was 26.2 µm, giving the 250.0 µm effective thickness in

this target assembly This effective thickness is relevant

to the stopping power of the 17 MeV→12 MeV proton

energy range in the metallic nickel

The characteristics of nuclear reactions induced in the Ni–Cu (substrate)–68Zn target system and their excitation functions were calculated (if not found in the literatures) and collected from the References 10, 14, 23,

29, 32 and presented in Table 2 and Figs 6 and 7 The reaction yields in Fig 8 were calculated based on the excitation function in Fig 7 and the stopping power was calculated using the data and equations in the literature.30,31

Obviously the current 68Zn target was designed for

67Ga production only As seen in Tables 2, 3 and 4, the

67Ga product was also contaminated with a significant

66Ga amount at E.O.B Because the half-life of 66Ga

(T1/2=9.4 hours) is shorter than that of 67Ga (T1/2=78.3 hours), the 66Ga contamination can be removed by standing the 67Ga product for the 66Ga decay Around

74 hours cooling time is required to decrease the 66Ga activity to the level of less than 0.2% (compared to the

67Ga radioactivity) as required by the British Pharmacopeia.33 However, this process causes the loss

of nearly 50% of the 67Ga activity too (compared to that

at E.O.B), which affects the production economics The applied proton energy range (17–29 MeV) is required to give a reasonably high yield of 67Ga from a productivity point of view However, the high radioactivity of the side reaction induced 66Ga, 67Cu, 56+57Ni, 55+56+57Co and 65+68Zn radionuclides was a problem for the radiochemical separation and radiation protection during the target disassembling and processing

Fig 6 The target assembly cross section and the radionuclides induced in the target currently used for the 67 Ga and 64 Cu production at NMC The

68 Zn, Nat Ni and Nat Cu layers were of 120.66 µm, 26.2 µm and 4.5 mm, respectively The stopping power and range of proton were calculated using

the data and equations in the literature 30,31