Applied Radiation and Isotopes 56 (2002) 41–46

Radiochemical analysis of 93Zr

A.G. Espartero*, J.A. Su!arez, M. Rodr!ıguez, G. Pi *na

Departamento de Fisi !on Nuclear, CIEMAT Avda. Complutense 22, 28040 Madrid, Spain

Abstract

The zirconium isotope 93Zr is a long-lived pure b-particle-emitting radionuclide, which is produced by nuclear fissionand neutron activation of the stable isotope 92Zr. This element is a constituent of the structural components of nuclear

reactor vessels. A selective liquid–liquid extraction method for radiochemical separation of Zr, based on liquid–liquidextraction with 1-(20-thenoyl)-3,3,3-trifluoroacetone in xylene and a subsequent stripping of 93Zr by an aqueous acidsolution, has been developed. The method was utilised to separate Zr from other pure b-particle and b2g emitters in

different kinds of samples. Decontamination factors higher than 99% for the pure b-particle and b2g emitters and anoverall chemical yield of 80% were obtained. The sensitivity of the method allows the determination of the isolated 93Zrby liquid scintillation counting and the minimum detectable activity value obtained was 0.067 Bq over a counting period

of 60 min. r 2002 Elsevier Science Ltd. All rights reserved.

Keywords: Radiochemical analysis; Zirconium determination; Liquid–liquid extraction; TTA–xylene; Radiological characterisation

1. Introduction

The zirconium isotope 93Zr is a long-lived pure b-particle-emitting radionuclide, with a maximum energyof 56 keV and a half-life of 1.53� 106 y. This radio-

nuclide is produced by both nuclear fission and neutronactivation of the stable isotope 92Zr, which is present indifferent quantities in the structural components of

nuclear reactor vessels and in the lining elements ofnuclear fuel elements.

The solution chemistry of zirconium is very compli-cated and different opinions exist regarding the ionic

species present in aqueous solutions. Zirconium ionsundergo extensive hydrolysis and polymerisation, formcomplexes and generate a great variety of colloid species,

strongly dependent on pH and concentration (Connickand McVey, 1949).

Its most important oxidation state is (+IV) and when

the acid concentration is high, zirconium exists as atetra-positive ion (Blumenthal, 1958; Zielen and Con-nick, 1956). Zirconium forms ionic complexes with

many ions, fluoride being the most stable complex. Italso forms chelate complexes which are soluble in

organic solvents and this is of great importance in thedevelopment of radiochemical separation procedures(Steinberg, 1960).

The aim of this work was to develop a selectiveradiochemical separation method based on liquid–liquidextraction using 1-(20-thenoyl)-3,3,3-trifluoroacetone

(TTA) in xylene and a subsequent stripping of 93Zr byan aqueous acid solution. The radiochemical method isapplied to different kinds of radioactive samples in orderto separate zirconium from other pure b-particle- and

b2g-emitting radionuclides, present in the samples suchas 54Mn, 55Fe, 60Co, 63Ni, 65Zn, 90Sr, 90Y, 94Nb, 133Ba,137Cs, 152Eu and 241Pu. The isolated 93Zr is determined

by liquid scintillation counting.

2. Materials and methods

2.1. Reagents

The reagents used included: hydroxylamine hydro-

chloride from Merck (Germany); a solution of 1 MAlCl3 that was obtained by dissolving AlCl3 � 6H2O from

*Corresponding author. Tel.: +34-91-346-6132; fax: +34-

91-346-6576.

E-mail address: amparo.espartero@ciemat.es

(A.G. Espartero).

0969-8043/02/$ - see front matter r 2002 Elsevier Science Ltd. All rights reserved.

PII: S 0 9 6 9 - 8 0 4 3 ( 0 1 ) 0 0 1 6 4 - 6

Merck (Germany) in 4 M HCl; a 95Zr/95Nb tracersolution from Amersham (UK); 0.50 M solution of

TTA in xylene was obtained by dissolving the TTAcompound from Merck (Germany) in xylene solvent; allother chemicals were of analytical grade.

2.2. Equipment

* g-ray spectrometer for g-ray measurements with a

Canberra System (Olen) multichannel analyser(MCA), a coaxial HP-Ge detector with relativeefficiency 20% and resolution 1.8 keV/1.33 MeV.

Volumetric samples of 25 ml, put in contact withthe HP-Ge detector, were used for g-ray measure-ments and detector calibration was carried out in the

same conditions using a multi-nuclide standardsolution.

* Liquid scintillation analyser, model Tri-Carb 2750TR/LL, with a minimum detection efficiency for 3H

of 60% and a minimum detection efficiency for 14C of95%. The analyser is equipped with a spectrumanalysis program.

2.3. Sample preparation and radiochemical procedure

Due to the different nature and composition of thesamples in which 93Zr determination will be done, it isnecessary to carry out a suitable sample preparation

treatment which involves digestion of the organic matter(Deconinck et al., 1990) and dissolution procedure ofthe solid samples and liquid samples containing un-

dissolved species (Espartero et al., 1998). The resultantsolution consists of 4 M HCl, in which the mainradionuclides present in the samples are in stable form.

To set up this radiochemical separation procedure for93Zr, a tracer solution of 95Zr/95Nb was used in order tofollow the behaviour of zirconium during the process byhigh-energy g-ray spectrometry through measurement of

the characteristic 95Zr line at 724 keV.Once the sample has been treated and dissolved, the

radiochemical separation procedure is applied to an

aliquot of the resultant solution. Because zirconiumforms hydrolytic and colloid species and in order toavoid difficulties in ensuring that radioactive and stable

zirconium are in the same chemical form, no inactive Zrcarrier was added.

The radiochemical procedure mainly consists of azirconium liquid–liquid extraction from a 4 M HCl/1 M

AlCl3 solution using TTA in xylene as extracting agent.The complete radiochemical separation process con-

sists of the following:

(1) Before the zirconium extraction with TTA inxylene, it is necessary to reduce the oxidation state ofPu(IV) to Pu(III) using hydroxylamine hydrochloride.

This is done in order to avoid Pu(IV) extraction withTTA (Cuninghame and Miles, 1956). (2) This is then

followed by liquid–liquid extraction of zirconium with

0.50 M TTA in xylene using same volumes of bothorganic and aqueous phases. (3) Due to the highconcentration of 55Fe presents in the samples and theeffective extraction of iron by TTA, it is necessary to

perform a stripping step to re-extract zirconium into anaqueous solution using 0.25 M HNO3–HF. This is donein order to achieve complete decontamination. (4)

Lastly, the final solution is measured by liquid scintilla-tion counting to determine the 93Zr activity.

In order to check the quality of the separation

procedure, the analysis was carried out with a samplecontaining 95Zr and with a sample free of zirconium.The flow diagram of the radiochemical procedure is

shown in Fig. 1.

3. Results and discussions

Table 1 shows the activity, in units of Bq, of the mainpure b-particle- and b2g-emitting radionuclides present

in the aliquots of samples where 93Zr must be analysed.Fig. 2 shows the g-ray spectrum of the initial solution, inwhich a known activity of 95Zr/95Nb tracer was added in

order to check the zirconium behaviour during theradiochemical procedure by g-ray spectrometry.

The decontamination factors shown in Table 2

correspond to the first zirconium extraction step using0.50 M TTA–xylene as the extracting agent, except that

Zr + Fe

Aliquot

Liquid-liquid

re-extraction

(Stripping)

0.25 M

HNO3+HF

Aqueous phase

Organic phase Zr

95Zr / 95Nb tracerNH2OH·HCl

Liquid-liquid

extraction

0.50 M

TTA-xylene

Determination by

LSCRadioactive

waste

Fig. 1. Flow diagram of the 93Zr radiochemical separation

procedure.

A.G. Espartero et al. / Applied Radiation and Isotopes 56 (2002) 41–4642

for 55Fe. In this case, the decontamination factor value

corresponds to that for a re-extraction step with 0.25 MHNO3–HF. These decontamination factors indicate thatthis radiochemical separation procedure is very selective

and effective since they are higher than 99% and theradioactive concentration of each radionuclide in thefinal solution was lower than the corresponding mini-

mum detectable activity (MDA) value.The g-ray spectrum of the TTA–xylene solution after

the zirconium extraction step (Fig. 3), where only thelines of 95Zr are present, corroborates the quality, in

terms of selectivity, of this extraction step.

Fig. 4 shows the b-particle spectrum of the TTA–-

xylene solution after the zirconium extraction step, of asample free of zirconium in which Pu(IV) has not beenpreviously reduced to Pu(III) by hydroxylamine hydro-

chloride. Although the separation of zirconium from thepure b-particle and b2g emitters is quantitative withhigh decontamination factors (Table 2), it may be

noticed that 55Fe and 241Pu are extracted effectively byTTA. The stripping step using 0.25 M HNO3–HF isnecessary in order to separate zirconium from iron. Theselectivity of this re-extraction step is shown in the b-

particle spectrum of the final aqueous solution of a

Table 1

Mean activity of the pure b-particle- and b2g-emitting radio-

nuclides present in the samples before the radiochemical

separation procedure

Radionuclide Main

emission

Mean

activity (Bq)

Uncertainty

(%)

54Mn X–g 4.86� 103 455Fe X 1.69� 104 1060Co b2g 3.75� 104 263Ni b 6.07� 104 565Zn b2g–X 1.20� 104 490Sr b 1.27� 102 590Y b 1.27� 102 594Nb b2g 4.93� 102 12133Ba g–X 2.00� 103 4137Cs b2g 1.13� 104 3152Eu b2g–X 4.62� 103 5241Pu b 2.59� 101 12

1.0E+0

1.0E+1

1.0E+2

1.0E+3

1.0E+4

1.0E+5

0 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500

Energy, keV

Co

un

ts

13

3B

a

15

2E

u

133Ba

13

7C

s9

5Z

r

95N

b5

4M

n 65Z

n

60C

o

15

2E

u

60C

o

40

K

15

2E

u

Fig. 2. g-ray spectrum of the initial solution with a known activity of 95Zr/95Nb tracer added.

Table 2

Decontamination factors for pure b-particle and b2g emitters

present in the samples, corresponding to the extraction step

using TTA–xylenea

Radionuclide Decontamination factorsa (%)

54Mn >99.80855Feb >99.13560Co >99.98463Ni >99.02565Zn >99.73990Sr >99.89090Y >99.89094Nb >99.987133Ba >99.269137Cs >99.525152Eu >99.321241Pu >99.613

a Decontamination factors ð%Þ ¼ ½1002ðAfinal=AinitialÞ100�:b Corresponds to the stripping step using 0.25 M HNO3–HF.

A.G. Espartero et al. / Applied Radiation and Isotopes 56 (2002) 41–46 43

sample free of zirconium, where only 241Pu is detected

(Fig. 5).The extraction of 241Pu is avoided by reducing its

oxidation state from Pu(IV) to Pu(III) using hydro-xylamine hydrochloride, as was indicated before. In this

way, the b-particle spectrum of the final solution, after

the complete radiochemical separation process from a

sample free of zirconium (Fig. 6), indicates that there isno interference whatsoever from other b-particle-emit-ting radionuclides present in the samples.

The chemical yield of the complete radiochemical

separation process is obtained from the activity of the

1.0E+0

1.0E+1

1.0E+2

1.0E+3

0 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500

Energy, keV

Co

un

ts

95Zr

95Zr

Fig. 3. g-ray spectrum of the organic solution after zirconium extraction using 0.50 M TTA–xylene.

Fig. 4. b-particle spectrum of the organic solution after the extraction step using 0.50 M TTA–xylene, of a sample free of zirconium in

which Pu(IV) has not been reduced to Pu(III).

A.G. Espartero et al. / Applied Radiation and Isotopes 56 (2002) 41–4644

Fig. 5. b-particle spectrum of the final aqueous solution after the stripping step of a sample free of zirconium in which Pu(IV) has not

been reduced to Pu(III).

Fig. 6. b-particle spectrum of the final aqueous solution after the complete radiochemical separation process from a sample free of

zirconium.

A.G. Espartero et al. / Applied Radiation and Isotopes 56 (2002) 41–46 45

95Zr tracer added and measured by g-ray spectrometryand is found to be around 80%.

This radiochemical method is very sensitive since thedecontamination factors of the pure b-particle and b2gemitters present in the samples are higher than 99%

(Table 2). The 93Zr MDA value for a typical sample is0.067 Bq over a counting period of 60 min. This MDAvalue is low enough to achieve the necessary thresholdsfor the radiological characterisation of radioactive

samples.

Acknowledgements

This work has been developed within theframework of the Association Contract CIEMAT-ENRESA.

References

Blumenthal, W.B., 1958. The Chemical Behavior of Zirconium.

Van Nostrand, Princeton, NJ.

Connick, R.E., McVey, W.H., 1949. J. Am. Chem. Soc. 71,

3182.

Cuninghame, J.G., Miles, G.L., 1956. J. Inorg. Nucl. Chem. 3,

54.

Deconinck, J.M., Van Iseghem, P., Boden, R., Daniels, A., De

Regge, P., Hurtgen, C., Huys, D., Klein, M., Verrezen, F.,

1990. Investigation on the determination of disposal critical

nuclides in waste from PWR power plants. UE Final

Report, Research Contract no. FI1W-0225-B(T).

Espartero, A.G., Su!arez, J.A., Rodr!ıguez, M., 1998. Appl.

Radiat. Isot. 49 (9–11), 1277–1282.

Steinberg, E.P., 1960. The Radiochemistry of Zirconium and

Hafnium. Nuclear Science Series, NAS-NS 3011. National

Academy of Sciences, National Research Council,

Washington, DC.

Zielen, A.J., Connick, R.E., 1956. J. Am. Chem. Soc. 78, 5785.

A.G. Espartero et al. / Applied Radiation and Isotopes 56 (2002) 41–4646