- Email: [email protected]
A novel method for large-area sources preparation for the calibration of β- and α-contamination monitors
Applied Radiation and Isotopes 56 (2002) 21–29
A novel method for large-area sources preparation for the calibration of b- and a-contamination monitors V. Tsoupko-Sitnikova,*, J.L. Picoloa, M. Carrierb, S. Peulonc, G. Moutarda a
Laboratoire National Henri Becquerel, BNM-LNHB, CEA/DAMRI Saclay, B.P. 52, 91191 Gif-sur-Yvette Cedex, France b ! Institut National des Sciences et Techniques Nucleaires, CEA/Saclay, 91191 Gif-sur-Yvette, France c ! Laboratoire de Chimie, Electrochimie Moleculaire et Chimie Analytique, CNRS/Universite! de Bretagne Occidentale, 6, avenue Le Gorgeu, BP 809, 29285 Brest Cedex, France Accepted 31 July 2001
Abstract A method is proposed for the preparation of large-area reference sources for the calibration of b- and acontamination monitors. It is based on the incorporation, by the ion-exchange mechanism, of the radionuclide in a thin film of a conducting polymer ion-exchanger preliminarily grown on a metal support. Conducting pyrrolebased polymer functionalized by carboxylic cation-exchange groups is used to prepare 60Co and 90Sr–90Y b-particle sources. Electrochemical polymerization of the corresponding monomer on different conducting supports is studied and a special electrochemical equipment developed permitting the preparation of large-area polymer films of controlled and reproducible thickness. The ion-exchanger obtained is characterized in terms of chemical affinity for cations Co2+ and Sr2+. Incorporation of the radionuclides in the large-area ion-exchanger films thus obtained is studied and optimized with respect to the uniform distribution of the radionuclide. The performance of the procedure is demonstrated using the example of circular sources 44 mm in diameter prepared on stainless steel supports. The sources obtained are characterized in terms of activity, b-particle flux, uniformity and source efficiency. r 2002 Published by Elsevier Science Ltd.
1. Introduction Radioactive surface contamination control is usually carried out by means of large-area gas-filled (proportional or Geiger–Muller) counters or scintillation detectors. The characteristics of the large-area reference source used for the calibration of these monitors are specified in the international standard ISO 8769 (ISO, 1988), for both a-emitters and b-emitters with maximum b-energy greater than 150 keV. In accordance with these requirements the active surface area of such sources must be at least 100 cm2
*Corresponding author. Fax: +33-169089529. E-mail address: [email protected] (V. TsoupkoSitnikov).
(rectangular shape of 100 150 mm2 is specially recommended). For Class I sources the particle flux (2000–10,000 s1) must be measured by the National Metrology Laboratory with standard uncertainty not exceeding 3%, and the radionuclide activity contained in the source is to be deduced by the manufacturer from the fabrication procedure in a way traceable to the national standards with an uncertainty lower than 10%. The non-uniformity of the particle flux over the surface of the source must be lower than 10%. For its calculation the active surface of the source is to be divided into equal individual elements (surface area p10 cm2), and the dispersion of the particle fluxes of all the elements is to be compared with the average value. The use of windowless 2p proportional gascirculation counters for the absolute measurement of the particle flux of Class I sources imposes a conducting source surface.
0969-8043/02/$ - see front matter r 2002 Published by Elsevier Science Ltd. PII: S 0 9 6 9 - 8 0 4 3 ( 0 1 ) 0 0 1 6 1 - 0
22
V. Tsoupko-Sitnikov et al. / Applied Radiation and Isotopes 56 (2002) 21–29
The conventional method of preparation of such reference sources is based on the incorporation of the radionuclide in the porous surface of an anodized aluminium foil. The radioactive material in such sources is usually concentrated in a relatively thick (3–10 mm) surface layer composed of aluminium and aluminium oxide, the activity depth distribution profile strongly depending on the minute details of the preparation procedure and often being difficult to control and to reproduce. The reproducibility of the depth distribution profile is of crucial importance to achieve the reproducible characteristics of the reference sources, especially for a- and low-energy b-particle emitters (see the detailed analysis made by Janen and Thieme, 2000). Its variation influences the source efficiency value and changes the energy spectrum of the emerging particles, the latter in its turn altering the response function of a secondary reference transfer instrument (e.g. a proportional counter with an entrance window). Thus two reference sources of the same radionuclide and emerging particles flux but of different spectrum shape will not create the same response when measured with such an instrument. The relatively ‘‘thick’’ reference sources described above do not well-represent a typical radioactive contamination where the radionuclides are deposited in a very thin and fragile layer on the surface. Finally, in many cases the aluminium supports of the sources are not representative enough of the heavier materials to be checked for contamination (stainless steel, copper, and zinc). Availability of reference sources fabricated on this kind of support is then desirable. The goal of this work was to develop a preparation method for large-area reference sources which would be more representative of the typical case of radioactive contamination (‘‘very thin sources’’) and of materials to be checked (ZE26). Special attention was paid to the reproducibility of the thickness of the active layer of the source. The results exposed in this article concern the radionuclides existing as bivalent cations in the solution, namely 60Co and 90Sr, as well as all other radioactive isotopes of these elements.
2. Conducting polymer ion exchanger as source substrate The source preparation procedure using this technique (Carrier et al., 1996) can generally be described by the following sequence: 1. A thin (100–350Zm) insoluble film of a conducting polymer ion exchanger is formed by electrochemical polymerization of an organic monomer on a largearea conducting support. The film thickness and thus
the future source efficiency value are electrochemically controlled. 2. A radionuclide is subsequently incorporated in the film from a relatively large volume (E1 ml cm2 of the surface) of aqueous solution of a certain composition by the ion exchange mechanism. 3. The aqueous solution is discarded (with a possible control of the residual activity), and the source, after appropriate rinsing and drying, is subjected to the particle flux absolute measurement. This technical approach has recently been developed by St!ephan and Carrier (1997) and St!ephan et al. (1995a, b) with a view to accelerate the small-size a-sources preparation in actinide environmental assay. To bind strontium and cobalt cations we used a polypyrrole functionalized by the carboxylic cationexchange group (hereinafter ‘‘poly-PYCOO’’, Fig. 1) which forms regular, well-adherent films when electrochemically prepared on small-area (1–2 cm2) electrodes of glass covered with semiconductor tin oxide (St!ephan et al., 1995a). The following aspects of this technique were studied in this work: (a) Possibility of the preparation of good-quality polyPYCOO films on metal surfaces (e.g. stainless steel) by anodic polymerization of PYCOO; feasibility of the preparation of a homogeneous coating on large metal surfaces. (b) Chemical affinity of the carboxylic ion-exchanger poly-PYCOO for the cations Sr2+ and Co2+ which must be strong enough to assure sufficient and reproducible transfer of the radionuclides from the solution to the film. (c) Feasibility of the homogeneous incorporation of the radionuclide from a solution into the large-area ion-exchange films. (d) Overall reproducibility of the procedure and of the characteristics of the sources obtained.
3. Experimental equipment The monomer PYCOO was synthesized according to the procedure described by Deronzier and Marques (1989). All other products were of analytical grade and used as received. Electrochemical experiments were performed using a Princeton Applied Research model 263A potentiostat/ galvanostat. Poly-PYCOO films for the affinity study were grown on commercially available (20 20 1) mm3 glass plates with vacuum-sputtered fluor-doped SnO2 coating on one side. Anodic polymerization on this kind of samples was carried out in a 20 mm glass spectrophotometric cell with a platinum plate as
23
V. Tsoupko-Sitnikov et al. / Applied Radiation and Isotopes 56 (2002) 21–29 COOH
COOH
PYCOO
poly-PYCOO H2C
2
-2 e N CH2
COOH
H2C -2X e
N N
+2H
H2C
COOH
XPYCOO
N
+ 2XH+
N H2 C
COOH
N x
H2C
COOH
Fig. 1. Scheme of the electrochemical polymerization of the 4-(pyrrol-1-yl)methyl)benzoic acid (hereinafter ‘‘PYCOO’’). The oxidized conductive form of the polymer produced (‘‘poly-PYCOO’’) contains approximately one positive charge for 4 pyrrole units.
counterelectrode under conditions described by St!ephan et al. (1995a). The polymer mass deposition rate on gold-coated electrodes as a function of total electrical charge was measured using a quartz microbalance coupled with the galvanostat. Film thickness on glass and polished stainless steel was measured profilometrically (Dektak 3030 T ST SLOAN Auto Surface Texture profiler). Radionuclide distribution experiments for the affinity study were performed in the system {glass/SnO2 20 20 mm2 sample carrying a poly-PYCOO film85 ml of liquid phase} in polyethylene vials shaken at a speed of 120 min1 during 24 h for equilibration. 60Co and 85Sr were used as tracers, available at Laboratoire National Henri Becquerel as 0.1 M HCl solutions with a certified specific activity of 800 kBq g1 at an uncertainty of 0.1%. Activities of the solid samples and the liquid phase residual activity were measured by HPGe gspectrometry. Residual activity of 90Sr and 90Y in the liquid phase was measured by liquid scintillation (LS) (GUARDIAN 1414, WALLAC). A conventional laboratory agitator was used for the horizontal orbital (18 mm) agitation of the cells for the radionuclide incorporation into the large-surface samples. For the agitation by vertical vibration (50 Hz) a home-made electromagnetic apparatus with variable vibration amplitude (0.1–1.5 mm) was applied. The uniformity of the large-area b-particle sources was calculated from their digital autoradiographies obtained with a commercial INSTANTIMAGER 20 24 cm2 microchannel array detector (Packard In-
strument Co.). For rectangular sources the calculation was performed using the standard rectangular analytical grill consisting of 8 8 mm2 individual elements, applied to the whole surface of the source. The circular sources (active surface 44 mm in diameter) were analysed using a specially composed grill (32 circular elements 7 mm in diameter, Fig. 5). The absolute measurement of the b-particle flux of the large-area sources was performed using a modified commercial large-area gas-flow proportional counter AB 710 (EURISYS MESURES) with an active area of 368 213 mm2 for particle detection. The aluminizedmylar window and the protective grid of the counter were removed and replaced by a stainless steel support which also holds the source in a reproducible position. The whole set-up was shielded by a 70 mm thick lead cladding on all sides. The counting gas was pure methane circulated at a flow rate of 1.70 l/h. The counter was operated at +3000–+4000 V in the b-plateau. Conventional electronics and software were used in the counting system; a deadtime of 15 ms was superimposed to assure the correction of the deadtime losses.
4. Results and discussion 4.1. PYCOO electrochemical polymerization For PYCOO polymerization on glass/SnO2 and goldcoated electrodes the electrochemical conditions proposed by Ste! phan et al. (1995a), were used: 0.005 M PYCOO solution in acetonitrile containing 0.01 M
24
V. Tsoupko-Sitnikov et al. / Applied Radiation and Isotopes 56 (2002) 21–29
HClO4 to keep the carboxylic group protonated and 0.05 M tetrabutylammonium perchlorate to assure sufficient conductivity of the solution. The polymerization was carried out at a constant current density of 0.2 mA cm2. Under these conditions the polymer mass deposited on a golden anode increases linearly with time up to the value of the charge transferred 150 mC cm2: m ¼ 0:91Q;
s ¼ 2:2%;
ð1Þ
where m is the mass of the polymer formed per unit of the anode area, in mg cm2, and Q is the electrical charge transferred per unit of the anode area, in mC cm2. The uncertainty stems from linear fitting of deposited massFelectrical charge experimental data. The average thickness of the polymer film formed on the glass/SnO2 electrodes also grows linearly with the electrical charge transferred: d ¼ 5:02 þ 6:91Q;
s ¼ 20%;
ð2Þ
d being expressed in Zm. The considerable uncertainty of this value is due to the insufficient uniformity of the film formed on the small-area anode totally immersed in the monomer solution. The approximate density of the polymer material rE1:32 g cm3 can be deduced from the two parameters above. The electropolymerization yield estimated from Eq. (1) is about 85% taking into account the presence of one ClO 4 ion per four monomer units in the conducting oxidized form of the polymer. The ion-exchange capacity of the polymer was determined from the isotherm of Sr2+ cations adsorption using 85Sr as tracer (Fig. 3). Its saturation corresponds to the specific quantity of the ion exchange sites z ¼ 4:3109 mol mC1
ð3Þ
or, combining with (1), 4.7 mol kg1. This value is in acceptable accord with the one obtained by saturation of the polymer with trisphenanthrolineiron: 3.8 109 mol mC1 (Ste! phan et al., 1995a). The electrochemical polymerization yield deduced from (3) is of approximately 90%. Under the conditions mentioned above, the polymerization of PYCOO on a polished stainless steel anode results in good-quality adherent polymer films very similar to those obtained on the glass/SnO2 anodes. Profilometrical measurements show that within the uncertainty of 20% the thickness of the films obtained on stainless steel is compatible with one of the films grown on glass/SnO2 plates. The considerable uncertainty is due to the B100 Zm roughness of the mirrorpolished steel surface. Hereinafter we assume the validity of the results (1)–(3) for the poly PYCOO films grown on stainless steel.
4.2. Preparation of uniform large-surface poly-PYCOO coatings The difficulties associated with the preparation of uniform electrochemical coatings on large-area conducting substrates are analysed by Graham (1971), Prentice and Tobias (1982), Shih and Pickering (1987) and Gerisher and Tobias (1994). On a metal plate completely immersed in an electrochemical bath the thickness of the coating obtained usually varies parabolically, increasing from the centre to the perimeter of the plate. This effect was also observed in PYCOO anodic polymerization on stainless steel plates. Thus on a 50 80 mm2 rectangular plate the thickness variation from the centre to the edge can reach 300%. Very uniform films on extended plane metal surfaces of various configurations can be obtained using a specially designed electrochemical cell (Fig. 2). The surface to be coated occupies the entire cross-section of the cell, being limited by a normally positioned insulator wall (Teflon). The edges of the stainless-steel plate are not exposed. The counterelectrode (cathode) is parallel to the treated surface and equals it in size. This configuration assures the homogeneous current distribution over the whole metal surface exposed to the electrolyte in the cell. The state of the stainless steel surface is of great importance for the uniformity of the coatings obtained. The flatness of the plate must be better than 0.1 mm. Very good results were obtained with mechanically mirror-polished and sand-blasted stainless steel supports. The following treatment of the plates was applied before the polymer deposition in the cell: (a) Washing in an ultrasonic bath in a 10% solution of an alkaline detergent at 601C, 120 min; (b) Washing in an ultrasonic bath in a 2 M NaOH solution at 601C, 60 min; (c) Rinsing with hot distilled water; rinsing with acetone and drying. All the above mentioned conditions fulfilled, coatings with non-uniformity lower than 10% can be obtained on large-area stainless steel plates (up to 150 cm2). The best criterion for a rapid and reliable uniformity check is the colouration of the coating due to light interference.
4.3. Poly-PYCOO affinity for cations Co2+ and Sr2+ Poly-PYCOO shows cation-exchange properties in alkaline solutions where the deprotonization of the carboxylic group takes place. Although the affinity of carboxylic ion exchangers for bicharged cations is relatively high, the background concentration of alkaline cations (Na+, NH+ 4 y) should be minimized.
25
V. Tsoupko-Sitnikov et al. / Applied Radiation and Isotopes 56 (2002) 21–29 counterelectrode
protective lid
teflon cage
source support
spacer
set screw
contact screw
O-ring
base of the cell
Fig. 2. Scheme of the cell for electrochemical polymerization.
Dilute ammonia solutions with pH varying from 8 to 10.5 present a good compromise in this respect. Table 1 shows the dependence of no carrier added 60 Co fixation on 50 mC cm2 poly-PYCOO films on ammonia concentration in aqueous solution. The best and the most reproducible yields are obtained at 0.001 M ammonia concentration. The radionuclide transfer in the film is sufficiently complete for practical application of the films. The affinity of poly-PYCOO for the cations of strontium is considerably higher and does not depend so drastically on ammonia concentration. A series of experiments performed showed that in the same configuration (ammonia concentration 0.01–0.02 M) the average strontium sorption yield amounts to 99.6% with variations of the order of 0.2%. In order to better quantify the process the respective adsorption isotherm was studied by increasing concentration of strontium carrier in the solution up to the complete saturation of the ion exchanger (Fig. 3). The distribution coefficient value deduced from the initial linear part of the isotherm is 9.0 106 l kg1. The demonstrated very high affinity of the ionexchanger for strontium will allow the quasi-quantitative transfer of 90Sr to the film without the necessity of a control measurement. 4.4. Radionuclide incorporation in large-area poly-PYCOO films The construction of a special extraction cell designed for this operation is depicted in Fig. 4. Due to the watertight junction between the Teflon cage and the stainless steel plate, only the ion-exchanger film is exposed to the radionuclide solution. The corked opening serves to fill and evacuate the cell and to add the radionuclide
Table 1 Adsorption of 60Co on poly-PYCOO films at different ammonia concentrations. System: {films 50 mC cm2 on 4 cm2 glass/SnO2 platesF5 ml of aqueous phase}. The uncertainties stem from the activities measurement by g-spectroscopy and do not correspond to the reproducibility of the ion-exchange equilibria Ammonia concentration (M)
Sorption yield
Uncertainty (%)
0
0.63 0.58 0.97 0.95 0.95 0.92 0.88 0.85 0.74
0.7 0.5 0.5 0.4 0.5 0.4 0.5 0.5 0.5
0.001
0.01 0.0167
solution from a weighted pycnometer. To assure the reliable contact of the liquid phase with the entire surface of the plate and good agitation conditions, the level of the liquid in the cell must be at least 5 mm high. The radionuclide surface distribution on the sources thus obtained strongly depends on the agitation mode applied, since the adsorption is controlled by the diffusion. The ‘‘static’’ incorporation (no agitation) gives very good results in this respect but does not assure the equilibrium incorporation yields in a reasonably short time (o24 h). High-speed horizontal orbital agitation (400 min1) assures the equilibrium incorporation yields but often
26
V. Tsoupko-Sitnikov et al. / Applied Radiation and Isotopes 56 (2002) 21–29
1E-7 4.0E-8 3.6E-8
Equilibrium Sr in the solid phase, mol
Sr retained in the ion-exchanger film, mol
ion-exchanger saturation
1E-8
3.2E-8 2.8E-8 2.4E-8 2.0E-8 1.6E-8 Fit Results:
1.2E-8
Equation: Y = 327.853 * X + 1.94013E-009 Number of data points used = 16
8.0E-9 4.0E-9
1E-11 2E-11 3E-11 4E-11 5E-11 6E-11 7E-11 8E-11 9E-11 1E-10
Equilibrium Sr in the aqueous phase, mol
1E-12
1E-11
1E-10
1E-9
1E-8
1E-7
1E-6
1E-5
Sr equilibrium quantity in the solution, mol Fig. 3. Isotherm of Sr2+ cations adsorption on poly-PYCOO films and linear approximation of its initial part (superimposed). System: glass/SnO2 20 20 mm2 sample carrying 50 mC cm2 poly-PYCOO film85 ml of 0.017 M ammonia solution.
feed hole teflon cage
radionuclide solution
source support + polymer film
spacer
set screw
contact screw
O-ring
base of the cell
Fig. 4. Scheme of the radionuclide incorporation cell.
brings reproducible non-uniform distributions due to static distribution of turbulences in the cell (Fig. 5), especially for the circular cell geometry. The inhomogeneity of the rectangular sources thus obtained is of the order of 11–13%, whereas that of circular ones 18–23%.
Sufficiently homogeneous radionuclide surface distributions were obtained independent of the source configuration using the agitation by vertical vibration of the cell. At 50 Hz and 1.0–1.5 mm vibration amplitude sources with a non-uniformity lower than 6% are reproducibly obtained (Fig. 5).
V. Tsoupko-Sitnikov et al. / Applied Radiation and Isotopes 56 (2002) 21–29
27
Fig. 5. Influence of the cell agitation mode on the radionuclide surface distribution for circular sources. (a) Horizontal orbital agitation with 18 mm amplitude, source uniformity 22%. (b) Vertical vibration 50 Hz/1.0 mm amplitude, source uniformity 4%. (c) Grid for source uniformity calculation.
4.5. Sources preparation and characterization The numerical results of the preparation of the largearea sources and their characterization are demonstrated using circular sources with an active deposit diameter of 44 mm, prepared on stainless steel supports of 50 mm diameter and 3 mm thick. Two series of four sources of 60Co (two on mirrorpolished and two on sand-blasted supports) and of three sources of 90Sr–90Y on sand-blasted supports were prepared. The thickness of the polymer film electrochemically grown on the supports was 30 mC cm2 (about 200 Zm). The liquid phase volume in the radionuclide incorporation cell was 12 ml (0.79 ml cm2 of the source surface). Ammonia concentration was 0.001 M for the preparation of the 60Co sources and 0.017 MFfor the 90Sr–90Y sources. The radionuclide solution (1–2 drops from a weighted pycnometer) was added directly into the incorporation cell just before starting the agitation. The cell was agitated during 15 h at 50 Hz with a 1 mm vertical amplitude. The liquid phase was then removed from the cell and the latter rinsed 2 times with 3 ml of acetonitrile. The cell was then disassembled and the source dried for 5 min at room temperature. Residual activity in the liquid phase and in the rinsing acetonitrile was finally measured. The ‘‘lost activity’’ value (residual activity on the Teflon cage of the cell) was estimated by smear tests. The 90Sr–90Y sources were stored during 30 days to assure the equilibrium between the two radionuclides. The results of the sources characterization are presented in Table 2. In the case of 90Sr–90Y sources preparation the LS control confirmed the low absolute value of the residual activity of the radionuclide not incorporated in the support (liquid phase+rinsing solvent+wall adsorption) and its sufficient reproducibility (0.3–0.6% of the activity initially introduced). The constant correction of 0.5% is considered to be adequate
for the correct source activity calculation and the routine LS control is deemed to be unnecessary in the case of 90Sr–90Y. The residual activity of 60Co is significantly higher and less reproducible (1.6–10.5%) due to the lower distribution coefficient of cobalt in the system polyPYCOOFwater solution and the significant sensitivity of the ion-exchange equilibrium to the ammonia concentration. Consequently, a systematic g-spectrometric control of the residual activity of 60Co in the liquid phase is recommended for the precise source activity calculation. The source efficiency values are generally high and reproducible for the same kind of source (Table 2). For 60 Co the efficiency of the sources prepared on mirrorpolished stainless steel supports is several per cent higher than on sand-blasted supports (74% and 70%, respectively). This difference is naturally explained by the increase of the absorption of the low-energy b-particles in the source and by a certain decrease of backscattering on sand-blasted supports in comparison with mirrorpolished ones. The absolute source efficiency values obtained (Table 2) are considerably higher than those of the commercially available large-area b-sources prepared by the anodized aluminium foil technique (AMERSHAM, 1997): B48% for 60Co and B63% for 90Sr–90Y. The values obtained are, nevertheless, compatible with the efficiency estimates of the ‘‘ideally thin’’ large-area sources fabricated on aluminium supports (Berger, 1998): 70% for 60Co and 69.6% for 90Sr–90Y. Even higher efficiency values in the case of the sources on stainless steel supports are due to the superposition of two factors: the increase of the number of the particles backscattered on stainless steel in comparison with aluminium (approximately +50%) (Siegbahn, 1979) and absorption in the 200 Zm thick source active layer composed of an organic matter, especially in the case of 60Co.
28
3.5 3.6 3.5 1.8 1.8 1.8 0.741 0.742 0.741 1.5 1.5 1.5 7567 5170 10655 1.0 1.0 1.0 5105 3485 7185 0.6 0.8 0.7 5.3 2.3 2.7 0.5 0.6 0.6 a
MP: mirror-polished; SB: sand-blasted stainless steel support.
5.9 5.3 5 6 6 6 13 13 13 1.0 1.0 1.0 90
Sr–90Y
51SB 52SB 53SB
5129 3505 7206
6 6 6 6 13 13 13 13 0.17 0.16 0.17 0.16 60
Co
54MP 55MP 57SB 58SB
6450 8155 6737 7457
Unc. (%) Activity ‘‘lost’’, (Bq)
252 116 343 763
2 3 1.5 1.3
1.5 2 F 3.9
10 15 F 10
6183 8024 6381 6677
0.2 0.2 0.2 0.25
4590 5935 4475 4710
1.5 1.5 1.5 1.5
0.742 0.740 0.701 0.705
1.5 1.5 1.5 1.5
5.7 4.9 4.2 4.4
5. Conclusion
Unc. (%) Activity introduced, (Bq) N1 and type of supporta Nuclide
Table 2 Preparation and characterization of +44 mm circular sources
Residual activity in liq. Phase (Bq)
Unc. (%)
Activity in rinsing solvent (Bq)
Unc. (%)
Source activity (Bq)
Unc. (%)
b flux, (s1)
Unc. (%)
Source b yield
Unc. (%)
Source uniformity (%)
V. Tsoupko-Sitnikov et al. / Applied Radiation and Isotopes 56 (2002) 21–29
The large-area b-sources fabricated using the technology described above meet the requirements imposed by the respective international standard (ISO, 1988) in terms of uncertainties of stated source activity and bparticle flux and of source uniformity. Thus the source activity can be deduced from the parameters of the fabrication procedure with an uncertainty not exceeding 1%, and the non-uniformity of the sources obtained is typically lower than 5% for the strict procedure of the uniformity evaluation applied. The developed procedure permits the source preparation on stainless steel, nickel, copper and some other metal supports. The sources thus obtained are characterized by a low intrinsic absorbance due to their particularly thin active layer (200–350 Zm of an organic matter). The reproducibility of the thickness controlled at a preliminary step of the source preparation assures the reproducibility of their efficiency value. At present the technical approach has been validated for the case of the radionuclides 60Co and 90Sr–90Y. The source efficiency values obtained for these radionuclides are considerably higher than those for the commercially available sources of the same radionuclides: 70–74% for 60 Co depending on the type of the metal support quality and 74% for 90Sr–90Y for the sand-blasted stainless steel supports. The described method, with certain modifications concerning the nature of ion-exchange groups present in the polymer, can be extended on other b- and a-emitting nuclides of metallic as well as non-metallic elements. This work was financially supported by Laboratoire Etalons d’Activit!e LEA/CERCA, Site du Tricastin, BP 75-26701 Pierrelatte Cedex, France. The authors express their gratitude to Philippe Cassette (LNHB) for useful discussions and consultations on LS measurements. References AMERSHAM, 1997. In: Calibration Standards and Instruments for Measuring Radioactivity. AMERSHAM Isotrak catalogue, 1st Edition. Amersham International, Buckinghamshire HP7 9NA. Berger, J., 1998. A method for determining the 2p-counting efficiency of beta-particle sources. Nucl. Instrum. Methods. B134, 276–286. Carrier, M., Burger, P., De Bruyne, T., 1996. Preparation de sources radioactives, notamment en vue d’analyse d’actinides en solution. Brevet franc-ais FR9403186. Deronzier, A., Marques, M.J., 1989. Electrodes modified by a Ni-dibenzotetraaza[14] annulene complex via reductive electropolymerization of an a-dibromobenzyl derivative or oxidative electropolymerization of a dipyrrole derivative. J. Electroanal. Chem. 265, 341–353.
V. Tsoupko-Sitnikov et al. / Applied Radiation and Isotopes 56 (2002) 21–29 Gerisher, H., Tobias, Ch.W., 1994. Advances in electrochemical science and engineering. In: Gerisher, H., Tobias, Ch.W. (Eds.), Current Distribution and Shape Change in Electrodeposition of Thin Films for Microelectronic Fabrication, Vol. 3. VCH Verlagsgesellschaft mbH, FRG. Graham, A.K., 1971. In: Kenneth Graham, A. (Ed.), Electroplating Engineering Handbook, 3rd Edition. Van Nostrand Reinhold Company, New York. ISO, 1988. Standard ISO 8769: Reference sources for the calibration of surface contamination monitorsFBeta emitters (maximum beta energy greater than 0.15 MeV) and alpha-emitters. International Standard Organization, Geneva. Janen, H., Thieme, K., 2000. On the determination of activity profiles in wide-area reference sources. Appl. Radiat. Isot. 52, 533–537. Prentice, G.A., Tobias, C.W., 1982. Simulation of changing electrode profiles. J. Electrochem. Soc. 129, 78–85.
29
Shih, H., Pickering, H.W., 1987. Three-dimensional modeling of the potential and current distribution in an electrolytic cell. J. Electrochem. Soc. 134, 551–558. Siegbahn, K., 1979. a-, b-, and g-Ray Spectroscopy, 5th Edition. North-Holland Publishing Company, Amsterdam-New York-Oxford. St!ephan, O., Carrier, M., 1997. Preparation of thin a-particle sources using polypyrrole films functionalized by alkylammonium groups. Radiochim. Acta 76, 29–36. St!ephan, O., Carrier, M., Le Bail, M., Deronzier, A., Moulet, J.C., 1995a. Ion binding by poly[4-(pyrrol-1-ylmethyl)benzoic acid] thin films. J. Chem. Soc. Faraday Trans. 91, 1241–1246. St!ephan, O., Carrier, M., Page, J., Frontier, J.P., Trouslard, P., 1995b. Particle induced X-ray emission (PIXE) and radiochemistry study of poly(pyrrole) doped with toluenesulphonate or polysterenesulphonate anions. Synth. Met. 75, 181–185.
A novel method for large-area sources preparation for the calibration of b- and a-contamination monitors V. Tsoupko-Sitnikova,*, J.L. Picoloa, M. Carrierb, S. Peulonc, G. Moutarda a
Laboratoire National Henri Becquerel, BNM-LNHB, CEA/DAMRI Saclay, B.P. 52, 91191 Gif-sur-Yvette Cedex, France b ! Institut National des Sciences et Techniques Nucleaires, CEA/Saclay, 91191 Gif-sur-Yvette, France c ! Laboratoire de Chimie, Electrochimie Moleculaire et Chimie Analytique, CNRS/Universite! de Bretagne Occidentale, 6, avenue Le Gorgeu, BP 809, 29285 Brest Cedex, France Accepted 31 July 2001
Abstract A method is proposed for the preparation of large-area reference sources for the calibration of b- and acontamination monitors. It is based on the incorporation, by the ion-exchange mechanism, of the radionuclide in a thin film of a conducting polymer ion-exchanger preliminarily grown on a metal support. Conducting pyrrolebased polymer functionalized by carboxylic cation-exchange groups is used to prepare 60Co and 90Sr–90Y b-particle sources. Electrochemical polymerization of the corresponding monomer on different conducting supports is studied and a special electrochemical equipment developed permitting the preparation of large-area polymer films of controlled and reproducible thickness. The ion-exchanger obtained is characterized in terms of chemical affinity for cations Co2+ and Sr2+. Incorporation of the radionuclides in the large-area ion-exchanger films thus obtained is studied and optimized with respect to the uniform distribution of the radionuclide. The performance of the procedure is demonstrated using the example of circular sources 44 mm in diameter prepared on stainless steel supports. The sources obtained are characterized in terms of activity, b-particle flux, uniformity and source efficiency. r 2002 Published by Elsevier Science Ltd.
1. Introduction Radioactive surface contamination control is usually carried out by means of large-area gas-filled (proportional or Geiger–Muller) counters or scintillation detectors. The characteristics of the large-area reference source used for the calibration of these monitors are specified in the international standard ISO 8769 (ISO, 1988), for both a-emitters and b-emitters with maximum b-energy greater than 150 keV. In accordance with these requirements the active surface area of such sources must be at least 100 cm2
*Corresponding author. Fax: +33-169089529. E-mail address: [email protected] (V. TsoupkoSitnikov).
(rectangular shape of 100 150 mm2 is specially recommended). For Class I sources the particle flux (2000–10,000 s1) must be measured by the National Metrology Laboratory with standard uncertainty not exceeding 3%, and the radionuclide activity contained in the source is to be deduced by the manufacturer from the fabrication procedure in a way traceable to the national standards with an uncertainty lower than 10%. The non-uniformity of the particle flux over the surface of the source must be lower than 10%. For its calculation the active surface of the source is to be divided into equal individual elements (surface area p10 cm2), and the dispersion of the particle fluxes of all the elements is to be compared with the average value. The use of windowless 2p proportional gascirculation counters for the absolute measurement of the particle flux of Class I sources imposes a conducting source surface.
0969-8043/02/$ - see front matter r 2002 Published by Elsevier Science Ltd. PII: S 0 9 6 9 - 8 0 4 3 ( 0 1 ) 0 0 1 6 1 - 0
22
V. Tsoupko-Sitnikov et al. / Applied Radiation and Isotopes 56 (2002) 21–29
The conventional method of preparation of such reference sources is based on the incorporation of the radionuclide in the porous surface of an anodized aluminium foil. The radioactive material in such sources is usually concentrated in a relatively thick (3–10 mm) surface layer composed of aluminium and aluminium oxide, the activity depth distribution profile strongly depending on the minute details of the preparation procedure and often being difficult to control and to reproduce. The reproducibility of the depth distribution profile is of crucial importance to achieve the reproducible characteristics of the reference sources, especially for a- and low-energy b-particle emitters (see the detailed analysis made by Janen and Thieme, 2000). Its variation influences the source efficiency value and changes the energy spectrum of the emerging particles, the latter in its turn altering the response function of a secondary reference transfer instrument (e.g. a proportional counter with an entrance window). Thus two reference sources of the same radionuclide and emerging particles flux but of different spectrum shape will not create the same response when measured with such an instrument. The relatively ‘‘thick’’ reference sources described above do not well-represent a typical radioactive contamination where the radionuclides are deposited in a very thin and fragile layer on the surface. Finally, in many cases the aluminium supports of the sources are not representative enough of the heavier materials to be checked for contamination (stainless steel, copper, and zinc). Availability of reference sources fabricated on this kind of support is then desirable. The goal of this work was to develop a preparation method for large-area reference sources which would be more representative of the typical case of radioactive contamination (‘‘very thin sources’’) and of materials to be checked (ZE26). Special attention was paid to the reproducibility of the thickness of the active layer of the source. The results exposed in this article concern the radionuclides existing as bivalent cations in the solution, namely 60Co and 90Sr, as well as all other radioactive isotopes of these elements.
2. Conducting polymer ion exchanger as source substrate The source preparation procedure using this technique (Carrier et al., 1996) can generally be described by the following sequence: 1. A thin (100–350Zm) insoluble film of a conducting polymer ion exchanger is formed by electrochemical polymerization of an organic monomer on a largearea conducting support. The film thickness and thus
the future source efficiency value are electrochemically controlled. 2. A radionuclide is subsequently incorporated in the film from a relatively large volume (E1 ml cm2 of the surface) of aqueous solution of a certain composition by the ion exchange mechanism. 3. The aqueous solution is discarded (with a possible control of the residual activity), and the source, after appropriate rinsing and drying, is subjected to the particle flux absolute measurement. This technical approach has recently been developed by St!ephan and Carrier (1997) and St!ephan et al. (1995a, b) with a view to accelerate the small-size a-sources preparation in actinide environmental assay. To bind strontium and cobalt cations we used a polypyrrole functionalized by the carboxylic cationexchange group (hereinafter ‘‘poly-PYCOO’’, Fig. 1) which forms regular, well-adherent films when electrochemically prepared on small-area (1–2 cm2) electrodes of glass covered with semiconductor tin oxide (St!ephan et al., 1995a). The following aspects of this technique were studied in this work: (a) Possibility of the preparation of good-quality polyPYCOO films on metal surfaces (e.g. stainless steel) by anodic polymerization of PYCOO; feasibility of the preparation of a homogeneous coating on large metal surfaces. (b) Chemical affinity of the carboxylic ion-exchanger poly-PYCOO for the cations Sr2+ and Co2+ which must be strong enough to assure sufficient and reproducible transfer of the radionuclides from the solution to the film. (c) Feasibility of the homogeneous incorporation of the radionuclide from a solution into the large-area ion-exchange films. (d) Overall reproducibility of the procedure and of the characteristics of the sources obtained.
3. Experimental equipment The monomer PYCOO was synthesized according to the procedure described by Deronzier and Marques (1989). All other products were of analytical grade and used as received. Electrochemical experiments were performed using a Princeton Applied Research model 263A potentiostat/ galvanostat. Poly-PYCOO films for the affinity study were grown on commercially available (20 20 1) mm3 glass plates with vacuum-sputtered fluor-doped SnO2 coating on one side. Anodic polymerization on this kind of samples was carried out in a 20 mm glass spectrophotometric cell with a platinum plate as
23
V. Tsoupko-Sitnikov et al. / Applied Radiation and Isotopes 56 (2002) 21–29 COOH
COOH
PYCOO
poly-PYCOO H2C
2
-2 e N CH2
COOH
H2C -2X e
N N
+2H
H2C
COOH
XPYCOO
N
+ 2XH+
N H2 C
COOH
N x
H2C
COOH
Fig. 1. Scheme of the electrochemical polymerization of the 4-(pyrrol-1-yl)methyl)benzoic acid (hereinafter ‘‘PYCOO’’). The oxidized conductive form of the polymer produced (‘‘poly-PYCOO’’) contains approximately one positive charge for 4 pyrrole units.
counterelectrode under conditions described by St!ephan et al. (1995a). The polymer mass deposition rate on gold-coated electrodes as a function of total electrical charge was measured using a quartz microbalance coupled with the galvanostat. Film thickness on glass and polished stainless steel was measured profilometrically (Dektak 3030 T ST SLOAN Auto Surface Texture profiler). Radionuclide distribution experiments for the affinity study were performed in the system {glass/SnO2 20 20 mm2 sample carrying a poly-PYCOO film85 ml of liquid phase} in polyethylene vials shaken at a speed of 120 min1 during 24 h for equilibration. 60Co and 85Sr were used as tracers, available at Laboratoire National Henri Becquerel as 0.1 M HCl solutions with a certified specific activity of 800 kBq g1 at an uncertainty of 0.1%. Activities of the solid samples and the liquid phase residual activity were measured by HPGe gspectrometry. Residual activity of 90Sr and 90Y in the liquid phase was measured by liquid scintillation (LS) (GUARDIAN 1414, WALLAC). A conventional laboratory agitator was used for the horizontal orbital (18 mm) agitation of the cells for the radionuclide incorporation into the large-surface samples. For the agitation by vertical vibration (50 Hz) a home-made electromagnetic apparatus with variable vibration amplitude (0.1–1.5 mm) was applied. The uniformity of the large-area b-particle sources was calculated from their digital autoradiographies obtained with a commercial INSTANTIMAGER 20 24 cm2 microchannel array detector (Packard In-
strument Co.). For rectangular sources the calculation was performed using the standard rectangular analytical grill consisting of 8 8 mm2 individual elements, applied to the whole surface of the source. The circular sources (active surface 44 mm in diameter) were analysed using a specially composed grill (32 circular elements 7 mm in diameter, Fig. 5). The absolute measurement of the b-particle flux of the large-area sources was performed using a modified commercial large-area gas-flow proportional counter AB 710 (EURISYS MESURES) with an active area of 368 213 mm2 for particle detection. The aluminizedmylar window and the protective grid of the counter were removed and replaced by a stainless steel support which also holds the source in a reproducible position. The whole set-up was shielded by a 70 mm thick lead cladding on all sides. The counting gas was pure methane circulated at a flow rate of 1.70 l/h. The counter was operated at +3000–+4000 V in the b-plateau. Conventional electronics and software were used in the counting system; a deadtime of 15 ms was superimposed to assure the correction of the deadtime losses.
4. Results and discussion 4.1. PYCOO electrochemical polymerization For PYCOO polymerization on glass/SnO2 and goldcoated electrodes the electrochemical conditions proposed by Ste! phan et al. (1995a), were used: 0.005 M PYCOO solution in acetonitrile containing 0.01 M
24
V. Tsoupko-Sitnikov et al. / Applied Radiation and Isotopes 56 (2002) 21–29
HClO4 to keep the carboxylic group protonated and 0.05 M tetrabutylammonium perchlorate to assure sufficient conductivity of the solution. The polymerization was carried out at a constant current density of 0.2 mA cm2. Under these conditions the polymer mass deposited on a golden anode increases linearly with time up to the value of the charge transferred 150 mC cm2: m ¼ 0:91Q;
s ¼ 2:2%;
ð1Þ
where m is the mass of the polymer formed per unit of the anode area, in mg cm2, and Q is the electrical charge transferred per unit of the anode area, in mC cm2. The uncertainty stems from linear fitting of deposited massFelectrical charge experimental data. The average thickness of the polymer film formed on the glass/SnO2 electrodes also grows linearly with the electrical charge transferred: d ¼ 5:02 þ 6:91Q;
s ¼ 20%;
ð2Þ
d being expressed in Zm. The considerable uncertainty of this value is due to the insufficient uniformity of the film formed on the small-area anode totally immersed in the monomer solution. The approximate density of the polymer material rE1:32 g cm3 can be deduced from the two parameters above. The electropolymerization yield estimated from Eq. (1) is about 85% taking into account the presence of one ClO 4 ion per four monomer units in the conducting oxidized form of the polymer. The ion-exchange capacity of the polymer was determined from the isotherm of Sr2+ cations adsorption using 85Sr as tracer (Fig. 3). Its saturation corresponds to the specific quantity of the ion exchange sites z ¼ 4:3109 mol mC1
ð3Þ
or, combining with (1), 4.7 mol kg1. This value is in acceptable accord with the one obtained by saturation of the polymer with trisphenanthrolineiron: 3.8 109 mol mC1 (Ste! phan et al., 1995a). The electrochemical polymerization yield deduced from (3) is of approximately 90%. Under the conditions mentioned above, the polymerization of PYCOO on a polished stainless steel anode results in good-quality adherent polymer films very similar to those obtained on the glass/SnO2 anodes. Profilometrical measurements show that within the uncertainty of 20% the thickness of the films obtained on stainless steel is compatible with one of the films grown on glass/SnO2 plates. The considerable uncertainty is due to the B100 Zm roughness of the mirrorpolished steel surface. Hereinafter we assume the validity of the results (1)–(3) for the poly PYCOO films grown on stainless steel.
4.2. Preparation of uniform large-surface poly-PYCOO coatings The difficulties associated with the preparation of uniform electrochemical coatings on large-area conducting substrates are analysed by Graham (1971), Prentice and Tobias (1982), Shih and Pickering (1987) and Gerisher and Tobias (1994). On a metal plate completely immersed in an electrochemical bath the thickness of the coating obtained usually varies parabolically, increasing from the centre to the perimeter of the plate. This effect was also observed in PYCOO anodic polymerization on stainless steel plates. Thus on a 50 80 mm2 rectangular plate the thickness variation from the centre to the edge can reach 300%. Very uniform films on extended plane metal surfaces of various configurations can be obtained using a specially designed electrochemical cell (Fig. 2). The surface to be coated occupies the entire cross-section of the cell, being limited by a normally positioned insulator wall (Teflon). The edges of the stainless-steel plate are not exposed. The counterelectrode (cathode) is parallel to the treated surface and equals it in size. This configuration assures the homogeneous current distribution over the whole metal surface exposed to the electrolyte in the cell. The state of the stainless steel surface is of great importance for the uniformity of the coatings obtained. The flatness of the plate must be better than 0.1 mm. Very good results were obtained with mechanically mirror-polished and sand-blasted stainless steel supports. The following treatment of the plates was applied before the polymer deposition in the cell: (a) Washing in an ultrasonic bath in a 10% solution of an alkaline detergent at 601C, 120 min; (b) Washing in an ultrasonic bath in a 2 M NaOH solution at 601C, 60 min; (c) Rinsing with hot distilled water; rinsing with acetone and drying. All the above mentioned conditions fulfilled, coatings with non-uniformity lower than 10% can be obtained on large-area stainless steel plates (up to 150 cm2). The best criterion for a rapid and reliable uniformity check is the colouration of the coating due to light interference.
4.3. Poly-PYCOO affinity for cations Co2+ and Sr2+ Poly-PYCOO shows cation-exchange properties in alkaline solutions where the deprotonization of the carboxylic group takes place. Although the affinity of carboxylic ion exchangers for bicharged cations is relatively high, the background concentration of alkaline cations (Na+, NH+ 4 y) should be minimized.
25
V. Tsoupko-Sitnikov et al. / Applied Radiation and Isotopes 56 (2002) 21–29 counterelectrode
protective lid
teflon cage
source support
spacer
set screw
contact screw
O-ring
base of the cell
Fig. 2. Scheme of the cell for electrochemical polymerization.
Dilute ammonia solutions with pH varying from 8 to 10.5 present a good compromise in this respect. Table 1 shows the dependence of no carrier added 60 Co fixation on 50 mC cm2 poly-PYCOO films on ammonia concentration in aqueous solution. The best and the most reproducible yields are obtained at 0.001 M ammonia concentration. The radionuclide transfer in the film is sufficiently complete for practical application of the films. The affinity of poly-PYCOO for the cations of strontium is considerably higher and does not depend so drastically on ammonia concentration. A series of experiments performed showed that in the same configuration (ammonia concentration 0.01–0.02 M) the average strontium sorption yield amounts to 99.6% with variations of the order of 0.2%. In order to better quantify the process the respective adsorption isotherm was studied by increasing concentration of strontium carrier in the solution up to the complete saturation of the ion exchanger (Fig. 3). The distribution coefficient value deduced from the initial linear part of the isotherm is 9.0 106 l kg1. The demonstrated very high affinity of the ionexchanger for strontium will allow the quasi-quantitative transfer of 90Sr to the film without the necessity of a control measurement. 4.4. Radionuclide incorporation in large-area poly-PYCOO films The construction of a special extraction cell designed for this operation is depicted in Fig. 4. Due to the watertight junction between the Teflon cage and the stainless steel plate, only the ion-exchanger film is exposed to the radionuclide solution. The corked opening serves to fill and evacuate the cell and to add the radionuclide
Table 1 Adsorption of 60Co on poly-PYCOO films at different ammonia concentrations. System: {films 50 mC cm2 on 4 cm2 glass/SnO2 platesF5 ml of aqueous phase}. The uncertainties stem from the activities measurement by g-spectroscopy and do not correspond to the reproducibility of the ion-exchange equilibria Ammonia concentration (M)
Sorption yield
Uncertainty (%)
0
0.63 0.58 0.97 0.95 0.95 0.92 0.88 0.85 0.74
0.7 0.5 0.5 0.4 0.5 0.4 0.5 0.5 0.5
0.001
0.01 0.0167
solution from a weighted pycnometer. To assure the reliable contact of the liquid phase with the entire surface of the plate and good agitation conditions, the level of the liquid in the cell must be at least 5 mm high. The radionuclide surface distribution on the sources thus obtained strongly depends on the agitation mode applied, since the adsorption is controlled by the diffusion. The ‘‘static’’ incorporation (no agitation) gives very good results in this respect but does not assure the equilibrium incorporation yields in a reasonably short time (o24 h). High-speed horizontal orbital agitation (400 min1) assures the equilibrium incorporation yields but often
26
V. Tsoupko-Sitnikov et al. / Applied Radiation and Isotopes 56 (2002) 21–29
1E-7 4.0E-8 3.6E-8
Equilibrium Sr in the solid phase, mol
Sr retained in the ion-exchanger film, mol
ion-exchanger saturation
1E-8
3.2E-8 2.8E-8 2.4E-8 2.0E-8 1.6E-8 Fit Results:
1.2E-8
Equation: Y = 327.853 * X + 1.94013E-009 Number of data points used = 16
8.0E-9 4.0E-9
1E-11 2E-11 3E-11 4E-11 5E-11 6E-11 7E-11 8E-11 9E-11 1E-10
Equilibrium Sr in the aqueous phase, mol
1E-12
1E-11
1E-10
1E-9
1E-8
1E-7
1E-6
1E-5
Sr equilibrium quantity in the solution, mol Fig. 3. Isotherm of Sr2+ cations adsorption on poly-PYCOO films and linear approximation of its initial part (superimposed). System: glass/SnO2 20 20 mm2 sample carrying 50 mC cm2 poly-PYCOO film85 ml of 0.017 M ammonia solution.
feed hole teflon cage
radionuclide solution
source support + polymer film
spacer
set screw
contact screw
O-ring
base of the cell
Fig. 4. Scheme of the radionuclide incorporation cell.
brings reproducible non-uniform distributions due to static distribution of turbulences in the cell (Fig. 5), especially for the circular cell geometry. The inhomogeneity of the rectangular sources thus obtained is of the order of 11–13%, whereas that of circular ones 18–23%.
Sufficiently homogeneous radionuclide surface distributions were obtained independent of the source configuration using the agitation by vertical vibration of the cell. At 50 Hz and 1.0–1.5 mm vibration amplitude sources with a non-uniformity lower than 6% are reproducibly obtained (Fig. 5).
V. Tsoupko-Sitnikov et al. / Applied Radiation and Isotopes 56 (2002) 21–29
27
Fig. 5. Influence of the cell agitation mode on the radionuclide surface distribution for circular sources. (a) Horizontal orbital agitation with 18 mm amplitude, source uniformity 22%. (b) Vertical vibration 50 Hz/1.0 mm amplitude, source uniformity 4%. (c) Grid for source uniformity calculation.
4.5. Sources preparation and characterization The numerical results of the preparation of the largearea sources and their characterization are demonstrated using circular sources with an active deposit diameter of 44 mm, prepared on stainless steel supports of 50 mm diameter and 3 mm thick. Two series of four sources of 60Co (two on mirrorpolished and two on sand-blasted supports) and of three sources of 90Sr–90Y on sand-blasted supports were prepared. The thickness of the polymer film electrochemically grown on the supports was 30 mC cm2 (about 200 Zm). The liquid phase volume in the radionuclide incorporation cell was 12 ml (0.79 ml cm2 of the source surface). Ammonia concentration was 0.001 M for the preparation of the 60Co sources and 0.017 MFfor the 90Sr–90Y sources. The radionuclide solution (1–2 drops from a weighted pycnometer) was added directly into the incorporation cell just before starting the agitation. The cell was agitated during 15 h at 50 Hz with a 1 mm vertical amplitude. The liquid phase was then removed from the cell and the latter rinsed 2 times with 3 ml of acetonitrile. The cell was then disassembled and the source dried for 5 min at room temperature. Residual activity in the liquid phase and in the rinsing acetonitrile was finally measured. The ‘‘lost activity’’ value (residual activity on the Teflon cage of the cell) was estimated by smear tests. The 90Sr–90Y sources were stored during 30 days to assure the equilibrium between the two radionuclides. The results of the sources characterization are presented in Table 2. In the case of 90Sr–90Y sources preparation the LS control confirmed the low absolute value of the residual activity of the radionuclide not incorporated in the support (liquid phase+rinsing solvent+wall adsorption) and its sufficient reproducibility (0.3–0.6% of the activity initially introduced). The constant correction of 0.5% is considered to be adequate
for the correct source activity calculation and the routine LS control is deemed to be unnecessary in the case of 90Sr–90Y. The residual activity of 60Co is significantly higher and less reproducible (1.6–10.5%) due to the lower distribution coefficient of cobalt in the system polyPYCOOFwater solution and the significant sensitivity of the ion-exchange equilibrium to the ammonia concentration. Consequently, a systematic g-spectrometric control of the residual activity of 60Co in the liquid phase is recommended for the precise source activity calculation. The source efficiency values are generally high and reproducible for the same kind of source (Table 2). For 60 Co the efficiency of the sources prepared on mirrorpolished stainless steel supports is several per cent higher than on sand-blasted supports (74% and 70%, respectively). This difference is naturally explained by the increase of the absorption of the low-energy b-particles in the source and by a certain decrease of backscattering on sand-blasted supports in comparison with mirrorpolished ones. The absolute source efficiency values obtained (Table 2) are considerably higher than those of the commercially available large-area b-sources prepared by the anodized aluminium foil technique (AMERSHAM, 1997): B48% for 60Co and B63% for 90Sr–90Y. The values obtained are, nevertheless, compatible with the efficiency estimates of the ‘‘ideally thin’’ large-area sources fabricated on aluminium supports (Berger, 1998): 70% for 60Co and 69.6% for 90Sr–90Y. Even higher efficiency values in the case of the sources on stainless steel supports are due to the superposition of two factors: the increase of the number of the particles backscattered on stainless steel in comparison with aluminium (approximately +50%) (Siegbahn, 1979) and absorption in the 200 Zm thick source active layer composed of an organic matter, especially in the case of 60Co.
28
3.5 3.6 3.5 1.8 1.8 1.8 0.741 0.742 0.741 1.5 1.5 1.5 7567 5170 10655 1.0 1.0 1.0 5105 3485 7185 0.6 0.8 0.7 5.3 2.3 2.7 0.5 0.6 0.6 a
MP: mirror-polished; SB: sand-blasted stainless steel support.
5.9 5.3 5 6 6 6 13 13 13 1.0 1.0 1.0 90
Sr–90Y
51SB 52SB 53SB
5129 3505 7206
6 6 6 6 13 13 13 13 0.17 0.16 0.17 0.16 60
Co
54MP 55MP 57SB 58SB
6450 8155 6737 7457
Unc. (%) Activity ‘‘lost’’, (Bq)
252 116 343 763
2 3 1.5 1.3
1.5 2 F 3.9
10 15 F 10
6183 8024 6381 6677
0.2 0.2 0.2 0.25
4590 5935 4475 4710
1.5 1.5 1.5 1.5
0.742 0.740 0.701 0.705
1.5 1.5 1.5 1.5
5.7 4.9 4.2 4.4
5. Conclusion
Unc. (%) Activity introduced, (Bq) N1 and type of supporta Nuclide
Table 2 Preparation and characterization of +44 mm circular sources
Residual activity in liq. Phase (Bq)
Unc. (%)
Activity in rinsing solvent (Bq)
Unc. (%)
Source activity (Bq)
Unc. (%)
b flux, (s1)
Unc. (%)
Source b yield
Unc. (%)
Source uniformity (%)
V. Tsoupko-Sitnikov et al. / Applied Radiation and Isotopes 56 (2002) 21–29
The large-area b-sources fabricated using the technology described above meet the requirements imposed by the respective international standard (ISO, 1988) in terms of uncertainties of stated source activity and bparticle flux and of source uniformity. Thus the source activity can be deduced from the parameters of the fabrication procedure with an uncertainty not exceeding 1%, and the non-uniformity of the sources obtained is typically lower than 5% for the strict procedure of the uniformity evaluation applied. The developed procedure permits the source preparation on stainless steel, nickel, copper and some other metal supports. The sources thus obtained are characterized by a low intrinsic absorbance due to their particularly thin active layer (200–350 Zm of an organic matter). The reproducibility of the thickness controlled at a preliminary step of the source preparation assures the reproducibility of their efficiency value. At present the technical approach has been validated for the case of the radionuclides 60Co and 90Sr–90Y. The source efficiency values obtained for these radionuclides are considerably higher than those for the commercially available sources of the same radionuclides: 70–74% for 60 Co depending on the type of the metal support quality and 74% for 90Sr–90Y for the sand-blasted stainless steel supports. The described method, with certain modifications concerning the nature of ion-exchange groups present in the polymer, can be extended on other b- and a-emitting nuclides of metallic as well as non-metallic elements. This work was financially supported by Laboratoire Etalons d’Activit!e LEA/CERCA, Site du Tricastin, BP 75-26701 Pierrelatte Cedex, France. The authors express their gratitude to Philippe Cassette (LNHB) for useful discussions and consultations on LS measurements. References AMERSHAM, 1997. In: Calibration Standards and Instruments for Measuring Radioactivity. AMERSHAM Isotrak catalogue, 1st Edition. Amersham International, Buckinghamshire HP7 9NA. Berger, J., 1998. A method for determining the 2p-counting efficiency of beta-particle sources. Nucl. Instrum. Methods. B134, 276–286. Carrier, M., Burger, P., De Bruyne, T., 1996. Preparation de sources radioactives, notamment en vue d’analyse d’actinides en solution. Brevet franc-ais FR9403186. Deronzier, A., Marques, M.J., 1989. Electrodes modified by a Ni-dibenzotetraaza[14] annulene complex via reductive electropolymerization of an a-dibromobenzyl derivative or oxidative electropolymerization of a dipyrrole derivative. J. Electroanal. Chem. 265, 341–353.
V. Tsoupko-Sitnikov et al. / Applied Radiation and Isotopes 56 (2002) 21–29 Gerisher, H., Tobias, Ch.W., 1994. Advances in electrochemical science and engineering. In: Gerisher, H., Tobias, Ch.W. (Eds.), Current Distribution and Shape Change in Electrodeposition of Thin Films for Microelectronic Fabrication, Vol. 3. VCH Verlagsgesellschaft mbH, FRG. Graham, A.K., 1971. In: Kenneth Graham, A. (Ed.), Electroplating Engineering Handbook, 3rd Edition. Van Nostrand Reinhold Company, New York. ISO, 1988. Standard ISO 8769: Reference sources for the calibration of surface contamination monitorsFBeta emitters (maximum beta energy greater than 0.15 MeV) and alpha-emitters. International Standard Organization, Geneva. Janen, H., Thieme, K., 2000. On the determination of activity profiles in wide-area reference sources. Appl. Radiat. Isot. 52, 533–537. Prentice, G.A., Tobias, C.W., 1982. Simulation of changing electrode profiles. J. Electrochem. Soc. 129, 78–85.
29
Shih, H., Pickering, H.W., 1987. Three-dimensional modeling of the potential and current distribution in an electrolytic cell. J. Electrochem. Soc. 134, 551–558. Siegbahn, K., 1979. a-, b-, and g-Ray Spectroscopy, 5th Edition. North-Holland Publishing Company, Amsterdam-New York-Oxford. St!ephan, O., Carrier, M., 1997. Preparation of thin a-particle sources using polypyrrole films functionalized by alkylammonium groups. Radiochim. Acta 76, 29–36. St!ephan, O., Carrier, M., Le Bail, M., Deronzier, A., Moulet, J.C., 1995a. Ion binding by poly[4-(pyrrol-1-ylmethyl)benzoic acid] thin films. J. Chem. Soc. Faraday Trans. 91, 1241–1246. St!ephan, O., Carrier, M., Page, J., Frontier, J.P., Trouslard, P., 1995b. Particle induced X-ray emission (PIXE) and radiochemistry study of poly(pyrrole) doped with toluenesulphonate or polysterenesulphonate anions. Synth. Met. 75, 181–185.