Journal of Environmental Radioactivity 62 (2002) 1–15
Mechanisms of desorption of 134Cs and 85Sr aerosols deposited on urban surfaces J. Reala,*, F. Persinb, C. Camarasa-Clareta a
Department of Environmental Protection, Laboratory of Experimental Radioecology, Institute of Protection and Nuclear Safety, CE/Cadarache Building 186, BP 1, 13108 Saint-Paul-Lez-Durance Cedex, France b " Bataillon, 34095 Montpellier Cedex 5, University Montpellier 2, IEM -UMR 5635, CC047, Place Eugene France Received 17 March 2001; received in revised form 19 July 2001; accepted 26 July 2001
Abstract The radioactive isotopes of cesium and strontium may be deposited on urban surfaces in the case of an accidental atmospheric discharge from a nuclear facility and thus imply a health hazard. In order to handle the decontamination of these surfaces, we have carried out experiments under controlled conditions on tiles and concrete and we have studied the physical and chemical mechanisms at the solid–liquid interface. The deposition of radionuclides was carried out in the form of aerosols indicating an accidental source term. Their desorption by rainwater is low in all cases, of the order of 5–6% for cesium for any material and 29 and 12% for strontium on tile and concrete, respectively. The low desorption values of cesium may be explained by the strong bonding that occurs with the silicates constituting the tile due to virtually irreversible processes of exchange of ions and by the formation of insoluble complexes with the C–S–H gel of concrete. The strontium-tile bonds are weaker, while strontium precipitates with the carbonates of concrete in the form of SrCO3. In view of these characteristics, washing solutions with high concentrations of chloride and oxalate of ammonium chosen for their ion-exchanging and sequestering properties were tested on these surfaces. The desorption of cesium improved strongly since it reached 70% on tile and 90% on concrete after 24 h of contact, which is consistent with our knowledge of the bonds between this element and the surfaces. Strontium, given the greater complexity of physical and chemical forms that it may take is less well desorbed. The ammonium chloride improves the desorption (50% and 40%, for tile and concrete, respectively) but the oxalate, while it does not affect desorption on the tiles, decreases that on the concrete since by strongly
*Corresponding author. Tel.: +33-4-42-25-75-43; fax: +33-4-42-25-64-44. E-mail address:
[email protected] (J. Real). 0265-931X/02/$ - see front matter r 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 2 6 5 - 9 3 1 X ( 0 1 ) 0 0 1 3 6 - 9
2
J. Real et al. / J. Environ. Radioactivity 62 (2002) 1–15
etching the concrete, it causes the release of carbonate ions that precipitate with strontium. r 2002 Elsevier Science Ltd. All rights reserved. Keywords: Cesium; Strontium; Aerosols; Desorption; Urban surfaces
1. Introduction The presence of radionuclides on urban surfaces implies a two-fold hazard: immediate for the residential populations in the form of external irradiation, and a longer term danger for man and the environment. This contamination may be transferred by run-off and by infiltration into other aquiferous or agricultural areas; it may enter into the food chain. Following the Chernobyl accident, numerous studies have focused on the decontamination of the urban surfaces, whether natural decontamination by urban rainfall (Nicholson & Hedgecock, 1991; Roed & Sandalls, 1989; Pioch & MadozEscande, 1995), forced decontamination in particular by the application of ammoniac solutions (De Witt, Goldammer, Brenk, Hille, & Jacobs, 1990; Sandalls, Stewart & Wilkins, 1986), or by hosing (Roed & Andersson, 1996; Gjrup, Hedeman Jensen, Lauridsen, Roed, & Warming, 1986; Roed & Jacob, 1990). Andersson and Roed (1999) carried out a survey of the simple methods of decontamination of a wide range of urban sites (streets, trees, roofs, walls, roads, lawns, and cultivated areas). He presented the decontamination rates obtained, the initial contamination being achieved by the deposit of 137Cs, either wet or dry. He concluded that the efficacity of decontamination techniques is much greater when the application of the procedure is carried out as soon as possible after contamination and the deposit occurs by a wet pathway. He noted an exception, however, for the decontamination of roofs by hosing which would be more effective if contamination had occurred by dry pathway. Finally, this author pointed out that if the application of the procedure is not undertaken immediately on tile roofs, the penetration of the contaminants into this type of material implies the use of high-pressure jets to achieve the same level of decontamination. The aim of all these studies on the decontamination of cesium, was to provide assistance to the authorities for decision-making in the event of a nuclear accident. Nevertheless, it would appear necessary in order to decontaminate in the best conditions, to have prior knowledge of the characteristics of the different phases between which the radionuclide is distributed, solid, liquid or gaseous, and their patterns of change over time. Knowledge of these parameters and of the physical and chemical mechanisms involved may indeed provide a better basis for choosing the most effective countermeasures for decontamination. During the application of washing techniques, the phenomena involved are numerous and complex. The washing solution may on the one hand penetrate into the material or flow over it, and on the other hand dissolve the solid phases, aerosols and materials.
3
J. Real et al. / J. Environ. Radioactivity 62 (2002) 1–15
If washing can a priori diminish the contamination of surfaces by carrying off dissolved fraction from aerosols, the penetration of the liquid into more or less porous material is likely to delay or inhibit decontamination. As concerns the chemical species generated during dissolution, they may as stated above be carried off by the washing water or participate in secondary reactions and remain on the material by sorption (Skoog, West & Holler, 1996), exchange of ions (Fujikawa & Fukui, 1997) or co-precipitation (Shrivastava, Verma, & Wattal, 1995) with other elements that constitute the matrix. In the latter case, decontamination may become difficult. The rates of decontamination by washing depend on a wide range of parameters and mechanisms that occur at the liquid–solid interface at the surface of or within the contaminated material. The present work mainly concerns the study of the relatively short-lived radionuclides 134Cs (T1=2 ¼ 2:06 yr) and 85Sr (T1=2 ¼ 65:2 d) on the basis of experiments under controlled conditions. Our choice was made on short life radionuclides for facilities of management of waste. These radionuclides are initially deposited in the form of aerosols by dry pathway on used urban type materials (clay tiles and concrete slabs). The aerosols are representative of those that would be released in the case of an accident in a pressurised water reactor with core fusion (29501C). The desorption of these two radionuclides by rainwater and by washing solutions was studied, and the phenomena that limit it were examined. 2. Materials and methods 2.1. Urban materials Used clay tiles and concrete slab samples were collected in an urban environment. The tiles are mainly made of calcium aluminosilicates (diopside, anorthite), silicates (wollastonite, quartz) with the addition of urban pollution in the form of a surface deposit of calcium sulphate and of organic matter but without chlorides. The concrete is composed of chalk, carbonates and to a lesser extent silicates, aluminates and magnesia. Mercury porosimetric analyses of the two materials (Table 1) show that the total porosity of the tiles is much higher than that of concrete (36.6% for used tile and 14.1% for concrete) in addition to a higher median diameter (1 and 0.16 mm, respectively).
Table 1 Results of analysis by mercury porosimetry of urban surfaces
Used tile New tile Concrete surface Concrete core
Porosity total (%)
Diameter median (mm)
Distribution of pores (mm)
Specific area (m2 g1)
36.56 29.36 14.06 15.62
1.05 0.303 0.164 0.079
0.2–1.4 0.06–0.5 0.01–2 0.01–1
3.0 5.8 6.9 13.4
4
J. Real et al. / J. Environ. Radioactivity 62 (2002) 1–15
Fig. 1. Photographs of surfaces of concrete and tile taken with a scan electron microscope.
The photos in Fig. 1, taken by using a scanning electron microscope, show the difference in the surface structure of the two materials. 2.2. The aerosols An experimental laboratory device known as POLYR can be used to reproduce at reduced scale the conditions that would occur within the containment shell of a pressurised water reactor of 900 MWe in the event of partial fusion of the core (Pioch, Cartier, Quinault, & Picat, 1992; Pioch, 1995; Maubert & Drouet, 1992). The aerosols are formed in this device from a mixture of stable elements representative of fission products (I, Cs, Sr, Te, Ru, Ce), reactor vessel structure materials and control rods (Fe, Zr, Cr, Ni, In, Sn, Cd) in proportions which take into account their relative abundance in the corium (Table 2). The nuclear fuel is not present in this mixture because of constraints of measurement. The mixture was placed in a graphite crucible and brought by electrical induction to a temperature of 29001C, during which phase, the elements are transformed into complex aerosols. The particle size ranged from 5 to 0.2 mm. The d25 2d75 diameter range was between 0.4 and 1 mm with a mean mass diameter (d50 ) of about 0.5 mm. An analysis by ICP/MS has shown that the aerosols are 92–93% by mass made of elements representative of structural materials and 7–8% of fission products. Scanning electron microscope observations and elementary electronic probe microanalyses have shown that the crystallised particles are most often constituted of several metals either oxidised or not, mainly Ag and/or Fe and/or In and/or Cr (Fig. 2). While strontium is found in these metal associations, cesium has been
J. Real et al. / J. Environ. Radioactivity 62 (2002) 1–15
5
Table 2 Composition of the mixture used for the production and composition of aerosols Elements
Mixture mass (mg)
Aerosols composition (%)
I Cs Te Sr Ru Ce Zr Sn Ag In Cd Fe Cr Ni
15 15 10 31 19 50 1833 20 190 36 12 2060 620 410
0.6–2.4 1.3–2.0 B1.2 2.8–3.0 F B0.1 0.2–1.7 0.5–0.6 9.7–11.8 5.6–6.0 1.4–2.0 71.0–72.9 0.6–0.8 0.3–0.4
Fig. 2. Photographs of two particles of POLYR aerosols taken with a scan electron microscope.
detected in spheres of Cs–Cd–Cl. However, part of these elements could be found in amorphous form. The aerosols generated were deposited on the urban materials by dry deposition (134Cs+stable Cs 150–350 mg m2, 85Sr+stable Sr 200–500 mg m2). 2.3. Experimental procedures The experiments were carried out in bottles with orbital shaking and at a constant temperature of 151C on discs of urban material of a diameter of 9.4 cm. All the sides
6
J. Real et al. / J. Environ. Radioactivity 62 (2002) 1–15
except the upper surface were covered with a coating that is waterproof and inert with regard to the elements studied. Their geometrical area of exchange and their thickness are, respectively, 75 cm2 and 1.8 cm for the tiles and 69 cm2 and 2.1 cm for the concrete. To simulate the rainwater, we chose water from Mont Roucous, because its physical and chemical characteristics are similar to the mean for rainwater in France (Table 3). The washing solutions tested all contain high concentrations of ammonium salt (NH4Cl 0.1 and 1 M or (NH4)2C2O4 0.05 and 0.1 M). They have an acidic character through the characteristics of the cation and exhibit the properties of ion exchangers. The ratios volume/area in ml cm2 are 6.7 for tile and 7.2 for concrete and the total concentrations in cesium and strontium in contaminating solution are 2.5– 5 mg l1. During sorption tests, a disc of urban material was introduced into a bottle containing the contaminating solution at time t0 : For desorption, 500 ml of solution (water or washing solution) was placed in contact with a disc that has already been contaminated by dry deposit of aerosols. The kinetics of sorption and desorption are monitored over a period of 21 days by sampling over time of aliquots of 10 ml. All samples are filtered at 0.45 mm to determine the distribution between the dissolved (including submicron particles, colloidal phase and cationic species which cannot be distinguished here) and particular fraction and then analysed by gamma-Ge high resolution spectrometry (Canberra GR 2020, 25.1% relative efficacy, 1.9 keV resolution for 1.33 MeV, connected with a 4096 channel Canberra Series 35 Plus analyser and calibrated with an Amersham standard).
Table 3 Chemical composition of Mont Roucous water PH
5.87
Conductivity SiO2 HCO 3 Cl SO2 4 NO 2 NO 3 H2PO 4 Na+ K+ Ca2+ Mg2+ NH+ 4
34 mS cm1 6.6 mg l1 mg l1 6 3 1 0.007 2.3 0.02 4 0.4 1.7 0.85 0.01
Cs+ Sr2+
mg l1 0.1 o2
meq l1 0.1 0.085 0.02 0.0002 0.037 0.17 0.01 0.085 0.07 0.0006 meq l1 0.0008 o0.05
7
J. Real et al. / J. Environ. Radioactivity 62 (2002) 1–15
Table 4 Patterns of change in the composition of solutions in major ions during tests of dissolution in the presence of Mont Roucous water or washing solutions with used tiles and concretea T0 Solution
pH
Mont Roucous water NH4Cl 1 mol l1 NH4Cl 0.1 mol l1 (NH4)2C2O4 0.1 mol l1
5.8 4.6 5.1 6.5
21 d contact time Used tiles Mont Roucous water NH4Cl 1 mol l1 NH4Cl 0.1 mol l1 (NH4)2C2O4 0.1 mol l1 Concrete Mont Roucous water NH4Cl 1 mol l1 NH4Cl 0.1 mol l1 (NH4)2C2O4 0.1 mol l1
Na+ (mg l1)
HCO 3 (meq l1)
SO24 (mg l1)
K+ (mg l1)
Ca2+ (mg l1)
3.5 2.3 3.0 3.0
0.4 0.4 0.4 0.4
1.3 1.4 1.7 1.0
6.0 6.5 6.5 7.5
15.5 23.5 19 20
5.8 182 38 75
76 270 133 F
0.9 2.5 1.2 100
125 875 112.5 *
6.4 8.0 7.9 8.8
14.2 82 41 36
18 500 188 200
94 262 259 F
2.6 2.5 2.6 141
25 40 87.5
0.1 0.2
NH3 (meq l1)
1
*
SiO2 (mg l1) 7.25 6.65 7.45
1
F 1 1 34
2.5 15 7.5
F 15.5 2.2 43
11 7.5 14
a
Note: ‘F’ Not detectable by the method used; * Dosage used impossible in presence of C2O2 4 by the method used.
The pH and the composition of the solution in certain major ions were followed and in Table 4 we give values of these parameters to time 0 and the end of the experiment i.e. 21 days.
3. Results and discussion 3.1. Penetration of water into the material As we have pointed out earlier, the penetration of the solution into the material may be a limiting phenomenon with regard to desorption of contaminants since it may be followed by the virtually irreversible fixation of the dissolved elements. We therefore feel that it is of interest to assess its extent. To achieve this, we determined by weighing the mass of water that has penetrated into the discs of material. It was 24 g for tile in 50 min and 10 g for concrete i.e. a water flux rate of 29 ml h1 and 12 ml h1, respectively. At saturation, the volume of water gave access to open porosity of the material which corresponds to the volume of pores accessible to water. It is 35–36% for the used tiles and 14–15% for the concrete after 21 days of contact. It is appreciably constant beyond 15 days. Only 50 min were required in the case of the tile to reach 50% of this value, it took 3 h to reach this figure for the concrete. These results are consistent with those obtained by
8
J. Real et al. / J. Environ. Radioactivity 62 (2002) 1–15
Sandalls (1987) on concrete, but diverge for the tiles (open porosity 15% against 35% in our experiments). The explanation for this divergence could be that the tiles used (specific area, size and distribution pattern of the pores) had different characteristics. Our results confirm that the large penetration of water into the materials studied depends strongly on the characteristics of the material. The efficiency of the decontamination technique could therefore be partly limited by the penetration of the more or less contaminated liquid and by phenomena induced within the material by the erosive, corrosive or clogging properties of the solution. 3.2. Sorption of radionuclides on the materials The contaminant solution used for this experiment comes from a direct deposit of the aerosols in water. I was used more than 6 days after the deposit bus of the tests showed that the total dissolution of cesium and strontium was effective only after a deadline of 6 days. After aerosols dissolution the dissolved species may not only penetrate within the material, carried by the water flux, but also be retained by electrostatic or covalent interactions. We have therefore determined the development in terms of sorption of these elements in the presence of the materials studied (Fig. 3). Sorbed fraction 1.0
0.8
0.6
0.4
0.2
0.0 0
5
10
15
20
Time contact (d)
Used tile Cs - test 1 Cs - test 2
Sr Sr
test 1 test 2
Concrete Cs test 1 Cs test 2
Sr test 1 Sr test 2
Fig. 3. Development in time of sorption of dissolved cesium and strontium on urban surfaces in presence of water.
J. Real et al. / J. Environ. Radioactivity 62 (2002) 1–15
9
The affinity of cesium for used tile is greater than that of the strontium (respectively, 90% and 45% at equilibrium). A comparison with the sorption results for new tiles obtained by Pioch (1992) in similar conditions shows that the sorption of the radionuclides studied on new tile (80%) is less efficient than on used tile. In agreement with Sandalls et al. (1986), we think that the low porosity of the new tiles (Table 1) is to a large extent responsible for this difference even if we cannot exclude an unquantifiable influence of the organic matter and calcium sulphate present on the used tiles. In addition, the presence of calcium sulphate could explain the low sorption of the strontium by surface coprecipitation. On the concrete, the sorption of the two radionuclides follows the similar kinetics with a slightly higher affinity for cesium. At equilibrium, the sorption rate is of 95% and 82%, respectively, for cesium and strontium. 3.3. Desorption by water The curves for the kinetics of desorption presented in Fig. 4 show marked differences from one material to another; nevertheless, the dissolved fractions are always low. In addition, Table 4 shows that if the pH varies little during the experimentation, the major ions on the other hand increase in the solution and in particular, SO2 4 in the case of the tiles and Ca2+ in the case of the concrete. For the used tile, the desorption of cesium is very low since the dissolved fraction reached the maximum value of 5% in 14 days, then appeared to decrease very slightly. The desorption kinetics of strontium is of the same rate, but the value is higher, reaching 29%. A comparison between the new tile and used tile shows that desorption is higher when the material is a new tile, the porosity being lower. The rate of desorption kinetics of the radionuclides initially present in the concrete differs from that observed in the tile. The dissolved fraction shows a maximum during the first hour of contact, then sharply decreases before levelling off. The maximum value is 16% for cesium and 23% for strontium, reached between 20– 50 min, then decreasing markedly to 6% for the first and 12% for the second. In order to try to better understand the phenomena, we have compared the patterns of change of the dissolved fraction during the first hour of contact between a system [aerosols–cellulose filter–rainwater] and the system [aerosols–urban material– rainwater] (Fig. 5). In this case, the filter is considered as an inert material with regard to the elements studied. In a system without urban material, the dissolution of the cesium and the strontium present in the aerosols is not total after 50 min and the dissolved fraction curves of the two radionuclides are virtually superimposable. This identical behaviour of the two elements could be linked to their strong fixation on metal oxides and in particular that of iron which constitutes a major fraction of the matrix of aerosols (Camarasa-Claret, 1997). On the other hand, the curves that we have obtained in the presence of the material shows that the dissolution of these elements is quite distinct, cesium in particular being less soluble than in the absence of material. Sorption phenomena are
10
J. Real et al. / J. Environ. Radioactivity 62 (2002) 1–15
Cesium 0.18 0.16 0.14 0.12 0.1 0.08 0.06 0.04 0.02 0 0
5
10
15
20
25
contact time (days) used tile
concrete
new tile
Strontium 0.5 0.45 0.4 0.35 0.3 0.25 0.2 0.15 0.1 0.05 0 0
5
10
15
20
25
contact time (days) used tile
concrete
new tile
Fig. 4. Desorption by rainwater of cesium and strontium deposited on urban surfaces.
thus responsible for these differences of behaviour. These phenomena are related to the increase in K+, Na+ and especially in SO2 which support the sorption of 4 cesium on materials (Lieser 89, Fujikawa 97) by the formation of complexes III (RS–SO4–Cs) or IV (RS–OH2–SO4–Cs). The sorption of strontium is favoured during the increase in Ca2+ by coprecipitation of surface (Nilsson 85, Fujikawa 97). When desorption occurs on the tiles, after 50 min of contact a marked difference in the dissolved fraction is apparent between the strontium (0.33) and the cesium (0.06).
J. Real et al. / J. Environ. Radioactivity 62 (2002) 1–15
11
Fig. 5. Developments in time of dissolved fraction for cesium and strontium during tests of dissolution and desorption of aerosols in the presence of water.
Since the results were obtained during the same desorption test, the difference observed cannot be linked to the difference in the properties of each of these elements with regard to the material. Numerous works (Onishi, 1981; Cremers, Elsen, de Preter, and Maes, 1988; Lieser, 1989; Erten, 1990) have shown that the sorption of cesium on clayed minerals involves a process of exchange of ions that is more or less reversible with three types of fixation sites with different selectivity. The predominant mechanisms of sorption of strontium are the exchange of ions but also the physical adsorption on the external surface of the solid particles (Onishi, Serne, Arnold, Cowan & Thompson, 1981; Nilsson, Skytte Jensen, & Carlsen, 1985; Lieser & . urk, Steinkopff, 1989; Erten, Eyle & Gokt . 1990) in which the binding energies are lower than those involved during the sorption of cesium. In the presence of concrete, the dissolved fraction during the first minutes is higher than that observed during the dissolution of aerosols alone, then the desorption kinetics are very low. It might be supposed that the dissolution of the matrix of the concrete (calcite) releases carbonate or even hydrogenocarbonate ions that may initially be favourable to the dissolution of the iron oxides of aerosols releasing a greater fraction of radionuclides. The second stage seems to be linked to coprecipitations. According to Curti (1999) for cesium, coprecipitation would involve compounds other than calcite, perhaps compounds of iron or the gel C–S–H of cement (Shrivastava, 1995). For strontium, the presence of carbonate and the low solubility of strontium carbonate (co-logarithm of the product of solubility of this compound 9.27) may suggest that reprecipitation of strontium is possible when the concentration of dissolved carbonate is high enough.
12
J. Real et al. / J. Environ. Radioactivity 62 (2002) 1–15
3.4. Desorption by ammoniac washing solutions In the presence of ammoniac washing solutions and after 2 days of contact, the dissolved fraction in cesium is higher than 0.7 whatever the material and solution. Except in the case of oxalate, the dissolved fractions in strontium obtained after a same contact time are all higher than 0.3 (Fig. 6). In all cases, after 21 days of contact these desorption values are always higher than those obtained in the presence of rainwater. These results are consistent with those of our previous studies (Camarasa-Claret, Persin & Real, 1997) which showed that the aerosols present a strong iron oxide component with which cesium and especially strontium are bonded. Stumm (1992) reports that the rate of dissolution of this type of oxide increases with the concentration in proton or acid species (such as ammonium ion) but also that of ligands capable of chelating iron(III) such as oxalate and hydrogenocarbonate. The ammonium ion is one of the cations that has a good affinity with ion exchanging clayey materials and in particular, tiles (Sandalls, 1987). Since the
Fig. 6. Desorption of cesium and strontium on surfaces by different ammoniac washing solutions.
J. Real et al. / J. Environ. Radioactivity 62 (2002) 1–15
13
phenomenon of exchange of ions intensifies when the concentration of exchanger ion is high, the utilisation of increasingly concentrated ammoniac solutions should favour the desorption of cesium ions. This hypothesis is confirmed by our experiments. Nevertheless, one cannot exclude the possibility that the other cations resulting from the dissolution of the matrix of the material (Table 4) may play no role in desorption. The results of analyses of solutions obtained after 21 days of contact show a strong increase in the potassium, calcium and sulphate content, which increases as the ammonium concentration increases. Desorption is not, however, total even in 0.1 and 1 M ammoniac medium. One notes that the dissolved fraction in cesium increases with the content of ammonium in solution, when the duration of the experiment goes beyond 5 days. This trend would appear to indicate the intervention of another slower mechanism that does not involve the anion, perhaps an exchange with the deeper sites. With regard to desorption of strontium on tiles, the dissolved fractions are lower than those of the cesium under the same conditions, the lowest being that obtained in particular in ammonium oxalate medium. Since the affinity of strontium for these exchange materials was less marked than that of cesium, a phenomenon than the exchange of ions must be involved to explain this decrease. In view of the solubility products of the oxalate and the sulphate of strontium (Bernard & Busnot, 1984) and of the pH conditions, the co-precipitation of the strontium salts (sulphate, oxalate) may be envisaged. In the case of the system concrete–aerosol–ammoniac solution, the desorption of cesium always shows a maximum, whatever the anion. This maximum corresponds to a more or less total dissolution in ammonium oxalate medium. In ammonium chloride medium, this maximum is all the greater when the ammonium ion is high. This strong concentration in ammonium ion implies the acidification of the solution that will thus become more and more aggressive with regard to the concrete. The analyses that we have carried out have enabled us to confirm the dissolution of the matrix and thus our hypothesis (Table 4). The presence of oxalate favours the dissolution of cesium in this complex system, concrete–aerosol–ammoniac solution. The chelating effect of the oxalate added to the acid effect of ammonium ions initially favours the attack on the surface of the concrete releasing a large quantity of calcium and of carbonate. The simultaneous presence of calcium, oxalate and hydrogenocarbonate might induce the precipitation of CaCO3 or even of CaC2O4 and the formation of a layer on the surface of the material that would thus lose its sorbant properties with regard to cesium. With regard to dissolution of the strontium in the absence of material, our experiments have shown that it was higher in the presence of oxalate than of chloride because of dissolution of the iron oxides favoured by the complexing action of the oxalate. But in the presence of concrete, this phenomenon is reversed, and the fraction dissolved is then distinctly lower (0.05) than that observed with ammonium chloride (0.5 to 0.35 depending on the concentration level). One may thus conclude that the carbonate ions and hydrogenocarbonates released during the attack on the concrete by the oxalate are responsible for this greater decrease since they cause the precipitation of the strontium released in the form of strontium carbonate. On the
14
J. Real et al. / J. Environ. Radioactivity 62 (2002) 1–15
other hand, since the dissolution of the matrix is lower with ammonium chloride, the resulting concentration in hydrogenocarbonate is lower and therefore insufficient to cause a greater precipitation of strontium in the form of strontium carbonate.
4. Conclusion The radionuclides cesium and strontium deposited on the used tiles and concrete in the form of aerosols representative of an accidental source term fix strongly. Their sorption depends on the porosity and the composition of the material as well as on their own physical and chemical properties with regard to the components of the material. The desorption of radionuclides on these surfaces is very low in the presence of rainwater (about 5% for cesium and 29% for strontium on tiles and, respectively, 6% and 12% on concrete). The cesium is little desorbed whatever the material, either because it is fixed in a virtually irreversible fashion by the silicates constitutive of the tile, or because it forms insoluble complexes with the C-S-H gel of the concrete. The relative desorption of strontium on the tile is due to the weak bonding between this element and the surface of the material. In contrast, in the presence of concrete, it precipitates with the carbonates released in the form of SrCO3. Washing solutions with a high concentration in chloride and ammonium strongly improve the desorption of the cesium radionuclides whether on tile or on concrete (70% to 90%). On tile, it is the aggressive character and the exchanger properties of the ammonium ion that intervenes; in concrete, it is the chelating behaviour of the oxalate ion that by favouring the precipitation of CaCO3 limits the sorption of the element. The desorption of strontium is improved in presence of ammonium chloride but in a lower proportion than cesium (50% for tile and B40% for concrete), little influenced by oxalate as regards tiles (28%) but strongly diminished in the presence of concrete (4%). The results may be explained by the precipitation of strontium in the form of strontium carbonate with the carbonate ions released during the attack on the concrete by the oxalate. Overall, this study suggests that decontamination by washing might be envisaged with a concentrated ammonium solution; however, it should be possible to perform this with a minimum volume of solution.
References Andersson, K. G., & Roed, J. (1999). A Nordic preparedness guide for early clean-up in radioactively contaminated residential areas. Journal of Environmental Radioactivity, 46, 207–223. Bernard, M., & Busnot, F. (1984). Usuel de chimie g!en!erale et min!erale, DUNOD, Paris, p. 636. Camarasa-Claret, C. (1997). Etude de la mise en solution des radionucl!eides c!esium et strontium d!epos!es sur des mat!eriaux urbains. Applications a" la d!econtamination. Thesis, University of Montpellier II, Montpellier, France, 322p. Camarasa-Claret, C., Persin, F., & Real, J. (1997). D!econtamination de surfaces urbaines expos!ees a" une pollution radioactive accidentelle simul!ee. Elimination du c!esium et du strontium. Proceedings of Regional Congress of IRPA. Avignon, France, pp. 175–178.
J. Real et al. / J. Environ. Radioactivity 62 (2002) 1–15
15
Cremers, A., Elsen, A., de Preter, P., & Maes, A. (1988). Quantitative analysis of radiocaesium retention in soils. Nature, 335, 247–249. Curti, E. (1999). Coprecipitation of radionuclides with calcite: estimation of partition coefficients based on a review of laboratory investigations and geochemical data. Applied Geochemistry, 14, 433–445. De Witt, H., Goldammer, W., Brenk, H. D., Hille, R., & Jacobs, H. (1990). Decontamination strategies in urban areas after nuclear accident. Proceedings of the Seminar on Methods and codes for assessing the off-site consequences of nuclear accident, Athens, Greece, EUR 13013, 2, pp. 615–632. . urk, Erten, H. N., Eyle, M. C., & Gokt . H. (1990). Sorption of radionuclides on some clays and soils. Biomovs, 457–468. Fujikawa, Y., & Fukui, M. (1997). Radionuclide sorption to rocks and minerals: effects of pH and inorganic anions. Part 1 sorption of cesium, cobalt, strontium and manganese. Radiochimica Acta, 76, 153–162. Gjrup, H. L., Hedeman Jensen, P., Lauridsen, B., Roed, J., & Warming, L. (1986). Deposition, retention, and decontamination of radioactive material on urban surfaces. Proceedings of Workshop on Methods for assessing the off-site radiological consequences of nuclear accidents, Luxembourg, EUR 10397, pp. 463–491. Lieser, K. H., & Steinkopff, T. (1989). Chemistry of radioactive cesium in the hydrosphere and in the geosphere. Radiochimica Acta, 46, 39–47. Nicholson, K. W., & Hedgecock, J. B. (1991). Behaviour of radioactivity from Chernobyl. Weathering from buildings. Journal of Environmental Radioactivity, 14, 225–231. Maubert, H., & Drouet, P. (1992). Chemical speciation of radionuclides. In: G. Desmet, & J. Sinnaeve (Ed.), Evaluation of data on the transfer of radionuclides in the food chain, EUR 12550, pp. 3–48. Nilsson, K., Skytte Jensen, B., & Carlsen, L. (1985). The migration chemistry of strontium. European Applied Research Report-Nuclear Science Technology, 7(1), 149–200. Onishi, Y., Serne, R. J., Arnold, E. M., Cowan, C. E., & Thompson, F. L. (1981). Critical review: radionuclide transport, sediment transport, and water quality mathematical modeling; and radionuclide adsorption/desorption mechanisms. NUREG/CA–1322, PNL-2901, RE: 8.1-24 8.64-82 8.20123 8.278-88. Pioch, M., Cartier, Y., Quinault, J. M., & Picat, Ph. (1992). Solubilit!e des a!erosols e! mis en cas d’accident nucl!eaire - Cons!equences sur les contre-mesures. Proceedings of International Seminar on Intervention levels and countermeasures for nuclear accidents, Cadarache, France, EUR 14469, pp. 313–325. Pioch, M., & Madoz-Escande, C. (1995). Effect of rainwater on the remobilization and dissolution of 134Cs and 85Sr contained in aerosols similar to those discharged in a nuclear accident and deposited in a urban environment. Journal of Environmental Radioactivity, 26, 51–61. Roed, J., & Andersson, K. G. (1996). Clean up of urban areas in the CIS countries contaminated by Chernobyl fallout. Journal of Environmental Radioactivity, 33, 107–116. Roed, J., & Jacob, P. (1990). Deposition on urban surfaces and subsequent weathering. Proceedings of the Seminar on Methods and codes for assessing the off-site consequences of nuclear accident, Athens, Greece, EUR 13013, 1, pp. 335–356. Roed, J., & Sandalls, F. J. (1989) The concentration levels of Chernobyl fallout on different surfaces in G.avle. Proceedings of the XVth regional congress of IRPA, Visby, Sweden, pp. 367–371 (in Swede). Sandalls, F. J. (1987). Removal of radiocaesium from urban surfaces contaminated as the result of a nuclear accident. Harwell Laboratory, Oxfordshire OX11ORA, AERE R 12355. Sandalls, F. J., Stewart, S. P., & Wilkins, B. T. (1986). Natural and forced decontamination of urban surfaces contaminated with radiocaesium. Proceedings of Workshop on Methods for assessing the offsite radiological consequences of nuclear accidents, Luxembourg, EUR 10397, pp. 511–532. Stumm, W. (1992). Chemistry of the solid-water interface (p. 428.). New York: Wiley. Skoog, D. A., West, D. M., & Holler, F. J. (1996). Chimie analytique, De Boeck University, Paris. Shrivastava, O. P., Verma, P., & Wattal, P. K. (1995). Intrinsic sorption potential of aluminiumsubstituted calcium silicate hydroxy hydrate for 137Cs. Advanced Cement Based Materials, 2, pp. 80– 83.