TRANSFER AND MOBILITY OF 137CS AND 60CO IN OXISOL AND ALFISOL

Maria Angélica Wasserman*, Daniel Vidal Pérez**, Claudio Carvalho Conti*; Flavia Bartoly*, Aline Gonzalez Viana* and Júlio César Wasserman ***

* Instituto de Radioproteção e Dosimetria/CNEN Av. Salvador Allende s /n°. Recreio

CEP: 22780-160, Rio de Janeiro, RJ. Brasil E-mail: angelica@ird.gov.br

** Centro Nacional de Pesquisa de Solos/EMBRAPA

Rua Jardim Botânico, 1024. CEP: 22460-000, Rio de Janeiro, RJ. Brasil.

*** Departamento de Geoquímica/UFF

Outeiro de São João Batista s/n°, Centro, Niterói, RJ, 24.020-007, Brasil.

ABSTRACT

One manner to assess potential mobility of radionuclides in soils is the use of sequential extraction procedures. These procedures intend to provide the radionuclide partitioning in geochemical phases of soil affected by changes in physico-chemical conditions. In this study a new sequential chemical extraction protocol was choose to evaluate 60Co and 137Cs mobility under a large range of physico-chemical soil properties. The results of sequential procedure was further compared with soil to plant transfer factors (TF) data for maize and radish and with soil properties, showing rather consistent results. The 137Cs distribution in soil showed that Fe oxides are the main sink for this element and after 14 years after contamination the 137Cs was still available for plants. The 60Co distribution showed that Mn oxides are the main sink for this element in Alfisol and 5 years after contamination no 60Co was detected as bioavailable neither detectable in maize. The knowledge of the bio-geochemical behavior of radionuclides in soil system can be useful for risk assessment studies, to be applied in the case of nuclear accident or contamination scenarios.

Keywords: mobility, bioavailability, sequential chemical extraction, cesium, cobalt.

I. INTRODUCTION The study of 137Cs in agricultural areas is of large concern because, once in the soil, it can recycle within the biota, similarly to potassium and as long as its radioactive half-life permits. Interactions between radionuclides and soil depends on the chemical form of element and some soil properties as pH, mineralogical composition, organic matter content and nutrient status [1-6]. The knowledge of the mobility of radionuclides in soils is of special interest to assess potential hazards from the radionuclide inputs into food chains. The plant uptake of radionuclides will be dependent on interactions occurring in soils as well metabolic and physiological characteristics of the species [1].The differences found between types of soils and vegetation engender a variety of measured soil-to-plant transfer factors (TF) [7]. This figure is still complicated by the use of fertilizers and other agricultural practices that modifies plant uptake by direct changes on some physico-chemical properties of soils [1,2,8,9].

One manner to assess potential mobility of radionuclides in soils is the use of sequential extraction procedures. These procedures intend to provide the radionuclide partitioning in geochemical phases of soil affected by changes in the physico-chemical conditions [10-12]. The results of sequential procedure for 137Cs are going to be discussed considering soil properties and soil to plant transfer factors (TF) for maize and radish obtained in an Oxisol soil accidentally contaminated in Goiânia, and in an Alfisol and Oxisol, both artificially contaminated. The bio-geochemical behavior of 60Co in Alfisol soils is going to be discussed considering soil properties and soil to plant transfer factors (TF).

II. MATERIALS AND METHODS Lysimeters measuring 1m2 of area and with a depth of 1 m were installed in a restricted area of the Institute for Radioprotection and Dosimetry (CNEN/Brazil). Six

lysimeters were filled with the Oxisol soil [13], contaminated with 137Cs in 1992. One lysimeter was filled with the Alfisol soil [13], contaminated with 60Co and 137Cs in 1996. Each lysimeter received a liter of solution containing approximately 60µCi/L of 137Cs or 60Cs. The contamination was done directly in the lysimeters. The solution was sprayed in every 2 cm of soil, layer by layer, up to 40 cm. Another Oxisol soil was collected in the city of Goiânia, in a site where the radiological accident occurred in 1987 [14]. Three lysimeters were filled with Goiânia soil. Samples of soil were collected at lysimeters filled with Oxisol, Goiânia soil and Alfisol to perform sequential extraction analyses. Sequential extractions were done as described in Table 1. Details of protocol can be found in Wasserman et al. [10]. All the extracts were analyzed by gamma ray spectrometry using an Ge detector ORTEC with a relative efficiency of 30%, for the determination of 137Cs and 60Co. Individual counting errors for the used geometry (vial of 250 mL) were less than 3%. The Soils Office of the Brazilian Agricultural Research Corporation (EMBRAPA-Soils) performed soil analyses and clay mineralogy determinations in order to characterize samples according to their standard manual [15].

TABLE 1. Protocol of sequential extraction method considering geochemical meaning by phase.

Slightly acidic phase: CH3COOH + CH3COONa 1:1; pH 4.7; At room temperature.

geochemical meanings: elements readily bioavailable. Easily reducible phase:

NH2OH.HCl (0.1 M); pH 2; At room temperature.

geochemical meanings: elements mainly bound to Mn oxides

Oxidizable phase: H2O2 (30%) + CH3COONH4 (1M);

pH 2; At room temperature. geochemical meanings: elements bound to labile organic

matter Alkaline phase: NaOH (0.1 M) ;

pH 12; At room temperature. geochemical meanings: elements bound to Fe

compounds Resistant phase:

Aqua regia. Heat to 50°C/ 30mim.

geochemical meanings: elements not available for transfer processes

The transfer factor (TF) was determined for maize (Zea mays, L.) and radish (Raphanus sativus, L.) following the IUR protocols to determine transfer factors [16]: TF =Ap/As

Where, Ap = Activity in the edible part of the plant (Bq.kg-1 dry weight) and As = Activity in the soil (Bq.kg-1 dry weight). Edible parts of plants were collected and washed. Plants and soils were dried, ground and sieved through a 2mm screen before direct measurements of the of 137Cs and 60Co activity by gamma ray spectrometry using a Ge detector ORTEC with a relative efficiency of 30%. Individual counting errors for the used geometry (pot of 250 g) were less than 3%.

III. RESULTS AND DISCUSSIONS The main chemical and physical properties of studied soils are presented in Table 2. The Oxisol soil has sandy clay loam texture, low cation exchange capacity (CEC) and low content of nutrient elements. The mineralogical analyses indicate the presence of gibbsite and kaolinite as main constituents of the clay mineral fraction. Fertilizers were applied in this soil, in formulations and sometimes alone, depending on the crops cultivated. The Goiânia soil is also an Oxisol, but it is an urban soil with contributions from urban residues, like lime and cement (from building construction). No fertilization procedure was applied during these experiments. The mineralogical analyses of the clay fraction indicate the presence of gibbsite and kaolinite (Table 2). The 137Cs activity in this soil is lower than that measured in Oxisol. The Alfisol is an acid silty clay loam soil, with low nutrient content and low CEC. The mineralogical analyses indicate the presence of hematite, goethite and traces of vermiculite. The 137Cs activity in this soil is comparable to that of Oxisol. Geochemical Partitioning of 137Cs and 60Co. The geochemical distribution of the radionuclides furnishes a first approach to the investigation of their potential mobility in the studied soils. Figure 1 shows the geochemical partition of 137Cs. The distribution of 137Cs in Oxisol and in the Goiânia soil in all considered phases were very similar (Figure 1). No significant differences were observed in the soil properties considering texture, pH, organic matter content and secondary minerals (Table 2). The 137Cs transfer factor for maize and radish was comparable in Oxisol and Goiânia as showed in the Figure 2. These results agree with the similar results for 137Cs obtained in the slightly acidic phase to Oxisol (8% readily bioavailable) and Goiânia soil (9% readily bioavailable). In the Alfisol, very small activity of 137Cs was detected in the slightly acidic phase as well TF in this soil was generally lower than in the other 2 soil, despite of time after contamination be lower than the other two soils and level of contamination be comparable to the Oxisol (Figure 2). The conditions for extraction used in the slightly acidic phase seem to be very consistent with the transfer factor results, corroborating that this phase represents the bioavailable fraction.

TABLE 2. Pedological analyses of artificially contaminated soils and Goiânia soil.

Parameters Oxisol N=6

Goiânia soil N=3

Alfisol N=1

pH (water) 6.3 7.8 4.9 pH (KCl 1M) 6.7 8.0 4.2

Sand (%) 71 68 10 Clay (%) 18 16 49

CEC (cmolc kg-1) 3.2 5.2 8.4 C (dag kg-1) 1.3 1.14 1.24 N (dag kg-1) 0.12 0.12 0.13 P (mg kg-1) 5 53 8

K+ (cmol kg-1) 0.13 0.25 0.75 SiO2 (dag kg-1) 3.3 5.8 15.7 Al2O3 (dag kg-1) 10.5 10.9 10.2 Fe2O3 (dag kg-1) 3.4 5.4 13.3 Ca++ (cmolc kg-1) 1.6 4.2 not determined Mg++ (cmolc kg-1) 0.9 0.4 2 Al+++ (cmolc kg-1) 0.0 0.0 0.05

H+ (cmolc kg-1) 0.5 0.4 5.5 137Cs (Bq/Kg dry wet) ± 1σ 5969 ± 47 2397 ± 37 6227 ±32 60Co (Bq/Kg dry wet) ± 1σ - - 3087 ± 25 Time after contamination 9 years 14 years 5 years Main clay mineral type

(in order of predominance) gibbsite, kaolinite gibbsite,

kaolinite hematite, goethite,

vermiculite The results obtained in the oxidizable phase (organic bound) for Oxisol, Goiânia and Alfisol soil were associated with the similar organic C content in these soils (Table 2). The results of this operational method mobilize 137Cs associated with iron oxides in the alkaline phase. It can be observed that iron oxides have an important role as a sink for 137Cs in these soils. In the Alfisol that is an iron rich soil, the iron oxides can be responsible for the retention of more than 60% of 137Cs. Such behavior corroborates the findings of Sheppard et al. [17]. These authors, using the sequential extraction method of Tessier [18], obtained in four Canadian soil more than 40% of the total Cs associated with Fe and Mn oxides.

Figure 1. Geochemical partitioning of 137Cs in tropical soils.

Elements present in the residual phase of the classical sequential extraction of Tessier [18], are those

present in the structure of primary minerals, or to a lesser extension, those trapped in the structure of secondary minerals. Release of elements bound to the lattice of minerals is not a natural process observable on a human time scale [19] and refers to the background levels for the region. Elements originated from human activities are generally distributed among other physico-chemical phases, according to their chemical form and behavior. Since 137Cs has no natural origin, some radiocesium found in the residual phase of classical methods suggest that this radionuclide could be progressively fixed in some clay structures (mainly 2:1 clay types). In the presence of 2:1 clay mineral type, the fixation of 137Cs in the internal faces of these clays can occur after contamination (less than 3 years), reducing transfer to plants [12,20-22]. Wasserman et al. [23] using the classical method of Tessier, noticed that no significant radiocesium was observed in the residual phase of the Oxisol and the Goiânia soil, where kaolinite and gibbsite are predominant in the clay fraction and there’s no 2:1 clay mineral types in these soil to fixation processes. The Alfisol presents traces of vermiculite (2:1 clay mineral type) and it was contaminated with 137Cs in 1996, that represents time enough for Cs fixation. This can partly explain why, despite of the similar concentration of 137Cs in Alfisol compared with Oxisol (Table2), the concentration in the bioavailable phase in the Alfisol was lower (Figure1). The absence of 2:1 clay mineral in Oxisol and Goiânia soil explain why many years after contamination the soil to plant transfer factor for 137Cs remains the same in the Goiânia soil (Figure 3). This also partially explain why transfer factor is higher in some tropical soil than IUR values determined to temperate soils. In the resistant phase, the fluorhydric acid or other strong condition able to break down mineral structures was

137Cs

0%

20%

40%

60%

80%

100%

Oxisol Goiânia Alfisol

Resistant

Alkaline

Oxidizable

Easilyreducible

Slightlyacidic

not used, so it does not account for elements present in the mineral structures. The resistant phase correspond mainly to elements associated to compounds not destroyed in the previous phases and possibly not available for transfer processes in the water cycle. The results obtained using this sequential extraction protocol showed that Oxisol and Goiânia soil present an important percent of 137Cs (about 30%) resistant to be mobilized under normal and cultural environmental conditions. In the Alfisol the percent is lower (11%) than the Oxisol (Figure 1). It is possible that in the old contamination, the association of radionuclides with soil compounds be more refractory.

Figure 2. 137Cs Transfer factor for radish and maize obtained in Alfisol, Oxisol and Goiânia soil.

Figure 3. 137Cs Transfer factor for radish obtained in Goiânia soil in 1989 data from [24], 1996, 2000 and IUR

data [16].

Figure 4. 60Co distribution in the Alfisol.

The 60Co distribution in the Alfisol is presented in the Figure 4. In this Figure it is clear that 60Co is mainly associated with Mn oxides, followed by organic compounds and Iron oxides. An important thing to note with this result is that no 60Co was observed in the bioavailable phase. This result was very consistent with plant results, since no 60Co was detected in the maize plants (grain or leaves) and very few was transferred to radish root: 3% of soil total concentration. All these findings were similar to those reported by Colle et al. [25] reviewing the Co behavior in soils and plants, supporting that no specific behavior is expected for Co in tropical soil.

IV. CONCLUSIONS Changes in bioavailability of 137Cs were mainly due to differences in some soil properties. The results for sequential extraction method used in this study shows that 137Cs is mainly associated with iron oxides in Al rich soil or Fe rich soil. Basic conditions can be more efficient to mobilize 137Cs present in soil. These results showed that part of 137Cs can be strongly retained in the soil particles independent of the presence of 2:1 clay mineral. In the Alfisol, part of 137Cs can be chemically fixed in the internal faces of 2:1 clay mineral type. The 60Co distribution showed that Mn oxides are the main sinks for this element. Four years after the contamination no 60Co was detected as bioavailable in soil and in the same way it was not detectable in plants. A slightly reducible condition seems able to mobilize important fraction of 60Co in Alfisol. A specific behavior for Co in tropical soil is not expected, since these findings have been reported in the specialized literature for temperate climate. Data obtained with the new method are coherent with transfer factor results and soil properties and seems be adequate for radiological risk assessment purposes or to remediate or to manage radiological contaminated sites.

ACKNOWLEDGMENTS This work received a financial support from the International Atomic Energy Agency (contract BRA10456/RO), and received student’s fellowship from the Ministério de Ciência e Tecnologia.

REFERENCES [1] CAWSE, P.A. & TURNER, G.S. The Uptake of Radionuclides by Plants: A Review of Recent Literature. Environmental and Medical Science Division, 1982. [2] COUGHTREY, P.J. & THORNE, M.C. Radionuclide Distribution and Transport in Terrestrial and Aquatic Ecosystems. A.A. Balkema, Rotterdam, 1983.

0,01

0,10

1,00

10,00

Alfisol Oxisol Goiânia soil

TF

Radish Maize

0,01

0,10

1,00

10,00

TF

Goiânia1989*

Goiânia1996

Goiânia2000

IUR

0%

20%

40%

60%

80%

100%

Co-60

Alkaline

Oxidizable

Easily reducible

[3] DELMAS, J.; GRAUBY, A.; PAULIN, R.; RINAULDI, C.; NDOYE, R. & SEKGASSAMA, S. Etude Expérimentale du Transfert du Sr-90 et du Cs-137 d'un Sol de la Région du Cap Vert à l'Arachide. Radioprotection, vol. 22, p. 334-340. 1987. [4] PAPASTEFANOU, C., MANOLOPOULOU, M., STOULOS, S., IOANNIDOU, A. & GERASOPOULOS,

E. Soil -to-Plant Transfer of 137Cs, 40K and 7Be. Journal of Environmental Radioactivity, vol. 45, p. 59-65. 1999. [5] SKARLOU, V., PAPANICOLAOU, E. P. & NOBELI, C. Soil to Plant Transfer of Radioactive Cesium and its Relation to Soil and Plant Properties. Geoderma, vol. 72, p.53-63. 1996. [6] WASSERMAN, M.A. & BELÉM, L.J. Valores de Transferência do 137Cs de Latossolos para Plantas comestíveis. in VI Congresso Geral de Energia Nuclear. Rio de Janeiro, Brasil, 1996. [7] YASUDA, H. & UCHIDA, S. Statistical Analyses of Soil-to-Plant Transfer Factors: Strontium and Cesium. J. Nuclear Science and Technology, vol. 31, p. 1308-1313. 1994. [8] CLAUS, B., GRAHMANN, B., HORMANN, V., KEUNEKE, S., LEDER, M., MÜLLER, H., PETERS, E., RIEGER, E.M., SCHMITZ-FEUERHAKE, I. & WAGSCHAL, F. Sr-90 Transfer Factors for Rye in Podzolic Soils: Dependence on Soil Parameters. Radiat. Environ. Biophys., vol. 29, p. 241-245. 1990. [9] SHAW, G. & BELL, J.N.B. Competitive Effects of Potassium and Ammonium on Caesium Uptake Kinetics in Wheat. J. Environ. Radioactivity, vol. 13, p. 238-296. 1991. [10] WASSERMAN, M.A., PEREZ, D.V., BARTOLY, F. & POQUET, I. Assessment of the mobility and Bioavailability of 60Co and 137Cs in contaminated soils. in V Regional Congress on Radiation Protection and Safety 8 pages. Recife - Pernambuco, Brasil, 2001. [11] YOUSFI, I., BOLE, J. & GEISS, O. Chemical extractions applied to the determination of radium speciation in uranium mill tailing: study of different reagents. Journal of Radioanalytical and Nuclear Chemistry, vol. 240, p. 835-840. 1999. [12] RIISE, G., BJORNSTAD, H. E., LIEN, H. N., OUGHTON, D. H. & SALBU, B. A Study on Radionuclide Association with Soil Components Using a Sequential Extraction Procedure. J. Radioanal. Nucl. Chem., vol. 142, p. 531-538. 1990. [13] USDA. Keys to Soil Taxonomy. USDA/NRCS, Washington, 1998.

[14] MENDONÇA, A. H. Radiological Accident in

Goiânia, Brazil: an Overview. in IVe Symposium International de Radioecologie de Cadarache A1-A8. CEA/IUR, Cadarache-France, 1988. [15] EMBRAPA. Manual de Métodos de Análise de Solo. EMBRAPA-CNPS, Rio de Janeiro, 1997. [16] INTERNATIONAL UNION OF RADIOECOLOGISTS. Report of the Working Group Soil-to-Plant Transfer Factors. Bilthoven, Netherlands, 1989. [17] SHEPPARD, M. I. & THIBAULT, D. H. Desorption and Extraction of Selected Heavy Metals from Soils. Soil Science Society of American Journal, vol. 56, 415-423. 1992. [18] TESSIER, A., CAMPBELL, P.G.C. & BISSON, M. Sequential Extraction Procedure for the Speciation of Particulate Trace Metals. Analytical Chemistry, vol. 51, p. 844-851. 1979. [19] MEGUELLATI, N. Mise au Point d'un Schèma d'Extractions Séletives des Polluants Métalliques Associés aux Diverses Phases Constitutives des Sédiments. 134. Université de Pau et de Pays de L'Adour, Pau, France, 1982. [20] OUGHTON, D.H. Radiocaesium Association with Soil Components: The Application of a Sequential Extraction Technique. in VIIth Report of Working Group Soil-to-Plant Factors p. 182-188 . IUR, Uppsala, Sweden., 1990. [21] ANDERSON, S.J. & SPOSITO, G. Cesium Adsorption Method for Measuring Accessible Structural Surface Charge. Soil Science Society of American Journal, vol. 55, p. 1569-1576. 1991. [22] OUGHTON, D.H., SALBU, B., RIISE, G., LIEN, H., OSTBY, G. & NOREN, A. Radionuclide mobility and bioavailability in Norwegian and Soviet Soils. Analyst, vol. 117, p. 481-486. 1992. [23] WASSERMAN, M. A., PEREZ, D.V. & BOURG, A. Behaviour of 137Cs in Some Brazilian Oxisols. Communications in Soil Science and Plant Analysis, vol. 33, p. 1335 - 1349. 2002. [24] LAURIA, D.C., SACHETT, I., PEREIRA, J. C. & ZENARO, R. Determination of Cs-137 Soil-To-Plant Concentrations Ratios for Vegetables in Goiânia, Brazil. Journal of Radioanalitical and Nuclear Chemistry, vol. 182, 91-96. 1994. [25] COLLE, C., ROUSSEL DEBET, S. & REAL, J. Transferts du radiocobalt en milieu terrestre. Radioprotection, vol. 31, p. 343-353. 1996.