JEQ Grow Your Career With ASA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 25 May 2007
Published in J Environ Qual 36:943-952 (2007)
DOI: 10.2134/jeq2006.0402
© 2007 American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Roig, M.
Right arrow Articles by Rigol, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Roig, M.
Right arrow Articles by Rigol, A.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Roig, M.
Right arrow Articles by Rigol, A.
Related Collections
Right arrow Radionuclides
Right arrow Ecological Risk Assessment
Right arrow Soil Pollution

TECHNICAL REPORTS

Ecological Risk Assessment

Prediction of Radionuclide Aging in Soils from the Chernobyl and Mediterranean Areas

M. Roiga, M. Vidalb,*, G. Rauretb and A. Rigolb

a Institut de Tècniques Energètiques, Universitat Politècnica de Catalunya, Av. Diagonal 647, 08028 Barcelona, Spain
b Departament de Química Analítica, Universitat de Barcelona, Martí i Franqués 1-11, 3a Planta, 08028 Barcelona, Spain

* Corresponding author (miquel.vidal{at}ub.edu)

Received for publication September 26, 2006.

   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
The aging of soil-pollutant interaction, which may lead to an increase in pollutant fixation, is the main driving force in the natural attenuation of contaminated soils. Here a test was evaluated to predict the aging of radiostrontium and radiocesium in soils from the Chernobyl and Mediterranean areas. After contamination, soils were maintained at various temperatures for up to 12 mo, with or without drying–wetting (DW) cycles. Changes in the quantity of radionuclide reversibly sorbed over time were monitored using an extraction test (1 mol L–1 NH4Cl; 10 mL g–1; 16 h). The fixed fraction could not be predicted from soil properties controlling the sorption step. Aging was not as relevant for Sr as for Cs. The time elapsed since contamination was the main factor responsible for the slight (up to 1.3-fold) decreases in Sr extraction yields. The additional effect of DW cycles was negligible. Instead, all factors accelerated Cs aging due to the enhancement of Cs trapping by clay interlayer collapse, with up to 20-fold increases in Cs fixation. The DW cycles also caused secondary effects on the Cs-specific sorption pool, which were beneficial or detrimental depending on the soil type. Extraction yields from laboratory aged samples agreed with those from field samples taken a few years after the Chernobyl accident. These results confirm the prediction capacity of the laboratory test and its usefulness in risk assessment exercises and in the design of intervention actions, particularly because neither fixation nor aging were related to the soil properties, such as clay content.

Abbreviations: DW cycles, drying–wetting cycles • RIP, radiocesium interception potential • FES, frayed edge sites • REC, regular exchange complex • HAS, high selectivity sites


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
AFTER a radioactive release, a two-step process governs the interaction of a radionuclide with the solid phase in soils. The first step is a fast, reversible sorption that can be described through the distribution coefficient (Kd) of the labile radionuclide between the solid phase and the soil solution. After the initial soil-radionuclide interaction, a fraction of the labile sorbed radionuclide is apparently no longer available for direct exchange with ions in the soil solution, and it remains fixed by the solid phase. This is a much slower process, and the fixed fraction may increase with time. While for a limited number of radionuclides (e.g., Cs, Sr, U) the labile Kd can be reasonably well predicted from soil parameters (Echevarria et al., 2001; Sanchez et al., 2002), attempts to predict the radionuclide fixed fraction have failed to date. This process is radionuclide and soil dependent, and little is known about how environmental factors (e.g., temperature; hydric regime) can increase the fixed fraction with time, with the subsequent decrease in the reversibly sorbed fraction. This sorption dynamic is often reported as an aging process in the literature and it is the main cause of the natural attenuation of radionuclides in soils. The natural attenuation process can lead to a decrease of the radionuclide mobility in the environment with a subsequent reduction in food contamination several years after a deposition event (Krouglov et al., 1997).

As Cs soil sorption is controlled by illitic clay minerals, with highly selective sites located at the edges of the illitic particles (Frayed Edge Sites, FES) (Sawhney, 1972; Cremers et al., 1988), its aging has been attributed to solid-state migration into specific sites in the wedge area closer to the collapsed core, where Cs is less exchangeable and even irreversible Cs trapping may occur as a result of clay collapse (Hird et al., 1996). There is little knowledge and consensus concerning Sr aging, since its sorption is less specific and it is controlled by the cationic exchange capacity (CEC) of the soil (Valcke, 1993; Hilton and Comans, 2001).

The existence of the aging process makes it difficult to correctly estimate the field Kd in contaminated scenarios from a Kd predicted solely on the basis of soil properties, since the sorption dynamics causes an increase in the initial labile Kd (Comans and Hockley, 1992; Sanchez et al., 2002). To date, the dynamic aspects of radionuclide interaction have only been partially included in modeling exercises of radionuclide mobility in the environment (Gillet et al., 2001). It is crucial to be able to predict immediately after a contamination event the natural attenuation potential of the radionuclide for improved risk assessment, particularly when making decisions on whether to rely on natural attenuation or to implement intervention actions.

The natural attenuation due to aging can be estimated from sorption experiments in which changes in the distribution coefficient can be monitored over time (Absalom et al., 1995), or from desorption tests based on sequential and single extractions, monitoring potential changes over time in the extraction yields (Rigol et al., 1999a). Both approaches are operational and highly dependent on the experimental conditions. However, they allow for the quantification of the aging process as long as they are applied to the same soil with the same experimental conditions over time. In this paper we have developed and applied a laboratory test to estimate the medium-term aging of radionuclides in soils, based on maintaining contaminated soils at various temperature regimes for a given time, with or without drying–wetting (DW) cycles. A similar approach was followed in the past for radionuclides (Rigol et al., 1999a) and heavy metals (Sastre et al., 2004). A general conclusion from previous experiments was that time alone appeared to be of lesser significance than changes in the hydric regime in soils, provoked by DW cycles. However, the individual role of each factor was not identified. Therefore, in this work we tested the factors affecting aging, including time since radionuclide addition to the soil, temperature, and hydric regime. The changes in the fixed fraction of Sr and Cs were examined in a set of soils that covered a wide range of soil types of contrasting properties (i.e., entisol; spodosol; mollisol; peat), with a single extraction test based on the use of a mild extractant solution. Where possible, the validity of this approach was checked through comparisons with soil samples aged under field conditions. Finally, secondary effects of aging in the experimental soils (i.e., changes in the Cs-specific sorption pool) were also examined.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Soil Characteristics
One organic and five mineral soils were included in this study, as described in Table 1. The organic soil (Peat) was a Histosol from an area close to the Chernobyl nuclear power plant (NPP). The mineral soils originated from the Mediterranean (Alfisol and Entisol) and Chernobyl areas (Spodosols 3 and 5, and Mollisol). The soils originating from the Chernobyl areas were taken from plowed plots 6 yr (Spodosol 3), 8 yr (Peat), and 9 yr (Spodosol 5 and Mollisol) after the Chernobyl fallout. All soils were air-dried and screened through a 2-mm sieve before analysis.


View this table:
[in this window]
[in a new window]

  		Table 1. Summary of main soil characteristics (na, not available).

 
Soil pH was measured in 0.1 mol L–1 KCl (Merck Pro-Analysis, Darmstadt, Germany) using a solution/soil ratio of 2.5 mL g–1. The organic matter (OM) content was determined by weight loss on ignition at 450°C for 16 h. The CEC values were measured by the silver thiourea method (Chabra et al., 1975) in the soils from the Chernobyl area, and by the sum of exchangeable cations displaced by CH3COONH4 and exchangeable acidity displaced by BaCl2–triethanolamine (Thomas, 1982), for the Mediterranean soils. Particle size distribution was determined in all samples (MAPA, 1994).

The concentration of major elements in soil solution was determined in all soils, except for the Mollisol. Dried soil samples were saturated with water, and centrifuged at 0.33 bar (75 x g; 30 min). The amount of water remaining in the soil at this pressure was determined by the weight loss at 105°C to estimate the soil field capacity. Subsequently, soil solution composition was determined following a method based on previous studies (Adams et al., 1980). After rewetting the air-dried soil samples up to field capacity, the wet soil was left for 24 h at 20°C, placed in the upper part of a cylinder with a porous plate, and subsequently centrifuged (7000 x g; 90 min). The solution, separated and recovered in the base of the cylinder, was filtered (0.45 µm), acidified, and stored in polyethylene vials. Calcium, Mg, and K were determined by inductively coupled plasma-optical emission spectrometry (ICP–OES) (Thermo-Jarrell Ash 25, Thermo Elemental, Franklin, NJ). Ammonium was determined by UV-Vis spectroscopy (PerkinElmer Lambda 19 UV/VIS, Wellesley, MA) by an automated colorimetric method based on the formation of indophenol blue (USEPA, 1979).

Soil Samples and Treatments
Soil samples were contaminated after rewetted to saturation (visual inspection), with soluble 85Sr and 137Cs, using almost carrier-free solutions supplied by Damri (85SR-ELSB45 and 137CS-ELSB45; LMRI, Gif-Sur-Yvette, France). The activity of the solutions was within the range of 105 to 106 Bq L–1, leading to mass activity density in soils within the range of 7 x 104 to 106 Bq kg–1, depending on the volume of added solution. The high mass activity density in soils ensured that in those soils affected by the Chernobyl fallout the quantity of Chernobyl-originated radionuclide was negligible. This defined the initial samples. Portions of initial soil samples were extracted following the desorption procedure indicated below.

Study of the Effect of Time and Temperature on Radionuclide Aging
Aliquots of initial samples were kept in closed vessels at four constant temperatures (70, 40, 10, and –20°C) and at a constant hydric regime, for 5 d (5D samples), 35 d (1M samples), 120 d (4M samples), 250 d (8M samples), and 370 d (12M samples). Portions of all generated samples were extracted following the desorption test indicated below.

Study of the Effect of Drying–Wetting Cycles on Radionuclide Aging
The extraction yields obtained at a constant temperature and hydric regime were compared to those obtained from soil samples submitted to DW cycles at four temperatures (70, 40, 10, and –20°C). Samples were maintained in closed vessels at a fixed temperature for approximately 2 d, and then dried in open vessels at 40°C, and again rewetted to saturation. The average duration of cycles was 5 d. Soil samples were submitted to DW cycles for 5 d (5D* samples), 35 d (1M* samples), 120 d (4M* samples), 250 d (8M* samples), and 370 d (12M* samples). Finally, aliquots of all samples were extracted following the extraction test described below.

Extraction Test
Desorption yields were determined by extraction with 1 mol L–1 NH4Cl (Merck Pro-Analysis, Darmstadt, Germany) for all samples generated from the laboratory experiments. Soils from the Chernobyl area were directly analyzed for their 137Cs extraction yields to estimate the Cs fixed fraction after aging under field conditions in the medium-term after the contamination event, thus defining the 6Y samples (Spodosol 3), 8Y samples (Peat), and 9Y samples (Mollisol and Spodosol 5). Soil samples were suspended in the extractant solution (10 mL g–1), set on an end-over-end shaker for 16 h (30 rpm), and centrifuged at 13 000 x g for 30 min. Radionuclide content was determined in the supernatant, and the extraction yield was referred to the initial activity concentration of the sample.

Determination of the Radiocesium Interception Potential (RIP)
The RIP values were determined for all initial soil samples and in some selected samples submitted to DW cycles. In short, after pre-equilibrating the samples (1 g) with 50 mL of a solution containing 100 mmol L–1 Ca and 0.5 mmol L–1 K (mK = 0.5), samples were equilibrated with the same solution, but labeled with 137Cs. Radiocesium distribution coefficients (Kd[Cs]) were obtained by measuring 137Cs activity in the supernatant before and after the equilibration for 24 h, using a solid scintillation detector (Canberra PACKARD MINAXI 5000 Series, Schwadorf, Austria). The calculated product Kd(Cs) mK defined the RIP value (Wauters et al., 1996a).

Gamma Spectrometry
The 137Cs activity concentration in samples contaminated with Chernobyl fallout was measured by high resolution gamma spectrometry, using an intrinsic Ge detector (ORTEC GMX 15200-P, Oak Ridge, TN), and a multichannel analyzer (ORTEC 5600, Oak Ridge, TN) with 8192 channels. Samples contaminated with radionuclide solutions (137Cs and 85Sr) were analyzed in 20-mL-capacity polyethylene vials by gamma spectrometry with a solid scintillation detector (Canberra PACKARD MINAXI 5000 Series, Schwadorf, Austria), equipped with a 3-in NaI (Tl activated) crystal, and using a mathematical correction for 137Cs contribution in the 85Sr channels. Measurement time was set to obtain relative standard deviation (RSD) < 0.5%.

Quality Control
The quality of the measurements performed with the Ge detector was ensured by the continuous analyses of the IAEA 327 reference material, which is a mineral soil with reference values of gamma-emitter radionuclides, such as 137Cs, and by the participation in intercomparison exercises organized by the IAEA for several environmental matrices. In addition, intermediate activity solutions of 85Sr and 137Cs were prepared by diluting weighed amounts of commercial solutions of each radionuclide with deionized water, and used as internal controls.


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Radiostrontium Aging
Joint Effect of Time and Temperature
Table 2 shows the extraction yields for Sr in the initial soil samples tested. Values ranged from more than 95% in the Entisol to around 45% in the Peat. The reversibly sorbed fraction decreased with increasing OM content and CEC, as has been reported in the literature (Nisbet and Shaw, 1994; Rigol et al., 1999a). Good correlations were derived between the Sr extraction yields of the initial samples and the CEC and OM contents (Pearson correlation coefficient of 0.92 for both correlations). To check if the reversibly sorbed fraction, estimated by the extraction yields, was related to the previous sorption step, the Sr reversible distribution coefficient, Kd(Sr), was calculated from soil properties. The solid-liquid partitioning of Sr may be rationalized with reference to the partitioning of sorption competitive ions characterized by similar sorption behavior, as is the case of Ca and Mg for Sr. Hence, Kd(Sr) can be predicted from the ratio of the level of Ca and Mg in the exchangeable complex (Ca+Mg)exch vs. the sum of Ca and Mg concentrations in the soil solution (Ca+Mg)ss (Hilton and Comans, 2001; Camps et al., 2004). As seen in Table 2, the Kd(Sr) varied in a narrow range from 2 L kg–1 in the Entisol, to around 8 L kg–1 in the Alfisol, and were weakly correlated to extraction yields (Pearson coefficient of 0.40).


View this table:
[in this window]
[in a new window]

  		Table 2. Radiostrontium extraction yields of initial samples and distribution coefficients (Kd[Sr]) predicted from soil properties (na, not available; exch, exchangeable; ss, soil solution).

 
Figure 1 shows the Sr extraction yields obtained from those soil samples contaminated with soluble 85Sr and submitted to aging at laboratory at constant temperature and hydric regime. For the Spodosol 3 and Peat soils, results of 90Sr aged for 6 yr under field conditions were available from the literature, and are also given in Fig. 1 (6Y samples) (Askbrant et al., 1996; Torres et al., 1996). In general, Sr yields varied more as a function of soil type than due to aging induced by the treatments, since the joint effect of the time and temperature did not result in a significant aging of Sr in most cases. Exceptions were the Spodosol 3 and samples kept at 70°C of the Alfisol, Entisol, and Peat soils, for which extraction yields were significantly lower than the initial values (in the Spodosol 3, the Sr extraction yields (with their standard deviation in brackets) decreased from 83 (± 1) to 63 (± 2)%; from 84 (± 1) to 76 (± 2)% in the Alfisol; from 96 (± 1) to 90 (± 1)% in the Entisol; and from 47 (± 2) to 40 (± 3)% in the Peat). This pattern indicated that Sr aging is in general a process of minor significance. Moreover, results showed the difficulty of predicting sorption dynamics even in similar soils, as observed in the comparison of Spodosols 3 and 5 (in this latter soil, extraction yields only changed from 76 (± 2) to 72 (± 3)%). Regarding the Spodosol 3, the distinctive behavior of this soil can be explained by its CEC, which was the lowest of all soils tested. These results are similar to previous results that suggested that Sr aging was more relevant for soils with low CEC values (Valcke, 1993). Extraction yields of 12M samples of the Spodosol 3 and Peat soils were similar to the values obtained from field 6Y samples, thus suggesting that the maximum degree of aging had been reached in the laboratory experiments.


Figure 1
View larger version (58K):
[in this window]
[in a new window]

  		Fig. 1. Radiostrontium aging: effect of temperature (–20, 10, 40, and 70°C) and time (5 d: 5D; 1 mo: 1M; 4 mo: 4M; 8 mo: 8M; 12 mo: 12M) on Sr extraction yields (n = 3; bars indicate one standard deviation) in all examined soils (Entisol; Spodosol 3; Alfisol; Spodosol 5; Mollisol; Peat). Initial extraction yields (I) are represented by a hatched area (mean value ± one standard deviation). For the Spodosol 3 and Peat soils, the extraction yields of the 6 yr (6Y) field samples are also indicated by a hatched area (mean value ± one standard deviation).

 
Additional Effect of Drying–Wetting Cycles on Radiostrontium Aging
Figure 2 shows the Sr extraction yields obtained from those soil samples contaminated with soluble 85Sr and submitted to DW cycles at various temperatures. For the Spodosol 3 and Peat soils, results for the 90Sr sample aged for 6 yr under field conditions are also included (Askbrant et al., 1996; Torres et al., 1996).


Figure 2
View larger version (59K):
[in this window]
[in a new window]

  		Fig. 2. Radiostrontium aging: effect of temperature (–20, 10, 40, and 70°C), time, and drying–wetting cycles (5 d: 5D*; 1 mo: 1M*; 4 mo: 4M*; 8 mo: 8M*; 12 mo: 12M*) on Sr extraction yields (n = 3; bars indicate one standard deviation) in all examined soils (Entisol; Spodosol 3; Alfisol; Spodosol 5; Mollisol; Peat). Initial extraction yields (I) are represented by a hatched area (mean value ± one standard deviation). For the Spodosol 3 and Peat soils, the extraction yields of the 6 yr (6Y) field samples are also indicated by a hatched area (mean value ± one standard deviation).

 
The inclusion of DW cycles did not significantly alter the trends observed for the extraction yields in the previous section. A slightly greater decrease in the extraction yields was observed in Spodosol 5 and Mollisol, while Spodosol 3 behaved similarly to the sample maintained at a constant hydric regime. One distinctive pattern observed was that the DW cycles applied at low temperature induced an increase in the 85Sr extraction yields in the Peat soil. This result can be explained by an increase in the solubilization of the organic acids under these conditions, due to the partial breaking of humic acids-clay complexes (Bartlett and James, 1980). As in the previous section, results from Spodosol 3 and Peat 6Y samples allowed us to examine the validity of the laboratory tests as an approach to predict aging under field conditions. The fact that the extraction yields of the 12M* samples were similar to those of the 6Y samples confirmed that aging was significant 1 yr after the contamination, but negligible from that time onward. Furthermore, conclusions on aging derived from the changes in the extraction yields agreed with soil-plant transfer pattern observed in the field. Data obtained from field experiments after the Chernobyl fallout showed that Sr transfer was quite constant a few months after the contamination when condensed deposition was the source of contamination (Nisbet and Shaw, 1994; Ehlken and Kirchner, 1996).

The type of interactions expected for Sr in soils are responsible for the low relevance of its aging in soils, and also for the minor role that the DW cycles played in increasing Sr fixation. As Sr is mainly sorbed at regular exchange sites, Sr aging will be governed by those factors affecting the soil CEC and pH regardless of the initial reversibility of the sorption stage. The effect of the DW cycles that induce aging is consequently more pronounced for those soils with the lower CEC values, where changes in specific interactions in clay and clay OM complexes, or isomorphic substitutions in Ca-bearing minerals due to temperature and DW cycles, are of a relatively greater significance (Valcke, 1993). In all, these changes take place only to a limited extent and are not an ongoing process that can last for years. Results obtained here agree with those obtained with a moist loam soil incubated for 125 d (Wang and Staunton, 2005). In this study, authors reported a twofold increase in the Kd obtained from isolating the soil solution of the incubated soil, potentially as a result of a slight decrease in pH during the incubation period. However, the Kd obtained from desorption with an electrolyte solution did not vary during the time examined, thus suggesting that the slight aging observed with the soil solution was not observed when soil was suspended in an electrolyte solution. Thus, it can be concluded that aging may not be relevant in the short-term period after a contamination and then natural attenuation cannot be expected as a significant process to be considered in the management of Sr-contaminated soils.

Radiocesium Aging
Joint Effect of Time and Temperature
A general overview of the extraction yields of initial samples (see Table 3) confirms the absence of a clear relationship between the reversibly sorbed fraction and soil properties for Cs (Rigol et al., 1999b). Values ranged from about 75% in the Spodosol 3 to about 5% in the Spodosol 5, thus illustrating that although these were similar soils from the pedological standpoint, they differed in respect to Cs interaction, and are an example of the more dissimilar extraction yields among the soils examined in this study. Extraction yields were compared with the Cs reversible distribution coefficient, Kd(Cs), calculated from soil properties. Radiocesium sorption is controlled by the specific frayed edge sites (FES) in illitic clays, and only partially affected by sorption sites in the regular exchange complex (REC) (Sawhney, 1972; Vidal et al., 1995). The approach based on the concept of RIP, defined as the product of the FES capacity and the trace Cs-to-K selectivity coefficient, enables us to describe Cs-specific sorption in soils (Sweeck et al., 1990). The RIP estimates the capacity of a soil to specifically sorb Cs and it can be determined by measuring the Kd(Cs) in a well-defined scenario (Wauters et al., 1996a). The RIP values differed by more than two orders of magnitude between the soils examined here (see Table 3), showing great differences between soils of similar properties, with no correlation to the mineral matter content. The Kd(Cs) at the specific sites (KdFES [Cs]) can be predicted by dividing the RIP value by the sum of K and NH4+ concentrations in the soil solution (competitive species for Cs sorption), the latter amplified by the NH4+-to-K trace selectivity coefficient (KC(NH4/K)) at FES (Sweeck et al., 1990; Wauters et al., 1996b). Here this parameter was experimentally obtained (details not shown) and its values were 6.4 for Alfisol, 4.8 for Entisol, 5.1 for Spodosol 3, 2.4 for Spodosol 5, and 5.2 for Peat soil. To better estimate the Kd(Cs), a second term must be added to account for the sorption in the regular exchange sites (KdREC [Cs]) by dividing the sum of the exchangeable K and NH4+ by the sum of K and NH4+ concentrations in the soil solution (Camps et al., 2004). The equation derived may be written as follows:

Formula 1[1]


View this table:
[in this window]
[in a new window]

  		Table 3. Radiocesium extraction yields of initial samples and distribution coefficients (Kd[Cs]) predicted from soil properties (na, not available; exch, exchangeable; ss, soil solution).

 
The calculated Kd(Cs) ranged from 23 L kg–1 in Spodosol 5 to 725 L kg–1 in Alfisol. Correlations between extraction yields and RIP and extractions yields with Kd(Cs) were weak (Pearson coefficient of 0.22 and 0.26, respectively), confirming that for Cs the reversibly sorbed fraction cannot be predicted from soil properties controlling the sorption step. A clear example of this was the distinctive behavior of Cs in Spodosols 3 and 5. While the RIP values of the former soil was approximately six times higher (indicating a higher soil capacity to specifically sorb Cs), its initial extraction yield was also about 25 times higher, thus indicating a more reversible sorption than in the Spodosol 5.

Figure 3 shows the Cs extraction yields obtained from those soil samples contaminated with soluble 137Cs and submitted to aging at laboratory at constant temperature and hydric regime. For the Spodosol 3, Spodosol 5, Mollisol, and Peat soils, results of samples aged under field conditions are also given. Extraction yields diminished in general over time, especially at high temperatures. For some soils, time seemed to be a more significant factor than temperature, although treatments at 70°C led to more significant decreases than at other temperatures. Aging was thus confirmed as a significant process for Cs, due to its diffusion to inner sites in interlayer positions of the clays (Evans et al., 1983; Comans et al., 1991). These sites can have a higher selectivity for Cs and can easily collapse at high temperatures trapping the sorbed cation (Wauters, 1994; Hird et al., 1996). Radiocesium diffusion to inner sites has been reported to be favored by an increase in the temperature (Evans et al., 1983; Comans et al., 1991), since at high temperature dehydration is facilitated, thus causing the Cs to become trapped and fixed by the clay mineral (Hird et al., 1995). Aging was the lowest among the soils tested for the Alfisol and Spodosol 5 soils. The aging was of a lesser significance in Spodosol 5 because it had already a low initial reversibly sorbed fraction. On the opposite, the Mollisol and Peat soils had the highest decrease in the extraction yields. The extraction yields after 12M at 70°C in the Peat soil were <2%. The large fraction of sorbed Cs at noncollapsed sites, which subsequently became trapped by interlayer collapse after dehydration, was the factor responsible for the large effect of the temperature in the Peat soil (Hird et al., 1996). The Cs trapping followed by the related aging, can be more significant for soils that initially have noncollapsed clay layers, as was the case here for Peat, where the humic acids in the interlayer sites hinder clay collapse (Maguire et al., 1992; Dumat et al., 1997). The information derived from field samples confirmed that the laboratory methodology was a promising strategy to predict aging under field conditions, since regardless the values of the initial extraction yields, the results after 12M, especially at 70°C, showed an excellent agreement with the yields obtained from samples aged for 6, 8, and 9 yr.


Figure 3
View larger version (48K):
[in this window]
[in a new window]

  		Fig. 3. Radiocesium aging: effect of temperature (–20, 10, 40, and 70°C) and time (5 d: 5D; 1 mo: 1M; 4 mo: 4M; 8 mo: 8M; 12 mo: 12M) on Cs extraction yields (n = 3; bars indicate one standard deviation) in all examined soils (Entisol; Spodosol 3; Alfisol; Spodosol 5; Mollisol; Peat). Initial extraction yields (I) are represented by a hatched area (mean value ± one standard deviation). For the Spodosol 3, Spodosol 5, Mollisol, and Peat soils, the extraction yields of the 6 yr (6Y), 9 yr (9Y), and 8 yr (8Y) field samples are also indicated by a hatched area (mean value ± one standard deviation).

 
Additional Effect of Drying–Wetting Cycles on Radiocesium Aging
Figure 4 displays the Cs extraction yields obtained from those soil samples contaminated with soluble 137Cs and submitted to aging at laboratory with DW cycles at various temperatures. For the Spodosol 3, Spodosol 5, Mollisol, and Peat soils, results of 137Cs of samples aged under field conditions are also given.


Figure 4
View larger version (46K):
[in this window]
[in a new window]

  		Fig. 4. Radiocesium aging: effect of temperature (–20, 10, 40, and 70°C), time, and drying–wetting cycles (5 d: 5D*; 1 mo: 1M*; 4 mo: 4M*; 8 mo: 8M*; 12 mo: 12M*) on Cs extraction yields (n = 3; bars indicate one standard deviation) in all examined soils (Entisol; Spodosol 3; Alfisol; Spodosol 5; Mollisol; Peat). Initial extraction yields (I) are represented by a hatched area (mean value ± one standard deviation). For the Spodosol 3, Spodosol 5, Mollisol, and Peat soils, the extraction yields of the 6 yr (6Y), 9 yr (9Y), and 8 yr (8Y) field samples are also indicated by a hatched area (mean value ± one standard deviation).

 
The inclusion of the DW cycles in the laboratory tests had a marked effect on the extraction yields. For all soils, yields for soils subjected to the DW cycles decreased with time in a similar manner, regardless of the temperature applied, as seen by the more narrow ranges of ratios of extraction yields between the initial and 12M* samples for the same soils without the DW cycles. In most cases, the aging process was accelerated, and values obtained from 8M* samples were not statistically different from those of 12M* samples. The application of DW cycles also provoked a higher decrease in the extraction yields for those soils exhibiting a minor aging in the absence of DW cycles, as was the case of Spodosol 3 and Alfisol. For these, DW cycles led to an almost twofold increase of the aging effect with respect to what was observed due to time and temperature (the Cs extraction yields decreased from 29 to 17% in Spodosol 3, and from 41 to 23% in Alfisol). For those soils with already low extraction yields (as Spodosol 5 and Peat soil at high temperature) yields remained low. It was thus confirmed that the clay collapse, already occurring at high temperatures and after a given time after the contamination, was enhanced by the application of the DW cycles. Therefore, the DW cycles accelerated the dehydration of the clay interlayers and subsequent interlayer collapse, which caused the Cs trapping and was responsible for the additional decrease in Cs desorption (Hird et al., 1995). In mineral phases, this process is also responsible for the generation of new selective sites, which may interact with Cs and being exposed to a new interlayer collapse (Maes et al., 1985; Hird et al., 1996; Degryse et al., 2004). An apparent maximum decrease in extraction yields was achieved after applying DW cycles for 8M, confirmed by the similarity with yields from samples aged under field conditions. Moreover, as also pointed out in the case of Sr, the aging pattern explained the dynamics of Cs soil-plant transfer observed under field conditions in similar soils regarding decreases in transfer monitored in the short-term (Squire and Middleton, 1966; Adriano et al., 1984; Noordijk et al., 1992; Haak and Lönsjo, 1996), and constant values reported a few years after the Chernobyl fallout, in which changes in transfer were of the same order of magnitude as seasonal variations (Nisbet and Shaw, 1994; Ehlken and Kirchner, 1996).

Secondary Effects of the Drying–Wetting Cycles on the Radiocesium Interception Potential Determination
The examination of changes in the RIP values after a given number of DW cycles may complete the understanding on the ongoing mechanisms governing Cs natural attenuation in soils, as RIP is the most sensitive parameter describing soil-Cs interaction. Table 4 summarizes changes in the RIP values in selected samples of the Spodosol 3, Mollisol, and Peat soils submitted to DW cycles for 5 d (5D*-RIP), and 4 mo (4M*-RIP). A general comparison between the RIP values obtained highlighted two patterns, one for the mineral soils and other for the Peat soil. Regarding the mineral soils, the RIP values increased after the application of the DW cycles, especially in the Mollisol. While increases in the RIP for the 5D* samples compared to the initial samples were minor, for the 4M* samples the RIP values were much higher than initial RIP values for the two soils at all temperatures tested. The increases in the RIP values were explained by the ongoing illitization of the clay particles due to clay collapse caused by the DW cycles (Sucha and Siránová, 1991). In this process, new selective sites are created by the continuous collapse of the interlayer sites, turning outer sites, which are regular exchange sites in the FES, into high selectivity sites (HAS) since they are wedge sites near to the collapsed central core (Wauters, 1994; Hird et al., 1996). The collapse may continue until the entire interlayer has been induced to collapse. The increase in the RIP values after DW cycles was also observed in clay materials, such as bentonites, and in soils from grasslands (Degryse et al., 2004). The decrease in the RIP values observed in the Peat soil, on the other hand, is an unexpected secondary effect of the DW cycles. In this soil, the DW cycles caused an initial collapse around the sites in which Cs was already sorbed, as confirmed by the decrease in the extraction yields. However, the existence of organic acids sorbed to positively charged sites caused by broken Al-OH groups would impede a progressive collapse of the clays and the creation of new selective sites for Cs sorption as occurring in mineral soils (Hird et al., 1996; Dumat et al., 1997). The final balance is a decrease in the number of the HAS sites in the wedge zone due to the clay collapse, which is not compensated by a sufficient creation of new high selectivity sites. Hence, in addition to the observed decrease in the reversibility of Cs sorption, the DW cycles are responsible for a secondary beneficial or detrimental effect on soils on the RIP values, depending on the soil type.


View this table:
[in this window]
[in a new window]

  		Table 4. Effect of the drying-wetting cycles on the radiocesium interception potential (RIP) values (n = 3; na, not available).

 

   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
Monitoring changes in the reversibly sorbed fraction of radionuclides through the quantification of the desorption yields with a single extraction test after submitting soil samples to treatments based on modifying temperature and water regime (DW cycles) proved to be a suitable laboratory approach to predict the fixation capacity and the radionuclide aging under field conditions. This approach is particularly useful considering that neither the fixation nor the aging could be predicted from soil properties controlling radionuclide sorption. The pattern derived from the extraction yields was validated with data on soil-plant transfer. These findings confirm aging as the main process controlling radionuclide natural attenuation in soils.

Radiostrontium showed only a marginal effect of aging in all the soils tested here, which depended basically on the time elapsed since the contamination event. The influence of the application of the DW cycles for Sr aging was close to negligible. Therefore, laboratory tests focused solely on this radionuclide may consider it not necessary to apply DW cycles, and thus a simpler test can be used. In contrast, Cs aging was a significant process, due to interlayer collapse by dehydration, thus causing the Cs cations to be entrapped. Although time and temperature played a significant role in decreasing the extraction yields, the DW cycles were the primary factor that accelerated aging. They are required to estimate, with only a few weeks of laboratory experiments, the aging process under field conditions a few years after an accidental release. Secondary effects on the soil RIP values, which characterize Cs sorption pattern, must also be considered when examining changes in the Cs interaction in the medium-term.

Although this laboratory approach was tested only for radionuclides, this test is recommended as a tool to examine the interaction not only of radionuclides but also heavy metals in soils from scenarios that may be potentially affected by an accidental release. This test may also help the decision-making step before the design of intervention actions in the frame of risk assessment exercises, especially when natural attenuation can be considered as the remediation strategy with the lowest cost and with a similar efficiency to other approaches.


   ACKNOWLEDGMENTS
 
This work was partially funded by a CICYT project (PPQ2002-00264). ICP-OES analyses were carried out in the Serveis Científico-Tècnics of the Universitat de Barcelona.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 





This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Roig, M.
Right arrow Articles by Rigol, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Roig, M.
Right arrow Articles by Rigol, A.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Roig, M.
Right arrow Articles by Rigol, A.
Related Collections
Right arrow Radionuclides
Right arrow Ecological Risk Assessment
Right arrow Soil Pollution

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Journal of Natural Resources
and Life Sciences Education		 Vadose Zone Journal
Soil Science Society of America Journal Journal of Plant Registrations The Plant Genome