HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via ISI Web of Science (1) Citing Articles via Google Scholar Google Scholar Articles by Álvarez-Benedí, J. Articles by García-Sinovas, D. Search for Related Content PubMed Articles by Álvarez-Benedí, J. Articles by García-Sinovas, D. GeoRef GeoRef Citation Agricola Articles by Álvarez-Benedí, J. Articles by García-Sinovas, D. Related Collections Other Contaminants Sorption/Exchange Soil Pollution

SPECIAL SECTION: ZNS'03 VADOSE ZONE RESEARCH

Adsorption–Desorption of Arsenate in Three Spanish Soils

^a Instituto Tecnológico Agrario de Castilla y León, Ctra. de Burgos, Km. 119, 47014 Valladolid, Spain
^b Dep. de Ingeniería Química, Universidad de Valladolid, Prado de la Magdalena s/n, 47005 Valladolid, Spain
^* Corresponding author (jabenedi{at}iq.uva.es)

Received 22 June 2004.

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS CONCLUSIONS REFERENCES

 
Adsorption–desorption processes of As in agricultural soils are of great importance because of the risk of entry of this contaminant into the food chain, As(V) being the main form of As in well-aerated environments. This study presents the dynamics of the adsorption of As(V) in three agricultural soils (Xerochrept, Xerofluvent, and Xerorthent) with loamy sand, sandy clay loam, and clay textures, respectively. Initially obtained kinetic data showed a relatively rapid first adsorption stage reaching pseudoequilibrium within a few hours. The adsorption–desorption isotherms revealed the presence of an unrecoverable fraction of As linked to the soil, which was further supported by a strong hysteresis subsequently observed in the desorption process. The temperature showed a relatively small effect on the isotherm within 10 to 25°C, although a decrease in the fraction of sorbed As(V) was found at 40°C as expected for an exothermic process. Finally, the effects of other ions typically present in productive agricultural systems were evaluated by quantifying their effects on the As(V) adsorption isotherm. The presence of phosphates significantly decreased the adsorption of As, thus increasing its bioavailability, while nitrate had the opposite effect. The presence of chlorides and sulfates did not present similar significant effects. Values obtained from modeling the sorption experiments in this study can provide a basis for the use of solute transport simulation models. Further application of such models will allow the establishment of future production strategies in zones having elevated As content in irrigation water.

Abbreviations: As(V), arsenate • As(III), arsenite • CEC, cation exchange capacity • XRD, X-ray diffraction

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS CONCLUSIONS REFERENCES

 
ARSENIC IS A TOXIC element with a highly variable distribution that is poorly correlated with geological formations, climate, or soil (Matschullat, 2000). This chemical element is found in the subsoil incorporated in a number of different compounds, mostly as metal sulfides. Arsenic has been widely used in medicines, cosmetics, paint manufacture, electronic components, and even in the construction of lasers. All these applications have contributed to its dispersion in the environment, with perhaps the greatest contribution coming from the extensive use of As in various inorganic forms in agricultural pesticides. This usage was extended from the 19th century to the last third of the 20th century, ultimately demonstrating phytotoxic effects in some crops (Woolson et al., 1971).

Arsenic occurs in –3, 0, +3, and +5 valence states, both in organic and inorganic species, typically subjected to oxidation–reduction, precipitation–dissolution, adsorption–desorption, organic methylation, and biochemical processes. Its most abundant form in aerobic environments such as the upper soil horizon is arsenate, As(V), which is a stable species under commonly encountered conditions of aeration (Deuel and Swoboda, 1972; Onken and Hossner, 1996). Under natural pH conditions As(V) exists in solution as H₂AsO₄^– and HAsO₄^–2 . Arsenic is retained by the soil in this oxidation state. Arsenite, As(III), is the most toxic form of As. It destroys all tissue by reacting with the sulfhydryl groups of proteins (Wauchope, 1983). At ambient pHs levels, it is also the most soluble and mobile form, making it the most available form for absorption by crops. Arsenite occurs in solution as H₃AsO₃ and H₂AsO₃^– . Due to a pH–pK_a interaction process and an increase of the negative potential in the plane of sorption, there is a decrease in As(V) sorption as pH increases which has been explained as preferential sorption of As(V) by some surface sites (Smith et al., 1999). Conversely, increasing pH leads to an increase in As(III) sorption, the As(III) being relatively immobile at elevated pH values (Smith et al., 2002). When other elements such as Al, Fe, and Mn are present, As(III) may form poorly soluble compounds, which also hinders the availability of As to plants. Finally, there are the organic forms of As, most of them volatile, toxic compounds.

Oxidation–reduction processes of As(III) and As(V) can take place in soils. These are kinetics processes that must, in principle, be considered in sorption studies. As mentioned above, As(V) is the most common form of As under experimental conditions in the field (agricultural studies on the upper horizon of the soil) and in the laboratory (Violante and Pigna, 2002), and also, As(V) is the most likely species to undergo adsorption by the soils. However, Goldberg (2002) showed that the rate of conversion of As(III) to As(V) in batch experiments reached about 5% in 2 h, while McGeehan (1996) performed the (reverse) reduction process under naturally reducing conditions during periods from 10 to 20 d. Thus, under aerated conditions, sorption of As(V) must not be affected by kinetics redox processes.

Concentrations of As of higher than the 50 µg L^–1 (as established by the European Community Directive 80/778/CE) were observed in the public water supply of the town of Íscar (Province of Valladolid, region of Castilla y León, Spain) in July 2000 (Fig. 1) . As the sampling area was progressively widened, the list of similarly affected locations began to rise. Finally, a zone of 1.700 km² was delineated in which land use is mainly agricultural. Arsenic concentrations in wells of the area have shown significant spatial and seasonal variability, in the worst cases reaching concentrations of 500 µg L^–1 (Calvo et al., 2003). The use of surface water from the Adaja and Eresma Rivers solved the problem of potable water supply in the affected area. However, the remaining problem was the use of groundwater for irrigation, which could have a direct effect on the safety of foods produced in agriculture. It was urgent, therefore, to characterize the dynamics of As in agricultural soils of the region. In this study, we evaluated adsorption–desorption processes of As(V) in agricultural soils. These processes control the mobility and bioavailability of As in soil–plant–water systems.

View larger version (109K): [in this window] [in a new window]   Fig. 1. Map of the surveillance zone (blue shaded region) including Eresma and Adaja rivers, and the cities of Valladolid and Iscar.

Given the known existence of the two oxidation states, As(III) and As(V), interest could arise in widening the study to include different conditions of pH and redox potential. Goldberg (2002) recently showed that the effect of pH was important in the adsorption of As(V) above pH 9.5, although this represented situations far from values found in the agricultural soils studied here. The study of the effect of pH would require the addition of strong acids or bases (e.g., HNO₃ and NaOH, respectively), which can alter the natural composition of the studied soils, and thus was not considered in this work. Similarly, redox potentials differing from natural values have been shown to produce important changes in the responses of soil oxides and hydroxides, which ultimately have effects on adsorption (McGeehan et al., 1998). The present study, however, is an attempt to characterize processes in our chemically unaltered agricultural soils; thus, the control or modification of the above cited conditions (e.g., by means of strong acid or bases) were not considered.

Our study objectives, based on observations made with three Spanish agricultural soils of differing textural classes, were (i) characterization of adsorption of As(V), including a kinetic study and definition of its adsorption–desorption isotherms at 25°C; (ii) determination of the effect of temperature on the adsorption isotherms, including two additional temperatures (10 and 40°C); and (iii) evaluation of the effect of previous application of agrochemicals FeSO₄, KNO₃, and K₃PO₄ on the adsorption isotherms of As(V) in a soil under agricultural practices. An additional objective was to extract practical conclusions from the results for future use in predictive transport models related to the behavior of As in the soils studied.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS CONCLUSIONS REFERENCES

 
On the basis of previous sampling surveys within the affected area, initially we selected two soils, a Typic Xerofluvent with a sandy clay loam texture and a Xerochrep with a loamy sand texture, both from the Ap horizon of agricultural soils from the surveillance zone (Fig. 1). Given the relatively low variability of the soil textures within the study area, a clay soil from the Ap horizon of an Xerorthent agricultural soil from Valladolid (Spain) was also selected. The three soils were air dried and passed through a 2-mm sieve before study. Table 1 shows texture and relevant chemical compositions including the total As content of the soils. The pH of the soil was measured using a 1:2.5 soil/water ratio. Organic matter content was determined by dichromate oxidation in sulfuric acid media. Soil texture was determined by the Hydrometer (Bouyoucos) method. The available P concentration was estimated by a colorimetric method using 0.5 M NaHCO₃ (Olsen method). Ammonium acetate (1 M) adjusted to pH 7 was used as extractant to determine exchangeable Na, K by atomic emission spectroscopy, and Ca and Mg by atomic absorption spectroscopy. The cation exchange capacity (CEC) was calculated after saturation of soil with Na⁺, washing with ethanol, and further displacement of Na⁺ with NH₄⁺ for quantification of displaced Na⁺ by atomic emission spectroscopy. All are normalized methods that can be found elsewhere (e.g., AFNOR, 2004).

Estimation of the Arsenic Concentration
The method developed for the determination of As in soils was based on the digestion of 0.5 g of sample with HNO₃/H₂O₂ (8 and 2 mL, respectively) in a CEM MarsX microwave oven (CEM Corporation, Matthews, NC) using a temperature ramp from ambient to 180°C over 20 min, followed by 10 min at 180°C. After the digestion, the excess oxidants were eliminated using HCl and urea. Then the sample was treated with potassium iodide and ascorbic acid (30/10%) to reduce the As(V) to As(III). This mixture was also used with water samples in the analysis of dissolved As. The amount of dissolved As was determined by continuous-flow-hydride generation and atomic absorption spectrometry using a 929 Solaar system equipped with a VP-90 hydride generator and EC 90 oven (Unicam, Cambridge, UK). A 1% sodium borohydride in 0.1% NaOH solution was used to transform As(III) into AsH₃ using a stabilization time of 50 s and a N₂ carrier gas flow of 180 mL min^–1. The AAS readings were made at 193.7 nm, using a deuterium lamp as a baseline corrector. This procedure was validated following UNE-ISO-EN 17025 quality criteria. The limit of determination was 0.2 mg kg^–1 and 1.0 µg L^–1 in soil and water, respectively. Based on a previous statistical tests performed with standards, the accuracy was estimated within ±10%. Simultaneously, a comparative method was developed in an external laboratory based on the generation of hydrides followed by atomic fluorescence spectroscopy, using the same sample preparation. No differences were found between the results from the two methods. Arsenic speciation tests were performed by means of pH control on the hydride generation. According the pH and Eh values of water samples, only As(V) was detected.

Characterization of Adsorption Kinetics
The objective of the kinetics study was to estimate the time required for the soil–water solute systems to reach equilibrium, or more precisely, pseudoequilibrium. This is the minimum time to be used in the determination of adsorption–desorption equilibria in the subsequent sections. Sealed 100-mL Erlenmeyer flasks were prepared, containing 25 g soil and 50 g of As(V) solution. The higher concentrations used in adsorption experiments (see next section) were selected for the kinetics experiments. Thus, an approximate concentration of 1.000 µg kg^–1 solution was used with the loamy sand and sandy clay loam soils, and an approximate concentration of 10.000 µg kg^–1 solution was used with the clay soil. The suspensions were allowed to equilibrate under controlled temperature conditions (25.0 ± 0.1°C) using magnetic stirring. Duplicate samples were obtained from the systems at 12, 24, 48, 72, 96, and 120 h. After these times, stirring was stopped and 30-g samples were taken of each supernatant suspension. The samples were placed in tubes and centrifuged for 10 min at 5500 rpm to obtain a clear supernatant liquid. Approximately 25 mL of each clear supernatant was stored at 4 ± 1°C until analysis. The amount of adsorbed As was determined by a mass balance equation:

[1]

where q_e represents the estimated amount of adsorbed As(V) (µg kg^–1), S represents the soil mass (kg), C₀ the initial concentration of As in the solution (µg kg^–1 solution), L is the mass of the solution (kg), and C is the measured concentration of As(V) in solution after the equilibration time. Because the soils contained a certain amount of As before the sorption experiments, the total As content at equilibrium (q) will be given by the sum of the estimated q_e in the mass balance (Eq. [1]), and the corresponding value of As concentration in soil before the sorption experiments (Table 1).

Given the possible influence of pH on redox processes and As(V) adsorption, variations of pH were monitored in separate control experiments. For this, duplicate trials with each of the three studied soils using the same experimental procedure of the kinetics study [i.e., the same soil and solution amounts and As(V) concentrations] were prepared, and the pH of the supernatant liquid was measured at intervals of approximately 1 h during the first 6 h. Further pH measurements were performed at t = 24, 30, 48, and 54 h after the start of the experiment.

Data on microbial degradation of arsenicals shows that losses during the equilibrium time selected for adsorption desorption experiments (24 h) can be assumed negligible. For example, Gao and Burau (1997) recovered 90% of applied As(V) after 70 d of incubation under favorable conditions for microbial activity (–0.03 MPa and 22°C). Arsenate was, in fact, the main stable metabolite produced from microbial degradation after application of two organic arsenicals used in their experiments (sodium cacodylate and Methanearsonic acid). Turpeinen et al. (2002) estimated microbial loses of As(V) to be 0.5% in contaminated soils under aerobic and anaerobic conditions. For this reason soil microbial activity was not inhibited in our experiments.

Adsorption Equilibrium
Based on measured concentrations of As(V) in local groundwater field wells, solutions of 0, 150, 250, 500, 750, and 1000 µg kg^–1 solution of As(V) were prepared. First experiments, however, showed a high degree of adsorption on the clay soil. For example, using an initial solution concentration of approximately 1000 µg kg^–1 solution, a concentration of 66.3 µg kg^–1 solution was obtained at pseudoequilibrium, which was considerable lower than the upper field-measured values. For this reason, additional experiments were performed with higher initial As(V) (e.g., C₀ = 5000 and C₀ = 10000 µg kg^–1 solution, respectively). These experiments allowed for the generation of adsorption isotherm points at appropriate dissolved equilibrium concentrations of As(V).

Materials and experimental conditions were as described above, using the same soil/solution ratio (1:2) and performing duplicates for each trial. The resulting suspensions were temperature controlled and stirred continuously for 24 h, which was the experimental time established after the kinetics study. After this time, the same process of centrifugation, removal of the supernatant liquid phase, and estimation of As concentration were performed.

The shape of the adsorption isotherm was conditioned by the existence of a certain quantity of adsorbed As(V) at the beginning of our tests. Also, in the study it was ascertained experimentally that a fraction of As(V) was apparently irreversibly bound to the soil after de sorption–desorption experiments. The adsorption isotherm therefore was not expected to pass through the origin. The equilibrium data were fitted to a modified linear isotherm given by

[2]

where q represents the quantity of adsorbed As at equilibrium (µg kg^–1 soil), K is the equilibrium coefficient, and C is the dissolved concentration (µg kg^–1 solution). The quantity of apparently irreversibly adsorbed As(V), q_i, was operationally defined as all "unrecoverable" As(V) by using a soil–water equilibration procedure. The concept of irreversibly bound solutes has been extensively used in the scientific literature to give a macroscopic description of adsorption and ion-exchange phenomena (e.g., Beek and van Riemsdijk, 1982). Experiments performed with labeled phosphate revealed that all of the apparent irreversibly bound solute could be exchanged, although in part very slowly, and that slowness of exchange was the physical explanation of an apparent irreversibility. Thus, for the purpose of modeling, q_i must be considered an empirical parameter that can be obtained by fitting of experimental sorption data to Eq. [2] or, alternatively, by independent estimation of the unrecoverable fraction of As(V) from extraction experiments, as in our case. Care must be taken, however, in the estimation from extraction experiments, as this quantity may depend on the extracting solvent (e.g., water or phosphate) and the extraction procedure.

The results of applying this model were compared with those obtained with a modified Freundlich, nonlinear model:

[3]

where K_f is the equilibrium coefficient and 1/n is a coefficient accounting for the nonlinearity of the isotherm.

In this work, the amount of irreversibly adsorbed As(V) at pseudoequilibrium with a concentration C = 0 (i.e., q_i), was independently determined by means of four-step extraction experiments. These experiments were performed with the untreated soils by duplicate (natural background As contents for these soils were presented in Table 1). For this, 24 g of each untreated soil were placed in contact with 48 g of ultra-pure water with constant stirring at 25°C for 24 h and then allowed to settle during other additional 24 h. Then 20 g of the supernatant water was analyzed for As. The same quantity of ultra-pure water as was removed was then returned to the system. The above equilibration procedure was repeated. This trial was performed in duplicate with equilibrations made for 48, 96, 144, and 192 h, which constitutes four periods of stirring (24 h) and subsequent decantation (24 h). The measured values of As(V) in the solution (C) and values of As(V) in soil (q) estimated from the mass balance (Eq. [1]) were plotted. As experiments were made by duplicate, eight data pairs were generated for each of the three studied soils. The value for q_i was then obtained from the ordinate at the origin of the data plot of q vs. C.

Desorption Equilibrium
Solutions of 0, 150, 250, 500, 750, and 1000 µg kg^–1 solution of As(V) were prepared to determine the desorption isotherms. Twenty grams of soil and 48 g of solution were prepared in duplicate (two trials per concentration level and soil). The suspensions were stirred at 25.0 ± 0.1°C for 24 h in Erlenmeyer flasks. After this time, the suspensions were allowed to settle (by gravity) for 24 h to obtain an initial separation of liquid from solid phase. Samples of approximately 24 mL were then taken by means of a syringe, placed in tubes, and centrifuged for 10 min at 5500 rpm to obtain a clear supernatant for analysis. Sediment remaining after centrifugation was recovered and returned to the Erlenmeyer flask, along with a quantity of distilled water to replace the initial amount of solution. This procedure was repeated four times for each flask, generating a point on the desorption isotherm at every step.

Effect of the Presence of Other Ions
The same experimental methodology was employed to investigate the effect of other ions on the adsorption of As(V) in soil. In this case, ionic solutions were added to the soils before the sorption experiments, reflecting ionic concentrations typically encountered in agricultural environments. In this part of the study, the sandy clay loam soil was selected. This type of soil was chosen based on its intermediate characteristics and capacity for adsorption, as well as the well-represented soil texture of the study area. Soil samples were treated with different ionic solutions selected among the used chemicals in the zone. A solution of 7.6 g L^–1 KNO₃ was used to obtain 0.7 kg m^–3 N in the soil, and a solution of 0.5 g L^–1 K₃PO₄ was used to obtain 0.2 kg P₂O₅ m^–3 soil. The objective of this study was to derive practical conclusions for As(V) sorption–desorption dynamics under agricultural scenarios. For this reason, we used K-based salts (used in field practices) instead of Na salts, which are recognized to be more inert with interlayer properties of 2:1 clay minerals. Iron salts can have also a potential influence on As(V) sorption. Solutions of 0.2 g L^–1 FeCl₃ and 0.2 g L^–1 FeSO₄ were used to obtain a final concentration of 0.044 kg Fe m^–3 soil.

Once these solutions were applied, the soils were allowed to equilibrate during 4 d in sealed containers. These soils were air dried before carrying out the As adsorption experiments. Solutions of 250, 500, and 1000 µg kg^–1 solution of As(V) were prepared to determine the desorption isotherms, and batch equilibrium experiments were performed following the method described in the adsorption equilibrium section.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS CONCLUSIONS REFERENCES

 
Adsorption Kinetics
Figure 2 shows the As(V) adsorption kinetics for the three soils at 25°C, which can be described following the classic model of two sequential kinetic processes (biphasic kinetics). There is an initial rapid adsorption phase, in which the soil retains most of the solute. In a second phase, the kinetic is much slower, with a smaller amount of the As(V) adsorbed, evolving toward an equilibrium point that cannot be reached within a reasonable experimental time. The 24-h period in which the pseudoequilibrium state was obtained was a key factor, providing an experimentally feasible time frame for our isotherm study. The amount of As adsorbed was considerably higher in the clay soil than in the other soils, which was to be expected given the greater specific adsorption surface of the clay soil. Moreover, due to the small differences on the clay compositions of the soils, the differences on the sorptive behavior of the soils must be attributed to differences in the clay content, which is corroborated by the correlation between CEC and clay content values shown in Table 1.

View larger version (22K): [in this window] [in a new window]   Fig. 2. Evolution of the concentration of dissolved As(V) in the kinetics experiments: clay soil (open circles), sandy clay loam soil (open squares), and loamy sand soil (filled circles). Dashed line represents the time at which the pseudoequilibrium was reached.

An evidence of the negligible role of microbial activity during the experiments can be seen in Fig. 2. Here, the decrease of the concentration after the initial fast sorption (24 h) is negligible when compared with the decrease due to the sorption process (0–24 h). This decrease at t > 24 h could be attributed to microbial activity, but also we expect intra-aggregate diffusion and kinetics sorption processes to take place during this time. Then, although assuming that microbial degradation is the unique mechanism for As(V) disappearance for t > 24 h and that this degradation process starts from t = 0 at an approximately constant rate equal to that observed from t > 24 h, the role of microbial activity on the decrease of concentration within the first 24 h of the experiment (time of the sorption–desorption studies) will be negligible, as was previously assumed.

The pH of the soil–water system progressively moved to slightly lower values during the experiments. However, pH changes were not significant, values being within the range of 6.9 to 6.5, 7.9 to 7.3, and 8.3 to 7.8 for the loamy sand, sandy clay loam, and clay soils, respectively. Although a general decrease on As(V) sorption with pH has been reported in other studies (Smith et al., 1999; Manning and Goldberg, 1996), this effect is only important for relatively high concentrations (beyond of the concentration limits used here) and also for variations of several pH units (Manful et al., 1989). The effect of pH on As(V) sorption in illite and kaolinite was studied by Goldberg (2002). Arsenate sorption on illite (dominant material in the clay fraction of the studied soils) decreased approximately 10% from pH = 6 to 8, while the effect of pH within the same range of variation was not so evident for the As(V) sorption in kaolinite. Based on these results, variations on pH during sorption experiments must not have a critical effect on the extent of sorption in our study. Also, compared with other soil properties, such as texture, the differences on pH of the three used agricultural soils are not large enough to support a role of pH in differing sorptive capacities of the soils.

Adsorption Equilibrium
The results of the adsorption trials at 25°C were fitted to modified linear and Freundlich isotherms (Eq. [2] and [3], respectively). Values of q_i obtained from independent four-step extraction experiments in untreated soils were as follows: q_i = 9234.9 ± 0.6 µg kg^–1 soil for the clay soil, q_i = 1627.2 ± 92.5 µg kg^–1 soil for the sandy clay loam soil, and q_i = 811.4 ± 3.7 µg kg^–1 soil for the loamy sand soil. Figure 3 shows the results for the loamy sand soil as an example. These results have direct practical implications for the modeling of As(V) in vadose zone transport because the model of the sorption isotherm must account for an unrecoverable fraction of solute.

View larger version (16K): [in this window] [in a new window]   Fig. 3. Equilibrium concentrations after the four-step water extraction experiments with the untreated soil for the estimation of qi for the loamy sand soil (811.4 m ± 3.7 µg kg–1). Duplicated experiments generated two points for each extraction step.

View larger version (20K): [in this window] [in a new window]   Fig. 4. Adsorption isotherms of As(V) in soils at 25°C: clay soil (open circles), sandy clay loam soil (open squares), and loamy sand soil (filled circles). The solid lines represent linear isotherms, and the dotted lines, only visible for the clay soil, represent the Freundlich isotherms.

View larger version (14K): [in this window] [in a new window]   Fig. 5. Desorption isotherms at 25°C for the three studied soils.

Table 3 summarizes the results after fitting of the desorption isotherms for the sandy clay loam and loamy sand soils to the nonlinear Freundlich model. Both parameters K_fd, and n_d showed a marked trend to increase with the initial concentrations of the solutions, thus decreasing the value of the exponent 1/n_d. Figure 6 shows the suggested tendency for both parameters, which allows for the establishment of modeling guidelines in solute transport models.

View larger version (18K): [in this window] [in a new window]   Fig. 6. Parameters after fitting the desorption isotherms as a function of the initial concentration. S.C.L., sandy clay loam soil; L.S., loamy sand soil. No appreciable desorption was found in the clay soil.

View larger version (19K): [in this window] [in a new window]   Fig. 7. Effect of temperature on the adsorption of As(V) in the sandy clay loam soil: 15°C (triangles), 25°C (circles), and 40°C (squares).

Effect of the Presence of Other Ions
The sandy clay loam soil was also used for this study. Table 4 shows the results after determination of the As(V) adsorption isotherms in the presence of ions from KNO₃, K₃PO₄, FeSO₄, and FeCl₃. It can be seen that FeCl₃ had a minimal effect on the adsorption of As(V), while KNO₃ had a considerably larger effect (Fig. 8) . Taking into account the flocculent properties of FeCl₃, and given that the quantity of As adsorbed was estimated on the basis of a mass balance, it is uncertain whether there is a true effect on adsorption or if there are precipitation processes taking place. This is interesting since Fe(III) hydroxide precipitates are known to adsorb and/or coprecipitate As(V). Conversely, the presence of K₃PO₄ produced a decrease in As sorption (Fig. 9) , probably due to the fact that As and phosphate compete for common adsorption sites within the soil (Peryea, 1991; Manning and Goldberg, 1996). Also, surface alteration by K⁺ adsorption on 2:1 clays can be responsible of variations in As(V) sorption. Competition between the phosphate and the As(V) as presented in Table 4 was recently described by Violante and Pigna (2002). They established that the quantity of As(V) adsorbed in the presence of phosphate increased with a decrease in the pH. Similarly, Table 4 shows that the presence of FeSO₄ appeared to interfere with the adsorption of As(V). Smith et al. (2002) showed that the presence of certain ions such as phosphates produced some decrease in the adsorption of As, although the effects of other ions including chlorides, nitrates, or sulfates did not appear to have significant effects. Surprisingly, the adsorption of As(V) increased with the presence of nitrate, which is unexplainable in terms of inter-ion competition or sorbent–sorbate interactions. The increase noted may be due to the effect of the nitrate ion on the properties of the solution, particularly its ionic strength. This effect, however should be also similar for other ions. Ionic strength of the salt solutions used in this study are 1.3 x 10^–3, 7.4 x 10^–3, 2.2 x 10^–2, and 7.5 x 10^–2 for FeSO₄, FeCl₃, K₃PO₄, and KNO₃, respectively. These values are two or three orders of magnitude higher than ionic strengths of the As(V) solutions used in the sorption experiments. The increase of the ionic strength is particularly important for KNO₃, which produced an increase on As(V) sorption. Phosphate increase of the ionic strength is also important, but in this case, the competitive sorption seems to be the dominant process influencing As(V) sorption.

View larger version (17K): [in this window] [in a new window]   Fig. 8. Adsorption isotherms of As(V) in the sandy clay loam soil (filled circles) and effect of the presence of KNO3 (open circles).

View larger version (16K): [in this window] [in a new window]   Fig. 9. Arsenate adsorption isotherms in sandy clay loam soil (filled circles) and effect of the presence of K3PO4 (open circles).

	CONCLUSIONS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS CONCLUSIONS REFERENCES

Studies of adsorption for other As species require maintaining the pH and redox potential of the solutions within appropriate intervals to reflect environmentally encountered adsorption conditions. However, control of both pH and Eh can be extremely difficult, and severe modifications of soil pH can influence the soil components involved in the sorption processes.

A prerequisite of this adsorption study was determination of the As(V) fraction irreversibly linked to the soil, operationally defined as the portion of soil As(V) that is not subjected to extraction by water. The isotherms obtained showed a high degree of linearity, and an expected dependence on the specific surface of the sorbent (soil texture), which is directly related to the particle size distribution or, more precisely, to the clay content of the soils. The phyllosilicate fraction of the three soils was composed of 80% illite, and thus, differences on the sorptive capacity of the soils must be explained by clay content, the clay composition playing a negligible role.

Desorption showed, in all cases, a marked hysteresis effect, which was different among soils of different textures. The extent of hysteresis was also different among soils with different textures, being highest in the clay soil. Temperature had a relatively small effect on As adsorption, particularly within the habitual range in the field (10–25°C). Adsorption decreased at the highest temperature (40°C), typical of an exothermic process.

Modeling of As(V) sorption dynamics in the studied soils can be successfully achieved by assuming isotherm linearity, although there is a need for accounting for an unrecoverable fraction of As(V) bounded to the soil q_i. Also, a concentration-dependent hysteresis must be considered for describing desorption in solute transport models. Finally, temperature showed an expected negative effect on sorption (exothermic process). Nevertheless, this effect was only evident in the experiments performed at 40°C, and thus, modeling the temperature effect on sorption must not be a critical factor of the performance of models accounting for As(V) dynamics.

The presence of other ions showed some interesting effects and a way to develop future production strategies in the study area. Nitrate presented an interesting effect in which a decrease in the bioavailability of the As was observed. The opposite effect was found for phosphate. It is unusual to note that the coexistence of raised levels of nitrates in groundwater may help mitigate the As problem, as the nitrate may promote accumulation of As by the soil. Under these conditions the As would be less available for absorption by plants. Conversely, our results also suggest that phosphates introduced in agricultural practice may promote lixiviation of As(V) in soils, and thus their bioavailability, resulting in risks of their being introduced into the food chain.

	ACKNOWLEDGMENTS

 
This study was financed by the Spanish Instituto Nacional de Investigación Agraria y Tecnología de los Alimentos (INIA), project CAL-01-029, and Junta de Castilla y León, Project JCYL 023/03. The authors acknowledge Dr. Louis DiSalvo for his advice with English language. Authors are also grateful to Prof. Jesús Medina (Dep. of Crystallography and Mineralogy, Univ. of Valladolid) for his valuable support in the mineralogical characterization of the soils.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS CONCLUSIONS REFERENCES

 

• AFNOR. 2004. Evaluation de la qualité des sols. Vol. 1. Methods ánalyse chimique. AFNOR, Paris.
• Calvo, C., J. Alvarez-Benedí, M. Andrade, P. Marinero, and S. Bolado. 2003. Arsenic contamination in groundwater in the Province of Valladolid: Seasonal Variations. p 91–98. In J. Álvarez-Benedí and P. Marinero (ed.) Studies on the vadose zone. Vol. VI. (In Spanish.) Inst. Tecnol. Agrario de Castilla y León, Valladolid, Spain.
• Beek, J., and W.H. van Riemsdijk. 1982. Interaction of orthophosphate ions with soil. p. 259–284. In G.H. Bolt (ed.) Soil chemistry. B. Physico-chemical models. Elsevier Science, New York.
• Deuel, L.E., and A.R. Swoboda. 1972. Arsenic solubility in a reduced environment. Soil Sci. Soc. Am. Proc. 36:276–278.
• Gao, S., and R.G. Burau. 1997. Environmental factors affecting rates of arsine evolution from and mineralization of arsenicals in soil. J. Environ. Qual. 26:753–763.[Abstract/Free Full Text]
• Goldberg, S. 2002. Competitive adsorption of arsenate and arsenite on oxides and clay minerals. Soil Sci. Soc. Am. J. 66:413–421.[Abstract/Free Full Text]
• Manful, G.A., M. Verloo, and F. De Spiegeleer. 1989. Arsenate sorption in soils in relation to pH and selected anions. Pedologie 39:55–68.
• Manning, B.A., and S. Goldberg. 1996. Modeling competitive adsorption of arsenate with phosphate and molybdate on oxide minerals. Soil Sci. Soc. Am. J. 60:121–131.[Abstract/Free Full Text]
• Matschullat, J. 2000. Arsenic in the geosphere–A review. Sci. Total Environ. 249:297–312.[CrossRef][Medline]
• McGeehan, S.L. 1996. Arsenic sorption and redox reactions: Relevance to transport and remediation. J. Environ. Sci. Health A31(9):2319–2336.
• McGeehan, S.L., S.E. Fenfodorf, and D.V. Naylor. 1998. Alteration of arsenic sorption in flooded-dried soils. Soil Sci. Soc. Am. J. 62:828–833.[Abstract/Free Full Text]
• McLaren, R.G., R. Naidu, J. Smith, and K.G. Tiller. 1998. Fractionation and distribution of arsenic in soils contaminated by cattle dip. J. Environ. Qual. 27:348–354.[Abstract/Free Full Text]
• Onken, B.M., and L.R. Hossner. 1996. Determination of arsenic species in soil solution under flooded conditions. Soil Sci. Soc. Am. J. 60:1385–1392.[Abstract/Free Full Text]
• Peryea, F.J. 1991. Phosphate-induced release of arsenic from soils contaminated with lead arsenate. Soil Sci. Soc. Am. J. 55:1301–1306.[Abstract/Free Full Text]
• Smith, E., R. Naidu, and A.M. Alston. 1999. Chemistry of arsenic in soils: I. Sorption of arsenate and arsenite by four Australian soils. J. Environ. Qual. 28:1719–1726.[Abstract/Free Full Text]
• Smith, E., R. Naidu, and A.M. Alston. 2002. Chemistry of inorganic arsenic in soils: II. Effect of phosphorus, sodium, and calcium on arsenic sorption. J. Environ. Qual. 31:557–563.[Abstract/Free Full Text]
• Turpeinen R., Pantsar-Kallio, M., and T. Kairesalo. 2002. Role of microbes controlling the speciation and production of arsines in contaminated soils. Sci. Total Environ. 285(1–3):133–145.[Medline]
• Violante, A., and M. Pigna. 2002. Competitive sorption of arsenate and phosphate on different clay minerals and soils. Soil Sci. Soc. Am. J. 66:1788–1796.[Abstract/Free Full Text]
• Wauchope, R.D. 1983. Uptake, translocation and phytotoxicity of arsenic in plants. p. 348–374. In L. Fensterheim (ed.) Arsenic: Industrial, Biomédica, Environmental Perspectives. Arsenic Symposium, Gaithersburg, MD. Van Nostrand Reinhold Company, New York.
• Woolson, E.A., I.J. Axleyand, and P.C. Kearney. 1971. Correlation between soil available soil arsenic estimated by six methods and response to corn (Zea mays L.). Soil Sci. Soc. Am. Proc. 35:101–105.

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via ISI Web of Science (1) Citing Articles via Google Scholar Google Scholar Articles by Álvarez-Benedí, J. Articles by García-Sinovas, D. Search for Related Content PubMed Articles by Álvarez-Benedí, J. Articles by García-Sinovas, D. GeoRef GeoRef Citation Agricola Articles by Álvarez-Benedí, J. Articles by García-Sinovas, D. Related Collections Other Contaminants Sorption/Exchange Soil Pollution