SSSAJ Grow Your Career with SSSA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 27 October 2006
Published in Soil Sci Soc Am J 70:2008-2016 (2006)
DOI: 10.2136/sssaj2005.0385
© 2006 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
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 Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Rocha, W. S. D.
Right arrow Articles by Alleoni, L. R. F.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Rocha, W. S. D.
Right arrow Articles by Alleoni, L. R. F.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Rocha, W. S. D.
Right arrow Articles by Alleoni, L. R. F.
Related Collections
Right arrow Pesticides
Right arrow Organic Compounds
Right arrow Soil Pollution

Soil Chemistry

2,4-D Residues in Aggregates of Tropical Soils as a Function of Water Content

Wadson S. D. Rochaa, Jussara B. Regitanob,* and Luís R. F. Alleonic

a Faculdades Integradas de Rondonópolis, União de Escolas Superiores de Rondonópolis, Rua Floriano Peixoto, 597, CEP 78700-040, Rondonópolis, Mato Grosso, Brazil
b Lab. de Ecotoxicologia, Centro de Energia Nuclear na Agricultura, Univ. de São Paulo, Caixa Postal 97, CEP 13400-970, Piracicaba, São Paulo, Brazil
c Dep. de Solos e Nutrição de Plantas, Escola Superior de Agricultura "Luiz de Queiroz", Univ. de São Paulo, Caixa Postal 09, CEP 13418-900, Piracicaba, São Paulo, Brazil

* Corresponding author (regitano{at}cena.usp.br)


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
The effects of soil moisture on reactions between pesticides and soil aggregates are little understood, particularly in soils from the humid tropics. Thus, the influence of soil water content on retention and mineralization of the herbicide 2,4-D [(2,4-dichlorophenoxy)acetic acid], as well as its distribution on the aggregate fractions (diameter >150, 53 to 150, 20 to 53, 2 to 20, and <2 µm) were evaluated. Soil samples from a Xanthic Dystrudox, a Xanthic Acrudox, and two Rhodic Acrudoxes were collected and moisture was adjusted to 25, 50, and 75% maximum water holding capacity. 14C 2,4-D was then applied and the flasks incubated in a semi-dark, climatized room at 22 ± 2°C, for 42 d. 2,4-D mineralization was determined weekly. After three extractions with a 0.01 mol L–1 CaCl2 solution, soil aggregates were sonicated using water as a dispersing agent. 2,4-D retention increased with soil moisture. Higher soil moisture favored 2,4-D retention in aggregates < 20 µm, ratifying the hypothesis of 2,4-D diffusion to less accessible soil sites. 2,4-D mineralization was also enhanced with soil moisture. For the Acrudoxes, in which the clay fraction was less dispersed in water, 2,4-D retention was higher in larger aggregates (>53 µm), having higher total organic C (OC) content and higher degree of humification of its organic matter (OM).

Abbreviations: LSS, liquid scintillation spectrometry • MWHC, maximum water holding capacity • OC, organic C • OM, organic matter • SSA, specific surface area • WDC, water dispersible clay


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
ASSOCIATIONS AMONG CLAY MINERALS, Fe and Al oxides, OM, and other soil components result in aggregates of different sizes (Tisdall and Oades, 1982), whose composition depends on soil management practices that directly affect the retention of pesticides (Celis et al., 1999; Alves et al., 2004; Nicolella et al., 2005). In most cases, pesticide retention is primarily associated with aggregates that contain larger amounts of OC (Nkedi-Kizza et al., 1983; Schulten and Leinweber, 2000; DeSutter et al., 2003). Soil OM directly affects formation and stability of soil aggregates (Ristori et al., 1992) and also affects pesticide sorption. Evaluation of pesticides fate in soils is usually performed under laboratory conditions, in soil aggregates with diameters < 2 mm (Wu et al., 2000; Koskinen et al., 2002; Landry et al., 2004), but not much is known about the effects of moisture on pesticide biodegradation and its redistribution into differently sized soil aggregates, especially in the humid environment of the tropics.

The use of water as a dispersing agent (Christensen, 1992) and soil physical fractionation are better tools to evaluate pesticide sorption in soils than chemical procedures, such as soil OM fractionation (Barriuso and Koskinen, 1996). The soil physical fractionation keeps the aggregate integrity during separation and, therefore, supplies a good insight on pesticide behavior. In addition, the less humified OM, usually associated with the coarser soil fraction, may be separated from the more humified OM, that is usually associated with the fraction < 50 µm (Barriuso and Koskinen, 1996). This smaller fraction contains aggregates more strongly bound by Fe and Al oxides (Tisdall and Oades, 1982) and has organic components resulting from more intense process of microbial decomposition, which generates high proportion of aromatic structures (Chefetz et al., 2002).

Soil water content affects primarily microbial activity, oxygen availability, and pesticide sorption and transport (or diffusion) in soils. As a consequence, these processes influence pesticide turnover rates in soils. In general, the presence of water favors microbial activity and pesticide diffusion and subsequent sorption into hydrophilic sites within the aggregates (Roy et al., 2000). Therefore, pesticide degradation is usually lower in soils with low water potentials (dry soils) than in wet soils, and it is mainly due to the low microbial activity under this circumstance (Griffin, 1981). The lack of water may hinder molecule transport to biodegrading soil microbes. At higher water potentials (wet soil), pesticide availability is usually reported as the rate-limiting factor for degradation. However, at high water potentials (saturated or very wet soil), the water excess may limit oxygen availability, precluding aerobic biodegradation (Helweg, 1987; Flint and Witt, 1997).

Historically, matric potential has been considered an important measure of soil water availability, but soil water content (measured as water-filled pore space) appears to be a more useful parameter when water availability is not limiting microbial activity (Skopp et al., 1990). Recently, however, some authors have advocated the use of conceptual models including soil type dependent expressions for solute and gaseous diffusivity to predict transport of organic chemicals (Moldrup et al., 2000; Olesen et al., 2001a, 2001b) and microbial activity (Schjønning et al., 2003). These models postulate that for sorbing organic chemicals, such as most pesticides, the effective diffusion rate is determined partly by the diffusion properties and partly by the sorption of the substrate in the gaseous and liquid phases (Olesen et al., 2001a).

Soil moisture at the time of application is another factor influencing pesticide sorption. Pesticide application on dry soils may result in strong sorption initially whereas application on wet soils leads to stronger sorption only after long soil-pesticide contact time (aging) (Gaillardon and Dur, 1995; Gaillardon, 1996). In the last case, diffusion should dictate molecule transport to soil micropores and/or to soil organic matrices, since it is a time-dependent mechanism (Brusseau et al., 1991). Alexander (1995) suggested that sorption enhancement with residence time is due to molecule sequestration in less accessible sorption microsites within the soil matrix. The effects of soil moisture are even more difficult to predict for ionizable molecules (Roy et al., 2000). It is believed that water promotes diffusion of anions to less accessible sorption sites within soil aggregates (Koskinen et al., 2003).

The herbicide 2,4-D is widely used to control broad leaf weeds in sugarcane (Saccharum officinarum L.). Sugarcane is grown on about 6 millions ha in Brazil and is highly important for the production of sugar and alcohol (a renewable energy source as a fuel). 2,4-D is a weak acid herbicide (pKa = 2.8; Johnson et al., 1995) which usually undergoes deprotonation in agricultural soils. Therefore, electrostatic (hydrophilic) forces also affect its sorption potential (Grover and Smith, 1974) in addition to hydrophobic interactions.

The objectives of this research were to evaluate the influence of soil moisture on the mineralization and retention of 2,4-D and to determine the distribution of its residues in the organomineral aggregates (diameter >150, from 53 to 150, from 21 to 53, from 2 to 20, and <2 µm) of four very representative soils of the tropics (all Oxisols), with different OM contents.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Soil samples from the 0- to 0.2-m layer of a Xanthic Dystrudox (LAd), located in Piracicaba, São Paulo (22°42' S and 47°38' W), a Xanthic Acrudox (LAw) and a Rhodic Acrudox (LVw-1), located in Uberlândia, Minas Gerais (19°54' S and 48°16' W), and another Rhodic Acrudox (LVw-2), located in Canoinhas, Santa Catarina (26°10' S and 50°24' W) were collected from sites with no history of 2,4-D application. Oxisols represent about 60% of the Brazilian territory (more than 5 million km2), in areas of great agricultural importance.

The soil OC was measured after oxidation of the soil OM with a potassium dichromate solution in the presence of sulfuric acid, and the excess of dichromate was titrated with ferrous ammonium sulfate. The content of Fe and Al oxides were measured after addition of 18 mol L–1 H2SO4 (Vettori, 1969). Particle-size analysis was performed according to the densimeter method, where the clay, silt, and sand particles were chemically dispersed with sodium hexametaphosphate and 0.01 mol L–1 NaOH (Camargo et al., 1986). Water dispersible clay (WDC) content was measured in the absence of electrolyte, with a Wiegner agitator. After sieving the sand, silt and clay fractions were separated by sedimentation (Camargo et al., 1986). The specific surface area (SSA) was quantified by the ethylene glycol monoethyl ether (EGME) method (Cihacek and Bremner, 1979), by oven-drying the soil at 110°C for 24 h. The oven-dry method was adopted because it is faster and linearly correlated to P2O5 method (r = 0.99, Ratner-Zohar et al., 1983). Available P was extracted by an ionic exchange resin (Murphy and Riley, 1962), and determined by phosphomolibidate colorimetry. The exchangeable cations (Ca, Mg, and K) were also extracted by an ionic exchange resin (Raij et al., 1986), but were quantified by absorption spectrophotometry and flame photometry. The exchangeable Al was extracted with 1.0 mol L–1 KCl (Barnihisel and Bertsch, 1982) and quantified by titration with 0.1 mol L–1 NaOH, whereas H+Al was extracted with 1.0 mol L–1 calcium acetate (pH 7.0), but also quantified by titillation with 0.1 mol L–1 NaOH (Camargo et al., 1986). Soil-pH was determined in deionized water, using a 1:2.5 soil/solution ratio. The zero point of salt effect (ZPSE) was determined by the titration method, using two concentrations of KCl (0.05 and 0.002 mol L–1). The ZPSE corresponded to the pH value when {Delta}pH = 0 (Camargo et al., 1986). The electric surface potential ({Psi}0) was calculated by the Nernst equation [{Psi}0 = 59.1 (ZPSE – pH)] (Singh and Uehara, 1999). Soil physicochemical attributes are presented on Table 1.


View this table:
[in this window]
[in a new window]
  		Table 1. Some attributes of the studied soils.{dagger}

 
For each treatment, duplicate 50-g soil samples (air dried) were placed into 300-mL Bartha flasks, and sufficient water was added to reach 25, 50, and 75% of maximum water holding capacity (MWHC). The MWHC represents the amount of water remaining in the soil after draining all free-water in soil pores and it was measured according to Luchese et al. (2001). The MWHC for the LAd, LAw, LVw-1, and LVw-2 corresponded to 0.220, 0.356, 0.349, and 0.305 g g–1, respectively. After 1 wk of incubation, a solution of 2,4-D (Acid-2,4-D, purity = 99% and 14C-2,4-D, purity > 98% and specific radioactivity = 2.51 MBq mg–1) was applied at the maximum rate recommended for sugarcane (2.7 kg a.i. ha–1, which corresponded to 2.25 µg a.i. g–1 of soil and to 0.69 kBq g–1 of soil) (Rodrigues and Almeida, 1995). Then, the treated soil samples were incubated for 42 d in flasks wrapped in aluminum foil to avoid undesired photolysis, at 22 ± 2°C. The rate of 2,4-D mineralization was monitored at 7-d intervals, by capturing 14C-CO2 into 10 mL of 0.2 mol L–1 NaOH (Koskinen et al., 2003). The radioactive concentration of this solution was determined by liquid scintillation spectrometry (LSS), using the PACKARD 1600 TR instrument (Packard Instrument, Meriden, CT) equipped with calibration curve for quench correction. The initial soil microbial activity was obtained in parallel flasks through the radiorespirometry method, using 14C-glucose (Freitas et al., 1979).

After incubation, the 50-g soil samples were placed into 250-mL centrifuge tubes and then extracted with 150 mL of 0.01 mol L–1 CaCl2. The tubes were agitated for 40 min at 140 rpm. Soil suspensions were then centrifuged at 12100 x g for 20 min, and 1-mL aliquots of the supernatants were taken to determine concentration of 14C-2,4-D residues in solution, by LSS. This extraction procedure was repeated twice more. The identity of the measured radioactivity was checked by an automatic TLC-Linear Analyzer (BERTHOLD, Berlin, Germany), and it showed that only 2,4-D was found (data not shown). The soil slurries were air dried and subsamples combusted in a Biological Oxidizer OX500 (HARVEY Instruments, Hillsdale, NJ) to determine the amount of 14C-2,4-D residues remaining sorbed in the soils. For the whole experiment, recovery corresponded to 93.5 ± 9.4% of the applied radioactivity.

For the physical fractionation, 40 g soil samples (air dried) and 160 mL distilled water were placed into 250-mL centrifuge tubes and agitated for 16 h in a horizontal shaker. The tubes were later placed in a sonication bath at 240 W, for 15-min (Christensen, 1992). After dispersion, the soil-suspensions were passed through 150- and 53-µm mesh sieves, separating aggregates larger than 150 µm and those with diameters from 53 to 150 µm. The remaining suspensions were transferred to a 1000-mL graduated cylinder, maintained in a semi-dark room at 22 ± 2°C. The further aggregate fractionation, in sizes from 20 to 53, from 2 to 20, and <2 µm, was based on sedimentation time estimated according to Stokes' Law (Christensen, 1992). After fractionation, the aggregates of different sizes were oven dried (40 ± 2°C), and the amounts of 14C-2,4-D residues were determined by LSS, after combustion and capture of the released 14CO2. The total C content in the aggregates was quantified in a LECO CR 412 (Leco corp., St. Joseph, MI) and assumed to be all organic.

The degree of OM humification in the soil samples and aggregates was determined by laser induced fluorescence (LIF). A laser beam (wavelength = 351 nm and power = 0.248 W) collided over the sample. The emitted fluorescence was captured by a photomultiamplifier (tension = 839 V). The intensity readings resulted in a spectrum whose area allowed calculating the degree of OM humification. The spectrum area was normalized to the OC content of the samples (Milori et al., 2002a).

Analysis of variance was performed with resulting data set to verify the treatment effects (soil type x aggregate size x moisture) on sorption, extractability, and mineralization of 2,4-D. If interactions were found between treatments, then mean comparisons were made by the Tukey test (P < 0.05).


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Organomineral Aggregates and Organic Carbon Distribution
The size of sand, silt, and clay fractions varied with the dispersing agent adopted. The ratio between particle contents dispersed in water and those dispersed in sodium hydroxide and sodium hexametaphosphate (WD/CD) was higher for the clay and smaller for the sand and silt fractions in the LAd when compared with the other soils (Table 2). It suggests that sand and silt particles of the LAd were less coated with clays. Conversely, the sand and silt fractions of the other soils were expected to be coated mainly with Fe and Al oxides, very abundant in the Acrudoxes (Table 1).


View this table:
[in this window]
[in a new window]
  		Table 2. Water dispersible (WD) and chemical dispersible (CD) soil fractions and respective ratios (WD/CD).

 
The smaller the Fe2O3 content (Yu, 1997) and the more negative the {Psi}0 (Alleoni and Camargo, 1994), the higher the amount of WDC, as was the case for the LAd. The negative correlation between {Psi}0 x Fe2O3 product and WDC content (r = –0.96*, P < 0.05) ratified this observation. In general, the higher the WDC content, the lower the sand particles coatings.

The amount of aggregates was higher in the fractions > 150 µm (Table 3); however, the second highest fraction of aggregates varied among the soils. The soils with higher OM contents had lower amounts of aggregates < 2 µm (r = –0.93, P < 0.10) (Tables 1 and 3). It reinforces that OM acts as a strong soil-cementing agent, reducing the dispersion of soil particles (Tisdall and Oades, 1982). The positive and significant correlation between the amounts of WDC and aggregates < 2 µm (r = 0.93, P < 0.10) suggested that the lower the soil OM content, the higher the amounts of WDC and aggregates < 2 µm.


View this table:
[in this window]
[in a new window]
  		Table 3. Size distribution of organomineral aggregates.

 
The OC content (g kg–1 of aggregate) was higher for aggregates < 20 µm, that is, for the silt and clay fractions (Table 4), as was also observed by Christensen and Sorensen (1985), Monrozier et al. (1991), Ristori et al. (1992), Barriuso and Koskinen (1996). In these smaller aggregates, soil OM likely remains protected from biological decomposition due to their incorporation inside the microaggregates (Bayer et al., 2002). On the other hand, aggregates > 150 µm had higher total amounts of OC (g kg–1 of soil) (Table 5), because the total mass of these particles was larger than that of the smaller fractions (Table 3). Soil OM content directly affects the amount of 2,4-D sorbed (Johnson et al., 1995; Prado et al., 2001). So, the aggregates > 150 µm should have more hydrophobic-sites available for 2,4-D retention.


View this table:
[in this window]
[in a new window]
  		Table 4. Organic C contents in the aggregate fractions (g kg–1 of aggregate).

 

View this table:
[in this window]
[in a new window]
  		Table 5. Total organic C contents in the aggregate fractions (g kg–1 of soil).

 
Degree of Organic Matter Humification in the Aggregates
The LAd showed a higher degree of OM humification than the LVw-1 and the LVw-2 (Table 6). The LAd was collected from a cropped field, where many macroaggregates in the surface layer were broken down due to the tillage. Under this condition, OM becomes more exposed to oxidation, resulting in further degradation and higher degree of humification (Tisdall and Oades, 1982). Soils in non-cropped areas, such as the LVw-2, usually have higher soil OM contents with a lower degree of humification (Table 6). In general, the OM turnover and its degree of humification are greater in intensively cultivated soils (Bayer et al., 2002; Milori et al., 2002b).


View this table:
[in this window]
[in a new window]
  		Table 6. Degree of organic matter humification in the aggregate fractions.

 
The position in the landscape may also influence the soil's degree of humification. The LAw and LVw-1 were collected from the apex of the South American Plateau, but the LAw was located at a subsidence area where water accumulation was frequent. Under these conditions, Fe can be reduced, solubilized, and transported into other areas, explaining the smaller Fe2O3 content in the LAw in relation to the LVw-1 (Table 1). Most likely, the lower content of Fe available to cement the aggregates in the LAw (Table 1) resulted in higher rate of OM oxidation and higher degree of humification (Table 6). In general, soil particles and Fe and Al oxides are strongly attached in highly weathered soils, reducing their exposure to humification (Tisdall and Oades, 1982). On the other hand, OM decomposition is faster in well drained soils from humid tropical regions, which may favor the breakdown of larger aggregates (Bayer et al., 2002).

Soil Water Content and 2,4-D Retention
Retention of 2,4-D was lower in the LAd (Table 7), which had lower OM content, lower SSA, and higher pH (Table 1). Conversely, 2,4-D retention was higher in the LVw-2, the soil with the highest OM and SSA. In general, the higher the OM content, the higher the 2,4-D retention (r = 0.94, P < 0.10). This confirms the importance of OM and hydrophobic interactions on the fate of this herbicide, even in tropical soils. In parallel, OM and SSA showed a positive and highly significant correlation (r = 0.98, P < 0.05) (Table 1), reinforcing the evidence that OM and pesticides bind mainly to the external physical surface of soil particles, while the role of the internal surface sites is less important (Schulten and Leinweber, 2000).


View this table:
[in this window]
[in a new window]
  		Table 7. Soil water contents, sorption and extraction of 14C-2.4-D residues.

 
In addition to OM and SSA, pH is also relevant on sorption of 2,4-D (Johnson et al., 1995), since 2,4-D is an ionizable organic herbicide that undergoes dissociation at pH values above pKa (2.8). Thus, dissociation of carboxylic groups from the 2,4-D should be greater in the LAd, which had a higher pH value (Table 1).These groups may undergo electrostatic repulsion because of the greater amount of negative charges present in the colloids of this soil (see {Psi}0, Table 1). This behavior is similar to that of other ionizable herbicides (Regitano et al., 2001, 2005; Rocha et al., 2002).

As the water content increased, the amount of 2,4-D extracted from the soils by CaCl2 decreased and the amount sorbed increased (Table 7). This trend was masked by the intense 2,4-D mineralization in the LAd, at 75% of MWHC, in which 62.2% of the applied amount was mineralized. Sorption depends on soil and pesticide properties, as well as on pesticide diffusion in the soil-water phase, and thus on water content of the soil (Olesen et al., 2001a). Therefore, soil water content becomes a limiting factor for sorption reactions when considering the same soil and the same pesticide molecule.

The extent of soil water content impact on 2,4-D retention depends on sorption rates at fast and slow phases (Kookana et al., 1992; Pignatello and Xing, 1996). Low soil moisture content (or dry soil) favors pesticide sorption initially (Gaillardon and Dur, 1995), which could be associated with fast phase sorption to readily available external sites. On the other hand, high soil moisture content (or wet soil) favors its sorption at longer run (Gaillardon, 1996), which could be associated with slow phase sorption due to diffusion-sequestration in micropores and/or organic matrices (Pignatello and Xing, 1996; Nam and Alexander, 1998). Diffusion seems to be the main process responsible for the increase in 2,4-D retention under higher moisture conditions (Gaillardon, 1996; Roy et al., 2000).

As for other ionizable pesticides, there must be sufficient water in the soil in order for 2,4-D molecules to diffuse to the retention sites (Roy et al., 2000). However, the magnitude of the diffusion process also depends on the physicochemical properties of the molecule (Gaillardon, 1996). At pH values commonly found in tropical soils (4.5–6.0), 2,4-D molecules are in the anionic form and must be transported to hydrophobic sites of OM and/or to positively charged soil sites, mainly to the Fe and Al oxides present in these soils.

Soil Water Content and 2,4-D Retention into Soil Aggregates
Haberhauer et al. (2000), de Jonge et al. (2000), and DeSutter et al. (2003) pointed out that pesticide sorption occurs mainly in smaller, silt- and clay-sized aggregates, with higher OC contents. Notwithstanding, this was only observed in the LAd, in which most of the 2,4-D was retained in aggregates < 2 µm (Table 8). In the Acrudoxes, most of the 2,4-D was retained in aggregates > 150 µm, followed by the fraction between 53 and 150 µm. It was already mentioned that the aggregates > 150 µm showed higher total OC amounts (Table 5). Differently from the OC content in the aggregates (Table 4), the total amount of OC expresses the total number of sites available for herbicide retention, especially the hydrophobic sites (de Jonge et al., 2000; Schulten and Leinweber, 2000; DeSutter et al., 2003). The discrepancy of our results was likely due to the different nature of the soils studied mainly that of the Acrudoxes, which represent the extreme degree of weathering and are very rich in Fe oxides, which assures strong cementation among soil particles. Another reason may be the adoption of a less harsh separation method, since water instead of sodium hydroxide and/or sodium hexametaphosphate solution was used as dispersant agent.


View this table:
[in this window]
[in a new window]
  		Table 8. Soil water contents and the distribution of 14C-2.4-D residues in the aggregate fractions.

 
Pesticide retention in the larger aggregates was also verified for atrazine in a soil with 41.1 g kg–1 OC and 210 g kg–1 clay (Barriuso and Koskinen, 1996) and for isoproturon in a soil with 21.8 g kg–1 OC and 267 g kg–1 clay (Benoit et al., 2000). Barriuso and Koskinen (1996) speculated that atrazine bioaccumulated in soil microorganisms, mostly in actinomycetes and fungi located in soil fractions larger than 50 µm (Tisdall and Oades, 1982). These authors even mentioned that certain fungi can absorb large amounts of atrazine, reducing its degradation within the fungus.

The total amount of OC in aggregates <2 µm and >150 µm were nearly the same in the LAd (Table 5). Nevertheless, 2,4-D retention was higher in the smaller aggregates (Table 8). This was expected since sorption of weak organic acids, such as 2,4-D, depends considerably on the SSA (Celis et al., 1999), which is larger in clay-sized aggregates. This higher SSA for smaller aggregates resulted in greater exposure of cationic and hydrophobic sites in which the herbicide can be sorbed. Haberhauer et al. (2000) reported that the sorption of certain acid pesticides (2,4-D, MCPA, mecoprop, and dichlorprop) was higher in the clay fraction than in the silt and sand fractions for sandy and medium-textured soils. It is well known that the smaller the OC content of a soil, the greater the importance of the clay fraction for organic molecule retention (de Jonge et al., 2000; Regitano et al., 2000). The contribution of the clay fraction becomes more evident when the ratio clay/OC is higher than 40 (Karickhoff, 1984; Green and Karickhoff, 1990).

In this study, it was also evaluated which soil attributes were correlated to 2,4-D retention in the soil aggregate fractions. A positive correlation between 2,4-D retention and the product between the total amount of OC (Table 5) and the degree of humification (Table 6) was observed for the Acrudoxes: LAw (r = 0.86, P < 0.10); LVw-1 (r = 0.99, P < 0.005); and LVw-2 (r = 0.93, P < 0.05). So, one can conclude that 2,4-D retention in these highly weathered soils was higher in aggregates containing higher amounts of OC and higher degree of humification. No correlation was observed for the LAd.

Soil water content affected 2,4-D retention in the aggregates (Table 8). In general, 2,4-D retention decreased in aggregates >150 µm and increased in aggregates <20 µm as soil water content increased (Table 8). This confirms the hypothesis that water at adequate quantities is crucial for 2,4-D molecules to diffuse into smaller aggregates, favoring its transport and sorption/sequestration in inaccessible microsites within the soil matrix, mainly the hydrophobic sites (Brusseau et al., 1991; Pignatello and Xing, 1996; Nam and Alexander, 1998). For ionizable molecules, the polarity of the ionizable moiety defines its sorption to hydrophilic (polar or electrostatic) sites, while the other structures of the molecule define its sorption to hydrophobic (apolar) sites. The dynamics of these interactions and the spatial orientation of these molecules are highly influenced by enhancing soil moisture (Regitano et al., 1997, 2001; Roy et al., 2000).

Soil Water Content and 2,4-D Mineralization
A negative correlation was observed between OM content and microbial activity of the soils (r = –0.97, P < 0.05) (Tables 1 and 9). In this case, the higher the microbial activity, the higher the OM decomposition and, consequently, the lower its accumulation in the soil. Thus, microbial activity was higher in the LAd and lower in the LVw-2 (Table 9), corresponding to the soils with the lowest and the highest OM contents, respectively (Table 1). The smaller microbial activity in the LVw-2 can be explained by the lower temperature in the region where it was collected.


View this table:
[in this window]
[in a new window]
  		Table 9. Soil water contents, amounts of 2.4-D mineralized and initial values of microbial activity.

 
This study showed the influence of soil chemical composition, or soil fertility, on soil microbial activity and therefore on pesticide biodegradation rate (Rigobelo and Nahas, 2004). For example, the amount of 2,4-D mineralized was positively correlated with pH values (r = 0.97, P < 0.05), sum of bases (r = 0.99, P < 0.01), and P contents (r = 0.91, P < 0.10), while a negative correlation was observed with Al saturation (r = –0.99, P < 0.001). Therefore, the higher the soil fertility (which is the case for the LAd), the higher the microbial activity and the mineralization of 2,4-D. The LAd was limed and fertilized in previous years, which can be verified by its higher pH, higher P and basic cation contents, higher base saturation (V%), and smaller Al saturation (m%) in relation to the other soils (Table 1). It is important to point out that 2,4-D serves as a source of OC to soil microorganisms and is readily biodegraded (Cupples et al., 2000).

In addition to chemical attributes, soil physical attributes can also influence 2,4-D mineralization. As a matter of fact, both volumetric soil-water and soil-air contents will influence aerobic pesticide degradation since it will depend on pesticide diffusion in the soil-water phase and on oxygen diffusion in soil-water and soil-air phases (Schjønning et al., 2003). The use of disturbed soil samples in this research ruled out, at least in great part, the importance of evaluating these parameters. However, it can be speculated that LAd will have the largest air-filled porosity and the best aeration since it has the highest sand content (Table 1), favoring oxygen transport and microbial activity. As a consequence, this soil showed higher mineralization rate than the other soils (62.2% of the applied amount at 75% of MWHC, Table 9). Its higher fertility corroborated it. Kristensen et al. (2001) verified that 76% of applied 2,4-D was also mineralized from a sandy soil.

Sorption reactions also tend to decrease pesticide availability, thus reducing its effective diffusion and its biodegradation rate in soils. For instance, Olesen et al. (2001a) showed that hydrophobic sorption decreased by two orders of magnitude the effective diffusion of lindane. This helps to explain the higher 2,4-D mineralization rate in the LAd, the soil with the lowest sorption potential, and also the lower 2,4-D mineralization rate in the LVw-2, the soil with the highest sorption potential (Table 7).

The rate of 2,4-D mineralization increased with soil water content (Table 9). The same was verified for cloransulam-methyl in a Typic Endoaquoll (Cupples et al., 2000) and for metolachlor in a Typic Haplaquoll and an Aquic Hapludoll (Rice et al., 2002). On one hand, it was not expected since higher soil water content enhanced 2,4-D sorption and it should have decreased its bioavailability. However, it may have been offset by the ease biodegrading nature and the low binding strength involved in the sorption of 2,4-D, which may favors desorption. On the other hand, solute diffusion in the soil-water phase should be the main process regulating aerobic microbial activity since oxygen diffusion in soil-air phase should not be limited in these soils even at higher water contents, considering that 75% of MWHC and disturbed soil samples were adopted. In summary, higher soil water contents should have assured favorable conditions for microbial proliferation and for transport, probably by diffusion, of 2,4-D molecules to regions where microbial colonies were located (Johnson et al., 1995). However, water must not be in excess in the soils because this could reduce or even eliminate available oxygen, which is essential for aerobic microorganism action (Flint and Witt, 1997).


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
In general, 2,4-D retention and its mineralization increased with soil moisture.

In the Acrudoxes (LAw, LVw-1, and LVw-2), where water dispersed clay contents were smaller, 2,4-D retention was higher in larger aggregates (>150 µm). In the LAd, where clay was more dispersed, retention was higher in the smaller aggregates (<20 µm).

Increased soil moisture allowed 2,4-D diffusion into smaller aggregates (<20 µm), especially in the Acrudox samples.

In the Acrudoxes, 2,4-D retention was higher in samples with higher amounts of OC and higher degrees of OM humification.


   ACKNOWLEDGMENTS
 
To the Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP), Brazil for the scholarship to the first author (Project # 00/08370-1). To the Embrapa Instrumentação Agropecuária, São Carlos-SP, Brazil to make available their research facilities.

Received for publication November 29, 2005.


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





This Article
Right arrow Abstract Freely available
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 Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Rocha, W. S. D.
Right arrow Articles by Alleoni, L. R. F.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Rocha, W. S. D.
Right arrow Articles by Alleoni, L. R. F.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Rocha, W. S. D.
Right arrow Articles by Alleoni, L. R. F.
Related Collections
Right arrow Pesticides
Right arrow Organic Compounds
Right arrow Soil Pollution

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