This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
The aim of this study was to evaluate the release kinetics, speciation, and fractionation of boron in some calcareous soils of western Iran. Ten surface soil samples were incubated with 100 mg B kg-1 for a week at field capacity moisture. After air drying of samples, the trend of B release was experimented using sequential extraction with 10 mM CaCl2. B speciation in soil solution was calculated for the first and the last steps of extraction by visual MINTEQ program. The distribution of B among five fractions including exchangeable (F1), specially adsorbed (F2), bound by Fe-Mn oxides (F3), organically bound (F4) and residual (F5), was determined in control and spiked soils. The results indicated that the release rates were initially rapid followed by a slower reaction and the main proportion of the added boron was extracted by CaCl2. The Release kinetics of B was described well with Elovich, Parabolic diffusion, Power function, and first order equations. The Speciation results revealed that the uncharged boric acid (H3BO30) was the dominant species in soil solutions. In control soils, B concentration in different fractions decreased in the following order: F5 > F1 > F2 > F3 > F4. In spiked soils, however, the largest and the smallest fractions were exchangeable and residual, respectively. This implies that B transformation from soluble to less mobile and non-labile forms is not a rapid process and requires longer time than a week. The significant relationship observed between kinetic parameters of power and parabolic equations and organically bound B fraction and OM content indicated that organic matter played an important role in B adsorption and release in calcareous soils.
Key words boron; kinetics; speciation; fractionation
Plants need boron (B) as one of the micronutrient elements in very small quantities. In general, there is a small concentration range between deficiency and toxicity of B as compared with other nutrient elements. As B concentration in soil solution is generally low, boron deficiency symptoms are common in humid or rainy regions due to the B leaching with high amounts of water (Goldberg 1997). Keren and Bingham (1985) reported that B deficiency could lessen in clay soils because of more adsorption of B in contrast to sandy soils. On the other side, it has been indicated that B toxicity occurs in arid areas because of accumulation of B in soil and presence of high amounts of B in the irrigation water (Nable et al. 1997).
As boron has stable redox status as well as it is unable to lose from soils by volatilization process, the chemical properties of B in soils are very simpler than the other nutrients. The boric acid as the dominant form of B in the most of soil solutions at 3 < pH < 9 is a very weak and monobasic acid, which behaves as a Lewis acid by accepting a hydroxyl ion to form the borate anion at pH > 9 (Goldberg 1997). Goldberg (1993) investigated the chemistry and mineralogy of boron in soils and found that the major controlling processes of B solubility in soils are B adsorption reactions not dissolution or precipitation of boron containing minerals whether very insoluble (tourmaline) or very soluble (hydrated B minerals) ones.
The water extractable boron that can easily uptake by plants refers to labile fraction in the soil solution. The less labile fraction comprises non-specifically adsorbed B on layer silicate clay minerals (Goldberg et al. 1993), specifically adsorbed B on variable charged oxides and (oxy) hydroxides (Goldberg and Glaubig 1985; Su and Suarez 1995), organic B complexes (Van Duin et al. 1985; Gu and Lowe 1990) and adsorbed B on carbonate minerals (Goldberg and Forster 1991). Furthermore, B can isomorphously substitute trivalent cations in (oxy) hydroxides or Si4+ in the tetrahedral sheets of phyllosilicate clays (Hingston 1964). This fraction in addition to B in primary minerals, such as tourmaline, or secondary minerals e.g. colemanite is considered to be a non-labile form (Hou et al. 1996). The main mechanism of specific adsorption of B on mineral and organic surfaces is ligand exchange between boron and OH- groups and formation of covalent bounds with the structural cations. As the sign of the net surface charge does not influence the specific adsorption (Hingston et al. 1972), both boron species involving B(OH)30 and B(OH)4¯ can include in this process. It has been reported that broken edges of clay particles are the most significant sites for B adsorption (Couch and Grim 1968; Bingham and Page 1971; Keren et al. 1981).
The Chemical fractionation schemes have been widely used to partition the trace elements in soils and sediments as different fractions. Tessier et al. (1979), Harrison et al. (1981), Shuman (1985), Miller et al. (1986), Kheboian and Bauer (1987), Shan and Chen (1993) and Salbu et al. (1998) have developed fractionation methods for trace elements. Every scheme involves segmentation of different fractions through successive extraction by specific reagents. Although the various methods often differ in the reagents used, the application order of extractants for the organic fraction, and other functional details, such as solid to solution ratios and treatment time (Hou et al. 1996), the distinct fractions are similar in various schemes and include soluble species, non-specifically and specifically adsorbed forms, organically bound forms and residual fractions. The least forceful reagents are used first in sequential extraction schemes and successive extractants have generally decreasing pH values. In other words, the considered elements are divided into different fraction in weak to strong order. However, these schemes have been widely criticized and the great variety of approaches that have been developed reflects the complexity of the problems including lack of uniformity in the procedures, lack of selectivity of the reagents used, lack of quality control, results highly dependent on the procedure used, and etc. Despite of these limitations, the sequential extraction schemes remain widely used and are considered as essential tools in establishing element fractionation in soils and sediments (Gleyzes et al. 2002).
The various soil chemical properties including soil pH, clay mineralogy and the contents of Al and Fe oxides and hydroxides, carbonates and organic matter have significant effects on mobility, transport and distribution of B in soils (Keren and Bingham 1985; Keren et al. 1985a). To assess the agricultural and environmental impacts of B, it is necessary to understand and quantify the different forms of B in soils and sediments. Since B release kinetics has not been still considered in calcareous soils and on the other hand, the evaluation of B fractionation in soils is of great practical interest in crop nutrition and in monitoring water quality near agricultural areas, this study has been conducted to evaluate release kinetics, fractionation and availability of B in some calcareous soils under different land use types.
Materials and Methods
Ten surface soils (0-30 cm) were collected from areas covered with different land uses in Hamadan, western Iran. Soil samples were air dried and ground to pass through a 2-mm sieve for laboratory experiments. Particle size distribution was determined by the hydrometer method (Rowell 1994). Soil pH and EC were measured in H2O using a 1:5 soil to solution ratio, after the soil suspension had been equilibrated for 1 h on a shaker. The organic matter (OM) was determined by dichromate oxidation (Rowell 1994). The equivalent calcium carbonate (ECC) and cation exchange capacity (CEC) were determined by neutralization with HCl and by using NH4OAc pH 7, respectively (Rowell 1994). The oxalate-extractable iron and aluminum were extracted with 0.175 M ammonium oxalate + 0.1 M oxalic acid adjusted to pH 3.0 (Pansu and Gautheyrous 2006). Iron and Al were determined in the extract using Varian, SpectrAA-400 and spectrocolorimetric method (Pansu and Gautheyrous 2006), respectively. Total concentration of boron was determined after digestion of 1 g of each soil sample in a mixture of 5 mL of nitric acid and 1 mL of perchloric acid in a conical flask and heated on a hot plate at 60°C for 6 h. After evaporation, 1 mL of 2 M HNO3 was added and the residue after dissolution was filtered into a 100 mL volumetric flask and diluted with distilled water (Tessier et al. 1979). The supernatants were analyzed for B using the colorimetric Azomethine-H method (Bingham 1982). pH values ranging from 4.5 to 7.3 have been reported for optimum color development. The time required for complete color development has been reported from less than 30 min to well over two hours (Bingham 1982). Furthermore, Saba and Singh (1997) indicated that temperature of sample solutions should be maintained constant between 15 and 20°C throughout the period of color development and absorbance measurement for higher sensitivity and accuracy of B determination. Therefore, we considered all of these points in B measurements by Azomethin-H method. The total B concentration in soil samples was also measured using fusion method (Pansu and Gautheyrous 2006). Since the difference between concentrations obtained from two methods was less than 1%, we reported the results based on Tessier et al. (1979) procedure.
Soil samples were placed in plastic bags and B was added at the rate of 100 mg kg-1 soil as boric acid and then they were incubated for one week at 22 ± 2°C. The appropriate amount of distilled water was added to bring the soil to the estimated field capacity. At the end of incubation, samples were air dried before analysis.
One gram of prepared soil samples in duplicate was added to polypropylene centrifuge tubes then suspended in 20 ml of 0.01 M CaCl2 solution for 15 minutes at 22 ± 2°C and after that the suspensions centrifuged for 5 minutes and filtered. This process sequentially repeated for 10 times. Four mathematical models were used to describe the kinetics of B release in studied soils:
Parabolic diffusion model q = a + bt1/2
Power function equation lnq = lna + blnt
Elovich equation q = a +blnt
First order equation ln(q*- q) = a - bt
where, q is the cumilative B released at time t, q* is maximum B released and a and b are constants. An important term in theses equation is b, which is an indicative of B release rate. The determination coefficients (R2) were obtained by least squares regression of measured versus predicted values. The standard error (SE) of the estimate was calculated by
where, q and q* represent the measured and predicted B released, respectively, and n is the number of data points evaluated.
Boron speciation in the first and last step of sequential extraction was determined by visual MINTEQ. The input file included equilibrium pH and concentration of Ca2+, Mg2+, Na+, K+, HCO3-, SO42-, Cl-, B as H3BO3 and dissoleved organic carbon (DOC).
Determination of boron distribution in soils was performed by the sequential chemical extraction scheme (Salbu et al. 1998) which was a modified version of Tessier et al. (1979). It is designed to separate heavy metals and trace elements into five operationally defined fractions.
Two gram of each soil sample was weighed and placed in a 50 ml polypropylene centrifuge tube and the following extractions were performed sequentially:
- Exchangeable fraction (F1): extraction with 20 mL of 1 M NH4OAc at pH 7 for 2 h at room temperature.
- Specifically adsorbed and carbonate-bound fraction (F2): extraction of the residue from F1 with 20 mL of 1 M NH4OAc at pH 5 for 2 h at room temperature.
- Fe-Mn oxides bound fraction (F3): extraction of the residue from F2 with 20 mL of 0.04 M NH2OH.HCl in 25% HOAc for 6 h in a water bath at 60 °C.
- Organically bound fraction (F4): extraction of the residue from F3 with 15 mL of 30% H2O2 at pH 2 for 5.5 h in a water bath at 80 °C. After cooling, 5 mL of 3.2 M NH4OAc in 20% HNO3 was added. Sample was shaken for 0.5 h, and finally diluted to 20 mL with distilled water. The mixture was centrifuged for 15 min. The supernatant was filtered into and the residue was used for the last stage.
- Residual fraction (F5): 5 mL of nitric acid and 1 mL of perchloric acid were added to 1 g of residue obtained from the step above in a conical flask and heated on a hot plate at 60°C for 6 h. After evaporation, 1 mL of 2 M HNO3 was added and the residue after dissolution was filtered into a 100 mL volumetric flask and diluted with distilled water.
The fractionation studies were conducted in control and spiked soils. All experiments were carried out in duplicate. The Statistical analysis was performed by SAS v.6.12 software.
Results and Discussion
Characterization of soil samples
The selected chemical and physical properties of the soil studied have been given in Table 1. The range of pH was 6.8 to 7.8 and the soils were low in EC. The contents of ECC varied from 4.8 to 20.3%. The clay contents in all soils averaged 29.1% and ranged from 14.5 to 43.2%. OM averaged 1.9% and ranged from 0.7 to 3.9% (Ranjbar and Jalali 2012). Total B in soils averaged 40.01 mg kg-1 and varied from 26.26 to 51.86 mg kg-1.
B release Kinetics
Fig. 1 shows the trend of boron release by 0.01 M CaCl2 in soil samples. The release process was rapid during 90 minutes and then became slower. The two phases of B release are characteristic of a diffusion-controlled process that have been observed for other ions such as potassium (K+) and ammonium (NH4+) (Feigenbaum et al. 1981; Steffens and Sparks 1997). The two regions can be considered as dominant sinks for B from which B is released by a different rate. The observed pattern of initial rapid release followed by a slower release has been also reported for phosphorous (P) (Horta and Torrent 2007; Nafiu 2009). Madrid and De Arambari (1985) and Saha et al. (2004) stated that the two phases of release can be due to the heterogeneity of adsorption sites with different adsorption affinities.
The cumulative B concentration released in studied soils ranged from 66.43 mg kg-1 in soil no.7 to 94.98 mg kg-1 in soil no.4. There was no relationship between total amount of B and cumulative B released in studied soils. This means that the other factors such as organic matter and clay contents of soils and distribution of boron between different fractions can influence the release kinetics of the this element.
The kinetic equations previously mentioned were used to describe the trend of B release with time. The fitness of measured data with four equations has been shown in Fig. 2. In general, all of the utilized models were fitted well to experimental data. It has been reported that the linear fit to the parabolic diffusion model indicates that the release process is a diffusion controlled process (Jardin and Sparks 1984). But since this is just a fitting procedure, it cannot be definitely concluded. The fitness of the data with power equation was nearly curvilinear indicating two phase release process (Fig. 2). The parameter b (Table 2) was less than 1 in all soils indicating boron release rates decreased with time. The values of constant b in Elovich equation which indicated the release rate of boron in studied soils, ranged from 14.38 to 18.05 mg kg-1 min-1 (Table 2). This narrow range indicates that B providing power of the studied soils is nearly same. The trend of B release was curvilinear in case of application of parabolic diffusion, so the fitted lines with this model had the lowest R2 (Table 3).
Based on the highest determination coefficient and the lowest standard error values obtained from different models (Table 3), it was concluded that first order equation and power function were the best of various equations to describe B release in studied soils.
The significant correlation found between parameters of different models (Elovich, power function and parabolic diffusion) used for describing kinetic release of boron and soil organic matter (Table 4) indicated the importance of this parameter in B release form soils.
The results of speciation indicated that almost 100% of dissolved B was in the form of H3BO30. This species has been reported to be dominant at pH values less than 7 up to pH ≈ 9.2 (Keren and Sparks 1994; Klochko et al. 2009; Majidi et al. 2010). The equilibrium pH (as one of the input parameters in speciation model) of the first and last extracts ranged from 6.56 to 7.62 and 6.42 to 7.5, respectively. The uncharged molecule H3BO30 is electrostatically unfavorable for adsorption by charged surface sites (Majidi et al. 2010). Above pH 7, H3BO30 is gradually deprotonated and borate ion formed is attracted much stronger by the positively charged surface sites. Since the pH values of studied soils ranged from 6.7 to 7.8, it was likely one of the major reasons for weak adsorption of applied boron in spiked soils. As explained in the next section, organic matter was mainly responsible for B adsorption in studied soils. The results of speciation in the first extraction indicated that the saturation indexes of minerals were all negative which meant to undersaturation status. The positive saturation indexes of Aragonite and Calcite in soils no. 4, 5, 7, 8, 9 and 10 in last step of extraction referred to supersaturation of these minerals. The mentioned soils had more amounts of ECC than the others.
Boron fractions in soils
The measured concentrations of boron (mg kg-1) in different fractions of control and spiked soils and their statistical evaluation have been shown in Table 5. The range of B concentration in control soils varied from 2.23 to 4.52 mg kg-1 in F1, 1.23 to 2.88 mg kg-1 in F2, 0.86 to 1.82 mg kg-1 in F3, 0.14 to 0.95 mg kg-1 in F4 and 24.56 to 43.27 mg kg-1 in F5. In all control samples, the proportion of exchangeable B (F1) (5.1-10.37%) was higher than the percentages of specially adsorbed (F2) (2.82-5.95%), oxide bound (F3) (1.95-4.7%) and organically bound (F4) (0.28-2.82%) fractions but not than the residual part. The same results were observed by Hou et al. (1996) in the synthetic soils containing mixtures of goethite, clay mica, humic acid, calcite and quartz, Datta et al. (2002) in 17 soils collected from different locations of India with considerable variations in physico-chemical properties and Lucho-Constantino et al. (2005) in some moderately alkaline agricultural soils of Mexico. The relative proportion of B in the residual fraction ranged from 79.16 to 88.13% of total B. These values are similar to those found by Xu et al. (2001) (87.4-99.7%) in 13 samples from china and Lucho-Constantino et al. (2005) (74-97.4%). In general, B concentration in the extracted fractions decreased in the following order: F5 > F1 > F2 > F3 > F4.
It was observed that the addition of boron resulted in increase of B concentration in all five fractions of spiked soils (Table 5). B concentration in different fractions of spiked soils ranged from 78.96 to 89.46 mg kg-1 in F1, 11.32 to 13.06 mg kg-1 in F2, 6.18 to 7.72 mg kg-1 in F3, 2.68 to 6.93 mg kg-1 in F4 and 27.35 to 45.68 mg kg-1 in F5. The main proportion of applied B was measured in exchangeable fraction easily extracted by 0.01 M CaCl2 in release kinetic experiment and the minimum part was observed in residual fraction. It can be concluded that B transformation is a time consuming process and requires longer time than one week. Distribution of applied B among various fractions followed this order: F1 (54.34-61.69%) > F5 (20.74- 30%) > F2 (7.62-8.79%) > F3 (4.11-5.46%) > F4 (1.76-5.26%).
The percentage of boron fixation was calculated using amounts of B released by sequential extraction with 0.01 M CaCl2 and total concentration of B in control and spiked soils. Soils no.4 and 7 which respectively adsorbed 8.07 and 36.22 mg B kg-1 of initial 100 mg B kg-1 added to soils indicated the minimum and maximum B fixation percentage, correspondingly. The results of B adsorption by 21 surface soils representing major soil series of Punjab in India (Arona and Chahal 2002) showed that the content of B adsorbed ranged from 15.5 to 34.4 mg kg-1. The calculated values of boron fixation indicated positive correlation with OM content (r = 0.932, P < 0.01). Datta and Bhadoria (1999) studied boron adsorption and desorption in some acid soils of West Bengal and reported that organic carbon had a beneficial effect on the B retention capacity of soils, which could be due to formation of a complex between dihydroxy organic compounds and B. The presence of organic materials can also occlude the B reactive adsorption sites on clays and soils (Yermiyahu et al. 2001). Boron adsorbed on soil surfaces is neither directly available nor toxic for plants (Keren et al. 1985a, 1985b; Ryan et al. 1977).
Correlations between B fractions with soil properties
The results of correlation coefficients (Table 6) indicated that the specially adsorbed B fraction (F2) in control and spiked soils had a significant positive correlation with pH and ECC. Calcite is probably an important sink for B in alkaline soils containing large amounts of fine-grained secondary carbonates (Hou et al. 1996). The oxide bound B fraction (F3) indicated a positive correlation with ammonium oxalate extractable Fe. Ammonium oxalate can solublize non-crystalline and some crystalline (oxy) hydroxides of Fe and Al from soils. So, the relationship between oxide bound B and ammonium oxalate extractable Fe is logical due to the adsorption of both species of boron including B(OH)30 and B(OH)4¯ on Fe2O3 via ligand exchange (Su and Suarez 1995). Boron associated with Mn oxides is not considerable because Mn oxides have a point of net zero proton charge less than 1.5 (Balistrieri and Chao 1990) and so the adsorption affinity of Mn oxides for B should be negligible.
The organically bound B (F4) indicated a direct relationship with organic matter in both control and spiked soils. Lemarchand et al. (2005) reported that significant B adsorption on humic acid was observed at 6 < pH < 12 with a maximum value at pH = 9.5-10. As pH values in all of studied soils ranged from 6.8 to 7.8 (Table 1), the organic matter could behave as an active adsorbing surface. The residual B fraction (F5) had a positive correlation with clay content. The importance of organic matter in B adsorption has been reported in former researches (Elrashidi and O'Connor 1982; Evans 1987; Harada and Tamai 1968). It confirms that the residual B is the structural constituent of clay. B adsorbed on organic matter, oxide minerals, clay minerals, and carbonates equilibrates with B concentration in solution phase (Goldberg 1993). Therefore, it can transform from non mobile to labile forms when is needed. Hou et al. (1996) observed direct relationships between goethite and mica contents in soils and oxide bound and residual fractions of boron, correspondingly. The corresponding correlation between oxide bound fraction and ammonium oxalate extractable Fe, organically bound fraction and organic carbon, and residual fraction and clay content has been also reported by Datta et al. (2002).
B release from spiked soils was initially rapid followed by a slower reaction. The major part of B added to soil was extracted with 0.01 M CaCl2. First order equation was the best model to describe the release of boron from soils. There was high negative significant correlation between kinetic parameters of power and parabolic equations indicating that with increasing OM in soil, the rate of B release will be reduced. Based on speciation results, boric acid was the major species of boron in the first and the last step of extraction that was in agreement with pH values less than 9. Since uncharged H3BO30 molecules do not adsorbed by charged surfaces, little amounts of B added to soils adsorbed specifically and the main part released in sequential extraction. This means pH plays an important role in adsorption and release of boron in soils even in calcareous soils that have capacity for B adsorption due to the presence of calcium carbonate as an adsorbing surface. So, in soils with pH values less than 9 and receiving different sources of boron, uncharged boric don not attract on charged surfaces. Therefore in these conditions, B leaching in soil profile and potential risk for contamination of ground waters is highly possible. The results of fractionation in control soils showed that the amount of B in the extracted fractions decreased in the following order: F5 > F1 > F2 > F3 > F4. Addition of B resulted in increase of B concentrations in all five fractions. The maximum and minimum proportion of added B in spiked soils was extracted in exchangeable and residual fractions, respectively. The results indicated that the soils could not adsorb and fix considerable parts of B added after one week and this process needed more time. It means that B transformation from soluble to less mobile and non-labile forms is not rapid and requires longer time in contrast to other nutrients such as phosphorus that transforms more quickly.