Soil Science Society of America Journal 65:283-290 (2001)
© 2001 Soil Science Society of America
DIVISION S-1-SOIL PHYSICS
An Unsaturated Transient Flow Method for Determining Solute Adsorption by Variable-Charge Soils
Hidetaka Katoua,
Koji Uchimurab and
Brent E. Clothierc
a Div. of Soil Sci., Natl. Inst. of Agro-Environ. Sci., Tsukuba, 305-8604 Japan
b Kagoshima Tea Exp. Stn., Chiran, Kagoshima, 897-0303 Japan
c Environment and Risk Management Group, HortResearch Inst., Private Bag 11-030, Palmerston North, New Zealand
Corresponding author (katouh{at}niaes.affrc.go.jp)
 |
ABSTRACT
|
|---|
The amount of reactive ions adsorbed during transport in variable-charge soils is often smaller than those predicted from batch adsorption experiments. This overestimation of adsorption for invading ions by the batch method is likely to be caused by excessive desorption of indigenous, strongly adsorbed ions. We propose an unsaturated transient flow method by which equilibrium adsorption of weakly reactive ions can be determined without causing appreciable desorption of indigenous ions. In our method, water is imbibed into horizontal columns packed with soil that has been mixed with a salt solution at different concentrations to attain adsorption equilibrium. We make use of the observation that during imbibition of water, the antecedent solution is pushed ahead by the invading solution and accumulates in the region beyond the plane of separation, with the solution concentration unchanged. From linear plots of the solute content vs. soil water content in this region, the amount of solutes adsorbed by soil and the equilibrium solution concentration prior to the water imbibition are obtained. The advantages of the method are that adsorption isotherms can be determined under conditions similar to those expected during transport processes in soil, and that it is exempt from uncertainty about attainment of adsorption equilibrium as in the miscible displacement methods. The proposed method is best suited for variable-charge soils where adsorption of weakly reactive ions is largely due to an increase in the exchange capacity.
Abbreviations: AEC, anion-exchange capacity • CEC, cation-exchange capacity
|
INTRODUCTION
|
|---|
PREDICTING THE TRAVEL VELOCITY of reactive solutes in soil requires information about the relationship between the amount of the solute adsorbed by soil and its concentration in the aqueous solution phase. Determination of such a relation, called the adsorption isotherm, is not always simple for variable-charge soils that have electrical charge dependent on pH and ionic strength of the bulk solution. This is particularly the case when the soils have an adsorbed phase composition as might be found in the field. For example, Wong et al. (1990) noted difficulties in measuring the effective positive charge responsible for the adsorption and retarded transport of NO-3 in Andisols and highly weathered tropical soils containing appreciable amounts of adsorbed SO2-4. This is in contrast to the case of soils presaturated with single ionic species in which the exchange capacities are kept constant. In the latter soils, conventional batch adsorption experiments have been found to produce exchange isotherms capable of predicting reactive ion transport (Bond and Phillips, 1990a, 1990b; Ishiguro et al., 1992).
In variable-charge soils, adsorption of electrolyte ions due to contact with a salt solution may be a result of the increase in cation- and anion-exchange capacities (CEC and AEC) in response to an increase in the ionic strength of the bulk solution, or ion exchange taking place with indigenous ions. For Andisols, Katou et al. (1996) suggested that if the initial electrolyte concentration of soil solution is low enough, the adsorption of monovalent anions (Cl- and NO-3) during invasion of a salt solution is largely due to the increase in AEC of the soil, and desorption of indigenous SO2-4 in exchange for the monovalent anions proceeds only to a limited extent. In such situations, batch methods that involve excessive desorption of strongly adsorbed indigenous ions are most likely to overestimate the adsorption of invading anions during transport process (Wong et al., 1990; Katou et al., 1996).
An alternative method that has been employed in assessing solute adsorption is the miscible displacement method. The method often gives a better estimate of effective adsorption capacity of the soil (Wong et al., 1990). But a serious difficulty arises when the observed solute adsorption is sensitive to the volume of influent loaded on the column. This dependency may be due to the absence of local adsorption equilibrium (Valocchi, 1985), or simply a result of unfavorable ion exchange (Bolt, 1982). Regardless, it makes the determination of adsorption isotherms difficult.
In principle, adsorption isotherms pertinent to transport of reactive ions through soil could be obtained if, in addition to the exchange selectivity coefficients between indigenous ions and the invading ions, the complicated interdependence among the CEC and AEC of the soil, the solution pH, and the solution phase and adsorbed phase composition were known. However, this does not seem practicable, for measurements of AEC using SO2-4, the dominant adsorbed anion species in many field soils (Kamewada, 1994), have been exceptional (Imai and Okajima, 1980; Kamewada and Takahashi, 1996), and the lack of experimental data is more serious for the selectivity coefficients for anion exchange in varible-charge soils. Thus, a simple and reliable method needs to be developed for determining solute adsorption in variable-charge soils.
In this paper, we propose a transient, unsaturated flow method by which adsorption of weakly reactive ions by a variable-charge soil can be determined without causing appreciable desorption of indigenous ions. The unsaturated flow method makes use of the piston-like displacement of antecedent solution by the invading water as observed during absorption of water into unsaturated homogeneous soils (Smiles and Philip, 1978; Clothier et al., 1988; Bond and Phillips, 1990a). In the method, water is imbibed into horizontal columns packed with soil that has been mixed with a salt solution. The amount of solute adsorbed by soil and the equilibrium solution concentration prior to the imbibition are obtained from plots of the solute content vs. water content in a region beyond the "plane of separation" where the antecedent water accumulates (Smiles and Philip, 1978). This allows adsorption isotherms to be determined at a relatively low water content without extracting aqueous solution from the soil. The method is applied here to Cl- adsorption by an Andisol, and the results are compared with those obtained by conventional batch and miscible displacement methods. Furthermore, we used the measured isotherm in a numerical model to test its ability to predict the observed pattern of solute displacement in the soil.
|
THEORY
|
|---|
Suppose water is imbibed into a horizontal, homogeneous soil column, at an initial volumetric water content of 
n, in which adsorption equilibria for reactive chemical species have been established between the adsorbed phase and the aqueous solution phase. Typical water content and solute content profiles are shown in Fig. 1
. For any reactive solute species i, its content per unit mass of soil, M (molc kg-1), is the sum of the contents in the adsorbed phase and in the aqueous solution phase such that
 | (1) |
where Q is the amount of the solute adsorbed per unit mass of soil (molc kg-1), C is the concentration in the liquid phase (molc m-3), is the volumetric water content (m3 m-3), 
is the bulk density of soil (kg m-3). Since we will carry out experiments for Cl-, we can drop the subscripting for the general case.
It has been well established that during absorption of water into an unsaturated homogeneous soil having small aggregate size (e.g., <0.35–2 mm), the antecedent water is pushed ahead by the invading water, and accumulates in a region beyond the plane of separation, x* (m) (Smiles and Philip, 1978; Clothier et al., 1988; Bond and Phillips, 1990a). This plane, which identifies the front of the invading water, is located by the material balance equation (Smiles and Philip, 1978)
 | (2) |
where x is the distance (m), and s is the water content at the proximal end of the soil column. If the displacement by the invading water is perfect, and the effect of solute dispersion can be neglected, the aqueous solution found in the region x > x* will be derived solely from the antecedent water pushed ahead. This is the key to our method, for if adsorption equilibria have been attained prior to the water imbibition, it follows that the solution composition at x > x* should not change during transport and should be the same as that of the antecedent solution. This means that solute adsorption by soil should also be constant for x > x*. So, any changes in the solute content with distance in this region (x > x*) are due solely to the changes in the water content due to accumulation of the antecedent solution (Eq. [1]). Thus, a plot of the solute content (M) vs. solution volume per unit mass of soil (/) for a reactive solute i in the region x > x* should give the straight line
 | (3) |
from which the initial concentration in the aqueous solution, Cn (molc m-3), and the initial adsorption of the solute by soil, Qn (molc kg-1), can be obtained (Fig. 1). Mixing soil with salt solutions at different concentrations will give different values of initial solute content Mn (molc kg-1) (and hence Cn and Qn), and will enable adsorption isotherms to be constructed from a series of column experiments. It should be noted that this method for estimating Cn and Qn can be applied not only to the added solutes, but also to native solutes present in the soil. If desired, exchange selectivity coefficients may be evaluated from the values of Cn and Qn determined for any set of two reactive anions or cations found in the soil.
|
MATERIALS AND METHODS
|
|---|
The soil material used in this study was the air-dried, sieved (<1 mm) subsoil of Kannondai Andisol (Hydric Hapludand) described by Katou et al. (1996). The clay fraction of the soil is dominated by allophane and other amorphous materials (National Institute of Agricultural Sciences, 1984), and the soil contains Ca2+ and SO2-4 as the predominant adsorbed ion species. Some relevant soil properties are given in Table 1.
Unsaturated Transient Water Absorption Experiments
One hundred grams of the air-dried soil, at a gravimetric water content of 
0.14 kg kg-1, was moistened with one of the salt solutions listed in Table 2, to give an initial gravimetric water content wn 0.29 kg kg-1. This value of wn was small enough to produce adequately large water content changes in the region x > x* following water absorption, while large enough to separate x* from the wetting front, thus allowing a sufficient number of soil samples taken from this region (Fig. 1). The concentration of salt solutions to be mixed with soil depends on the concentration range that adsorption isotherms of interest should cover, and the soil's ability to adsorb the incorporated ions from the aqueous solution. This may be chosen by making use of a first-guess adsorption isotherm of an arbitrary form together with Eq. [1], or simply by trial and error. The moistened soil was thoroughly mixed in a sealed plastic bag and left overnight so as to establish adsorption equilibria between the adsorbed phase and the aqueous solution phase.
The soil was then repacked into sectionable acrylic columns of 2.1 cm in internal diameter to a bulk density of 0.791 (±0.007) Mg m-3 so that n = 0.225 (±0.008) m3 m-3. The experimental setup was basically the same as that used in the laboratory method for determining soil water diffusivity (Klute and Dirksen, 1986). One-dimensional, horizontal, free-water absorption experiments were carried out by supplying distilled water from a Marriotte bottle to the proximal end of the column. After terminating each experiment, the column was rapidly sectioned and the soil samples immediately weighed and air-dried. The water content was calculated using the previously established air-dry water content of the soil. From the water content profiles (x) after an elapsed time t (s), the sorptivity S (m s-1/2) given by (Philip, 1969)
 | (4) |
and x* defined in Eq. [2] were calculated for each column.
The anion contents in soil were determined by the method described by Katou et al. (1996). One gram of soil (oven-dry basis) was shaken for 15 min with 100 mL of 0.01 mol L-1 NaOH, and after centrifugation the supernatant solution was analyzed for Cl-, NO-3, and SO2-4 ion chromatography. The anions extractable by this method comprise those present both in the adsorbed phase and in the liquid phase, and so they correspond to Mi in Eq. [1].
Determination of Anion Adsorption Isotherm
Once the water and anion content profiles were obtained, the anion contents, M(x), in samples taken from the region x > x* were plotted against the solution volume per unit mass of soil, (/), for each column. From linear regression analysis with (/) as an independent variable (Eq. [3]), the initial liquid phase concentration, Cn, and the initial adsorption by soil, Qn, were determined for Cl- and SO2-4. Chloride contents in the proximity of x* were often found to be lowered as a result of solute dispersion. The data from these samples were excluded from the regression analysis. The Cl- adsorption isotherm was constructed from the sets of Cn and Qn obtained from a series of transient water absorption experiments in which salt solution at different concentrations had been mixed with the soil.
|
RESULTS AND DISCUSSION
|
|---|
Anion Content Profiles and the Chloride Adsorption Isotherm
Figure 2
shows water and anion content profiles upon one-dimensional absorption of water into Kannondai subsoil premixed with 0.1 mol L-1 CaCl2 (Column LKS3). The plane of separation was found to be located at x* = 12.1 cm. The soil had an initial Cl- content of 29.6 mmolc kg-1, essentially all of which had come from the incorporated CaCl2 solution. The indigenous SO2-4 content was 94.0 mmolc kg-1. Upon absorption of water, Cl- was nearly completely removed from soil near the column inlet and accumulated in a region ahead of x* and behind the wetting front. Nevertheless, the front of Cl- being removed lagged behind x*, the location of the displacement front expected for an inert solute. This suggests that some portion of the added Cl- had been adsorbed.
In Fig. 3
, the Cl- contents in the region x > x* were plotted against (/) for the same column. In accord with the proposed theory, the plot yielded a straight line. From linear regression analysis performed on the Cl- and water content data from this region, estimates (±SE) of Cn = 31.6 (±0.5) mmolc L-1 and Qn = 20.2 (±0.3) mmolc kg-1 were obtained (r2 = 0.997***). These results show that when 0.1 mol L-1 CaCl2 was mixed with Kannondai subsoil having a negligible amount of indigenous Cl-, approximately two-thirds of the added Cl- was adsorbed by the soil (Table 2). This Cl- once adsorbed was, however, easily desorbed by invasion of water, even in the absence of exchange with other anions. This high susceptibility to leaching by water is characteristic of weakly reactive anions in variable-charge soils.
In contrast, there was no significant transport of indigenous SO2-4 observed during absorption of water. From a plot of SO2-4 content vs. (/) in the region x > x*, estimates of Cn = 1.6 (±1.0) mmolc L-1 and Qn = 93.6 (±0.6) mmolc kg-1 were obtained for SO2-4 (Fig. 3). These results indicate that effectively all the indigenous SO2-4 remained adsorbed after mixing the concentrated CaCl2 solution with the soil. This confirms the observation that indigenous SO2-4 had a much greater affinity for the soil than Cl-. Thus, adsorption of Cl- by the soil was mostly due to the increase in AEC (or total anion adsorption) in response to an increase in the ionic strength of the bulk solution, rather than through anion exchange with SO2-4 (Katou et al., 1996).
Figures 4 and 5
show anion content profiles upon absorption of water into Kannondai subsoil premixed with 0.2 mol L-1 CaCl2 (Column LKS6) and 0.05 mol L-1 CaCl2 (Column LKS7), respectively. Again, removal of Cl- by the invading water from soil near the column inlet was almost complete in both columns. But the front of Cl- being removed was closer to x* in Column LKS6 where a larger amount of Cl- had been incorporated into soil. This difference in the retardation of the Cl- front relative to x* stemmed from the fact that the adsorption of Cl- by the soil is nonlinear, with a larger retardation expected for a lower liquid phase Cl- concentration (Katou et al., 1996). Plots of Cl- content vs. (/) in the region x > x* yielded straight lines (Fig. 6)
, from which Cn = 93.4 (±1.7) mmolc L-1 and Qn = 32.1 (±1.0) mmolc kg-1, and Cn = 14.0 (±0.3) mmolc L-1 and Qn = 11.1 (±0.2) mmolc kg-1 were deduced for Columns LKS6 and LKS7, respectively. We see again in Fig. 4 and 6 that desorption of indigenous SO2-4 upon contact with 0.2 mol L-1 CaCl2 was negligible, and that the indigenous SO2-4 was highly resistant to leaching with water.
View larger version (34K):
[in this window]
[in a new window]
|
Fig. 4. Profiles of the anion content, M, the liquid phase concentration, C, and adsorption by soil, Q, upon absorption of water into Kannondai subsoil premixed with 0.2 M CaCl2. Estimated C and Q profiles and simulated profiles are based on the Cl- adsorption isotherm inferred from the transient flow experiments
|
|
View larger version (34K):
[in this window]
[in a new window]
|
Fig. 5. Profiles of the anion content, M, the liquid phase concentration, C, and adsorption by soil, Q, upon absorption of water into Kannondai subsoil premixed with 0.05 M CaCl2. Estimated C and Q profiles and simulated profiles are based on the Cl- adsorption isotherm inferred from the transient flow experiments
|
|
Estimates of Cn and Qn for Cl- obtained from five column experiments listed in Table 2 were then used to infer the Cl- adsorption isotherm. In principle, an adsorption equation of arbitrary form may be used to describe the Q–C relation, but here we adopt the Langmuir equation
 | (5) |
where Qmax is the maximum adsorption per unit mass of soil (molc kg-1), and K is an empirical constant (m3 molc-1). In a linear form, the equation can be written as (Sposito, 1984, p. 26–28)
 | (6) |
where KD (=Q/C) is the distribution coefficient (m3 kg-1). In Fig. 7
, the KD prior to the water imbibition (KD = Qn/Cn) is plotted against Q (=Qn) for Cl-. From a linear regression analysis with Q as an independent variable, the coefficients -K and KQmax were obtained, from which we deduced the adsorption parameters K = 0.0238 (±0.0031) m3 molc-1 and Qmax = 0.0461 (±0.0094) molc kg-1.
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 7. Comparison of Cl- adsorption isotherms obtained by the unsaturated transient flow method and the ponded steady-state leaching method. KD = distribution coefficient; Q = the amount of Cl- adsorbed by soil; K = empirical constant; Qmax = maximum adsorption in the Langmuir equation
|
|
Estimates of Cn and Qn for SO2-4 are also listed in Table 2. It should be mentioned that adsorption isotherms for the native ions cannot be obtained from such estimates, for possible desorption through exchange with the added ions (i.e., Cl- in the present case) would result in a decrease in Qn and an increase in Cn for the native ions such that Qn and Cn were negatively correlated. Poor correlation between Qn and Cn found for SO2-4 (r2 = 0.046) is additional evidence that most of the Cl- adsorption was not through anion exchange with the indigenous SO2-4. Larger relative errors for the estimates of Cn stemmed from the fact that nearly all SO2-4 was in the adsorbed phase. The results suggest that the unsaturated transient flow method is not well suited for strongly adsorbing solutes, for in such a case where Mn Qn >> (n/)Cn, the estimate of Cn will be sensitive to small experimental errors in the determination of the solute content.
Liquid Phase and Adsorbed Chloride Profiles Estimated from the Inferred Isotherm
Using the adsorption equation inferred above (Eq. [5] with K = 0.0238 m3 mol-1c and Qmax = 0.0461 molc kg-1) together with the mass balance equation (Eq. [1]), the Cl- concentration in the aqueous solution phase, C(x), and the Cl- adsorption by soil, Q(x), were estimated from the measured profiles of water content, (x), and the total Cl- content, M(x), for each column. The estimated C(x) and Q(x) profiles are also shown in Fig. 2, 4, and 5. As expected from the theory, Cl- adsorption was essentially constant in the region x > x*. Desorption of Cl- in the region 0 < x < x*, triggered by the invasion of water, was due to the decrease in AEC (or total anion adsorption) in response to the decrease in the ionic strength of the bulk solution.
Numerically Simulated Profiles Based on the Inferred Isotherm
An advantage of the transient flow method is that the adsorption isotherm deduced from M vs. (/) plots in the region x > x* can be tested for its ability to reproduce the solute content profiles in the region 0 < x < x*. In Fig. 2, 4, and 5, numerically simulated M(x), C(x), and Q(x) profiles based on the anion transport model of Katou et al. (1996) and the parameters K and Qmax evaluated above are also presented for comparison. In the simulation, the water flow equation
 | (7) |
subject to initial and boundary conditions of
 | (8) |
and the solute transport equation
 | (9) |
subject to
 | (10) |
were solved simultaneously to give (x, t) and C(x, t), from which Q(x, t) and M(x, t) were calculated using Eq. [1] and [5]. The soil water diffusivity function D() (m2 s-1) was assumed to be of the exponential form of Brutsaert (1979). We found interdependent constants ß = 8 and 
= 1.434 x10-3 in the D() function gave a satisfactory reproduction of (x) when Eq. [7] subject to Eq. [8] was numerically solved. For the solute dispersion coefficient Ds (m2 s-1), we assumed that the tortuosity factor 
= 0.44 and the dispersivity 
= 2 mm, found appropriate for an unsaturated repacked fine sand (Clothier et al., 1988, 1991), hold in the Kannondai subsoil (Katou et al., 1996). The value of Cn in Eq. [10] was chosen such that the initial Cl- content, Mn, in the numerical simulation was equal to Mn(x>x*), the initial content deduced from the estimates of Cn and Qn based on the M vs. (/) plot in the region x > x*. By solving Eq. [1] and [5] for C with M = Mn(x>x*) and = n, Cn to be used in the simulation was obtained as Cn = 32.2, 94.1, and 13.6 mmolc L-1 for Columns LKS3, LKS6, and LKS7, respectively. The value of C0 in Eq. [10] was taken as 0.01 mmolc L-1. The simulated M(x), C(x), and Q(x) profiles were, respectively, in good agreement with the measured M(x) and the estimated C(x) and Q(x) profiles in each column. These results demonstrate that the inferred Cl- adsorption isotherm was appropriate to the Cl- transport process in the unsaturated Kannondai Andisol.
If examined more closely, the agreement between the simulated and the measured M(x) profiles was somewhat poorer in Column LKS6 where CaCl2 solution at the highest concentration had been mixed with soil. This could be due to the effects of nonlinear Cl- adsorption not fully accounted for by the Langmuir equation, or alternatively, the effects of density difference (Mansell et al., 1998) between the invading water and the antecedent solution. It should be stressed, however, that possible occurrence of physical nonequilibrium in the region 0 < x < x* will not affect the adsorption isotherms obtained by the unsaturated transient flow method because the isotherms are deduced from the M vs. (/) plots in the region x > x* where the adsorption equilibrium established prior to the water imbibition is not perturbed.
Comparison with Conventional Methods
Table 3 presents adsorption of Cl- by Kannondai subsoil as determined by a modified repeated washing method of Schofield (1949). Two grams of soil was placed in a 50-cm3 centrifuge tube and washed with 25 cm3 of 1 mol L-1 NaCl and 0.5 mol L-1 CaCl2, five times in each, and then equilibrated with 0.005, 0.025, or 0.125 mol L-1 CaCl2. Anions remaining in soil after final equilibration were extracted with 0.01 mol L-1 NaOH, and the adsorption by soil obtained by subtracting the amount entrained in the equilibrated solution. The nonexchangeable SO2-4 in Table 3 refers to SO2-4 that was not replaced by washing with CaCl2 solutions, but was extracted with 0.01 mol L-1 NaOH. It is evident that owing to exhaustive desorption of indigenous SO2-4 in exchange for Cl-, a process hardly expected during transport process in soil, the repeated washing method greatly overestimates Cl- adsorption and the distribution coefficient, KD, as observed in column transport experiments. Omission of washing with the concentrated NaCl and CaCl2 solutions (i.e., washing with dilute CaCl2 solutions only) would leave some of the exchangeable SO2-4 in soil and result in a smaller adsorption of Cl-. However, the extent of SO2-4 desorption (and hence overestimation of Cl- adsorption) should then depend on the selectivity coefficient for Cl-–SO2-4 exchange as well as such experimental conditions as the number of washing, and be difficult to predict. Such excessive desorption of SO2-4 is avoided in our unsaturated transient flow method.
In Fig. 7, Cl- adsorption data obtained by the ponded, steady-state miscible displacement method (Katou et al., 1996) are also shown for comparison. In this method, 5.7 to 7.4 pore volumes of CaCl2 solution was applied to the top of the columns. The amount of Cl- adsorbed was obtained by subtracting the amount in the liquid phase from the 0.01 mol L-1 NaOH-extractable Cl- content in the soil. The adsorption isotherm obtained by the ponded displacement method (K = 0.0230 m3 molc-1 and Qmax = 0.0516 molc kg-1) was only slightly above that inferred from the transient flow method. We consider, however, that the fairly good agreements in the adsorption parameters obtained by the two methods were rather fortuitous. In the ponded displacement method, only data from soil samples that did not show appreciable desorption of SO2-4 were used. As mentioned above, the desorption of SO2-4 was accompanied by an increased adsorption of Cl-, the extent of which cannot easily be predicted. The ponded displacement experiments are essentially column-transport experiments, rather than adsorption experiments, so that it is uncertain a priori whether the adsorption equilibrium has really been attained.
The unsaturated transient flow method developed in the present study is in fact a kind of variant of the batch method in that adsorption equilibrium has been established by mixing a salt solution with soil. The method is thus exempt from uncertainty about the attainment of adsorption equilibrium as in the conventional miscible displacement methods. Desorption of strongly adsorbed indigenous ions, which is unrealistic but common in the conventional batch adsorption experiments, is minimized by a large soil/solution ratio when mixing the solution with soil. The accumulation of the antecedent solution caused by the piston-like displacement by the invading water allows the solute adsorption and the equilibrium solution concentration to be estimated simultaneously by making use of Eq. [3]. Difficulties associated with the extraction of aqueous solution from a relatively dry soil, which may necessitate the use of a dense water-immiscible liquid (Phillips and Bond, 1989; Bond and Phillips, 1990a), are thus circumvented in the unsaturated transient flow method. As a shortcoming, the method is not well suited for strongly adsorbing solutes, where the estimate of aqueous solution phase concentration is subject to a relatively large error. Determination of adsorption isotherms will not be straightforward if, contrary to the experimental design of the method, ion exchange between the added ions and the native ions takes place to a considerable extent. The proposed method is best suited for determining adsorption isotherms for weakly reactive ions in variable-charge soils where the adsorption is largely due to an increase in the exchange capacity.
|
CONCLUSIONS
|
|---|
A laboratory method was developed for determining solute adsorption isotherms under chemical conditions similar to those expected during transport in the field. In the method, water is imbibed into horizontal columns packed with soil that has been mixed with a salt solution to attain adsorption equilibrium. The amount of solute adsorbed and the equilibrium solution concentration prior to the imbibition are obtained from the plot of solute content vs. water content in the region beyond the plane of separation, where the antecedent solution accumulates with the solution concentration unchanged. Mixing a salt solution at different concentrations enables the adsorption isotherm to be constructed from a series of column experiments. The method is exempt from uncertainty about the attainment of adsorption equilibrium, and has the advantages that adsorption equilibria at a relatively low water content can be measured without extracting pore solution and that the inferred adsorption isotherm can be tested for its ability to reproduce independently measured solute content profiles behind the plane of separation. Minimizing desorption of indigenous, strongly adsorbed ions, the transient flow method is best suited for weakly reactive ions in variable-charge soil where the adsorption is largely due to an increase in the exchange capacity.
|
ACKNOWLEDGMENTS
|
|---|
Discussions with Nobiru Kozai of Kumamoto Agric. Res. Cent., D.R. Scotter of Massey Univ., and R.S. Kookana of CSIRO Land and Water are greatly acknowledged.
|
NOTES
|
|---|
This work was partly supported by a fellowship under the OECD Co-operative Research Programme: Biological Resource Management for Sustainable Agricultural Systems.
* , **, *** Significant at the 0.05, 0.01, and 0.001 levels of probability, respectively.
Received for publication December 15, 1999.
|
REFERENCES
|
|---|
- Bolt, G.H. 1982. Movement of solutes in soil: Principle of adsorption/exchange chromatography. p. 285–348. In G.H. Bolt (ed.) Soil chemistry B. Physico-chemical models. 2nd revised ed. Elsevier, Amsterdam, the Netherlands.
- Bond, W.J., and I.R. Phillips. 1990a. Ion transport during unsteady water flow in an unsaturated clay soil. Soil Sci. Soc. Am. J. 54:636–645.[Abstract/Free Full Text]
- Bond, W.J., and I.R. Phillips. 1990b. Cation exchange isotherms obtained with batch and miscible-displacement techniques. Soil Sci. Soc. Am. J. 54:722–728.[Abstract/Free Full Text]
- Brutsaert, W. 1979. Universal constants for scaling the exponential soil water diffusivity? Water Resour. Res. 15:481–483.
- Clothier, B.E., T.J. Sauer, and S.R. Green. 1988. The movement of ammonium nitrate into unsaturated soil during unsteady absorption. Soil Sci. Soc. Am. J. 52:340–345.
- Clothier, B.E., T.J. Sauer, and D.R. Scotter. 1991. Redistribution of water and solute following infiltration from a surface drip source. Water Resour. Res. 27:2091–2097.
- Imai, H., and H. Okajima. 1980. Studies on the nutrient retention power of soils. III. Nitrate adsorption. (In Japanese.). J. Sci. Soil Manure Jpn. 51:102–106.
- Ishiguro, M., K.-C. Song, and K. Yuita. 1992. Ion transport in an allophanic Andisol under the influence of variable charge. Soil Sci. Soc. Am. J. 56:1789–1793.[Abstract/Free Full Text]
- Kamewada, K. 1994. Vertical distribution of anions (Cl-, NO-3, SO2-4) in upland soils. (In Japanese.). Jpn. J. Soil Sci. Plant Nutr. 65:255–265.
- Kamewada, K., and Y. Takahashi. 1996. Increase in cation adsorption induced by surface complexation of sulfate on Andisols and prediction by "Four-plane model". Soil Sci. Plant Nutr. 42:785–799.
- Katou, H., B.E. Clothier, and S.R. Green. 1996. Anion transport involving competitive adsorption during transient water flow in an Andisol. Soil Sci. Soc. Am. J. 60:1368–1375.[Abstract/Free Full Text]
- Klute, A., and C. Dirksen. 1986. Hydraulic conductivity and diffusivity: Laboratory methods. p. 687–734. In A. Klute (ed.) Methods of soil analysis. Part 1. 2nd ed. Agron. Monogr. 9. ASA and SSSA, Madison, WI.
- Mansell, R.S., J.H. Dane, D. Shinde, and H.H. Liu. 1998. Density-coupled water flow and contaminant transport in soils. p. 371–387. In H.M. Selim and L. Ma (ed.) Physical nonequilibrium in soils: Modeling and application. Ann Arbor Press, Chelsea, MI.
- National Institute of Agricultural Sciences. 1984. Outlines of soil properties and fertilizer experiments in NIAS fields. (In Japanese.) Dep. of Soils and Fertilizers Misc. Publ. no. 3. Natl. Inst. Agro-Environ. Sci., Tsukuba, Japan.
- Philip, J.R. 1969. Theory of infiltration. Adv. Hydrosci. 5:215–296.
- Phillips, I.R., and W.J. Bond. 1989. Extraction procedure for determining solution and exchangeable ions on the same soil sample. Soil Sci. Soc. Am. J. 53:1294–1297.[Abstract/Free Full Text]
- Schofield, R.K. 1949. Effect of pH on electrical charges carried by clay particles. J. Soil Sci. 1:1–8.
- Smiles, D.E., and J.R. Philip. 1978. Solute transport during absorption of water by soil: Laboratory studies and their practical implications. Soil Sci. Soc. Am. J. 42:537–544.[Abstract/Free Full Text]
- Sposito, G. 1984. The surface chemistry of soils. Oxford Univ. Press, New York.
- Valocchi, A.J. 1985. Validity of the local equilibrium assumption for modeling sorbing solute transport through homogeneous soils. Water Resour. Res. 21:808–820.
- Wong, M.T.F., R. Hughes, and D.L. Rowell. 1990. Retarded leaching of nitrate in acid soils from the tropics: Measurement of the effective anion exchange capacity. J. Soil Sci. 41:655–663.
This article has been cited by other articles:
|
|
|
|
 
R. Cichota, I. Vogeler, N. S. Bolan, and B. E. Clothier
Simultaneous Adsorption of Calcium and Sulfate and Its Effect on Their Movement
Soil Sci. Soc. Am. J.,
April 5, 2007;
71(3):
703 - 710.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
T. E. Ochsner, B. M. Stephens, W. C. Koskinen, and R. S. Kookana
Sorption of a Hydrophilic Pesticide: Effects of Soil Water Content
Soil Sci. Soc. Am. J.,
October 27, 2006;
70(6):
1991 - 1997.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
R. Ahmad, H. Katou, and R. S. Kookana
Measuring Sorption of Hydrophilic Organic Compounds in Soils by an Unsaturated Transient Flow Method
J. Environ. Qual.,
May 11, 2005;
34(3):
1045 - 1054.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
H. Katou
A pH-Dependence Implicit Formulation of Cation- and Anion-exchange Capacities of Variable-charge Soils
Soil Sci. Soc. Am. J.,
July 1, 2002;
66(4):
1218 - 1224.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
