Printer Friendly

Effect of [K.sup.+] on Na-Ca exchange and the SAR-ESP relationship.


In Australia, many wastewaters, including winery wastewater, are reclaimed for irrigation use. These waters are rich in sodium ([Na.sup.+]), and the prevention of soil physical degradation is of paramount importance, particularly in the semi-arid grape-growing regions of South Australia. A high [Na.sup.+] concentration in winery wastewater results from the widespread use of sodium hydroxide for cleaning and sterilisation during winery operations (Kumar and Christen 2009). The [Na.sup.+] concentration in many winery wastewaters that are applied to soils is ~7.7 [mmol.sub.c]/L (Kumar and Christen 2009).

Where saline waters rich in [Na.sup.+], such as municipal wastewater, are routinely used for irrigation, increases in soil exchangeable [Na.sup.+] and reduced soil hydraulic conductivity have been widely documented (e.g. Halliwell et al. 2001; Menneer et al. 2001; Stevens et al. 2003). The sodium adsorption ratio (SAR) of wastewater is often used as an indicator of the sodicity risk. The SAR is defined by the concentrations of [Na.sup.+] and divalent cations [Mg.sup.2+] and [Ca.sup.2+], where concentrations of cations are expressed as [mmol.sub.c]/L:

SAR [(mmol/L).sup.05] = [[Na.sup.+]]/ [square root of ({[[Ca.sup.2+]]+[[Mg.sup.2+]]}/2)] (1)

Winery wastewaters also contain elevated concentrations of potassium ([K.sup.+]), which come primarily from grape lees, spent juice, and potassium hydroxide cleaners (Kumar and Christen 2009). The [K.sup.+] concentration in winery wastewater fluctuates in response to winery operations and is highest during vintage periods when grapes are crushed, ranging between 3.8 and 8.0 [mmol.sub.c]L (Arienzo et al. 2009).

Based on the large hydrated ion size and its affinity for clay minerals, high levels of exchangeable [K.sup.+] in soil have the potential to cause clay swelling and dispersion (Levy and Feigenbaum 1996). The soil structural effects of [K.sup.+] in irrigation wastewaters have received less attention than the effects of [Na.sup.+] due to the typically low abundance of [K.sup.+] in most waters (Arienzo et al. 2009). Less widely adopted than the SAR is the potassium adsorption ratio (PAR) defined by the concentrations of [K.sup.+] and [Ca.sup.2+] and [Mg.sup.2+]:

PAR[(mmol/L).sup.0.5] = [[K.sup.+]]/ [square root of ({[[Ca.sup.2+]]+[[Mg.sup.2+]]}/2)] (2)

Cation selectivity coefficients based on mass action law, such as the Vaneslow (1932) and Gapon (1933) equations, are frequently used to describe cation equilibria between exchangeable and solution phases based on knowledge of the SAR or PAR. From the analysis of saturation extracts obtained from 59 soils from the western United States, the United States Salinity Laboratory (USSL) found the relationship between SAR and exchangeable sodium percentage (ESP) to fit the model equation (Richards 1954):

ESP = 100(-0.0126 + 0.01475 SAR)/ 1 + ( 0.0126 + 0.01475 SAR) (3)

This model has subsequently been applied throughout the world to estimate soil ESP in a range of different soil types. In South Australia, guidelines relating to the application of winery wastewater to soil utilise the USSL SAR-ESP model (Richards 1954) to devise recommended SAR threshold values for the prevention of sodic conditions in irrigated soils (South Australian EPA 2004).

In some soils, the relationship described by the USSL SAR-ESP model has been shown to vary in response to changes in ionic strength, clay mineralogy, and pH (Lieffering and McLay 1996; Kopittke et al. 2006). Cation composition of the soil solution has also been reported to influence [Na.sup.+] selectivity. In South Australia, for instance, Laurenson (2010) reported that the concentration of exchangeable [K.sup.+] was nearly double that of [Na.sup.+] (EPP 18.4, ESP 11.3) in a Barossa Luvisol (FAO 2006) soil irrigated with winery wastewater. This occurred despite the higher [Na.sup.+] concentration in the irrigated wastewater (Na:K molar ratio 2 : 1). Exchangeable [Ca.sup.2+] and [Mg.sup.2+] remained unchanged, as did the proportion of monovalent cations relative to divalent cations. In a Na-Ca binary system, Levy and Feigenbaum (1996) observed a close correlation between measured and predicted ESP, based on the USSL SAR-ESP model. In a ternary system containing K-Na-Ca, however, ESP was significantly lower than in the binary system and was poorly correlated with the USSL SAR-ESP model. This suggests that monovalent cation dynamics in soils irrigated with winery wastewater may differ from that observed under more widely used saline waters and wastewaters due to the [K.sup.+] content of winery wastewater.

From the same soils used to develop the SAR-ESP model, the USSL reported a similar relationship between PAR and exchangeable potassium percentage (EPP), which is described by the PAR-EPP model equation (Richards 1954):

EPP = 100(0.036 + 0.1051 PAR)/ 1 + (0.036 + 0.1051 PAR) (4)

Using this model, Feigenbaum et al. (1991) reported close correlation between measured and predicted EPP in a binary K-Ca system and a ternary K-Na-Ca system. This showed that the presence of [Na.sup.+] in a ternary system had limited influence on the resulting EPP; however, the influence of [K.sup.+] on ESP was not investigated.

Given the influence of [K.sup.+] on the binding of [Na.sup.+], it seems necessary to ascertain whether the widely accepted USSL SAR-ESP model is adequate in predicting sodicity hazard in soils irrigated with winery wastewater with high concentrations of both [K.sup.+] and [Na.sup.+]. The aim of this study was, therefore, to determine (i) the relative affinity of [K.sup.+] and [Na.sup.+] in binary K-Ca and Na-Ca systems, and (ii) the influence of PAR on the binding affinity of [Na.sup.+] at various SAR and PAR concentrations in a ternary K-Na-Ca system. The results of this study will enable better determination of the sodicity hazard in soils irrigated with [K.sup.+]-rich winery wastewater.

Materials and methods

Soil characterisation

Soil used in this study was collected from the Viticultural Research Station in Nuriootpa, Barossa Valley, South Australia. This soil is classified as a Calcic Red Sodosol (Isbell 2002), Chromi-Calcic Abruptic Luvisol (FAO 2006). The soil profile is distinctly duplexed between the A and B horizon. The B horizon soil had substantially higher clay content and bulk density (61%, 1.45Mg/[m.sup.3]) than the overlying A horizon soils (17%, 1.2 Mg/[m.sup.3]). In South Australia, salts tend to accumulate near the A-B interface of soil profiles irrigated with saline waters including wastewater (e.g. Stevens et al. 2003). The risk of chemically induced soil dispersion is therefore greatest at this depth in the profile. For this reason, soil samples were collected from 600-700 mm depth, ~100 mm below the point of transition between the A and B horizon interfaces. In the laboratory, soils were air-dried by force-draft oven at 40[degrees]C for 48 h then sieved through 2-mm mesh.

In characterising the soil, soil p[H.sub.1:5]) and electrical conductivity (E[C.sub.(1:5])) were determined from a 1:5 soil water suspension using a calibrated pH probe (TPS SmartCHEM pH Electrode) and calibrated EC electrode (TPS SmartCHEM Conductivity sensor) (Rayment and Higginson 1992). Exchangeable cations were determined by adding 20mL of 0.1 M Ba[Cl.sub.2/0.1 M] N[H.sub.4]C1 to 2.0g of soil (Gillman and Sumpter 1986). A low soil EC(t: 5) indicated that unbound or 'free' cations in the soils comprised only a small fraction of the total cation mass; therefore, no pretreatment of soil was carried out before the displacement of exchangeable cations (Gillman and Sumpter 1986). Soils were shaken end-over-end for 2h and then centrifuged at 3450G (Heraeus Multifuge[R] 3 S-R) for 10 min. The supernatant was filtered (0.45gm, Millipore) and analysed for total cations using inductively coupled plasma-optical emission spectrometry (ICP-OES) (Perkin Elmer Optima 5300V). The effective cation exchange capacity [(CE[C.sub.c]).sub.+] was determined as the sum of exchangeable [Ca.sup.2+], [Mg.sup.2+], [K.sup.+], and [Na.sup.+]. Calcium was the dominant exchangeable cation, while ESP was ~13 and EPP was 6. Soil [pH.sub.(1:5)] is near neutral, while EC and organic carbon were both low (Hazelton and Murphy 2007).

Total organic carbon was determined using the chemical wet oxidation method of Walkley and Black (Rayment and Higginson 1992). Particle size was determined using the hydrometer method of Gee and Bauder (1986). Clay mineralogy was determined by X-ray diffraction (XRD) patterning and collected on a PANanalytical X'pert Pro system using iron-filtered cobalt K[alpha] radiation and X'Celerator silicon strip X-ray detector. Clay fractions were dominated by feldspar clay minerals and lower yet equal proportions of illite and kaolinite (Table 1).

Preparation of solutions of varying SAR and PAR concentration

To investigate [Na.sup.+] and [K.sup.+] cation dynamics in Na-Ca and K-Ca binary systems, 12 solutions of varying SAR and PAR concentration were prepared using chloride salts of [Na.sup.+], [Ca.sup.2+], and [K.sup.+], at an ionic strength of 2.5 x 100 [sup.2] M, which equates approximately to a total cation concentration of 18-25 [mmol.sub.c]/L (Kopittke et al. 2006). This ionic strength is within the range observed in winery and municipal wastewater; however, it does represent the upper range. Concentrations were chosen in order to avoid experimental difficulties with persistent dispersion at higher SAR or PAR. At a lower ionic strength, the affinity of [Ca.sup.2+] to the charged soil surface relative to [Na.sup.+] and [K.sup.+] is likely to increase (Kopittke et al. 2006). Solutions prepared for investigating the cation dynamics in a ternary system (Na-K-Ca) were also prepared using chloride salts of [Na.sup.+], [Ca.sup.2+], and [K.sup.+]. For ternary system investigations, 36 solutions containing pre-determined ratios of [Na.sup.+] and [K.sup.+] to give the desired SAR to PAR ratio were prepared, at an ionic strength of 2.5 x [10.sup.-2] M. All samples for both the binary and ternary system were run in duplicate.

Soil-solution equilibration procedure for measurement of selectivity coefficients

A 4-g soil sample was placed into a pre-weighed 50-mL centrifuge tube with a 40-mL aliquot of the corresponding equilibration solution. The suspensions were initially mixed using a vortex shaker for ~5 min, then shaken end-over-end for 16 h. Samples were then centrifuged at 3450G for 20 min and drained, and a fresh 40-mL aliquot of the corresponding equilibration solution was added to each tube (Kopittke et al. 2006). This process was repeated three times with the supernatant collected after the final equilibration step and analysed for [Ca.sup.2+], [Mg.sup.2+], [K.sup.+], and [Na.sup.+] using ICP-OES.

Once the equilibration procedure was completed, tubes containing soil were weighed to determine the volume of the entrained solution. Free cation masses of the solution were then calculated based on the composition of the final equilibration solution. Exchangeable cations retained in the soil were determined by extraction with 0.1M Ba[Cl.sub.2]/0.1M N[H.sub.4]Cl (Gillman and Sumpter 1986), as described above, and by subtracting the free cation mass of the entrained solution.

Calculation of selectivity coefficients in binary and ternary systems

To assess the relationship between soil ESP and EPP in binary Na-Ca and K-Ca systems, Vanselow selectivity coefficients were calculated based on the approach used by Marsi and Evangelou (1991) and Kopittke et al. (2006) and described as follows:

[K.sub.v] = [X.sub.m] [[a.sup.0.5].sub.Ca]/ [[X.sup.0.5].sub.Ca] (5)

[X.sub.m] = [m.sub.ex]/ [m.sub.ex] + [Ca.sub.ex] (6)

[] = [Ca.sub.ex]/[m.sub.ex] + [Ca.sub.ex] (7)

where [K.sub.v] is the Vanselow exchange selectivity coefficient, [] is mole fraction of exchangeable [Ca.sup.2+], [X.sub.m] is mole fraction of the exchangeable monovalent cation (i.e. [K.sup.+] or [Na.sup.+]), [Ca.sub.ex] is exchangeable [Ca.sup.2+] concentrations (mol/kg soil), [m.sub.ex] is exchangeable monovalent cation concentration ([K.sup.+] or [Na.sup.+]) (mol/kg soil), [] is activity of solution-phase [Ca.sup.2+] cation, and [a.sub.m] is activity of solution-phase monovalent cation ([K.sup.+] or [Na.sup.+]). Cation activities in solution were calculated using Phreeqcl 2.15.0 (Parkhurst 2003).

To address the selectivity of[Na.sup.+] v. [K.sup.+] in a ternary system, the K-selectivity coefficient presented by Shainberg et al. (1987) was adopted. This is described as:

K = [X.sub.k] x [a.sub.Na]/[X.sub.Na] x [a.sub.k] (8)

where K is selectivity coefficient [[(L/mol).sup.0.5]], [X.sub.K] and [X.sub.Na] are mole fraction of exchangeable [Na.sup.+] and [K.sup.+], and [a.sub.Na] and [a.sub.K] are activity of solution-phase monovalent cation (i.e. [K.sup.+] or [Na.sup.+]) calculated using Phreeqcl 2.15.0 (Parkhurst 2003). Values of K >1 indicate preference for [K.sup.+], while values <1 indicate preference for [Na.sup.+].

Statistical analyses

Comparison of predicted and measured ESP and EPP curves was carried out using paired sample t-tests in PASW Statistics 18 (SPSS 2009). For measured values, non-linear regression analysis was performed using PASW Statistics 18 (SPSS 2009) fitting curves of the form Y=A + [B.sup.(CX)] (Mitscherlich's model), where Y is ESP or EPP and X is SAR or PAR. Significant differences between measured ESP and EPP in binary and ternary systems were determined by analysis of variance (ANOVA) and general linear model (GLM) analysis using PASW Statistics 18 (SPSS 2009) to determine the significance of interaction between SAR and PAR in ternary systems.


Adsorption of [Na.sup.+] and [K.sup.+] in binary Na-Ca and K-Ca systems

In the Na-Ca binary system, the ESP of the Barossa soil increased with increasing SAR of the equilibrating solution (Fig. 1). This relationship was not linear but best described by a exponential function of the form ESP = A + [Be.sup.C(SAR)], where A, B, and C are adjustable parameters and where A is the asymptote to which ESP tends at high SAR; for the Na-Ca binary system, this is 40. This would suggest a decreasing affinity for [Na.sup.+] with increasing SAR and corresponding saturation of available binding sites on the clay surface. This result is not unexpected, as several studies have reported similar trends in [Na.sup.+] binding with increasing SAR (e.g. Shainberg et al. 1987; Levy and Feigenbaum 1996; Kopittke et al. 2006).


The data in our study were incorporated into the USSL SAR-ESP model, whereupon the curve of measured ESP had a greater asymptote relative to the curve of predicted values. This model has been shown to vary in response to the ionic strength of the soil solution and clay mineralogy (Shainberg et al. 1980; Kopittke et al. 2006). In pure clay systems, for instance, Kopittke et al. (2006) reported a significantly greater initial slope and higher asymptote of [Na.sup.+] adsorption with increasing SAR in pure specimen kaolinite clays relative to the USSL SAR-ESP model, which was attributed to preferential binding of [Na.sup.+] over Ca2+ on external planar surfaces. In an illite clay system, however, adsorption behaviour was similar to that predicted by the USSL SAR-ESP model. Amrhein and Suarez (1991) reported no effect of clay mineralogy on [Na.sup.+] [Ca.sup.2+] selectivity in soils dominated by smectite, vermiculite, illite, and kaolinite, despite a wide contrast in surface charge densities. Measured ESP in vermiculite and montmorillonite clays was, however, higher than predicted from the USSL SAR ESP model, while for illite there was close correlation. The presence of both kaolinite and illite clay minerals in the Barossa soil is thought to account for the variation in the curve of ESP from that predicted by the USSL SAR-ESP model. Although ionic strength has been shown to play an important role in the relationship between SAR and ESP (Kopittke et al. 2006), this was kept constant at 2.5 x [10.sup.-2]M.

In the K-Ca system (Fig. 1), EPP increased with increasing PAR. The relationship between EPP and PAR was also best described by an exponential function with asymptote of PAR 70. This, again, suggests a decreasing affinity of the clay surface for [K.sup.+] with increasing PAR. In the K-Ca system, however, the initial slope of the curve was steeper than in the Na-Ca system and the point of maximum adsorption was significantly (P<0.05) greater. Again, predicted EPP provided by the USSL PAR-EPP model is also presented and is significantly (t(11)=9.731, P<0.001) lower than measured values. The initial slope and magnitude (i.e. asymptote) of [K.sup.+] adsorption was also significantly (P< 0.05) greater than that observed for [Na.sup.+] in the Na-Ca system, thereby suggesting higher binding affinity of [K.sup.+] in these soils. Illite is a 2:1 clay mineral with tetrahedron-octahedron-tetrahedron layering of 10 [Angstrom] thicknesses (McBride 1994). Frayed edges of illite clay layers preferentially adsorb [K.sup.+] by 'weak force fixation' (Sawhney 1972; Rajec et al. 1999). Of the major exchangeable cations ([Ca.sup.2+], [Mg.sup.2+], [Na.sup.+], and [K.sup.+]), [K.sup.+] is the least strongly hydrated (Rajec et al. 1999) and attracted most strongly to sorption sites of the frayed edge. The smaller hydrated ion size enables [K.sup.+] to occupy interlayer spacing of 2:1 clay layers while other larger cations are excluded. Higher binding affinity of [K.sup.+] at low PAR may also reflect the preferential binding of [K.sup.+] with illite clays until the exhaustion of these specific sorption sites. A similar decrease in [K.sup.+] affinity in illite clays equilibrated with solutions of increasing PAR was reported by Sawhney (1970) and Levy and Feigenbaum (1996). Kaolinite, on the other hand, is a 1:1 clay mineral that comprises layered tetrahedral-tetrahedral facing surfaces and is dominated by external, non-specific binding sites. Other studies have shown preferential exchange selectivity for [K.sup.+] in illites over [Ca.sup.2+], [Mg.sup.2+], and [Na.sup.+], particularly at low PAR (Sawhney 1972; Tucker 1985), and as shown by Tucker (1985), a large portion of [K.sup.+] is specifically bound.

Binding affinity of [Na.sup.+] and [K.sup.+] in a binary system

The magnitude of the [K.sub.v] selectivity coefficient (Fig. 2) indicates affinity of the clay surface for either [Na.sup.+] or [K.sup.+] in each of the binary systems. A value >1 indicates preference for the monovalent cation (i.e. [K.sup.+] or [Na.sup.+]) over [Ca.sup.2+], while a value <1 indicates preference for [Ca.sup.2+]. The [K.sub.v] values for both the K-Ca and Na-Ca systems (Fig. 2) were generally >1, indicating preference for the monovalent cation over [Ca.sup.2+] . In both systems, [K.sub.v] values decreased with increasing saturation of either exchangeable [Na.sup.+] or [K.sup.+] (i.e. increasing ESP or EPP), indicating that these cations are preferentially bound to soils when saturation of exchangeable sites is low. It has often been observed that binding strength of solutes, including metal cations, decreases with increasing saturation of solutes as the negative charge potential is neutralised (Kinniburgh and Jackson 1981; Bolan et al. 1999).

In clays dominated by external surfaces (i.e. kaolinite), monovalent cations tend to show binding preference over divalent cations at low ionic strength (i.e. <0.2 x [10.sup.-2] M) (Shainberg et al. 1980). Based on diffuse double layer (DDL) theory, charge density along the surface of the clay particle exerts an attractive force on the layer of cations counter-balancing soil surface charge, thereby compressing the DDL (Sposito 1983; Bolan et al. 1999). Dehydrated ions are subsequently drawn close to the soil surface, entering an inner region of the DDL (Stern layer) where they are bound through specific electrostatic bonding to form inner layer complexes (Sposito 1983; Bolan et al. 1999; Sparks 2003; Greathouse and Sposito 2006). These cations remain exchangeable between solid and solution phases. In both binary systems (Na-Ca and K-Ca), affinity declined with an increase in monovalent cation adsorption on the exchange complex as the electric charge potential on the clay surface was reduced (Bolan et al. 1999). Other studies report a similar decline in [K.sub.v] with increasing monovalent cation saturation in binary systems involving Na--Ca (Shainberg et al. 1980; Kopittke et al. 2006) and K-Ca (Levy et al. 1988).


In the Na-Ca system, [Ca.sup.2+] is preferentially bound to soil at an ESP of 40, whereby [K.sub.v] declined to 0.77. This preference towards [Ca.sup.2+] adsorption suggests some heterogeneity of binding sites (Shainberg et al. 1980). In the presence of high ESP, expansion of 2 : 1 clay platelet layers can expose internal surfaces that have greater charge density and express strong affinity for [Ca.sup.2+] (Shainberg et al. 1980; Pils et al. 2007). It is possible, given the high ESP, that structural change has caused increased affinity for [Ca.sup.2+] in the Na--Ca binary system. Amrhein and Suarez (1991) and Kopittke et al. (2006) demonstrated a similar preference towards [Ca.sup.2+] at high ESP in pure 2 : 1 illite clays, and this was attributed to clay platelet expansion. With increasing monovalent dominance, however, clay platelets may separate and give rise to smaller domains of clay platelets or, if completely dispersed, single platelets (van Olphen 1977; McBride 1994; Pils et al. 2007).

The process of clay domain breakdown invariably increases the area of external planar surfaces, thereby increasing the binding affinity of the monovalent cation over [Ca.sup.2+] ions (Pils et al. 2007). Changes in the structural assemblage of the clay domain caused by a high monovalent cation concentration may have increased the exposed surface area of the Barossa soil, raising the affinity for monovalent cations. This would invariably moderate the rate by which the [K.sub.v] value declined with increasing exchangeable monovalent cation dominance. It should also be noted that mechanical shaking of soils end-over-end during the experimental procedure may, however, contribute to an expansion of clay platelets and increase exposure of internal surfaces (Rengasamy 2002). Soil structural changes and therefore cation exchange equilibria in undisturbed soils are likely to differ where the lack of mechanical force may prevent separation of clay platelets to the same extent.

Selectivity for [K.sup.+] in the K-Ca system was substantially greater than for [Na.sup.+] in the Na-Ca system, yet declined rapidly with increasing saturation. Other studies have shown high affinity for [K.sup.+] at low concentrations due to specific adsorption of [K.sup.+] within interlayer regions of the 2 : 1 clay platelet (Tucker 1985; Shainberg et al. 1987; Levy et al. 1988; Agbenin and Yakubu 2006). Although the Barossa soil contains a mix of kaolinite and illite, the greater selectivity for [K.sup.+] at EPP <40, followed by rapid decline with increasing [K.sup.+] saturation, suggests initial binding at specific charge sites, after which [K.sup.+] will compete for non-specific surface sites along the external basal planar surface of the clay.

Although the Barossa soil contains similar amounts of kaolinite and illite, the high affinity for [K.sup.+] and [Na.sup.+] appears to be more influenced by the illite fraction due to a greater surface charge density and high abundance of binding sites specific to [K.sup.+]. Processes that lead to the expansion of clay platelets in 2 : 1 illite clays also appear to be influential to the binding behaviour of [Ca.sup.2+], [Na.sup.+], and [K.sup.+] cations in the binary system, thereby influencing [K.sub.v].

Interaction of Na-K in a ternary system

The binding affinity of both [Na.sup.+] and [K.sup.+] in a ternary system was investigated in soil solutions containing Na-K-Ca. Curves describing the relationships SAR ESP and PAR--EPP are shown in Figs 3 and 4, respectively. In each Figure, the dotted black line describes the respective relationship in a binary system, as shown in Fig. 1.

In the ternary system, there was a significant (P<0.05) variation in the SAR--ESP relationship relative to the binary system in response to increasing PAR treatment. General linear model analysis indicated a highly significant interaction (F = 1312, d.f. = 20, P < 0.001) between SAR and PAR. Although the influence of [K.sup.+] on [Na.sup.+] binding was most pronounced at high K : Na ratios, the effect was also evident at very low PAR values, i.e. PAR 1 (Table 2). Similar observations were reported by Levy and Feigenbaum (1996), who reported an overestimation of ESP when using the USSL SAR--ESP model in a ternary Na-K--Ca system.



Investigations by Robbins (1984) indicated that when the Na : K ratio is large (i.e. 20 : 1), the presence of [K.sup.+] has limited influence on [Na.sup.+] binding and therefore ESP is adequately predicted by the USSL SAR--ESP model. When the Na:K ratio is <4:1, [K.sup.+] has an influential role on [Na.sup.+] binding, whereby ESP at a given SAR is less than predicted. In the Barossa soil, the influence of [K.sup.+] on [Na.sup.+] adsorption shows a similar trend to that described by Robbins (1984), yet was evident at a concentration ratio of 1:20 (i.e. SAR 20:PAR 1). This strong affinity of [K.sup.+] towards these soils, particularly at low concentration, may reflect specific binding to the illite clay minerals present in these soils (Shainberg et al. 1980).

There was a significant variation in curves describing the PAR--EPP relationship with increasing SAR of the solution. A highly significant interaction (F = 89, d.f. = 20, P < 0.001) between PAR and SAR was evident; however, this was less pronounced than the influence of PAR on the SAR--ESP relationship. Relative to the binary system, [Na.sup.+] had no significant (P<0.05) influence on the PAR--EPP relationship at the lowest SAR, i.e. 1 (Table 3). At SAR values >5, however, [K.sup.+] binding was significantly reduced in the presence of [Na.sup.+].

Investigating the influence of PAR on the SAR--ESP relationship, Levy and Feigenbaum (1996) reported that [Na.sup.+] had no effect on [K.sup.+] binding and suggested that EPP could be adequately predicted based on binary systems. In the Barossa soil, although the influence of [Na.sup.+] on the PAR--EPP relationship was small (Fig. 4), there was an effect of [Na.sup.+] on [K.sup.+] sorption (Table 3). This may reflect both a decrease in [K.sup.+] binding affinity with increasing PAR, as was observed in Fig. 2, and greater competition for external sites on the clay planar surface due to the increase in [Na.sup.+] concentration.

On a practical level, these results imply that in wastewaters containing a mix of both [Na.sup.+] and [K.sup.+], the predicted adsorption of [Na.sup.+] based on commonly used indices, i.e. USSSL SAR--ESP model (Richards 1954), will be overestimated due to the influence of [K.sup.+], which has the effect of lowering ESP predicted from the SAR of the water source. Rengasamy and Marchuk (2011) recently proposed the 'cation ratio of structural stability' (CROSS) equation for irrigation water, analogous to SAR or PAR. However, CROSS incorporates the differential dispersive effects of[Na.sup.+] and [K.sup.+] and the differential flocculating power of [Ca.sup.2+] and [Mg.sup.2+]. Jayawardane et al. (2011) applied this approach for assessing soil structural changes following irrigation with solutions that were of similar cation concentrations to winery wastewater:

CROSS [(mmol/L).sup.0.5] = [[Na.sup.+]] + 0.56[[K.sup.+]]/[square root of [[Ca.sup.2+]] + 0.6 [[Mg.sup.2+]] (9)

This equation, however, does not account for the preferential binding of [K.sup.+] over [Na.sup.+] that this current study suggests will occur. The sodicity risk may therefore be overestimated by the CROSS equation in soils irrigated with waters of high PAR and SAR.


Selectivity coefficient for [Na.sup.+] and [K.sup.+] in a Na K-Ca ternary system

In a ternary system, [K.sup.+] selectivity between [Na.sup.+] and [Ca.sup.2+] is shown in Fig. 5 and [K.sub.v] selectivity for [K.sup.+] and [Ca.sup.2+] in the same system is shown in Fig. 6. Similar to the binary system, selectivity for [Na.sup.+] over [Ca.sup.2+] was highest when the solution SAR concentration was low and affinity for [Na.sup.+] decreased as ESP increased. Affinity for [Na.sup.+] relative to [Ca.sup.2+] also decreased as the PAR of the soil solution increased, presumably due to the preferential binding of [K.sup.+] over [Na.sup.+]. At PAR 1, the selectivity for [Na.sup.+] was considerably greater than at PAR [greater than or equal to] 5. Furthermore, it was markedly greater than the [Na.sup.+] selectivity over [Ca.sup.2+] in the binary system. It is possible that a higher concentration of exchangeable monovalent cations in the ternary system, in combination with high [Na.sup.+] retention at low PAR concentrations, encourages some clay platelets to disperse. This in turn will increase the number of external binding sites and ultimately lead to greater [Na.sup.+] retention, similar to that observed. By increasing the PAR, however, the retention of [Na.sup.+] is subsequently reduced and breakdown of the clay domain is prevented. Expansion of the clay platelets may, however, occur and will ultimately encourage preferential retention of [Ca.sup.2+] (Kopittke et al. 2006).

The [K.sub.v] value for [K.sup.+] (over [Ca.sup.2+]) (Fig. 6) was higher than for [Na.sup.+] (over [Ca.sup.2+]). This is most evident at a low PAR concentration and further suggests preferential binding, presumably at selective sites within the interlayer and edge region of the clay (Sawhney 1972; Sparks and Carski 1985). Interestingly, at PAR < 10 there is a progressive increase in [K.sup.+] selectivity with increasing ESP. This further suggests some form of clay platelet separation and an increase in the external surface area. Both [Na.sup.+] and [K.sup.+] will compete for the newly exposed sites; however, given the greater affinity for [K.sup.+], noted in Fig. 2, this cation will be preferentially retained. This may also explain the decrease in [Na.sup.+] selectivity at PAR 1.



At PAR > 10, no change in [K.sup.+] selectivity over [Ca.sup.2+] was evident with increasing ESP, indicating homogeneity of binding sites and presumably, therefore, no dispersion of clay platelets. Essentially, this suggests that dispersion of the clay domain is a function of increasing ESP rather than EPP. As shown in Fig. 5, the soil ESP reduced with increasing PAR, and therefore it is likely that the degree of platelet dispersion will be less.

Binding affinity between [K.sup.+] and [Na.sup.+] in a Na--K--Ca ternary system

The selectivity coefficient describing the binding affinity between [K.sup.+] and [Na.sup.+] is shown in Fig. 7. Here, an overall binding preference of [K.sup.+] relative to [Na.sup.+] is again evident. The decline in [K.sup.+] selectivity with increasing PAR of the equilibrating solution is also evident and is similar to the trend of declining [K.sub.v] observed in the binary system associated with a lowering of the surface charge potential at high PAR (Sposito 1983; Bolan et al. 1999).

When PAR equals 1, a sharp increase in the selectivity of [K.sup.+] over [Na.sup.+] is evident with increasing ESP of the soil. This is consistent with evidence suggesting dispersion of the clay platelets within the illite fraction of the soil which occurs as a result of high ESP. It is also evident that when platelets separate, the exposed surfaces preferentially retain [K.sup.+] over [Na.sup.+]. Essentially, this implies that [Na.sup.+] is the principal cation responsible for the breakdown of the clay domain; however, upon breakdown, [K.sup.+] retention is likely to increase.


When wastewaters containing a high concentration of both [Na.sup.+] and [K.sup.+] are used to irrigate soils, interactions between cations need consideration when predicting the resulting impact on soil physical properties. This study supports our previous findings and those of others which show that when present in equal molar concentrations, the binding affinity of [K.sup.+] to soils exceeds that of [Na.sup.+]. In soils irrigated with winery wastewater, the relationship between SAR and ESP is moderated due to the higher [K.sup.+] affinity, particularly at high [K.sup.+] concentrations. It is expected, therefore, that soil ESP will be lower than predicted from the commonly used USSL SAR--ESP model when winery wastewater is used for irrigation. When dispersed, soil EPP is likely to increase due to preferential binding of [K.sup.+] on exposed clay platelets even in the presence of high [Na.sup.+] concentration. Ascertaining the influence of mechanical shaking on the exchange equilibria and subsequent separation of clay platelets would, however, be of value. Finally, it is important to note that elevated [K.sup.+] in grapes has a negative impact on wine quality, and therefore the translocation of [K.sup.+] in vines irrigated with [K.sup.+]-rich winery wastewater requires further attention.


This research was supported by the Co-operative Research Centre for Contamination Assessment and Remediation of the Environment (CRC-CARE) and the University of South Australia.


Agbenin JO, Yakubu S (2006) Potassium--calcium and potassium--magnesium exchange equilibria in an acid savanna soil from northern Nigeria. Geoderma 136, 542-554. doi:l0.1016/j.geoderma.2006.04.008

Amrhein C, Suarez DL (1991) Sodium--calcium exchange with anion exclusion and weathering corrections. Soil Science Society of America Journal 55, 698-706. doi:10.2136/sssaj1991.03615995005500030010x

Arienzo M, Christen EW, Quayle W, Kumar A (2009) A review of the fate of potassium in the soil plant system after land application of wastewaters. Journal of Hazardous Materials 164, 415-422. doi: 10.1016/j.jhazmat.2008.08.095

Bolan NS, Naidu R, Syers JK, Tillman RW (1999) Surface charge and solute interactions in soils. Advances in Agronomy 67, 87-140. doi:10.1016/ S0065-2113(08)60514-3

FAO (2006) 'World reference base for soil resources.' 2nd edn. World Soil Resources Reports, No. 103. (Food and Agriculture Organisation for the United Nations: Rome)

Feigenbaum S, Bar-Tel A, Portnoy R, Sparks DL (1991) Binary and ternary exchange of potassium on calcareous montmorillonitic soils. Soil Science Society of America Journal 55, 49-56. doi: 10.2136/sssaj1991.03615995005500010008x

Gapon EN (1933) On the theory of exchange adsorption in soils. Journal of General Chemistry of the U.S.S.R. 3, 144-152.

Gee GW, Bauder JW (1986) Particle-size analysis. In 'Methods of soil analysis. Part 1: Physical and mineralogical methods'. (Ed. A Klute) pp. 383-410. (American Society of Agronomy: Madison, WI)

Gillman GP, Sumpter EA (1986) Modification of the compulsive exchange method for measuring exchange characteristics of soils. Australian Journal of Soil Research 24, 61-66. doi:10.1071/SR9860061

Greathouse JA, Sposito G (2006) Electrical double-layer at particles. In 'Encyclopedia of surface and colloid science'. (Eds P Somasundaran, AT Hubbard) pp. 1955-1971. (Taylor and Francis: Boca Raton, FL)

Halliwell DJ, Barlow KM, Nash DM (2001) A review of the effects of wastewater sodium on soil physical properties and their implications for irrigation systems. Australian Journal of Soil Research 39, 1259-1267. doi: 10.1071/SR00047

Hazelton P, Murphy B (2007) 'Interpreting soil test results--What do all the numbers mean?' (CSIRO Publishing: Melbourne)

Isbell RF (2002) 'The Australian Soil Classification.' Revised edn (CSIRO Publishing: Melbourne)

Jayawardane NS, Christen EW, Arienzo M, Quayle WC (2011) Evaluation of the effects of cation combinations on soil hydraulic conductivity. Soil Research 49, 56-64. doi: 10.1071/SR09222

Kinniburgh DG, Jackson ML (1981) Cation adsorption by hydrous metal oxides and clay. In 'Adsorption of inorganics at solid--liquid interfaces'. (Eds MA Anderson, AJ Rubin) pp. 91-160. (Ann Arbor Science: Michigan, MI)

Kopittke PM, So HB, Menzies NW (2006) Effect of ionic strength and clay mineralogy on NaSa exchange and the SAR--ESP relationship. European Journal of Soil Science 57, 626--633. doi:10.1111/j.1365-2389.2005.00753.x

Kumar A, Christen E (2009) Developing a systematic approach to winery wastewater management. Report CSL05/02. Grape and Wine Research Development Corporation/CSIRO Land and Water Science, Adelaide, S. Aust.

Laurenson S (2010) The influence of recycled water irrigation on cation dynamics in relation to the structural stability of vineyard soils. PhD Thesis, Centre for Environmental Risk Assessment and Remediation, University of South Australia, Adelaide, S. Aust.

Levy GJ, Feigenbaum S (1996) The distribution of potassium and sodium between the solution and the solid phase in a ternary (K NaSa) system. Australian Journal of Soil Research 34, 749-754. doi: 10.1071/SR9960749

Levy GJ, van der Watt Hv H, Shainberg 1, du Plessis HM (1988) Potassium--calcium and sodium--calcium exchange on kaolinite and kaolinitic soils. Soil Science Society of America Journal 52, 1259--1264. doi:10.2136/sssaj1988.03615995005200050008x

Lieffering R, McLay C (1996) Effects of high pH solutions with large monovalent cation concentrations on cation exchange properties. Australian Journal of Soil Research 34, 229-242. doi: 10.1071/SR9960229

Marsi M, Evangelou VP (1991) Chemical and physical behavior of two Kentucky soils: I. Sodium--calcium exchange. Journal of Environmental Science and Health. Part A. Environmental Science and Engineering and Toxicology 26, 1147-1176.

McBride MB (1994) 'Environmental chemistry of soils.' (Oxford University Press: New York)

Menneer J, McLay C, Lee R (2001) Effects of sodium-contaminated wastewater on soil permeability of two New Zealand soils. Australian Journal of Soil Research 39, 877-891. doi:10.1071/SR99082

Parkhurst D (2003) Phreeqcl, 2.15.0 edn. United States Geological Survey. Available at: (accessed December 2009).

Pils JRV, Laird DA, Evangelou VP (2007) Role of cation demixing and quasicrystal formation and breakup on the stability of smectitic colloids. Applied Clay Science 35, 201-211. doi:10.1016/j.clay.2006.09.014

Rajec P, Sucha V, Eberl DD, Srodon J, Elsass F (1999) Effect of illite particle shape on cesium sorption. Clays and Clay Minerals 47, 755-760. doi:10.1346/CCMN.1999.0470610

Rayment GE, Higginson FR (1992) 'Australian laboratory handbook of soil and water chemical methods.' (Inkata Press Pty Ltd: Melbourne)

Rengasamy P (2002) Clay dispersion. In 'Soil physical measurement and interpretation for land evaluation'. (Eds N McKenzie, K Coughlan, H Cresswell) pp. 200-210. (CSIRO Publishing: Melbourne)

Rengasamy P, Marchuk A (2011) Cation ratio of soil structural stability (CROSS). Soil Research 49, 280-285. doi:10.1071/SRl0105

Richards LA (1954) 'Diagnosis and improvement of saline and alkali soils.' United States Department of Agriculture Report No. 60. (USDA: Washington, DC)

Robbins CW (1984) Sodium adsorption ratio--exchangeable sodium percentage relationships in a high potassium saline-sodic soil. Irrigation Science 5, 173-179. doi: 10.1007/BF00264606

Sawhney BL (1970) Potassium and cesium ion selectivity in relation to clay mineral structure. Clays and Clay Minerals 18, 47-52. doi:10.1346/CCMN.1970.0180106

Sawhney BL (1972) Selective sorption and fixation of cations by clay minerals: A review. Clays and Clay Minerals 20, 93-100. doi:10.1346/CCMN.1972.0200208

Shainberg I, Oster JD, Wood JD (1980) Sodium/calcium exchange in montmorillonite and illite suspensions. Soil Science Society of America Journal 44, 960--964. doi: 10.2136/sssaj1980.03615995004400050017x

Shainberg I, Alperovitch NI, Keren R (1987) Charge density and Na--K--Ca exchange on smectites. Clays and Clay Minerals 35, 68--73. doi: 10.1346/CCMN.1987.0350109

South Australian EPA (2004) Guidelines for wineries and distilleries. South Australian Environmental Protection Agency, Government of South Australia, Adelaide, S. Aust.

Sparks DL (2003) Sorption phenomena on soils. In 'Environmental soil chemistry'. (Elsevier: London)

Sparks DL, Carski TH (1985) Kinetics of potassium exchange in heterogeneous systems. Applied Clay Science 1, 89-101. doi:10.1016/0169-1317(85)905654

Sposito G (1983) On the surface complexation model of the oxide aqueous solution interface. Journal of Colloid and Interface Science 91, 329-340. doi: 10.1016/0021-9797(83)90345-4

SPSS (2009) 'PASW Statistics 18.0. Rel. 18.0.0.' (SPSS Inc.: Chicago, IL)

Stevens DP, McLaughlin MJ, Smart MK (2003) Effects of long-term irrigation with reclaimed water on soils of the Northern Adelaide Plains, South Australia. Australian Journal of Soil Research 41, 933-948. doi: 10.1071/SR02049

Tucker BM (1985) The partitioning of exchangeable magnesium, calcium and sodium in relation to their effects on the dispersion of Australian clay subsoils. Australian Journal of Soil Research 23, 405-416. doi: 10.1071/SR9850405

van Olphen H (1977) 'An introduction to clay colloid chemistry.' 2nd edn (John Wiley Sons: New York)

Vaneslow AP (1932) Equilibria of the base exchange reaction of bentonite, permutites, soil colloids, and zeolites. Soil Science 33, 95-114. doi: 10.1097/00010694-193202000-00002

Manuscript received 26 March 2011, accepted 12 August 2011

S. Laurenson (A,C), E. Smith (A), N. S. Bolan (A), and M. McCarthy (B)

(A) Centre for Environmental Risk Assessment and Remediation, University of South Australia, Mawson Lakes, SA 5095, Australia.

(B) South Australian Research and Development Institute, Viticulture Division, Nuriootpa, SA 5355, Australia.

(C) Current address: AgResearch Invermay, Private Bag 50034, Mosgiel 9053, New Zealand.

(D) Corresponding author. Email:

10.1071/SR11192 1838-675X/11/060538
Table 1. Selected soil properties of the Barossa B-horizon soil
Values are average f standard deviation

Soil parameter                                   Value

[pH.sub.(1:5)]                              6.8 [+ or -] 0.5
EC.sub.(1:5)] (dS/m)                     0.132 < [+ or -] 0.001
Total organic carbon (g 100/g)             0.33 [+ or -] 0.04
[CEC.sub.e] ([cmol.sub.c]/kg soil)        17.46 [+ or -] 0.30
[Ca.sup.2+]                               12.29 [+ or -] 0.20
[Mg.sup.2+]                                1.85 [+ or -] 0.04
[K.sup.+]                                  1.09 [+ or -] 0.01
[Na.sup.+]                                 2.24 [+ or -] 0.05
Particle size analysis                     34% sand, 5% silt,
                                                61% clay
Textural class                                 Heavy clay
Dominant clay minerals (wt %)          10% illite, 10% kaolinite,
                                       20% orthoclase, 20% albite

Table 2. Exchangeable sodium percentage (ESP) in soils equilibrated
with solutions of varying sodium and potassium adsorption ratios
(SAR and PAR) in a ternary Na-K-Ca system

The ESP is expressed as percentage of exchangeable [Na.sup.+]
comprising the effective cation exchange capacity. For each SAR value,
ESP values followed by the same letter are not significantly different
(P>0.05) in response to PAR treatment. In all instances n=5

SAR                                PAR
          0            1            5            10           15

1        3.72a        1.92b        1.40c        1.10d        0.82e
5       13.29a        9.92b        7.75c       6.50cd       5.30de
10      21.13a       16.86b       13.54e       11.23d        9.42e
15      27.39a       22.93b       18.10c       15.38d       13.00e
20      32.79a       27.58b       22.15c       18.84d       15.73e

SAR      PAR        F, d.f.

1        0.49f      1167, 5
5        4.64e       160, 5
10       8.10e       364, 5
15      10.97f      2097, 5
20      13.75e       283, 5

Table 3. Exchangeable potassium percentage (EPP) in soils
equilibrated with solutions of varying sodium and potassium
adsorption ratios (SAR and PAR) in a ternary Na-K-Ca system

The EPP is expressed as percentage of exchangeable [Na.sup.+] or
[K.sup.+] comprising the effective cation exchange capacity. For each
PAR value, EPP values followed by the same letter are not
significantly different (P>0.05) in response to SAR treatment. In all
instances n=5

PAR                       SAR                                  F, d.f.
         0         1         5        10       15       20

1     15.42a    16.15a    15.09ab   14.03b   12.24e   11.43c    82, 5
5     38.84a    37.43ab   35.66b    33.01c   30.04d   27.29e   157, 5
10    52.86a    52.51a    49.56b    45.77c   41.38d   38.61e   360, 5
15    60.44a    61.39a    57.36b    53.67c   49.04d   46.01e   730, 5
20    65.90a    66.13a    63.27b    59.01c   54.05d   51.61e   461, 5
COPYRIGHT 2011 CSIRO Publishing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2011 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Laurenson, S.; Smith, E.; Bolan, N.S.; McCarthy, M.
Publication:Soil Research
Article Type:Report
Geographic Code:8AUST
Date:Sep 1, 2011
Previous Article:Factors affecting potassium release in calcareous soils of southern Iran.
Next Article:Micromorphology of sepiolite occurrences in recent lacustrine deposits affected by soil development.

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters