INTRODUCTION

Potassium (K) is an essential macronutrient for plants. Physicochemical properties of the soil and soil management are the most important factors affecting the composition of K forms in the soil. Regarding the significant role played by the K in the growth of crops and humans, deficiency of this element in agricultural farmlands has become a global problem. Nowadays, due to intensive farming and insufficient application of K-supplying fertilizers, many soils are being exposed to K deficiency [3]. Availability of K for plants is generally governed by different forms of K, e.g. water-soluble K, exchangeable K, fixed K and mineral K. Soluble and exchangeable K forms are known to be rapidly released into the soil, making them readily available for plants. A recent study showed that these forms of K are considerably more abundant in the surface, rather than the subsurface, soil layer [14]. The distribution of K forms in the soil depends on particular properties of the soil such as the soil texture, content and type of clay minerals in the soil, content of organic matter (OM), land use, the soil moisture and temperature regimes, and the soil development [38]. Various factors can affect the composition and distribution of K forms in a soil sample, including the soil physical and chemical properties such as particle size distribution, the contents of carbonates and OM, cation exchange capacity (CEC) and electrical conductivity, the type and amount of K-bearing minerals of the soil such as mica and feldspar, the content of K‑stabilizing minerals of the soil such as vermiculite, smectite and weathered mica, the soil moisture and temperature conditions, the soil evolution, land use, and the soil physiographical status in the ground board, as studied by various researchers [17, 23, 30]. Non-exchangeable form of K has been less available as it occurs between the layers of mica; intensive crop production and insufficient application of K fertilizers has further reduced the availability of water-soluble and exchangeable K forms, thereby eliminating the non-exchangeable form of K from the layers of mica. Wang et al. [40] proposed a threshold for the K concentration in soil solution, above which the release of non-exchangeable K from clay minerals was inhibited. This threshold depended on the type of dominant clay mineral in the soil and the presence of calcium (Ca) and sodium (Na) ions. The presence of different cations in the soil solution affects the conversion of different forms of K into one another. With higher levels of hydration, energy and size than those of K, Ca ions cannot replace the interlayer K ions, as is evident in illite-bearing soils [37].

Clay minerals have a variety of sites for K storage, including surface, edges, and inner layers. The release of K from different sites is determined by the type of extractant. The K can be easily removed from the surface sites through an ion exchange mechanism, while the K ions on the edges and inner layers of the minerals require more energy to be released [7, 19]. In many cases, the soils in arid and semi-arid regions contain large amounts of non-exchangeable form of K as a result of their biotite and muscovite contents. The K ions trapped between the inner layers of clay minerals provide important source of K for plant growth in most soils. So far, various organic and mineral extractants have been used to investigate the mechanism and rate of K release from clay minerals in the soil. Due to its abundance in calcareous soils, the Ca ion has been the most common ion for exchanging the inter-layer K particles. Therefore, successive extraction of K with Ca has been an appropriate method for kinetic evaluation of non-exchangeable K release in these soils. The organic acid content of the soil contributes to weathering of the soil minerals by forming organic complexes [39]. Oxalic acid is an organic acid that largely contributes to the availability of nutrients in the soil. As the simplest dicarboxylic acid with two pKa values (i.e., 1.23 and 4.19), it has been found in sediments, forests and agricultural soils, especially in the rhizosphere of plants [6]. There are several kinetic models to describe the K release process, including Elovich, parabolic diffusion, power function, and zero-order and first-order equations. In general, the use of different equations to examine the K release kinetics and determination of the best equation depend on the main mechanism affecting the K release, the chemical and mineralogical composition of the soil, the type of laboratory method used, and pretreatment of the soil before starting the experiment [36]. Today, optimal and accurate management of fields and fertilizer recommendations require a comprehensive set of information on the mineral composition of the soil, various forms of K, and the relationship between the minerals and K stabilization and release. This sheds light on the need for comprehensive studies on the state of K forms in terms of their distribution in the soil, non-exchangeable K release, and the relationship among K forms, soil mineralogy, and soil properties across long-cultivated agricultural fields. Therefore, the present study aimed to investigate the composition and distribution of K forms, the kinetics of non-exchangeable K release using both organic and inorganic extractants, and their relationship with soil properties and clay mineralogy in highly cultivated calcareous soils in the southern of Iran.

MATERIALS AND METHODS

Physicochemical Properties of Soils

Based on previous soil survey maps, fifteen pedons with a wide range of different soil properties that were developed under different climatic conditions (aridic, xeric and ustic soil moisture regime) and physiographic positions (Alluvial fans (A.F), the Piedmont plains (P.P), and the Lowlands (L.L)), with calcareous parent materials were selected from northern and northwestern of Fars province, southern Iran (Fig. 1). To minimize the effect of potassium fertilizers, sampling was performed in autumn and before cultivation. Pits were dug, described and classified according to Keys to Soil Taxonomy [35]. All soil horizons were sampled (a total of 49 samples), air-dried and sieved (<2 mm) for laboratory analyses. Physical, chemical, and mineralogical properties of 30 surface and subsurface soils (15 pedon) of intensively cultivated regions in Fars Province used in the present study were determined. Soil particle-size distribution was determined using the hydrometer method [4]. Calcium carbonate equivalent (CCE) was measured by neutralization with HCl [15], and organic carbon (OC) was determined by wet oxidation method [24]. Soil pH was measured using the saturated paste extract, electrical conductivity (EC) was measured in soil saturation extract using conductometer, was determined using the saturated paste extract, cation exchangeable capacity (CEC) was analyzed by replacing exchangeable cations by NaOAc and exchanging Na+ with NH4OAc [5], and gypsum was measured by precipitation with acetone [29].

Fig. 1.
figure 1

Location of studied area in Fars Province, southern Iran.

Full size image

Potassium Forms

The content of potassium present in different forms was measured by methods outlined by Helmeke and Sparks [8]. Total K was determined following digestion (383°K) of 0.5 g soil sample with 10 mL of 48% HF and 1 mL of aqua regia.1 Solution K was measured in the saturated extract. Exchangeable K was determined by extraction of 5 g soil sample with 20 mL 1M ammonium acetate (pH 7) for 5 min. HNO3-extractable K was measured by extraction of 2.5 g soil sample with 30 mL of boiling 1.0 M Nitric acid for 1 h. Non-exchangeable K was calculated as the difference between HNO3-extractable K and NH4O Ac-extracteable K. Potassium was measured in filtered extracts using flame photometer. Three replications were conducted for all the analyses and the results were presented as mean.

For assessing the cumulative non-exchangeable K release with time, we firstly removed the native K with saturation by equilibrating 10 g of soil with 1 M CaCl2 three times for 48 h. We then washed out the extra Cl– by alcohol and deionized water. Samples were dried at 65°C. For each soil sample (30 in total), 2 g of Ca-saturated soil was placed in two centrifuge tubes (in duplicate) with different media including 10 mmol CaCl2 and 10 mmol oxalic acid solutions (20 mL in each tube) for 1200 h (2, 6, 12, 24, 48, 72, 72, 120, 168, 168, and 336 h) at 25 ± 1°C. Subsequently, they were shaken for 1 h before the suspensions were centrifuged for 10 min. At each stage, the supernatant liquid was determined for K using the flame photometer (Elico Model CL-360, India) [16].

Mineralogical Analysis

Clay mineralogy was performed on the first and second horizon (total of 30 samples) of each pedon. Prior to mineralogical analysis, samples were repeatedly washed to remove gypsum and soluble salts. Carbonates were removed using 1N sodium acetate, buffered at pH 5; this reaction was performed in a water bath at 80°C. Organic matter was oxidized by treating the carbonate-free soils with 30% H2O2. Iron oxides were removed from the samples by the dithionate citrate bicarbonate method [18]. The clay fraction was separated by centrifuge at 750 rpm for 5.4 min according to Kittrick and Hope [13] and Jackson [10]. After separation of clay fraction, samples were saturated with Mg2+ and K+, using 1 N MgCl2 and 1 N KCl, respectively. The Mg2+ and K+ saturated samples were saturated by ethylene glycol and heated at 550°C, respectively. In addition, samples were treated with 1 N HCl to discriminate kaolinite and Fe chlorite. Relative peak positions identify clay minerals, and peak intensities are the basis for semiquantitative estimates of mineral percent. The semiquantitative percentage of the clay minerals were estimated from the relative first order XRD peak areas of ethylene glycol solvated samples according to Johns et al. [12].

RESULTS AND DISCUSSION

Soil Classification across the Study Area

Based on Soil Survey Staff [35] and considering the surface and subsurface horizons and the moisture and temperature conditions across the study area, the studied soils were classified under five orders, namely Entisols (soils no. 1, 6, and 11), Inceptisols (soils no 7, 9, 12, 14, and 15), Alfisols (soils no. 8, and 13), Aridisols (soils no. 2, 3, 4 and 5) and Vertisols (soils no. 10). The mentioned orders were belonged to three physiographical units, including the Alluvial fans (AF), the Piedmont plains (PP), and the Lowlands (LL) (Table 1). Given the climatic conditions across Fars Province, local precipitation level is less than enough to affect the leaching of the soluble salts, transportation of carbonates, sulfates and fine clays, accumulation of OM, and, in general, the soil evolution. Indeed, the local soil properties largely follow the composition of parent materials, with the high content of calcium carbonate inhibited the soils evolution. Among the soil-forming factors, the topography and climate have imposed the largest impacts on the soil particle size distribution, type and distribution of OM, type of diagnostic horizons, and degree of profile evolution. Table 1 presents physical and chemical properties of the studied soils. According to the table, the pH in the studied soils was in the neutral and weak alkaline range (7 to 8.4), with the highest pH values observed in the soils of the aridic moisture regime. The highest electrical conductivity (EC) values were observed in dry lands across the piedmont plains and lowland physiographic units. The studied soils were calcareous with CaCO3 contents in the range of 9.9 to 74.3% (mean 47.8% in surface soils and 52.6% in subsurface soils). The highest contents of CaCO3 were associated with the Aridisols developed under aridic moisture regime. The organic carbon content of the soil samples ranged from 0.1% to 5.2%. The highest OM content was measured in the surface horizons dominated by the xeric moisture regime, while the lowest OM content was found in the soils taken from arid regions. On average, the OM constituted 1.4 and 0.7% of the surface and subsurface soils, respectively. A study of the particle size distribution indicated a heavier soil texture as one moved from more arid parts of the study area to more humid parts. The clay content of the soil varied from 8 to 62% across the study area. On average, lower clay contents were observed in the Aridisol, while the samples taken from Inceptisol and Alfisol orders that were dominated by a xeric regime exhibited higher clay contents. Across the study area, the CEC changed from 5.2 cmol(+) kg–1 in the soil of Aridisol order dominated by aridic regime to 29.4 cmol(+) kg–1 for the soil of Vertisol order under xeric moisture regime.

Table1.   General characteristics and physico-chemical properties and Semi-quantitative analysis of clay minerals, <0.002 mm in studied area
Full size table

Clay Fraction Mineralogy

X-ray diffraction analysis was performed to evaluate the clay fraction (<0.002 mm) in surface and subsurface soils, with the results presented in Table 1. The semi-quantitative analysis of clay fraction showed that the most important minerals in the clay fraction of the soils included illite, chlorite, smectite, and palygorskite, with trace amounts of vermiculite, kaolinite, quartz, and mixed minerals further detected. Illite was the most abundant mineral in all studied soil samples. Its wide availability and increasing content with depth in most profiles and the abundance in parent materials of the soil suggest that significant part the illite in the studied soils was of a hereditary origin and due to the presence of some of illite in the surface horizons; the process of illitization also seems probable. The highest amount of illite was observed in the P.P followed by A.F and L.L units. Smectite was the second dominant mineral in the clay fraction of the studied soil samples, constituting 10 to 45% (average: 24%) of the surface soils and 14 to 48% (average: 28%) of the subsurface soils. The smectite was seemingly come from different sources depending on the weather conditions, the parent materials, and the physiography of the studied soil. Based on the results of the semi-quantitative analysis of the clay minerals, the amount of smectite was negligible in younger soil samples with no profile development. The presence of this mineral in the parent materials of the studied soils indicated that this mineral was inherited from the parent material. It was also observed that the content of this mineral increases with the soil development (Alfisol), demonstrating the pedogenesis of this mineral‎. Smectite content was highest in P. P followed by L.L, and A.F units. On the other hand, the content of this mineral was increased across the lands of inappropriate drainage conditions (lowlands in humid areas) or shallow groundwater levels. This might be attributed to the pedogenesis of this mineral from chlorite and mica, as confirmed by the reduction of these minerals across the studied area, or to neoformation of soil solution, a proof of which is the contents of carbonated and Mg- and Ca-rich groundwater across these lands. Following the smectite, the chlorite and palygorskite minerals were the most abundant minerals in the clay fraction of surface and subsurface soils of the studied areas, comprising 21% and 11% of the clay fraction, respectively. On average, 13% and 16% of the surface and subsurface horizons were composed of palygorskite, respectively, with the highest levels observed in drier areas (mostly in pedons 1 to 5) and negligible levels in more humid areas. Only small amounts of vermiculite were found in the soil pedons 8, 9 and 10, with the other pedons exhibiting no amount of this mineral. Given the soil condition, the vermiculite is probably of an inheritance source. The illite and chlorite existed in all soil samples though at different contents. Some soil samples showed small amounts of kaolinite, with its average content being generally below 1%. The chlorite and kaolinite contents of the soils across the study area are of inherited origins, resembling the soils covering other southern parts of Iran. Azadi et al [2] reported the presence of chlorite, illite and kaolinite in the soils of arid areas and considered the hereditary origin as the main cause of their presence in the soil. Mixed minerals (i.e., illite-smectite and chlorite-smectite mixtures) constituted a small portion of the clay fraction (~1%). Najafi and Abtahi [22], Owliaie et al. [26], and Shakeri and Abtahi [32] reported similar results for calcareous soils in southern Iran. To sum up, based on their contents in the surface and subsurface horizons, the clay minerals were sorted in the following order illite > smectite > chlorite > palygorskite > vermiculite > kaolinite.

Potassium Forms

Water-soluble and exchangeable K. The water-soluble (or simply soluble) K is the form that can be absorbed by microorganisms and plants. Any form of K in the soil must be converted to a solution before it can be taken up by a plant. This indicates the importance of soluble K in plant nutrition [9]. As can be seen in Table 2, amount of soluble K contents ranged within 0.87–37.11 mg kg–1 and 0.49–4.1 mg kg–1 in surface and subsurface horizons, respectively. The maximum value of soluble K was observed in the surface soil covering the lowlands dominated by the xeric moisture regime, and the minimum value was seen in the subsurface soil covering the piedmont alluvial plains dominated by ustic moisture regime (Higher values of water-soluble K were observed for L.L followed by P.P and A.F units). The higher levels of soluble K in the surface soils could be attributed to the transfer of soluble salts, including K, from the depth to the surface as a result of evaporation of the irrigation and precipitation moisture, greater weathering of K-bearing minerals on the surface, higher levels of OM on the surface, K uptake by the plant roots from depth and its release to the surface upon the plant depth, and application of chemical fertilizers to the soil surface [22]. In their study on the soils of southern Iran, Najafi Ghiri et al. [23] obtained high levels of soluble K in the soils of arid areas dominated by aridic moisture regime, hyperthermic temperature regime and lowland soils. Despite this, in the end, not significant difference was observed between different units and moisture regimes (Fig. 2). The soluble K form has found to be significantly associated with OM content (0.522**), exchangeable K form (0.503**) and non-exchangeable K form (0.291**), but with neither structural K form nor other physicochemical properties of the soil. In general, a very small part of the total K of the studied soil samples (~0.1%) was composed of soluble K in both surface and subsurface horizons.

Table 2.   Amounts of different K forms and cumulative non-exchangeable K released in different media of extraction in representative Pedons
Full size table
Fig. 2.
figure 2

Potassium fractions for various physiographic units and soil moisture regimes.

Full size image

The exchangeable form of K can be easily used by the plant, exchanged with other cations, electrostatically bonded with OM and clay mineral surfaces. In the studied soil samples, the contents of exchangeable K ranged within 93.174–792.61 and 121.07–502.85 mg kg–1 for surface and subsurface horizons, respectively, comprising a total of 4.2% and 2.7% of the total K content in the surface and subsurface soils, respectively (2.9% on average). The highest amount of exchangeable K was observed in surface horizons of profile 8 (Alfisols, 791.62 mg kg–1, on average). These surface horizons contained large amounts of OM and smectite. The lowest amount of exchangeable K was observed in profile 5 (Aridisols, lowlands). In their studies, Shakeri et al. [31] observed the lowest amount of exchangeable K in the soils that were dominantly composed of illite, chlorite, and palygorskite, which is consistent with the mineralogical results of the Aridisol soil in this study. Regarding the effect of physiographic units on the distribution of exchangeable potassium forms, its higher values ​​were observed in the P.P followed by A.F and L.L physiographic units (Fig. 2). The P.P unit in xeric moisture regime has more exchangeable potassium due to its high clay content and high levels of illite and smectite. Based on the findings, the CEC and the contents of CaCO3, clay, illite, smectite, and vermiculite were significantly correlated to the content of the exchangeable K at 0.491*, 0.729**, 0.384*, 0.449*, 0.445** and 0.534*, respectively, in all pedons and horizons. No significant relationship was observed between the OM content and the content of exchangeable K in the studied soil samples. This highlighted the more important role of the clay fraction composition rather than the OM content in the retaining of this form of K in the studied soils. Sinha and Biswas [34] reported that soluble, exchangeable and non-exchangeable K forms have significant positive relationships with the clay content and CEC in the soils of western Bengal, India.

Non-exchangeable and structural potassium. Non-exchangeable K provides a major source of K for plants. The content of non-exchangeable K can be calculated by subtracting the exchangeable K content extracted by 1 M NH4OAc solution from the experimentally extracted amount of K by nitric acid. Non-exchangeable K is somehow different from structural (mineral) K [36]. The content of non-exchangeable K changed within 246.81–1446.14 and 219.42–1131.52 mg kg–1 in the surface and subsurface horizons, respectively, making up 8.9% and 8.3% of the total K in the surface and subsurface soils, respectively. The highest levels of available (soluble and exchangeable forms) and non-exchangeable K forms were observed under the xeric moisture regime. The results showed that, like the exchangeable K, the content of this form of K was higher in the surface soils rather than the subsurface horizons. The highest amount of non-exchangeable K form (1446.14 mg kg–1) was observed in the surface soil of the Alfisols order, and the lowest level was associated with the Inceptisols order. In their study on the soils covering Fars Province, Azadi et al. [1] stated that, except for the soluble K form, all K forms were maximal in more evolved soils of Alfisol with smectite and illite as dominant minerals. They further reported significant relationships among all forms of K, except the soluble K form, with illite as the predominant mineral. Raheb and Heidari [27] found that, due to the temperature and moisture conditions that favored the weathering of the primary minerals, the levels of exchangeable and non-exchangeable K forms in the surface soil were higher than those of the in-depth soil. Hosseinpour and Kavusi [9] suggested that the content of non-exchangeable K level was seemingly higher in heavy-texture soils (with higher clay contents) that were predominantly composed of smectite. Despite that P.P units had highest NEK, but no significant difference was observed between the physiographic units, while the xeric moisture regime contained the highest amount NEK between three moisture regimes (Fig. 2).

In all samples, the content of non-exchangeable K form exhibited significant correlations to CEC (0.354 *), CaCO3 (–0.788**), clay content (0.474**), smectite (0.481**), illite (0.492**) and vermiculite (0.534**). The significant positive relationship between the clay content and the non-exchangeable K form in all soils shows the importance of the clay content in the content of this form of K [1, 31].

Structural K, which is found in the structure of K‑bearing minerals such as mica and feldspar, is unavailable for the plant. However, by means of weathering, it can be converted to other forms of K that can be taken up by the plant in long run. On average, some 90% of total K in the studied soil samples from the surface and subsurface horizons was composed of structural K, indicating limited weathering of the soils of arid and semi-arid areas that led to higher content of hereditary illite and lower profile evolution. However, the contribution of the structural K form decreased in humid areas, indicating greater shares of exchangeable and non-exchangeable K forms due to further conversion of the illite to smectite and vermiculite. The surface and subsurface horizons contained exchangeable K form in the ranges of 30.35–13109.92 (mean: 6574.7) and 3692.99–15836.65 mg kg–1, respectively (Table 2), forming 86.8 to 88.9% of total K. As can be seen, the level of structural K in the surface horizons was less than that in subsurface ones. Environmental conditions at surface (seasonal day-night temperature difference, higher moisture due to precipitation and irrigation, and microorganism activities) contribute to instability and weathering of minerals in the soil surface. As a result, more K ions are removed from primary and secondary minerals at surface. This has further led to the conversion of the structural K into exchangeable and non-exchangeable forms in surface horizons [31]. The highest level of structural K (15 837 mg kg–1) was observed in pedon 8 (Alfisols, dominated by the xeric moisture regime). Like the NEK, this form of potassium, considering that it contained the highest amount in the P.P unit, but there was no significant difference between the studied units, but in examining the effect of moisture regime on Distribution of potassium fractions, xeric moisture regime also contained the highest amount (Fig. 2). As shown in Table 3, all around the study area, the structural K form content was significantly correlated to exchangeable K (0.758**), non-exchangeable K (0.812**), CEC (0.602**), CaCO3 (–0.872**), clay content (0.472**), illite (0.664*), smectite 0.612**), vermiculite (0.616**), and total of smectite and vermiculite (0.699*). These results suggest that the highly layered loads of smectite, vermiculite, and illite play important roles in the content of structural K in the studied soils. Sharma et al. [33] reported that all different forms of K are directly associated with the clay and silt contents and indirectly related to the sand content. The reason for the negative correlation of K forms to the contents of CaCO3 and sand is the mass storage of K in the form of clay and the decrease in the K form content with increasing the sand content. In line with the results reported by Nabiollahi et al. [21], positive significant correlations were obtained between the contents of different K forms in this study (Table 3), indicating some dynamic equilibrium among the forms [25].

Table 3.   Pearson correlation coefficients between soil K forms, clay minerals and physical-chemical properties
Full size table

Kinetic Modeling of K Release (with CaCl2 and Oxalic Acid as Extractants)

The pattern of successive extraction of K from some representative soil samples using CaCl2 and oxalic acid as extractants is plotted in Fig. 3. The release was initially rapid followed by a slower trend. The trend of the cumulative K release was non-linear and almost similar with the two extractants. The results showed that the maximum content of non-exchangeable K (1446.14 mg kg–1 of soil) was found in the surface horizon of Alfisols (pedon 8) with a high clay content (29%) and relatively low calcium carbonate content (mean = 15.4%), whereas the minimum content of non-exchangeable K (82 mg kg–1 of soil) was measured in the subsoil of Aridisols (pedon 4) with a low clay content (8%) and high calcium carbonate content (66.4%). With both extractants, the release rate of non-exchangeable K was higher in surface soil rather than the subsurface soil (Table 2). CaCl2 and oxalic acid released more non-exchangeable K in the surface soil (24%) rather than the subsurface soil (21%) (Table 2). The trend of K release was characterized by an initial fast reaction (2–164 h), followed by a slow reaction (164–1200 h), showing different mechanisms controlling the K release process (Fig. 3). Using CaCl2 and oxalic acid as extractant, one could release K at 104–286 mg kg–1(mean = 168 mg kg–1) and 87–233 mg kg–1(mean = 155 mg kg–1) from the surface soil samples and at 98–241 mg kg–1 (mean = 155 mg kg–1) and 82–211 mg kg–1 (mean = 135 mg kg–1) from the subsurface soil samples, respectively (Table 2). The results indicate that cumulative K release varied greatly between the different physiographic unit and moisture regimes. The highest values of the percentage of nonexchangeable K released were attributed to P.P unit and xeric moisture regime (Fig. 4), In spite of containing a larger amount of illite, NEK and clay content (Table 1). Therefore, it can be stated that the process of NEK release with both extractants in addition to some important factor such as clay mineralogy (mica content), soil development and soil depth has been affected by moisture condition and physiographic position. Rezapour et al. [28], presented that Plateau and Piedmont plains which occurred on a more stable landform position had a higher content of all forms of K along with clay. Different mathematical models were tested to describe the kinetics of non-exchangeable K release with CaCl2 and oxalic acid as extractants over 1200 h. Statistical analysis was performed by fitting the experimental data to kinetic models and evaluating the coefficients of determination (R2) and standard error of estimation (SE) (Table 4, Fig. 3). Identified by the highest R2 values, the best fits for surface and subsurface soils were obtained by applying the Elovich and power function equations, respectively.

Fig. 3.
figure 3

The trend of cumulative K release by CaCl2 and oxalic acid for selected (a, b) topsoils and (c, d) subsoils.

Full size image
Fig. 4.
figure 4

Potassium released in various physiographic units and soil moisture regimes.

Full size image
Table 4.   Coefficient of determination, R2, standard error of the estimates, SE, and a and b constants describing the kinetics of non-exchangeable K for soils
Full size table

Mousavi et al. [20] referred to the power function equation as the best equation for modeling the K release with CaCl2 and organic extractants such as oxalic and citric acids. Shakeri et al. [32] reported the power function, first-order and Elovich equations as the best ones for several selected soils in Kohgiluyeh and Boyer-Ahmad Province, Iran. Among other equations that were used to justify the K release (the power function, Elovich, and the first-order equations) with either extractant, the first-order equation returned the lowest initial rate of non-exchangeable K release (a), possibly because it provided fewer initially available K ions for the plant. At the other end of spectrum, the power function equation had the highest value of the initial release rate (a), indicating abundant available of K to the plant in the initial stage of release. Comparing the average K release rate (slope of equations) with both CaCl2 and C2H2O4 extractants, the Elovich and first-order equations set the scene for the highest and lowest K release rates, respectively. The Elovich and first-order equations were expected to provide for the highest and lowest release rates (b), respectively, compared to other equations considered herein. The highest and lowest K release rates for the power function, Elovich, and the first-order equations were related to the Alfisol (soil profile 8) and the Aridisol (soil profile 5) orders, respectively. The correlation coefficients obtained with the selected equations, soil properties, and clay mineralogy with both CaCl2 and C2H2O4 extractants are given in Table 5, respectively. When CaCl2 was used as extractant, the release rates estimated by the power function equation were significantly correlated to a number of chemical and physico-chemical properties, including non-exchangeable K, clay content, CaCO3 content, and the contents of illite and smectite content. When the Elovich equation was used instead of the power function, the release rate was significantly correlated to the clay content, CEC, and the contents of illite and smectite (Table 5).

Table 5.   Relationship between constant b values of kinetic models and cumulative K release, CaCl2 and Oxalic acid, with clay mieralogy and some soil propertices
Full size table

Due to the dilution effect, the release of the K from soils with CaCl2 and C2H2O4 as extractants showed a significant negative correlation to the content of CaCO3 at R = –0.472 ** and R = –0.493**, respectively, and significant positive correlations to non-exchangeable K, OM, CEC, and the contents of illite and smectite at (0.381*, 0.366**, 0.432*, 0.557**, 0.674**) and (0.541**, 0.272*, 0.479*, 0.566**, 0.694**), respectively. The results indicated that the clay composition, content, and particle size impose significant effects on the K release rate, so that the samples hosting the K release at the highest rates involved more amounts of smectite-bearing clay than the other samples. The smectite forms 2 : 1 clay with high CEC values and variable inter-layer spacing. Indeed, an isomorphic substitution induced a permanent negative charge on these clays, with the charge being neutralized by cations such as K. Due to their expandability and lower layer charge (especially in low-layer charge smectites), the smectite easily releases inter-layer K. Srinivasarao et al. [38] reported that the higher rate of non-exchangeable K release in soils dominated by smectite indicates the facilitation of the K release by the natural expansion of smectite on the edge of the mineral and wedged areas. The high correlation of OM content to the K release according to Elovich’s equation could be linked to the mineral degradation and conversion (e.g., illite to smectite) under the effect of the OM. Comparing the non-exchangeable K release rate between organic and inorganic extractants, the results showed lower non-exchangeable K release by C2H2O4 rather than CaCl2 in almost all samples. Shakeri and Abtahi [31] obtained similar results, and stated that the C2H2O4 occurs in the form of non-separated molecules at pH values below 3.5; accordingly, one of the hydrogens of the hydroxyl group is released at a pH between 3.5 and 4.5 and both hydrogens are released at pH values beyond 4.5, where negative free charges will become available for absorbing such cations as K, Ca and Na.

With C2H2O4 as extractant, the non-exchangeable K may be released through either mineral destruction or ion exchange [31] The CaCO3 of the soil was high in the study area. Since the solubility constant of CaCO3 is higher than that of calcium oxalate, the soil contained large amounts of Ca for a given amount of hydroxyl ion. As a result, the hydrogen released from the C2H2O4, which likely destroys the minerals in acidic soils, is used to neutralize the hydroxyl ions of the soil instead of attacking the minerals. The Ca in the soil solution forms calcium oxalate deposits with CaC2O4 ions and prevents the reaction of K with oxalate. Since calcareous soils are rich in Ca ions and have a high buffering capacity, they cancel out the effect of the oxalic acid on minerals and release less K than both the acidic soils and soils with less CaCO3 contents. High buffering capacity of the soil contributes to lower K release in the soils of higher OM and CaCO3 contents, as compared to other soils [32, 41]. Comparing the two extractants (i.e., CaCl2 and C2H2O4), Jalali and Zarabi [11] stated that C2H2O4 provides for higher K release rates, possibly due to the lower availability of CaCO3 in the studied soils (about 5–20%). Mousavi et al. [20] compared CaCl2 with organic extractants including C2H2O4 and citric acid (C6H8O7) on pure feldspar and muscovite minerals. They concluded that the organic acids tended to release the K at a 2.5–3 times faster rate than the CaCl2. This may be attributed to the use of pure minerals instead of soil. C2H2O4 has been able to easily release K by destroying or replacing the minerals in absence of disturbing ions such as Calcium (Ca).

CONCLUSIONS

Based on the results, the contents of all forms of K other than the soluble K form were found to be significantly correlated to the contents of illite, smectite, CEC, clay content and inverse associations with the CaCO3 content. Also, the results showed that, in addition to mineralogical composition and soil physicochemical properties such as the soil cation exchange capacity, calcium carbonate equivalent and particle size distribution, other factors such as moisture regime, soil order and physiographic units can affect the K distribution through the soil, Due to the similarity of the parent material in all three regions, two factors include: topography and climate, are effective in the amount of different forms of potassium. As shown in Table 2, the amount of different forms of potassium in the xeric moisture regime (pedons 6 to 10) is higher than in other regions. Therefore, climate is the most effective factor in determining potassium forms. However, in each region where the climate conditions are constant, different amounts of potassium are different in different physiographic units (geographical units). Therefore, with study the relationship between clay mineralogy and potassium forms in different soil moisture regimes and physiographic unit and the processes and mechanisms of potassium release we can apply controlled management and finally we expect to increase efficiency and productivity. on the other hand, Various factors affect the distribution of potassium (K) forms in calcareous soils. Knowledge of the presence and distribution of different forms of K in the soil can help optimize the application of the K fertilizers in terms of dosage and management procedure.