Analysis of Silicon in Soil, Plant and Fertilizer

Abstract

The first soil testing of plant-available silicon (Si) was not conducted until 1898 on Hawaiian soils. However, numerous procedures have since been developed for determination of Si content in a wide variety of materials including soils, plants and fertilizers. This chapter reviews current analytical procedures that are widely used for analysis of both total Si in soils, plants and fertilizers and plant-available Si in soils and fertilizers.

2.1 General

Although silica bodies were isolated from various plant species in the early nineteenth century (Davy 1819; Struve 1835), the first soil test for plant-available Si was not conducted until 1898 on Hawaiian soils (Maxwell 1898). Since then, numerous procedures for determination of Si content in a wide variety of materials including soils, plants and fertilizers have been developed (Snyder 2001; Sauer et al. 2006). However, apart from the review articles by Snyder (2001) and Sauer et al. (2006), no reviews have been published with focus on comparing the published extraction techniques or methodologies for chemically and physically analysing plant materials for Si and for determining total as well as plant-available Si in soils and fertilizers. The classical methods for determining total Si content of various materials involve conversion of insoluble silicates into sodium silicate through high-temperature fusion with sodium carbonate or hydroxide or other sodic bases (Snyder 2001) followed by determination of Si by a variety of methods including gravimetry, colorimetry and absorption/emission spectroscopy. In general, all the analytical methods involve two major steps; the first one is to dissolve Si contained in the insoluble silicates and extract or isolate Si from the materials, and the second one is to gauge Si based on gravimetric methods, spectrometric methods or microscopic observation. X-ray fluorescence spectrometry (XRF) as a nondestructive technique for multielemental analysis of soil and plant materials showed even higher measurement accuracy for Si over the destructive methods based on alkaline fusion or acid digestion (e.g. Ramsey et al. 1995; Reidinger et al. 2012); this technique, however, has not been routinely used so far. In this chapter, current methods that can be widely used are reviewed.

2.2 Analysis of Total Silicon in Soil

2.2.1 Gravimetric Methods

Gravimetric methods are classic accurate methods that are suitable for determination of total Si in a wide variety of solid materials including soils, waste materials, sewage sludge, plants, organic manures and fertilizers, etc. They require basic common and simple laboratory equipment except for the expensive platinum wares. On the other hand, they are time consuming and tedious and thus not suitable for routine analysis of a large number of samples especially if only an analysis of Si is required (Snyder 2001). Therefore, rapid and robust methods suitable for routine analyses are needed in modern laboratories. The principle and detailed procedure are listed in Table 2.1.

Table 2.1 The principle and procedure for gravimetric methods

2.2.2 Spectroscopic Methods

The rapid development of modern spectroscopic techniques has made it possible to use more rapid and robust techniques suitable for routine analysis of large samples. However, conversion of the silicates contained in the solid samples into a soluble form of Si by using sodium carbonate (Na₂CO₃), lithium metaborate (LiBO₂) or sodium hydroxide (NaOH) is needed before using such spectrometric techniques (Snyder 2001). The principle and detailed procedure are listed in Table 2.2.

Table 2.2 The principle and procedure for spectroscopic methods

2.3 Analysis of Total Si in Plant Material

2.3.1 Gravimetric Methods

To analyse Si in an organic matrix, such as plant material, the organic matter should be first removed by oxidation at a high temperature (550 °C), and then the non-Si elements are solubilized by 6 M HCl followed by filtration through ashless filter paper that retains the Si precipitates. The paper is then ignited and weighed. Afterwards, HF is used to evolve Si so that the weight loss is assumed to be Si (Snyder 2001). Yoshida et al. (1976) used a gravimetric method to determine Si and other mineral elements in rice straw following the removal of organic matter by acid digestion.

2.3.2 Spectrometric Methods

Spectrometric methods used widely for Si determination normally include colorimetric, atomic absorption spectrometric (AAS) and inductively coupled plasma spectrometric (ICP) assays of Si following alkaline- or acid-based digestion of plant materials. More recently, XRF, particularly portable X-ray fluorescence spectrometry (P-XRF), has been reported to be a much faster, safer, nonconsumptive and potentially more accurate method to determine Si concentrations in plant material (Reidinger et al. 2012). XRF works on the principle of excitation of inner orbital electrons by an X-ray radiation source. As the excited electrons relax to the ground state, they fluoresce, thereby ejecting photons of energy and wavelength characteristic of the atoms present. Despite several advantages of XRF over digestion-based techniques, such as its nonconsumptive nature and its higher measurement accuracy, XRF is still not routinely used for measurement of Si in plants partly due to the higher purchasing price of XRF instruments than that for equipment typically used in digestion-based techniques such as AAS or ICP (Reidinger et al. 2012).

2.3.2.1 Procedures for Sample Preparation (Table 2.3 )

Table 2.3 Selected procedures for sample preparation

2.3.2.2 Procedures for Si Determination

2.3.2.2.1 Colorimetric Methods

Silicon (Si) can be determined colorimetrically, using light absorption spectrometer either by the blue Si molybdenum method at lower Si concentrations or by the yellow Si molybdenum method at higher Si concentrations. Si in solution reacts with ammonium molybdate forming a yellow silicomolybdate complex. Silicomolybdate complex can be finally converted into a reduced silicomolybdate complex which is blue, using a reducing solution such as ammonia ferrous sulphate, ascorbic acid or a mixture of sodium sulphite, sodium bisulphite and 1-amino-2-naphthol-4-sulfonic acid. Oxalic acid or tartaric acid is added to remove phosphate interferences. These two methods are similar in principle; however, the blue molybdenum method is more popular because of its higher sensitivity (Table 2.4 ).

Table 2.4 Procedures and reagent preparation for colorimetric determination of Si using molybdenum blue method

2.3.2.2.2 AAS and ICP Methods

In addition to colorimetric determination of Si using a light absorption spectrometer, dissolved Si can also be determined by AAS using a nitrous oxide–acetylene flame (Eaton et al. 1995) or by ICP (Jones and Dreher 1996).

2.4 Analysis of Soluble Silicon in Soil

As the total Si content is not related to the concentration of soluble Si in soils and can provide little information on soil Si availability to plants, various chemical extractants (Table 2.5; Berthelsen and Korndörfer 2005) are developed to assess Si-supplying power or plant-available Si. Currently, the chemical extracts reported to be useful in assaying plant-available Si are water (at varying soil: water ratios), sodium acetate buffer (pH 4.0), ammonium acetate (pH 4.8), dilute HCl or H₂SO₄, ammonium oxalate (pH 3.0), dilute citric acid or acetic acid, dilute CaCl₂, dilute phosphate plus acetate (pH 3.5), sodium phosphate buffer (pH 6.2), dilute Na₂CO₃, etc. (Gillman and Bell 1978; Fox et al. 1967; Fox and Silva 1978; Haysom and Chapman 1975; Imaizumi and Yoshidai 1958; Haysom and Kingston 2001; Kato and Sumida 2000; Korndörfer et al. 2001; Nayer et al. 1977; Nonaka and Takahashi 1988, 1990). In general, the most successful extractants are acid rather than neutral or alkaline solutions (Imaizumi and Yoshidai 1958), and dissolution is further increased by chelating agents due to decreased Si sorption resulting from the lower concentration of Al and Fe in solution (Berthelsen and Korndörfer 2005). Other factors such as the method of equilibration, soil–solution ratio, temperature and pH of extractant solution are also important.

Table 2.5 Methods used to determine soluble and extractable soil Si

2.4.1 Acetate Buffer Method

The acetate buffer method proposed by Imaizumi and Yoshida (1958) is one of the most widely used analytical methods to assay the Si-supplying capacity in soils (Table 2.6).

Table 2.6 The procedure of acetate buffer method

By comparing various chemicals including hot hydrochloric acid, ammonium oxalate buffer at pH 3.0, 2 % sodium carbonate, 0.002 N sulphuric acid at pH 3.0, saturated carbonic acid water at pH 3.8 and sodium acetate buffer at pH 4.0 as extractants, Imaizumi and Yoshida (1958) found that more soil Si was extracted by acid than by alkaline chemicals. They further concluded that CO₂-saturated water (pH 3.8) was best suited for estimating the Si uptake by rice plants followed by the acetate buffer, and the amount of SiO₂ extracted by CO₂-saturated water was correlated nicely with that extracted by acetate buffer (pH 4.0). Considering that preparing the acetate buffer is easier, and a more stable pH can be reached during extraction, and that the acetate-extractable SiO₂ is well correlated with the amount of Si taken up by rice, Imaizumi and Yoshida (1958) suggested the use of acetate buffer as an extractant to estimate Si-supplying power in soil. Up to now, this acetate buffer extractant has been used to assess the Si-supplying capacity in soils in Japan, Chinese Taiwan (Lian 1976), China (Qin 1979; He et al. 1980; Zang et al. 1982; Ma et al. 1985; Liang et al. 1994), South Korea (Park 2001), Korea (Lian 1976), Malaysia and Thailand (Kawaguchi 1966), Ceylon (Takijima et al. 1970), India (Nayer et al. 1977), Australia (Haysom and Kingston 2001) and Brazil (Korndörfer et al. 2001).

As the acetate buffer method involves extracting Si in soil for 5 h at 40 °C and is thus time consuming and laborious, the modified procedure in Korea involves extracting Si for 90 min at 60 °C, which is reported to be more feasible for routine analysis of soil-available Si in laboratories (Park 2001).

However, it is also reported that the acetate buffer method overestimated the Si-supplying power in soils previously fertilized with slag-based calcium silicate fertilizers (Takahashi 1981; Takahashi and Nonaka 1986; Nonaka and Takahashi 1990) and in calcareous soils (Liang et al. 1994), because this acetate buffer is strong enough to dissolve some non-available Si from the residual calcium silicate fertilizers, and it extracts additional portions of Si which may be carbonate bound but not plant available in the calcareous soils (Liang et al. 1994).

2.4.2 Water Extraction Method

To overcome overestimation problems, Nonaka and Takahashi (1988, 1990) proposed a method for measuring water-soluble Si in rice paddy soils (Table 2.7).

Table 2.7 The procedure of water extraction method

For soils previously fertilized with slag-based calcium silicate fertilizers, Si extracted by this new method was generally correlated better with Si in rice straw than the Si extracted by the acetate buffer method. However, this method has not been widely used for routine lab testing or for commercial use since the 2-week period between soil sampling, and reporting of results is a serious disadvantage (Savant et al. 1997; Snyder 2001). Sumida (1992) developed two additional soil incubation methods. One involves incubation of soil under flooded conditions for 4 weeks at 30 °C using a soil–water ratio of 1:4. The other method requires 5 days of incubation of soil with the addition of external silicic acid in a range from 0 to 100 mg SiO₂ kg⁻¹ at 30 °C using soil–water ratio of 1:10. They serve the purpose to investigate the impacts of silicate fertilizer management on the characteristics of Si dissolution and adsorption of various paddy soils. Although these methods provide the most suitable the Si-supplying capacity indices in paddy soils previously amended with calcium silicate or organic manure and in the soils with varying clay minerals and textures, they are not suitable for use in a routine soil testing laboratory because of their longer time requirement and/or complexity (Savant et al. 1997; Snyder 2001).

2.4.3 CaCl₂ Extraction Method

Water extracts have often been used to assay water-soluble or readily soluble Si (Elawad et al. 1982; Fox et al. 1967; Fox and Silva 1978; Gillman and Bell 1978; Menzies and Bell 1988; Nonaka and Takahashi 1988, 1990; Takahashi and Nonaka 1986). It is reported that the water-extractable Si was significantly correlated with Si uptake by plants (Fox et al. 1967; Medina-Gonzales et al. 1988; Takahashi and Nonaka 1986). However, some researchers find that the incubation method is generally not a suitable method since the low ionic strength of the solution may cause dispersion (Elgawhary and Lindsay 1972; Lindsay 1979). Generally, as drying of soil samples results in changes in equilibrium between soluble and solid Si compounds, and as soluble Si compounds (mainly monosilicic acids) adsorbed on soil particles are dehydrated during soil drying, extracting Si with water as an extractant requires a longer period of time (Table 2.5).

Considering the resulting dispersion effect caused by the low ionic strength in the water incubation solution, Elgawhary and Lindsay (1972) recommended the use of 0.02 M CaCl₂ as the reactive media to equalize ionic strengths and facilitate ready flocculation of colloidal Si. According to the soil chemical equilibrium principle, Si extracted with diluted CaCl₂ may correspond more closely to the levels of Si(OH)₄ expected from solubility predictions (Lindsay 1979). Accordingly, a diluted CaCl₂ solution was recommended by Haysom and Chapman (1975) as an extractant to assess Si-supplying power in soils under sugarcane (Table 2.8).

Table 2.8 The procedure of water extraction method

The critical value is 100 mg Si kg⁻¹ soil, below which Si deficiency for sugarcane is expected. They concluded that although Si extracted with either 0.5 M NH₄OAC or 0.005 M H₂SO₄ was well correlated with cane yield, 0.01 M CaCl₂ was the best extractant. This method was also adopted by Wickramasinghe (1994) with some modifications. After having compared a number of extraction methods over a wide range of soil types from north Queensland in Australia, Berthelsen (2000) concluded that 0.01 M CaCl₂-extractable Si provided a measure of the readily available Si present in the soil solution, while NH₄OAC- and acetic acid-extractable Si fractions were likely to contain rather simple polymers affected by changes in pH, CEC and the ratio of soluble Si–Al in the soil solution. Berthelsen et al. (2003) also reported that the Si contents in sugarcane were significantly related to readily soluble soil Si extracted by 0.01 M CaCl₂ but were not correlated with the level of soil Si extracted by a stronger acid extractant (e.g. 0.005 M H₂SO₄).

2.4.4 Citric Acid Extraction Method

Ueda and Yamaoka (1959) measured soil Si-supplying capacity in degraded paddy soils where nutrients such as bases, Fe, Mn and Si were leached out by using 0.025 M citric acid, 2 % Na₂CO₃, oxalic acid and oxalate buffer and 1 M acetate buffer to extract plant-available Si. Their results showed that the straw Si was significantly correlated with citric acid-extractable Si and with acetate buffer-extractable Si, but was not well correlated to the Si extracted by other extractants. They thus proposed using citric acid as an extractant to assess the soil Si-supplying power. Later, numerous researches all over the world confirmed that 0.025 M citric acid was the most suitable extractant to assay the plant-available Si in soils. For example, by comparing 1.0 M acetate buffer extractant with three other extractants, i.e. distilled water, 0.2 M HCl and 0.025 M citric acid, Nayer et al. (1977) reported that the extracting power was in the order of 0.2 M HCl > 0.025 M citric acid > 1.0 M acetate buffer at pH 4 > distilled water, with only some exceptions in some soils. They further found that citric acid-extractable Si had the strongest correlation with the Si taken up by rice plants. Zhang and Zang (1982) also compared the differences and similarities in the analytical methods for determining plant-available Si in 28 Si-fertilized paddy soils derived from coastal marine deposits, granite, basalt, red sandstone, quaternary red earth and alluvial deposits in Guangdong, Jiangxi, Zhejiang and Jiangsu provinces of South China. They used acetate buffer, 0.01 M H₂SO₄, 0.2 M HCl and 0.025 M citric acid with intermittent shaking at 30 °C for 5 h or 1 h and at 15 °C for 15 min to extract the plant-available Si in all the soils used. The results showed that the Si amounts extracted by these extractants were all significantly correlated with the straw Si content and with the relative increase of rice yield in Si-fertilized plots. However, it was roughly calculated that the amount of Si extracted by the acetate buffer and 0.01 M H₂SO₄ was less and by 0.2 M HCl was more than the amount of Si removed by rice during the growing season and that the amount of Si extracted by citric acid was approximately equal to that removed by rice.

2.4.5 Sulphuric Acid-Extractable Method

More recently, Hong et al. (2011) also studied the effect of five extracting methods (0.025 M citric acid with continuous shaking at 30 °C for 2 h, 1 % citric acid with intermittent shaking at 30 °C for 5 h, 1 M acetate buffer with intermittent shaking at 40 °C for 5 h, 0.01 M H₂SO₄ with continuous shaking at 20 °C for 30 min and 0.2 M HCl without shaking at 40 °C for 16 h) on extraction of Si in neutral and alkaline paddy soils sampled from Heilongjiang province, Northeast China. They found that, on the average, the concentration of Si measured by 0.2 M HCl extraction was 39-fold higher than by the other methods, and the concentration of Si determined by 0.01 M H₂SO₄ extraction was almost equal to that by 0.025 M citric acid extraction. They concluded that both 0.01 M H₂SO₄ and 0.025 M citric acid were the most suitable extractants to assess the Si-supplying power in neutral and especially in alkaline soils and argued that the ‘too strong’ acidity and ‘too long’ incubation period might be the major reasons for the unexpectedly higher content of Si solubilized by the 0.2 M HCl extractant.

Kanamugire et al. (2006) compared the correlation coefficient between the amount of Si accumulated by sorghum and sugarcane in 5 acid soils and the amount of Si extracted by six different extractants (0.025 M H₂SO₄, 0.01 M H₂SO₄, 0.5 M acetic acid, 0.01 M CaCl₂, water and 0.5 M NH₄OAC) and found that the correlations between the total Si taken up by sorghum and sugarcane and Si extracted from soils by the extraction methods used were all statistically significant, with the 0.025 M H₂SO₄ extractant showing the best correlation with total Si uptake by plants. They also proposed that a positive response to the application of Si to be likely when the soil test level using 0.025 M H₂SO₄ is below 45 mg Si kg⁻¹ for sandy soils, below 65 mg Si kg⁻¹ for loamy sands to sandy clay loams and 100 mg Si kg⁻¹ for clayey soils.

2.4.6 Acetic Acid Extraction Method

Barbosa et al. (2001) developed a soil testing method using acetic acid as an extractant for determining plant-available Si in paddy and sugarcane soils in South Florida, and this method has been well accepted (Table 2.9). Critical values for soil Si test ranges of low (<7 mg Si L⁻¹), medium (7–24 mg Si L⁻¹) and high (>24 mg Si L⁻¹) were established (Snyder 2001).

Table 2.9 The procedure of acid extraction method

2.4.7 Phosphate Extraction Method

Khalid et al. (1978) proposed extracting water-soluble Si by shaking 3 g of soil with 30 mL distilled water for 4 h to measure the ‘intensity factor’ and using a phosphate extractant to gauge plant-available Si as the ‘capacity factor’. Briefly, 3 g of soil are shaken for 4 h with 30 mL 0.1 M acetic acid containing 50 mg P L⁻¹ as Ca(H₂PO₄)₂ at pH 3.5 (adjusted with NH₄OH). The results obtained from 5-year field trials showed that the amount of Si taken up by rice seedlings was higher than extracted by the acetic acid plus phosphate (Khalid et al. 1978). Kato and Sumida (2000) proposed a phosphate buffer method for measuring plant-available Si, in which 5 g soil is shaken for 24 h with 50 mL of 0.04 M phosphate buffer solution at pH 6.2 made by titrating 0.04 M Na₂HPO₄ with 0.04 M NaH₂PO₄ to pH 6.2. After centrifugation and filtration, the Si concentration in the supernatant is measured colorimetrically. Kato and Sumida (2000) reported a better correlation between the amount of Si taken up by rice with soil Si extracted by the phosphate buffer than by the acetate buffer method. They concluded that the phosphate buffer method was better for assessing the Si-supplying power in paddy soils than the traditional acetate buffer method because the proposed method did not overestimate Si availability in soils previously treated with calcium silicate fertilizers.

2.4.8 Alkaline Extraction Method

Alkaline wet chemical dissolution extracts such as NaOH, a strong base, and Na₂CO₃, a weak base, are frequently used to extract Si in soils. The NaOH- or Na₂CO₃-extractable Si is considered to represent the fraction of amorphous Si in soils (Foster 1953; Follett et al. 1965) and amorphous biogenically derived Si in sediments (DeMaster 1981; Mortlock and Froelich 1989; Koning et al. 2002). The principle of these alkaline-extractable methods is based on the fact that the solubility of amorphous Si increases with increasing pH values (Iler 1979). While NaOH is more often used in soils than Na₂CO₃, the latter is the most commonly used method to extract amorphous biogenically derived Si in aquatic sediments (Conley 1998). Foster (1953) used 0.5 M NaOH to extract amorphous soil Si at soil–solution ratio of 1 g : 50 mL in a boiling water bath for 4 h. Since then, this NaOH-based extraction has become a standard method to assay amorphous Si in soils. As NaOH also partly dissolves the silicate minerals, Hashimoto and Jackson (1960) proposed using 0.5 M NaOH to extract amorphous Si within 2.5 min to reduce the dissolution of silicate minerals.

Some extractants can dissolve or extract different components from certain siliceous soils. For example, dithionite–citrate–bicarbonate (DCB) (Weaver et al. 1968) and ammonium oxalate extracts (Wang and Schuppli 1986) which are originally prepared for determining crystalline and amorphous Fe also are often used for analysis of crystalline and amorphous Si.

Although numerous extractants have been proposed to assay the soil Si-supplying power all over the world, the most widely used extract is 1.0 M acetate buffer at pH 4.0 proposed by Imaizumi and Yoshidai (1958) as there are threshold or critical values below which plant response to Si fertilizers can be expected. The critical value of acetate buffer-extracted Si, below which positive response of rice to slag-based Si fertilizers can be expected, is 105 mg SiO₂ kg⁻¹ in Japan, 100 mg SiO₂ kg⁻¹ in Korea, 40 mg SiO₂ kg⁻¹ in Chinese Taiwan (Lian 1976) and 95 mg SiO₂ kg⁻¹ in China (Zang et al. 1982).

In fact, no extraction has been found to work equally well on all soils (Snyder 2001). The Si content extracted by all of the discussed extractants can only be considered as a reference index to estimate the Si-supplying power in soils, but it does not measure the exact amount of plant-available Si and should be used to make diagnosis for Si deficiency in plants together with plant Si content, the truely important variable. Research is also warranted on testing whether the Si extracted by any of the established extractants is well correlated with Si uptake by plants.

2.5 Analysis of Plant-Available Silicon in Fertilizers

Procedures for determination of total Si in soils such as alkaline fusion method can apply to analysis of Si in fertilizers. However, total Si content in Si fertilizers, which is not related to Si availability, can provide little information on plant-available Si, which is most important to assessing the quality and Si-supplying power of the fertilizers. Therefore, a powerful method for assaying plant-available Si content is needed to assess the Si-supplying capacity of the fertilizer of interest. Unfortunately, it is difficult to select a universally suitable method or extractant for assessing the Si-supplying capacity of all types of materials, all soils and soil conditions (Gascho 2001; Berthelsen and Korndörfer 2005). While a number of chemical extractants have been developed to estimate soluble or extractable Si in silicate materials, often the correlations between the Si contents in a fertilizer and a response of the Si concentrations in plants on fertilized fields are rather low (Berthelsen and Korndörfer 2005). Actually, a lot of physical and chemical properties of the materials such as pH, molar ratio of CaO–SiO₂ and mineral particle size can greatly influence Si solubility and plant availability (Ma and Takahashi 2002).

Up to recently, many researchers have used different extractants to assay the plant-available Si in solid fertilizers. These extractants include HCl + HF, Na₂CO₃ (10 g L⁻¹) + NH₄NO₃ (16 10 g L⁻¹), citric acid (50 g L⁻¹), HCl (0.5 mM) and ammonium citrate (Buck et al. 2011; Korndörfer and Pereira 2011; Sebastian et al. 2013). Based on the correlation coefficients, the best extractant for available Si in solid fertilizers was Na₂CO₃ + NH₄NO₃, while for liquid fertilizers, HCl + HF was found to be superior for total Si assessment (Buck et al. 2011; Korndörfer and Pereira 2011). It is shown that the 5-day Na₂CO₃–NH₄NO₃ soluble Si extraction method can be applied to quantify the plant-available Si in solid fertilizer products at levels ranging from 0.2 to 8.4 % Si, and the single-laboratory validation of the 5-day Na₂CO₃–NH₄NO₃ soluble Si extraction method has been approved by the Association of American Plant Food Control Officials for testing nonliquid Si fertilizer products (Sebastian et al. 2013) and the Association of American Plant Food Control Officials (AAPFCO) (Korndörfer and Pereira 2011).

Given in Table 2.10 are quantities of total and extractable Si in the waste materials tested (Haynes et al. 2013). The data listed in Table 2.10 clearly indicate that the total Si was over 20 % in fly ash and Ca silicate and between 14 % and 18 % in the BF slags, while it ranged from 5.4 % in the steel slag to 6.8 % in the processing mud. The water/exchange resin method extracted 11.7 % of total Si from Ca silicate, but ammonia acetate and HCl extracted < 1 %. The two Na₂CO₃–NH₄NO₃ methods extracted the greatest proportions of total Si from the Ca silicate (i.e. 24 % and 27 %). In the BF slags, approximately 17 %–20 %, 28 %–44 %, 1 %–5 %, 0.17 %–3.1 % and 0.67 %–17 % of total Si were extracted by ammonium acetate, HCl, water/exchange resin, Na₂CO₃–NH₄NO₃ and Na₂CO₃–NH₄NO₃ plus a 5-day equilibration, respectively. Although the total Si was lower in steel slag and processing mud than in the BF-slag samples, the proportions of ammonia acetate- and HCl-extractable Si were much greater in steel slag and processing mud than in the BF-slag samples. This was also true for the water/exchange resin extraction for steel slag. For the Na₂CO₃–NH₄NO₃ method, adding 5 days of equilibration prior to Si analysis could extract greater amount of Si from all the materials tested. Haynes et al. (2013) draw a conclusion that the BF slags tested are the most effective as sources of Si fertilizer and that, in slag-amended soils, CaCl₂ and NH₄OAC are the most reliable soil test extractants.

Table 2.10 Quantities of total and extractable Si in the waste materials tested^a

2.5.1 Acid Extraction Methods

The earliest widespread commercial Si fertilizer used for rice crop in Japan is slag-based calcium silicate (Ma and Takahashi 2002). Si extraction with 0.5 M HCl at 30 °C for 1 h is proposed as an ‘official’ method of gauging Si availability in slags in Japan (NIAES 1987; Ma and Takahashi 2002). In China, this procedure is also a commonly used method to gauge the plant-available Si in Si materials (Wang et al. 1995; Li et al. 2004). Wang et al. (1995) reported that both 0.5 M HCl and 2 % citric acid could be satisfactorily used to assess plant-available Si content in fertilizers. The amount of Si extracted by 0.5 M HCl was higher than that by 2 % citric acid. The amounts of plant-available Si extracted from different slag-based calcium silicates by 0.5 M HCl and 2 % citric acid at 30 °C were strongly correlated with the increment of the plant-available Si in both alluvial deposit- and red soil-derived paddy soils fertilized by slag-based calcium silicates. Considering that plant roots release organic acids to the rhizosphere, Wang et al. (1995) believe that citric acid is recommended as a more suitable extractant to estimate plant-available Si in fertilizers. It was reported that, however, Si extraction with 0.5 M HCl at 30 °C for 1 h was not an appropriate procedure to predict Si uptake by rice as the amount of Si extracted from the Si fertilizer was not related to the Si uptake by plants (Takahashi 1981; Kato and Owa 1997; Snyder 2001).

2.5.2 Acetate Buffer Method

The 1.0 M acetate buffer at pH 4 method proposed by Imaizumi and Yoshidai (1958) has also been recommended for assessing Si availability in slags (NIAES 1987). However, this method has been found unsuitable for gauging plant-available Si in soils fertilized by slag-based calcium silicate (Takahashi 1981; Takahashi and Nonaka 1986; Nonaka and Takahashi 1990).

2.5.3 Acidic Cation Exchange Method

Kato and Owa (1997) also demonstrated that the acid extraction method commonly used in Japan was unsuitable for gauging available Si content in slag-based calcium silicate fertilizers. They developed a procedure using a weakly acidic cation exchange H⁺-resin (Table 2.11).

Table 2.11 The procedure of acidic cation exchange H-resin method

This method is reported to provide the best indicator of plant-available Si in the fertilizer, and the Si extracted is correlated well with both the indirect chemical extraction results and also soil and plant Si and yield in the pot experiments (Kato and Owa 1997). Pereira et al. (2003) also tested this ‘resin’ method and found it to give the highest correlation between Si extracted from 12 different sources of Si materials and Si uptake by rice.

2.5.4 ‘Column’ Method

Snyder (2001) and his colleagues also developed a ‘column’ method to compare mineral Si sources on the basis of plant availability of Si, maintenance of neutral solution pH, low Ca concentration near the Si source and sufficiently low dissolved Si concentration to minimize polymerization (Table 2.12). However, Si laboratory analysis to identify promising mineral Si sources is required to correlate with both the crop responses and crop Si uptake through greenhouse and field trials to ultimately assess the Si-supplying capacity of the sources.

Table 2.12 The procedure of column method

2.5.5 Sodium Carbonate–Ammonium Nitrate Method

More recently, Sebastian et al. (2013) have developed a 5-day method for determining the soluble Si concentrations in nonliquid fertilizer products using a Na₂CO₃ + NH₄NO₃ extractant followed by visible spectroscopy with heteropoly blue analysis at 660 nm. This 5-day Na₂CO₃–NH₄NO₃ soluble Si extraction method has recently been approved by the Association of American Plant Food Control Officials (AAPFCO) for testing nonliquid Si fertilizer products (Sebastian et al. 2013) (Table 2.13).

Table 2.13 The procedure of 5-day Na₂CO₃–NH₄NO₃ soluble Si extraction method