Preparation of Slow-Release Fertilizer from Fly Ash and Its Slow-Release and Metal Immobilization Properties

Abstract

In this study, SRPF with metastable KAlSiO₄ as the main slow-release substance was prepared by the sintering method using fly ash and K₂CO₃ as raw materials, and an orthogonal experiment was conducted to optimize the raw material ratio and preparation parameters of SRPF. The optimum parameters for preparing SRPF are as follows: the potassium carbonate content is 15%; the sintering temperature is 1100 °C; heat preservation time is 60 min; cooling mode is furnace cooling, and the particle size of raw materials is not less than 150 μm. Initial leaching rates in water and 2% citric acid solution were 4.64% and 47.07%, respectively, and cumulative leaching rates at 28 days were 11.17% and 85.86%, respectively, showing that the SRPF prepared from fly ash and K₂CO₃ meets the standard GB/T 23348-2009 of China. A soil column leaching test was carried out to study the slow-release behavior of SRPF in soil. When the leaching medium is water, the 70-day cumulative leaching rate of SRPF in soil is about 4%, while when the leaching medium is citric acid, depending on the soil type, the 70-day cumulative leaching rate of SRPF can reach 21.2% and 43.5%. The results of the soil column leaching test showed that the total potassium content in the soil was negatively correlated with the slow-release rate of SRPF. Finally, the immobilization effect and mechanism of SRPF on lead ion immobilization was studied, and it was found that SRPF still had a considerable effect on lead ions immobilization. The BET results showed that, compared with fly ash, the BET surface area of SRPF was reduced by 48.3%, the total pore volume was reduced by 16.0%, and the average pore diameter had a small change. The decrease in total pore volume was mainly concentrated in the micropore volume and mesopore volume, which are reduced by 50% and 20% respectively, while the macropore volume hardly changes. In summary, fly ash can be used to prepare SRPF with a good release effect and similar heavy metal ions immobilization ability.

1. Introduction

According to a report by the International Energy Agency [1], global electricity demand grew by 6% in 2021 (over 1500 TWh in absolute value), of which the total thermal electricity generation increased by 980 TWh. As the country with the largest thermal power generation, China’s thermal electricity generation in 2021 was 5805.87 TWh [2]. The absolute growth of 475.62 TWh [2,3] contributed almost half of the world’s thermal power growth and produced more than 600 million tons of fly ash annually [4,5]. Restricted by many factors, such as market demand and the immaturity of technology, the utilization of fly ash is no more than 60% and mainly concentrated in some low-value-added fields, such as building materials. Thus, a large amount of fly ash has to be stockpiled, which causes many environmental problems, such as land pollution and occupation [6]. Therefore, how to expand the application of fly ash and enhance its functionality is of great significance for the large-scale resource utilization of fly ash.

Many studies have been conducted to expand the resource utilization of fly ash [7,8,9,10,11], and the attempts in the field of agriculture are worthy of attention [12,13,14,15,16,17]. Since fly ash is rich in calcium-silicon minerals with a pozzolanic nature, adding fly ash into the soil can improve soil bulk density, porosity, water holding capacity, and available water [18], which can increase the yield of crops [19,20,21,22]. In addition, fly ash contains almost all macro- and micro-nutrients necessary for plant growth [23,24]. In this sense, fly ash is suitable for use as fertilizers. The study in [25] has showed that applying 10 tons of fly ash per hectare soil could provide organic carbon (27 kg), phosphorus (32 kg), potassium (25 kg), calcium (33 kg), magnesium (17 kg), iron (127 kg), manganese (2.8 kg), zinc (238 g) and copper (178 g).

In fact, the use of solid wastes to prepare fertilizers or soil remediation agents has been extensively studied, and some fertilizers made from solid wastes, such as silicon fertilizers from steel slag, have been applied [26,27]. In the last two decades, due to the increasingly prominent environmental problems caused by the large-scale application of fertilizers, the research and development of slow-release fertilizers (SRFs) has attracted more and more attention [28,29].

SRFs refer to the compounds containing nutrients that release slowly in the soil or can be controlled to a certain extent for continuous absorption and utilization by crops [30]. SRFs can be synthesized by sintering the mixture of solid wastes and salts containing element such as phosphorus and potassium as nutrients. The release rate of the nutrients depends on the nature of the polymer bonds, stereochemical structure, hydrophilicity, crosslinking degree, and degradation difficulty [31]. Xi Ma et al. [32] prepared SRFs with K₂MgSi₃O₈ as the main slow-release substance by the sintering method using biotite acid-leaching residues, K₂CO₃ and Mg(OH)₂, as raw materials. The release behavior and performance of the SRFs were studied by water leaching and acid leaching tests, and the results showed that K₂MgSi₃O₈ had a potential as SRFs. Lijun Xiao et al. [33] prepared slow-release-type silicon potash fertilizers, in which K₂MgSi₃O₈ was the main slow-release substance, using blast furnace slag and potassium sulfate. The effect of preparation conditions on the slow-release performance of SRFs was studied through water leaching tests, and the initial leaching rate and 28-day cumulative leaching rate were approximately 10% and 20%, respectively.

Fly ash, which is rich in alumina and silica, is suitable to prepare high-quality SRFs. However, there are few studies on the preparation of SRFs using fly ash to date, and it is worthwhile to explore the fabrication of compounds from fly ash that have a sustained-release function. Since the ratio of raw materials and preparation parameters will affect the microstructure of the slow-release substance and then affect its leaching performance, it is necessary to design an orthogonal experiment to explore the relationship between these factors and the performance of the SRFs, and the preparation conditions of the SRFs can be optimized. Moreover, the release behavior of SRFs in soil is crucial for the application of SRFs; however, it was not explored in previous studies. Therefore, leaching tests of soil columns for the prepared SRFs that simulate the release behavior of SRFs in soil are necessary.

In addition, fly ash, as an alkali with high porosity and large specific surface area, has a strong adsorption performance for heavy metals, which may be used to improve heavy metal pollution in soil. Existing studies have indicated that the application of steel slag and fly ash could reduce the phytoavailability of heavy metals in soil, reducing the uptake of heavy metals by rice [34,35]. Some other studies [9,12,36] have also proved that modified fly ash with a zeolite-like structure had a good immobilization effect on heavy metals, but whether the SRFs made of sintered fly ash still have the ability to passivate heavy metal ions was not reported and needs to be explored in this study. There are a few related studies on the preparation of composite materials, either by using a slow-release function or by metal immobilization using fly ash, which is also the focus of this study.

In this study, a slow-release potassium fertilizer (SRPF) was prepared by sintering a mixture of fly ash and potassium salt, and, for the first time, a slow-release fertilizer with both slow-release and soil remediation functions was prepared. In order to obtain the best slow-release effect, an orthogonal test with potassium salt content, sintering temperature, heating time, cooling mode, and material particle size as variables was designed to determine the reasonable preparation parameters. Its slow-release behavior and performance in static water has been studied. In addition, soil column tests were carried out to study the release behavior of the prepared SRPF in soil. Furthermore, in order to better study the effect of organic acids released during plant growth on the SRPF, soil column tests using 2% citric acid solution for leaching were also carried out. Finally, the soil remediation performance of the SRPF was investigated, which made up for the lack of research on the multifunctional compounds of the SRFs.

2. Materials and Methods

In this study, the particle size of raw materials, fly ash, and analytical pure K₂CO₃, was less than 75 μm. The fly ash was collected from a power plant in Ordos, Inner Mongolia, China. Raw materials were mixed according to a certain ratio and pressed forming. The ratio of raw materials was calculated by the content level of K₂O shown in

Table 1. Level table of factors in orthogonal experiment.

	Factor	K2O Content A/%	Sintering Temperature B/°C	Heat Preservation Time C/min	Cooling Modes D	Particle Size E/μm
Level
1		15	900	0	Furnace cooling	<75
2		20	1000	30	Air cooling	75–150
3		25	1100	60	Forced air cooling	>150

. The pressed samples were placed inside corundum crucibles and sintered using a box electric resistance furnace (NBD-M1500, Henan Nobody Materials Science and Technology Ltd., Zhengzhou, Henan, China). The heating rate is 5~10 °C per minute and the sintering temperature is 900~1100 °C. During the heating process, all samples are kept at 100 °C for 30 min to ensure that all free water in the material is removed. The preparation flow chart is shown in Figure 1. Analyses of the physicochemical properties of fly ash and SRPF were carried out by means of X-Ray Diffraction (XRD, Ultima Ⅳ, Rigaku, Japan), X-Ray Fluorescence Spectrometer (XRF, XRF-1800, Shimadzu, Japan), Scanning Electron Microscope (SEM, S-4800, Hitachi, Japan), Fourier Transform Infrared Spectroscopy (FTIR, Nicolet iS 5, Thermo Scientific, Waltham, Massachusetts, USA), leaching toxicity, and corrosion tests, etc. In addition, the rational preparation process of SRPF was determined by orthogonal experiment, and the leaching effect and dissolution mechanism of the potassium element of SRPF were investigated by water leaching tests. At the same time, the soil remediation ability and mechanism of SRPF were tested by metal immobilization tests.

2.1. Samples Preparation

In order to obtain the fertilizers sample with optimal slow-release effect, an orthogonal experiment was designed with five factors of K₂O content, sintering temperature, temperature keeping time, cooling mode, and particle size. The level of each factor was determined according to the relevant references [37,38,39] and shown in Table 1, and a L₁₈ (3⁷) orthogonal experiment table was adopted. In addition, the optimal process parameters are determined based on the dissolution rate of potassium in deionized water and 2% citric acid solution.

First, 5 g of the prepared sample was placed in a 150 mL conical bottle, and 100 mL of deionized water or 2% citric acid solution was added to the conical bottle for 24 h at 25 °C. The concentration of potassium in each conical bottle was measured using an atomic absorption spectrophotometer (AAS, AA-6300). With references to existing research [40,41,42], the leaching rate is the ratio of the amount of leached potassium to the total potassium content of the sample. It can be expressed as Equation (1):

$${\mathrm{K}}_{\mathrm{f}}=\frac{\mathrm{C}\mathrm{V}}{1000\times \mathrm{K}}\times 100\mathrm{\%}$$

(1)

where K_f represents the initial leaching rate of potassium; C indicates the concentration of potassium in the leaching solution, mg/L; V is the volume of the leachate, L; and K is the content of potassium in the samples, g.

2.2. Performance Test of Samples

2.2.1. Nutrients Release Characteristic Test

The definition of “slow-release fertilizers” in GB/ T23348-2009 of China [30] is described as the initial leaching rate of fertilizers in the static water leaching test at 25 °C less than or equal to 15%, and a cumulative 28-day release rate under the same conditions not exceeding 80%. In order to simulate the actual situation of organic acid release from plant roots, the cumulative release rates of potassium in deionized water and 2% citric acid solution for 1, 2, 3, 4, 7, 14, 21, 28, 42, 56, and 70 days were calculated, respectively. The cumulative leaching rate can be expressed as Equation(2) [40,41,42]:

$${\mathrm{K}}_{\mathrm{i}}=\frac{({\mathrm{C}}_{\mathrm{i}}\mathrm{V}+\sum {\mathrm{C}}_{\mathrm{i}-1}{\mathrm{V}}_0)}{1000\times \mathrm{K}}\times 100\mathrm{\%}$$

(2)

where K_i represents the cumulative leaching rate of potassium element after the ith sampling and C_i is the concentration of potassium element in the leaching solution when the ith sampling, mg/L; V is the volume of the leaching solution, L; V₀ indicates the volume of solution taken out at each sampling, L; K represents the content of potassium in samples, g.

Soil column leaching tests were carried out in order to study the effect of the prepared SRPF in soil. Three kinds of soil from different regions, shown in

Table 2. Three types of soil physical and chemical properties required for soil columns.

	Rate of Water Content (%)	pH	Conductivity μs/cm	Organic Matter (%)	Available Potassium (g/kg)	Total Potassium (g/kg)	Ions Exchange Capacity (cmol/kg)
No. 1	8.003	8.1	82.5	0.45	0.249	11.9	8.5
No. 2	1.993	8.34	250	3.7703	0.25	17.9	14
No. 3	10.507	8.03	165.8	2.92	0.238	25.8	24.7

, were air-dried and passed through a 2-mm screen. Then, 100 g of each kind of soil was placed in a plexiglass column [43], shown in Figure 2, with 25 g of quartz sand at the bottom. Then, a 100 g mixture of soil and fertilizers containing 2 g of SRPF or 2 g of ordinary potassium sulfate fertilizers was put into the plexiglass column with the same bulk density as the test group. Twelve groups of soil column leaching tests were conducted in three kinds of soils that were saturated with deionized water or 2% citric acid solution, and then the potassium content in the leaching solution was measured periodically to calculate the cumulative release rate of potassium. Citric acid was used to investigate the effect of organic acids secreted by plants on the release rate of SRPF in the actual environment.

Data regression and kinetic models were used to study the release behavior of potassium in fertilizers and common model equations are shown in Equations (3)–(6) [44,45,46,47].

$$\begin{array}{ll}\text{First order kinetic model}:& \mathrm{y}=\mathrm{a}-\mathrm{a}·{\mathrm{e}}^{(-\mathrm{b}\mathrm{x})}\end{array}$$

(3)

$$\begin{array}{ll}\text{Double constant model}:& \mathrm{y}=\mathrm{a}{\mathrm{x}}^{\mathrm{b}}\end{array}$$

(4)

$$\begin{array}{ll}\text{Diffusion model}:& \mathrm{y}=\mathrm{a}+\mathrm{b}{\mathrm{x}}^{0.5}\end{array}$$

(5)

$$\begin{array}{ll}\text{Elovich model}:& \mathrm{y}=\mathrm{a}+\mathrm{b}\mathrm{l}\mathrm{n}\mathrm{x}\end{array}$$

(6)

where, a and b are constants obtained from regression curves, y represents the cumulative leaching rate and x is the leaching time.

2.2.2. Metal Immobilization Test

Soil No. 4 with a relatively low pH value was used to investigate whether the SRPF possessed the capacity to immobilize heavy metal-like fly ash. The physical and chemical parameters of No. 4 soil and the main heavy metal content are the average values of multiple measurements, as shown in

Table 3. The basic physicochemical properties of soil No. 4.

Rate of Water Content /%	pH	Conductivity μs/cm	Organic Matter /%	Available Potassium (g/kg)	Total Potassium (g/kg)	Ions Exchange Capacity (cmol/kg)
12.084	7.25	152.3	2.52	0.269	26.6	24.13

and

Table 4. The contents and standard values of common heavy metals in No. 4 soil.

	Pb Content (mg/kg)	Cd Content (mg/kg)	Cr Content (mg/kg)	Zn Content (mg/kg)	Cu Content (mg/kg)
No. 4 soil	48.58	0.414	201.14	147.2	83.45
Standard value	120	0.6	300	250	100

. The soil used in this paper meets the standard of soil environmental quality (GB15618-2018 [48]). Several common heavy metal contents are lower than the second standard value of soil inorganic pollutants and are in accordance with the standard of farmland. The soil was air-dried and passed through a 2 mm sieve after the removal of debris such as branches and stones. A certain volume of the above-mentioned soil was uniformly mixed with the same amount of lead ions solution, and the mixture was left standing for 3 days. Finally, the dried mixture was filtered with a 2-mm mesh and used as contaminated soil.

The contaminated soil was mixed with SRPF or fly ash according to the experimental plan shown in

Table 5. Metal immobilization test plan.

	No. 1	No. 2	No. 3	No. 4	No. 5	No. 6	No. 7
Soil/g	100	90	80	70	90	80	70
SRFs/g	0	0	0	0	10	20	30
Fly ash/g	0	10	20	30	0	0	0
Total weight/g	100	100	100	100	100	100	100

and placed in a beaker. Deionized water was added to the mixture of soil and fertilizer at the ratio of 2.5:1 between soil and water and the pH value of the soil was measured according to the standard (NY/T 1377-2007 [49]). Then, the available lead was extracted using the DTPA extraction method and the effective lead content per unit weight was calculated by Equation (7).

$$y=\frac{1000CV}{m\left(1-b\right)a}$$

(7)

where y is the content of effective lead in the soil, C is the content of lead in the leaching solution, V is the volume of leaching solution, and m is the extractive soil mass of each time, b is soil water content; a is the soil mass percentage in the soil fertilizer mixture.

The specific surface area, pore size distribution, and nitrogen adsorption-desorption isotherm of fly ash and SRPF were analyzed using the BET method. The isothermal adsorption and desorption tests of the samples were carried out using an ASAP-2460 2.01 automatic nitrogen adsorption instrument made in the United States. In the test, the fly ash and the prepared SRPF were degassed at 250 °C for 4 h, and then the nitrogen adsorption of the fly ash and SRPF were measured at −196 °C. Specific surface area, pore size distribution, and pore volume of fly ash and SRPF were calculated using the BET method and DFT method.

3. Results and Discussion

3.1. Physical and Chemical Properties of Fly Ash

The chemical composition of the fly ash used in this study is shown in

Table 6. XRF test results of fly ash.

Constituent	SiO2	Al2O3	CaO	Fe2O3	TiO2	MgO	P2O5	K2O	Else
Content/%	45.53	42.94	3.56	3.37	1.97	0.54	0.53	0.47	1.09

, in which the results show that the main chemical components of fly ash were SiO₂ and Al₂O₃, and these two components accounted for more than 80% of the total composition. The XRD result in Figure 3a with a dispersing diffraction peak indicates the presence of a large amount of amorphous material. The peaks around 16.417°, 25.986°, and 26.202° prove that the crystalline material in the fly ash was mainly a mullite phase (PDF#84-1205). Peaks at 25.539°, 35.099°, 43.289°, and 57.411° belong to Al₂O₃ (PDF#84-1205), and peaks at 20.752° and 26.524° belong to SiO₂ (PDF#89-8937). Figure 3b shows the microscopic morphology of fly ash. The particle size distribution of fly ash particles was not uniform. Some of the particles are spherical, while others are irregular. The dispersibility of fly ash particles is relatively good, and there is no obvious agglomeration phenomenon with spherical particles. Figure 3c is a higher magnification electron micrograph of fly ash, from which spherical particles of fly ash with a large number of pores on the surface can be clearly observed. This is basically consistent with the microscopic morphology of fly ash generation.

The results of the corrosion test for the fly ash show that the pH value of fly ash was 11.52, which is less than 12.5 according to GB5085.1-2007 [50]. The main leaching toxicity test was carried out using the horizontal oscillatory leaching methods (HJ557-2010 [51]). The results and the corresponding leaching toxicity index are shown in

Table 7. Leaching toxicity test results.

	Cr (mg/L)	Pb (mg/L)	Cd (mg/L)
Fly ash	≤0.1	≤0.1	≤0.1
Leaching toxicity index	15	5	1

, indicating that the content of the main heavy metals in fly ash was much less than the standard value of leaching toxicity identification (GB5085.3-2007 [52]). Therefore, the fly ash required for this study was considered to have no obvious corrosion and leaching toxicity and can be used for fertilizer preparation.

3.2. Orthogonal Test Results

According to the orthogonal experiment scheme, the initial leaching rate of each group is shown in

Table 8. Results of orthogonal experiment.

Test Number	A/%	B/°C	C/min	D	E/μm	Kwf/%	KCf/%
1	15	900	0	Furnace cooling	<75	9.83	44.36
2	15	1000	30	Air cooling	75–150	6.26	38.6
3	15	1100	60	Forced air cooling	>150	2.66	18.92
4	20	900	0	Air cooling	75–150	34.53	77.7
5	20	1000	30	Forced air cooling	>150	5.89	25.05
6	20	1100	60	Furnace cooling	<75	3.76	23.54
7	25	900	30	Furnace cooling	>150	19.99	21.26
8	25	1000	60	Air cooling	<75	13.01	51.96
9	25	1100	0	Forced air cooling	75–150	6.15	25.08
10	15	900	60	Forced air cooling	75–150	3.06	33.23
11	15	1000	0	Furnace cooling	>150	3.86	26.67
12	15	1100	30	Air cooling	<75	1.51	25.13
13	20	900	30	Forced air cooling	<75	7.56	55.85
14	20	1000	60	Furnace cooling	75–150	2.56	24.68
15	20	1100	0	Air cooling	>150	3.71	13.66
16	25	900	60	Air cooling	>150	15.94	26.23
17	25	1000	0	Forced air cooling	<75	16.49	42.99
18	25	1100	30	Furnace cooling	75–150	2.46	25.24
Kw1	4.53	15.15	12.43	7.08	8.69
Kw2	9.67	8.01	7.28	24.98	9.17
Kw3	12.34	3.38	6.83	13.94	8.68
Extreme difference	7.81	11.78	5.59	17.91	0.02
F	2.40	3.02	2.20	12.09	1.89
Superior scheme	A1	B3	C3	D1	E3
KC1	31.15	43.11	38.41	27.63	40.64
KC2	36.75	34.99	31.86	77.76	37.42
KC3	32.13	21.93	29.76	67.04	21.97
Extreme difference	5.59	21.18	8.65	39.41	18.67
F	1.87	2.11	1.90	10.71	2.07
Superior scheme	A1	B3	C3	D1	E3

Notes: K_wf is initial leaching rate in deionized water, K_Cf is initial leaching rate in citric acid of 2%, K_wi (i = 1, 2, 3) represents the average value of K_wf at different levels of factors, K_Ci (i = 1, 2, 3) represents the average value of K_Cf at different levels, where i is the number of levels.

. The XRD analysis of each group of products is shown in Figure 4.

From the results of the orthogonal experiment, it can be seen that the optimal scheme obtained by the deionized water leaching test and the 2% citric acid solution leaching test is the same and as follows: the potassium oxide content is 15%; the sintering temperature is 1100 °C; heat preservation time is 60 min; cooling mode is furnace cooling; and the particle size is not less than 150 μm. The order of factors affecting the initial leaching rate of K ions in deionized water is: cooling mode > furnace temperature > potassium oxide content > temperature keeping time > particle size, while the order of factors affecting the initial leaching rate of K ions in citric acid of 2% is: cooling mode > furnace temperature > particle size > temperature keeping time > potassium oxide content. Although the order of factors affecting of them is not exactly the same, there is no doubt that cooling mode and furnace temperature are the two most important factors. The effect of the cooling mode on the results is most significant when α is 0.05, which indicates that the initial leaching of potassium is mainly influenced by the mineral crystallization degree of the reaction product.

From Figure 4, it can be seen that the reaction products of the samples under different preparation processes are basically the same, mainly the mullite remaining after the reaction of fly ash and KAlSiO₄. However, the XRD peak strength of the mineral structure is different, indicating that the crystallization degree of the product is different and in good agreement with the leaching results of the orthogonal test. Therefore, based on the above test results and analysis, the optimal preparation process for SRPF was determined as follows: the potassium oxide content is 15%; the sintering temperature is 1100 °C; the temperature keeping time is 60 min; the cooling mode is furnace cooling; and the particle size of raw materials is not less than 150 μm.

3.3. Identification of Metastable KAlSiO₄ in the Sample of SRPF

The SRPF was prepared according to the process determined by the orthogonal experiment, and its characteristics were investigated. The XRD results of SRPF is shown in Figure 5a, the infrared spectra of fly ash and SRPF are shown in Figure 5b, and the SEM images are shown in Figure 5c,d.

From Figure 5a, it can be observed that KAlSiO₄ and mullite are the main composition of prepared SRPF. Shuangqing Su et al. [53] reported that KAlSiO₄ has multiple polymorphic forms, such as kalsilite and metastable KAlSiO₄, among which metastable KAlSiO₄ is more suitable for SRFs due to its relatively poor stability. In addition, the corrosion test results of the prepared SRPF show that its pH value is 9.43, indicating that the SRPF has no obvious corrosion.

It can be seen from Figure 5c,d that the SRPF is in an agglomerated state, loose and porous without obvious boundary contours, and irregular in shape, which belongs to the morphology characteristics of metastable KAlSiO₄. Although the microscopic morphology of SRPF and fly ash is quite different, they still have some similar structural features, such as high porosity and large specific surface area, which is conducive to improving the adsorption performance of the material. In addition, the bands at 978 cm⁻¹, 692 cm⁻¹, and 473 cm⁻¹ in Figure 5b belong to Si(Al)-O asymmetric tensile vibration, symmetrical tensile vibration of the Si(Al)-O skeleton, and Si(Al)-O skeleton bending vibration, respectively. These absorption bands are shared by two kinds of KAlSiO₄ structures. The vibrational band at 613 cm⁻¹, which is due to the substitution of Al, Si for other atoms in the crystal structure of K₂MgSi₃O₈, is the unique absorption peak of metastable KAlSiO₄; therefore, the slow-release substance of the prepared SRPF is identified as metastable KAlSiO₄ [32,53].

The absorption band at 1101 cm⁻¹ of fly ash in Figure 5b is caused by the asymmetric vibration frequency of aluminosilicate (Si-O-Si or Si-O-Al), and that at 564 cm⁻¹ and 458 cm⁻¹ represented O-Si-O vibration. The shift in the peak value of the SRPF is due to the polymerization reaction between the Si-O-Al chains [32,53].

3.4. Behavior and Effect of Slow Release of the Samples

3.4.1. Water Leaching and Acid Leaching Test

The nutrients release characteristics in deionized water and 2% citric acid are shown in Figure 6a, in which the horizontal and vertical coordinates are the logarithmic values of time and cumulative leaching rate, respectively, and the logarithmic values and percentages of the accumulative leaching rates of potassium on the 1st and 28th days were also marked. The initial leaching rate of SRPF potassium ions in water is 4.64%, which is less than 15%, and the cumulative leaching rate of 28 days is 11.17%, well under 80%. The above results indicated that the prepared SRPF meets the definition of “slow-release fertilizers” (GB/T 23348-2009 [30]). And compared with the SRFs in other studies [40,54], the slow-release performance of the SRPF prepared in this study is equivalent or slightly better.

The leaching rate of SRPF in 2% citric acid was significantly different from that in water. The initial leaching rate and 28-day cumulative leaching rate of SRPF were 47.07% and 85.86%, respectively, which indicated that citric acid greatly accelerated the release rate of potassium ions in SRPF. In fact, the organic acids secreted by plants during growth will play a similar role. This feature enables the release of SRFs to be synchronized with the growth of plants, thereby greatly reducing the ineffective release of SRFs and the loss of fertilizers. To a certain extent, the experiment simulates the working environment of fertilizers in actual use, providing experimental support for the possibility of practical application possibility of this kind of SRPF.

XRD and FTIR results of leached and unleached samples are shown in Figure 6b,c, respectively. It can be seen from Figure 6b that compared with the original XRD results of SRPF, the peak intensity and content of KAlSiO₄ in SRPF after water leaching and acid leaching were significantly reduced, indicating that the crystal structure of KAlSiO₄ in fertilizer was obviously destroyed after leaching. In addition, the main peak of KAlSiO₄ in fertilizer still existed after water leaching with a decrease in intensity, while the main peak of KAlSiO₄ disappeared completely after acid leaching, indicating that the crystal structure of KAlSiO₄ was more thoroughly destroyed in citric acid solution, which was beneficial for the release of fertilizers.

From Figure 6c, it can be observed that the FTIR peak value of SRPF changes obviously after leaching, and the degree of change is more obvious after acid leaching. After leaching, the absorption bands of 978 cm⁻¹, 692 cm⁻¹, and 473 cm⁻¹ were all shifted or disappeared, and the absorption bands of metastable KAlSiO₄ at 613 cm⁻¹ disappeared completely. These results indicate that the Si(Al)-O bond has broken and the crystal structure of KAlSiO₄ has decomposed.

The SEM images in Figure 7a–c can corroborate the structural changes in SRPF. The unleached SRPF shown in Figure 7a has a relatively complete structure and particle morphology. There are no obvious traces of dispersion or structural damage. In the SRPF after water leaching, shown in Figure 7b, it can be seen that the particles are more dispersed and some flocculus appears. These flocculi may be residues produced by the destruction of the SRPF structure. The structure of the acid-leached SRPF shown in Figure 7c is more thoroughly destroyed. Particles are almost completely gone and the residue is almost entirely flocculent. The above evidence shows that SRPF is also decomposing, which is consistent with the results of XRD and FTIR analysis. Therefore, the release of SRPF is mainly due to the replacement of potassium ions in metastable KAlSiO₄ by hydrogen ions in the solution, which triggers the disintegration of the spatial structure of the metastable KAlSiO₄ and promotes the release of potassium [32,55,56,57]. The specific release processes of fertilizers in water and acid can be expressed, respectively, by Equations (8) and (9) [32,57].

$$2{\mathrm{K}\mathrm{A}\mathrm{l}\mathrm{S}\mathrm{i}\mathrm{O}}_4+\left(\mathrm{n}+1\right){\mathrm{H}}_2\mathrm{O}={2\mathrm{K}}^{+}+{\mathrm{A}\mathrm{l}}_2{\mathrm{O}}_3·2{\mathrm{S}\mathrm{i}\mathrm{O}}_2·\mathrm{n}{\mathrm{H}}_2\mathrm{O}+2{\mathrm{O}\mathrm{H}}^{-}$$

(8)

$$2{\mathrm{K}\mathrm{A}\mathrm{l}\mathrm{S}\mathrm{i}\mathrm{O}}_4+\mathrm{n}{\mathrm{H}}_2\mathrm{O}+2{\mathrm{H}}^{+}={2\mathrm{K}}^{+}+{\mathrm{A}\mathrm{l}}_2{\mathrm{O}}_3·2{\mathrm{S}\mathrm{i}\mathrm{O}}_2·(\mathrm{n}+1){\mathrm{H}}_2\mathrm{O}$$

(9)

The fitting results of the kinetic equation of SRPF in solution are shown in

Table 9. Kinetic Model of K release from SRPF.

Solution	Deionized Water			2% Citric Acid Solution
	a	b	R2	a	b	R2
First-order kinetic model: y = a(1 − e−bx)	12.389	0.175	0.807	84.608	0.495	0.741
Double constant model: y = axb	4.268	0.292	0.986	50.844	0.152	0.977
Diffusion model: y = a + bx0.5	3.508	1.406	0.975	50.061	6.032	0.910
Elovich model: y = a + blnx	3.313	2.487	0.960	47.938	11.212	0.990

. According to the correlation coefficient, the kinetic model of potassium release in deionized water is the double constant model y = 4.268x^0.292, while the kinetic model of potassium release in 2% citric acid is the Elovich model y = 47.938 + 11.212lnx. Therefore, the organic acids released by plants have a significant release-promoting effect on the release of SRFs.

3.4.2. Column Test Results

The kinetic conditions provided by the water leaching test are very different from those in the soil, which can limit the test and make the results less practically significant. Therefore, a soil column leaching test in which SRPF was mixed with soil and leached was carried out to better study the release behavior of SRPF in soil. The results of the soil column leaching test are shown in Figure 8. For all three soils, the release rate of traditional potassium sulfate fertilizers was faster under acid leaching than water leaching. Furthermore, its release efficiency can reach about 90% in 10 days. In comparison, the release rate obtained in the soil column leaching tests was much slower than that obtained in the water leaching tests, showing better sustained release performance.

When the leaching medium is water, the release efficiency of SRPF in the three soils at 70 days was approximately 4%, which was significantly lower than that obtained in the static water leaching test. The addition of citric acid will greatly improve the release efficiency of SRPF, and the release efficiency can increase from 21.2% to 43.5% in 70 days, depending on the soil type. At the same time, it can be seen that the addition of citric acid increased the difference in the release rate of potassium ions in SRPF in different soils. Combining the property parameters of the three soils in Table 4, the rate of potassium ion release seems to be related to the total potassium content in the soil. When the total potassium content in the soil was high, the potassium ion concentration difference between the SRPF and the soil was reduced, hindering the potassium ion diffusion behavior driven by the concentration difference. The release rate of SRPF was slower when the content of potassium ions in the soil was higher. This phenomenon is in good agreement with the research conclusions of Mao et al. [41].

The soil column leaching test had different results from the water leaching test, and SRPF had a slower release rate in the soil column leaching experiment. Since the soil column leaching test can provide a similar environment to the soil, the results are considered to be more in line with the release behavior of SRPF in soil. In addition, through the soil column leaching test, the law of the influence of soil properties, such as the ion content in the soil, on the release of SRPF was revealed. This will provide experimental support for the application of the prepared SRPF.

3.5. Soil Remediation Performance

The immobilization ability and effect of the prepared SRPF on heavy metal ions are also worth evaluating, which is also a crucial part of the development of multifunctional compound SRFs. In this study, the soil remediation performance and immobilization mechanism of prepared SRPF on lead ions in soil were studied. The specific settings of the experiments are shown in Table 5. Among them, Group 1 is the control group, without adding any SRPF or fly ash to the soil. In Group 2 to 4, fly ash, which accounted for 10%, 20%, and 30% of the total mass, was added to the soil, respectively, while in Group 5 to 7, the prepared SRPF was added with 10%, 20%, and 30% of the total mass, respectively. The changes in soil available lead content and soil pH value after 4 weeks are shown in Figure 9a–c, respectively.

Figure 9a shows that, compared with the control group (group 1), adding fly ash (group 2–4) or SRPF (group 5–7) has an obvious immobilization effect on lead ions in contaminated soil. On the one hand, with the increase in time, the available lead content in the soil will gradually decrease. The decline was most pronounced between the first and second week, which was consistent across all experimental groups. In addition, increasing the amount of fly ash or SRPF can also accelerate the reduction of available lead content in soil. The experimental results also showed that, under the condition of using the same amount of SRPF and fly ash, the change trend and magnitude of the effective lead content in the soil were similar. Therefore, the above results can intuitively show that the SRPF prepared in this study not only has the function of slow-release potassium fertilizers but also has the function of passivating heavy metals. Furthermore, the heavy metal immobilization performance of fly ash and SRPF are similar.

As shown in Figure 9b, the pH values of the soil in each test group fluctuated little over time. Figure 9c shows that the pH of Group 2–7 is higher than that of Group 1 and will increase with the addition of fly ash or SRPF, which should be attributed to the alkaline nature of fly ash and SRPF. In addition, the larger the pH value was, the more obvious the immobilization effect of lead ions was, which was directly related to the surface precipitation mechanism and chemical adsorption in the metal immobilization mechanism of fly ash [58].

The nitrogen adsorption-desorption isotherms of fly ash and SRPF are shown in Figure 9d. From Figure 9d, the adsorption isotherms of fly ash and SRPF belong to capillary condensation adsorption [59]. The nitrogen adsorption amount of fly ash in the early stage increased rapidly with the increase in pressure, but the duration of this stage was small, and the adsorption increment was not large, which showed that the content of micropores in fly ash is not much. In the next stage, with the increase in relative pressure, the adsorption capacity of fly ash increased gradually, but the rate of increase had slowed considerably. Finally, when the relative pressure was greater than 0.5, the adsorption capacity of fly ash increased steeply with the increase in relative pressure due to capillary agglomeration, which indicated that there are a large number of mesopores in fly ash. Therefore, it can be concluded that the fly ash is dominated by mesopores and macropores, and the number of micropores is small.

The adsorption-desorption curve of SRPF was basically the same as that of fly ash, but the adsorption range was smaller and the adsorption increment was less in the early stage of SRPF, which also showed that the number of micropores in SRPF is small. The steep rise in the adsorption curve of SRPF also indicated that there were more mesopores and macropores in SRPF. However, it was found that the “tail” phenomenon of fly ash was stronger than that of SRPF, which indicated that the capillary condensation of fly ash is stronger in the process of adsorption, and that the mesopores of fly ash were more developed and the specific surface area was larger.

The pore size distribution curves of fly ash and SRPF are shown in Figure 9e and the structural parameters of fly ash and SRPF are listed in

Table 10. Texture parameters of fly ash and SRPF.

Sample	SBET (m2/g)	Dave (nm)	Vt (cm3/g)	Vmi (cm3/g)	Vme (cm3/g)	Vma (cm3/g)
Fly ash	2.6990	8.40417	0.00237	0.00012	0.00170	0.00055
SRPF	1.3966	8.29840	0.00199	0.00006	0.00136	0.00057

Notes: S_BET indicates the specific surface area of BET; D_ave is the average pore size of the pores; V_t is the total pore volume; V_mi is the micropores volume; V_me is the mesopores volume; and V_ma is the macropores volume.

. The results show that the specific surface areas of SRPF are 48.3% less than that of fly ash, but the average pore size is basically the same. The total pore volume of SRPF is reduced by 16%, which is mainly concentrated in the micropore volume (by 50%) and mesopore volume (by 20%). For macropores volume, the values of SRPF and fly ash are very close. The very small increase may be due to a measurement error. Overall, the adsorption characteristics, pore size, and distribution of SRPF and fly ash are basically the same with a small difference.

However, as far as the results of soil remediation experiments are concerned, the effect of these pore volume reductions on its metal immobilization ability is not very significant. This shows that the contribution of physical adsorption in SRPF is small, and supports the above conclusion that the adsorption mechanism of SRPF is mainly chemisorption.

4. Conclusions

Fly ash has excellent physical and chemical properties, and can be sintered with potassium carbonate to prepare SRPF that meets the national standard, while maintaining a certain ability to immobilize heavy metal ions.

(1) According to the results of orthogonal experiments, the optimum process for preparing SRPF from fly ash is as follows: the potassium carbonate content is 15%; the sintering temperature is 1100 °C; heat preservation time is 60 min; cooling mode is furnace cooling, and the particle size of raw materials is not less than 150 μm.

(2) The prepared SRPF has excellent sustained-release properties. The initial leaching rates of SRPF in water and 2% citric acid solution were 4.64% and 47.07%, respectively, and the 28-day cumulative leaching rates were 11.17% and 85.86%, which met the standard GB/T 23348-2009.

(3) The characterization results show that the main slow-release substance of the prepared SRPF is metastable KAlSiO₄. After leaching, the structure of SRPF is destroyed, and the characteristic features of metastable KAlSiO₄ are also weakened or disappear. The release of potassium ions is mainly due to the disintegration of the spatial structure of metastable KAlSiO₄ caused by the replacement of potassium ions by hydrogen ions in solution.

(4) The results of the water leaching test and column test showed that there was a large gap between the release efficiency of SRPF in deionized water and 2% citric acid solution, and the participation of citric acid was more conducive to the release of potassium ions. Therefore, the prepared SRPF not only has an excellent slow-release performance but also has a good release-promoting effect with the participation of organic acids secreted during plant growth, which can meet the needs of plant growth. Citric acid amplified the difference in potassium release from SRPF in different soils, which may be caused by changes in total potassium content in the soil.

(5) In soil remediation tests, the application of fly ash and SRPF can significantly reduce the content of available lead in soil compared with the control group. In the samples with the same amount of fly ash and SRPF, the change trend of the available lead was similar, and the difference in the content was small. Therefore, the adsorption properties of the prepared SRPF and fly ash compared to heavy metal ions are similar. The BET results show that, compared with fly ash, the BET surface area of SRPF is reduced by 48.3%, the total pore volume is reduced by 16.0%, and the average pore diameter has a small change. The decrease in total pore volume is mainly concentrated in the micropore volume and mesopore volume, which are reduced by 50% and 20%, respectively, while the macropore volume hardly changes. However, as far as the results of the soil remediation experiments are concerned, the effect of these pore volume reductions on its metal immobilization ability is not very significant. This shows that the SRPF immobilizes heavy metals mainly by chemical adsorption.