Study on Mechanism of MSWI Fly Ash Solidified by Multiple Solid Waste-Based Cementitious Material Using the Rietveld Method

Abstract

A novel multiple solid waste-based cementitious material (MSWCM) was developed to immobilize municipal solid waste incineration (MSWI) fly ash. The compressive strength of MSWCM with different ratios of MSWI fly ash reached the standard requirements after curing for 28 days. X-ray powder diffraction (XRD) in combination with the Rietveld method was employed to investigate the content and phase transformation of hydration products. The main hydration products of pure MSWCM paste were C-S-H, hydroxyapatite, ettringite and C-A-S-H. With increases in curing time, the content of ettringite and C-A-S-H increased significantly. The main hydration products of MSWCM paste with MSWI fly ash were C-S-H and Friedel’s salt. The contents increased markedly with increased curing time from 21.8% to 28.0% and from 8.53% to 16.7%, respectively. Additionally, a small amount of PbHPO₄ (0.51–0.96%) and lead phosphate Pb₃(PO₄)₂ (0.14–0.51%) were detected, indicating that phosphate had an effective curing effect on lead ions. The results showed that most of the hydration reactions had started at the initial stage of curing and reacted quickly to form a large number of hydration products. The quantitative analyses of hydration products provide essential information for understanding the immobilization mechanism of MSWI fly ash in MSWCM paste.

1. Introduction

Urbanization, population growth and industrialization have led to a rapid increase in municipal solid waste. By 2025, cities worldwide are expected to generate 2.2 billion tons of municipal solid waste [1]. Landfill and incineration are the two main waste disposal methods. Compared with landfill, the incineration method can save land resources. After incineration, the volume of municipal waste can be significantly reduced, and the heat generated by the complete incineration of organic matter can be used for heating and power generation, realizing the resource utilization of waste [2,3,4,5]. Therefore, the incineration method has been commonly used in the treatment of municipal waste. However, it tends to produce a large amount of municipal solid waste incineration (MSWI) fly ash. The level of MSWI has reached 146 million tons in China in 2020 and the production of MSWI fly ash has exceeded 7 million tons [6]. MSWI fly ash is produced during waste incineration. It contains a large amount of hazardous substances, such as leachable heavy metals, dioxins, soluble salts and other harmful components, which may cause harm to the environment and human health [7,8,9,10]. It can used in landfill only after solidification or stabilization [11,12,13].

Cement-based stabilization or solidification technology has been considered as the main treatment option for MSWI fly ash [14,15,16,17]. Ordinary Portland Cement (OPC) is a widely used cementitious binder for the treatment of MSWI fly ash [18,19]. However, OPC will form a strong alkaline environment that is not conducive to the solidification of heavy metals. The compressive strength of OPC–fly ash blocks may decrease significantly in the case of high waste incineration fly ash content. [20,21,22]. In addition, during the production process of Portland cement, some problems will become more prominent, such as pollutant discharge, excessive consumption of energy resources and damage to the balance of the ecological environment [23,24]. Therefore, it is necessary to develop a low carbon environment-friendly binder material.

Recently, due to excellent mechanical properties and good environmental protection, the cementitious material produced from multiple solid wastes, such as metallurgical slags and industrial solid wastes, has been widely used to treat various hazardous wastes including MSWI fly ash [25,26]. Blast furnace slag (BFS) and steel slag (SS) are commonly used metallurgical slags for solid waste-based cementitious material, and the main components are CaO, SiO₂ and Al₂O₃ [27]. BFS and SS can be used as an alternative material for alkaline activator because they are alkaline, which can provide Ca²⁺ ions and OH⁻ ions for the hydration reaction [28]. Desulfurization ash (DA) is a solid waste generated during the dry flue gas desulfurization process, with CaO and SO₃ as the main components [29,30]. Phosphoric acid sludge (PAS) is an industrial solid waste obtained from the production of wet phosphoric acid [31,32]. PAS can be used to immobilize MSWI fly ash due to the widespread use of phosphate as a curing agent [33,34,35]. Additionally, Pb can easily react with PO₄³⁻ and HPO₄²⁻ to form insoluble phosphate precipitation [33]. Therefore, phosphate shows a good solidification effect on Pb. These industrial solid wastes can be used as the raw materials of multiple solid waste-based cementitious material (MSWCM).

During the hydration process, heavy metals react with MSWCM through physical cementation, chemical adsorption, isomorphic replacement and complex precipitation. Finally, they settle in the cement hydration product in the form of hydroxide or complex [33]. The quantitative and qualitative analyses of mineral phases provide essential information for understanding the immobilization mechanism of MSWI fly ash in MSWCM paste. However, the study of assessing the mineral composition of fly ash and hydration products by conventional X-ray powder diffraction (XRD) has been hampered due to the complex phases and structures, which result in strong overlap of diffraction peaks [36]. Therefore, it is necessary to find a method that can determine the relative content of the mixture phase and hydration products accurately, effectively and conveniently in MSWCM.

XRD in combination with the Rietveld refinement method, is a powerful technique for the characterization of mineral phases. The Rietveld method is basically a profile-fitting method based on the least square approach, which can deal with complex diffraction patterns with strong overlapping peaks [37]. However, few studies have focused on using the Rietveld methods to study changes in the content of hydration products and mineral phases during hydration process.

In this study, blast furnace slag (BFS), steel slag (SS), desulfurization ash (DA) and phosphoric acid sludge (PAS) are used as raw materials to produce a novel multiple solid waste-based cementitious material (MSWCM). The development of the mechanical properties of MSWCM with different ratios of MSWI fly ash, including compressive strength and leachability at different curing times, is investigated. The Rietveld method is employed for the qualitative and quantitative analysis of the MSWI fly ash and of the hydration products in pure MSWCM pastes and pastes containing MSWI fly ash. Meanwhile, the changes in the hydration products and mineral phase content during the hydration process and the immobilization mechanism of MSWI fly ash in MSWCM pastes are also discussed.

2. Materials and Methods

2.1. Materials

Raw materials for the preparation of multiple solid waste-based cementitious material (MSWCM) included blast furnace slag (BFS), steel slag (SS), desulfurization ash (DA) and phosphoric acid sludge (PAS). The BFS, SS and DA were provided by Wuhan Metal Resources Company in Hubei Province, China. The PAS was collected from China City Environment Protection Engineering Limited Company. The chemical compositions and heavy metal content of the four raw materials were summarized in a previous study [27]. CaO, SiO₂ and Al₂O₃ were the major chemical compositions. Furthermore, the desulfurization ash contained 33.28% SO₃, the steel slag contained 26.12% Fe₂O₃ and the phosphoric acid sludge contained 21.80% P₂O₅. It had a high content of Pb (2412.81 mg/kg), which was the main heavy metal.

2.2. Orthogonal Experiment and Analysis

2.2.1. Orthogonal Experimental Design

Orthogonal experimental design is a method to study multi-factor and multi-level experiments. An orthogonal experiment was designed to optimize the mass ratio of the raw materials for the production of the MSWCM. The mass proportions of SS, DA and PAS were set as the three factors in the orthogonal experiment. The level settings of each factor are shown in

Table 1. Factors and levels for orthogonal test.

Levels	Factors
	Mass Proportions
	SS/%	DA/%	PAS/%
1	28	8	20
2	32	10	26
3	36	12	32
4	40	14	38

. The mass proportion of BFS was calculated by [100% − (SS% + DA% + PAS%)]. The standard orthogonal experimental design table L16 (43) was used and a total of 16 themes were tested. The orthogonal experimental design is displayed in

Table 2. Experimental design schemes of orthogonal test.

Sample ID	Levels			Mass Proportion of BFS/%
	SS	DA	PAS
O1	1	1	1	44
O2	1	2	2	36
O3	1	3	3	28
O4	1	4	4	20
O5	2	1	2	34
O6	2	2	1	38
O7	2	3	4	18
O8	2	4	3	22
O9	3	1	3	24
O10	3	2	4	16
O11	3	3	1	32
O12	3	4	2	24
O13	4	1	4	14
O14	4	2	3	18
O15	4	3	2	22
O16	4	4	1	26

.

2.2.2. Range Analysis Method

In this study, the range analysis method was used to determine the factors’ sensitivity to compressive strength according to the orthogonal experiment [38]. The range method is to subtract the minimum value from the maximum value in the data. In the calculation, the average value k_i and the influence degree R are as follows [39]:

$${\mathit{k}}_{\mathit{i}}=\frac{{\mathit{K}}_{\mathit{i}}}{\mathit{x}}$$

(1)

$$\mathit{R}=\mathit{m}\mathit{a}\mathit{x}\left\{{\mathit{k}}_1,{\mathit{k}}_2, {\mathit{k}}_3\cdots \right\}-\mathit{m}\mathit{i}\mathit{n}\left\{{\mathit{k}}_{\mathit{1}},{\mathit{k}}_{\mathit{2}}, {\mathit{k}}_{\mathit{3}}\cdots \right\}$$

(2)

where, k_i is the arithmetic mean value of the corresponding test results at level i; K_i is the sum of the corresponding test results; and x is the number of tests on level i. The influence degree R represents the influence of each factor on the index. The higher the R value, the greater the influence of the factor on the experiment results.

2.3. Compressive Strength Test

Compressive strength was measured in an electro-hydraulic universal testing machine (WAW-1000B) in compliance with the GB/T17671 [27] China standard. The measurements were performed on specimens for 3, 7 and 28 days at room temperature and at least three representative samples were tested to obtain the average values.

2.4. Leaching Test

The leaching toxicity of the heavy metals was evaluated by the Solid Waste-Extraction Procedure for Leaching Toxicity—Horizontal Vibration Method (HJ 557-2010). The samples were first crushed to below 3 mm, then 100 g was taken and put into a 2 L volumetric flask, and we added 1000 mL of deionized water. After that, the mixtures were put into a horizontal vibrator for 8 h with a frequency of 110 times per minute. After standing still for 16 h, the concentrations of heavy metals in the leachate were determined by inductively coupled plasma mass spectrometry (ICP-MS).

2.5. X-ray Powder Diffraction (XRD) Analysis

The phase analysis of the samples was determined by the X-ray powder diffraction (XRD) method. The XRD patterns were collected by an Rigaku D/Max-B diffractometer with an X’Celerator detector and using Cu Kα radiation (λ = 0.154056 nm), 40 kV and 40 mA. The measured range of the powder specimens was set from 5 to 80° 2θ, with a 0.04° step size.

2.6. Rietveld Method

The Rietveld method is an advantageous technique for the refinement of crystal structures and phase quantitative analysis. This method was employed to determine the major phases of the MSWI fly ash and the hydration products during solidification in this study. The principle of the Rietveld method is to minimize the difference between the observed and calculated intensities by a least square approach (Equation (3)). The BGMN program was used for the Rietveld refinement [40,41]:

$${\displaystyle \sum }_{\mathit{i}}{\mathit{w}}_{\mathit{i}}{\left[{\mathit{y}}_{\mathit{i}}\left(\mathit{obs}\right)-{ \mathit{y}}_{\mathit{i}}\left(\mathit{calc}\right)\right]}^2=\mathrm{Minimum}$$

(3)

where yi (obs) and yi (calc) are the observed and calculated intensities at point i, and w_i is the weight assigned to each intensity. The intensity at each point is calculated by summing the contributions from all the neighboring Bragg reflections and background scattering:

$${\mathit{y}}_{\mathit{i}}=\mathit{S}{\displaystyle \sum }_{\mathit{k}} \left[{\left|{\mathit{F}}_{\mathit{hkl}}\right|}^2· \mathit{G}(\mathrm{\Delta }2\mathrm{\theta })\mathrm{·} \mathit{Lp}\left(\mathit{\theta }\right)·{ \mathit{M}}_{\mathit{hkl}}·{ \mathit{T}}_{\mathit{f}}\left(\mathit{\theta }\right)· \mathit{A}·{ \mathit{Po}}_{\mathit{hkl}}\right]+{\mathit{y}}_{\mathit{b}\mathit{i}}$$

(4)

where ${\mathit{y}}_{\mathit{i}}$ intensity of the angular position i in the powder pattern;

$\mathit{S}$—scale factor of the phase;

${\mathit{F}}_{\mathit{hkl}}$—structure factor;

G($\mathrm{\Delta }2$θ)—profile shape function;

$\mathit{Lp}\left(\mathit{\theta }\right)$—Lorentz and Polarization factor;

${\mathit{M}}_{\mathit{hkl}}$—multiplicity factor;

${\mathit{T}}_{\mathit{f}}\left(\mathit{\theta }\right)$—temperature factor;

$\mathit{A}$—absorption factor;

${\mathit{Po}}_{\mathit{hkl}}$—correction factor of preferred orientation;

${\mathit{y}}_{\mathit{bi}}$—background intensity.

3. Results and Discussion

3.1. Characterization of MSWI Fly Ash

Municipal solid waste incineration (MSWI) fly ash was collected from MSWI plants in Wuhan, China. The main chemical composition and heavy metal contents are given in

Table 3. Chemical composition and heavy metal content of MSWI fly ash.

Oxide (%)	CaO	SiO2	Al2O3	Cl	SO3	Fe2O3	Na2O	K2O	MgO
	39.08	1.94	0.31	29.17	7.37	1.46	7.03	10.58	0.56
Heavy metal (mg/kg)	Pb	Zn	As	Cr	Hg
	1769.9	5636.2	45.7	34.3	2.4

. According to the results, CaO accounted for the largest proportion of MSWI fly ash, accounting for 39.08%, followed by Cl (21.5%) and Na₂O (7.03%). This is probably due to the addition of alkali to neutralize the acidic gases generated during incineration, such as HCl and SO₂ [42]. The high content of chloride ion may be attributed to the incineration of plastics and halite in food residues, e.g., NaCl [43,44]. The most abundant heavy metals are Pb and Zn, with a content of 1769.9 mg/kg and 5636.2 mg/kg, respectively.

The phase content of the MSWI fly ash was determined by the Rietveld method. Figure 1 presents the Rietveld refinement plot for the MSWI fly ash. The measured patterns showed good agreement with the calculated patterns. The R_wp value was 10.25%. According to the results, the main phases in the MSWI fly ash were NaCl (34.2%), CaClOH (24.6%), Ca(OH)₂ (16.5%) and KCl (14.3%) (

Table 4. Results of the Rietveld refinement of MSWI fly ash.

	Halite	Sylving	Calcium Chloride Hydroxide	Portlandite	Calcite	Gypsum	Rwp/%	Rexp/%
	NaCl	KCl	CaClOH	Ca(OH)2	CaCO3	CaSO4·2H2O
Phase content/w.t.%	34.2 (2) a	14.3 (1)	24.6 (2)	16.5 (1)	7.6 (5)	2.8 (7)	10.25	5.26

^a Standard deviation.

). Moreover, CaCO₃ and CaSO₄ were also detected.

3.2. Orthogonal Experiment Results Analysis

The range analysis method was adopted for the orthogonal experiment results. The R value (Range) is used to divide the impact degree of different factors on the experimental objectives. Figure 2 and

Table 5. The result of range analysis.

Levels	Compressive Strength/Mpa
	SS	DA	PAS
K1	81.8	74.9	95
K2	83.7	80.7	81.4
K3	73.1	81.3	71.7
K4	75.8	77.5	66.3
k1	20.45	18.73	23.75
k2	20.93	20.18	20.35
k3	18.28	20.33	17.93
k4	18.95	19.38	16.58
Range	2.65	1.6	7.17
Ranking	PAS > SS > DA
Optimum theme	SS2 DA3 PAS1

present the results of the orthogonal experiment and range analysis. According to the K value, the optimal combination of factors and levels is SS₂ DA₃ PAS₁, and their corresponding mass proportions are 32%, 12% and 20%, respectively. The mass proportion of BFS is calculated as 36%. As shown in Table 5, the order of the R value is PAS > SS > DA, indicating that the mass ratio of PAS is the dominant factor affecting the compressive strength, followed by the mass ratios of SS and DA.

Figure 3 shows the effect curves of different raw materials on the compressive strength. Based on the results, the influence of the PAS ratio on the compressive strength is the greatest. The compressive strength decreases with increases in the PAS mass ratio. However, the compressive strength shows a tendency to first increase and then decrease with increases in the DA mass ratio. The effect of the SS mass ratio on the compressive strength is fluctuating.

3.3. Effect of MSWI Fly Ash Proportions on Compressive Strength

The compressive strength of multiple solid waste-based cementitious material (MSWCM) with different amounts of MSWI fly ash is plotted in Figure 3. The compressive strength increased with curing time for all specimens and, after curing for 3, 7 and 28 days, the compressive strength varied from 1.2 to 33.4 MPa, which all met the standard requirements of Chinese technical standards for the solidification/stabilization treatment of MSWI fly ash. The compressive strength of the MSWCM pastes without MSWI fly ash had reached 27.2, 28.6 and 33.4 MPa after curing for 3, 7 and 28 days, respectively. It was noticed that the 7-day compressive strength was 81.4% of the 28-day strength, indicating a rapid early strength development.

According to Figure 4, the compressive strength showed a tendency to decrease with increases in MSWI fly ash content. In comparison with the MSWCM pastes after a 28-day curing period, the compressive strength of the MSWCM pastes decreased by 4.2%, 17.3%, 27.5%, 45.5%, 57.5%, 72.2%, 82.6% and 91.3%, respectively, with an increase in MSWI fly ash content from 10% to 80%. The addition of MSWI fly ash had a negative impact on the compressive strength of the MSWCM pastes, mainly due to the lower activity of MSWI fly ash and the high level of heavy metals, which inhibited the hydration reaction and weakened the compressive strength [42,45,46].

3.4. Leaching Toxicity

The leaching concentrations of As, Cr, Pb and Zn of the MSWCM pastes with MSWI fly ash at 3, 7 and 28 curing days are given in Figure 5. With increases in MSWI fly ash content, the leaching concentration of As, Cr and Pb increased. The pure MSWCM pastes showed a low leaching concentration of these four heavy metals. The maximum leaching concentrations of As, Cr, Pb and Zn were 22.6 µg/L, 12.5 µg/L,113.0 µg/L and 75.6 µg/L, respectively, which were far below the Toxicity Characteristic Leaching Procedure (TCLP) standard requirement and the threshold value specified in Standard GB16889-2008. This indicated that MSWCM can effectively immobilize these heavy metals.

MSWCM after curing with MSWI fly ash can be used as a supplementary cementitious material, replacing or partially replacing traditional cement and other cementitious materials. This type of cementitious material can be widely used in various applications, such as roadbed materials, landfill closure and ground hardening.

3.5. Characterization of MSWCM Hydration Products with and without MSWI Fly Ash

3.5.1. XRD Analysis of Hydration Products

The XRD patterns of the pure MSWCM pastes (MF0) and the MSWCM pastes with 30% MSWI fly ash (MF1) at a curing time of 3, 7 and 28 days are shown in Figure 6 and the main mineral phases have been labelled. According to the diffraction pattern, the diffraction peak with the strongest intensity in the pure MSWCM pastes belonged to calcium carbonate, and the main hydration products were hydroxyapatite, ettringite and calcium silicate hydrate (C-S-H and C-A-S-H).

In contrast, strong peaks for chlorine salts such as halite and sylvite were observed in the samples mixed with 70% MSWCM and 30% MSWI fly ash. Additionally, some new hydrates such as Friedel’s salt (3CaO·Al₂O₃·CaCl₂·10H₂O), PbHPO₄ and Pb₃(PO₄)₂ were formed. The diffraction peaks intensity of the Friedel’s salt increased significantly with increases in curing time. Meanwhile, the diffraction peaks of the halite (NaCl) and sylvite (KCl) gradually weakened with the increase of Friedel’s salt. It can be deduced that chloride salts might be partially transformed into Friedel’s salt, which indicated that part of the Cl in MSWI fly ash could be encapsulated in the MSWCM solidified body. The generation of PbHPO₄ and Pb₃(PO₄)₂ in sample MF1 indicated that the Pb took part in a chemical reaction with phosphate and then formed insoluble lead phosphate minerals. This conclusion was consistent with [42] and confirmed that phosphate shows a good solidification effect on Pb.

Furthermore, calcium-containing hydration products (C-S-H and C-A-S-H) were found in both the MF0 and MF1 samples, and the intensity of these diffraction peaks in sample MF1 is significantly stronger than that in sample MF0. However, no distinct peaks of portlandite were observed in the MF0 and MF1 samples. It is indicated that these hydration products were further hydrated with the components in the MSWI fly ash.

3.5.2. Immobilization Mechanism Analysis using Rietveld Method

Although the main mineral composition can be obtained by traditional XRD analysis, it is difficult to identify minor minerals and their content, and the accuracy of the results is not high. The qualitative and quantitative phases of the MSWCM pastes with MSWI fly ash are crucial for understanding the hydration process and the hydration mechanism. Therefore, the Rietveld method was employed in this work to investigate the mineral composition and content and the mineral phase transformation of the hydration products.

The mineralogical compositions and quantitative analysis of the hydration products in the MSWCM pastes (MF0) without or with MSWI fly ash (samples MF0 and MF1) at a curing time of 3, 7 and 28 days are presented in Table 6 and Table 7. The quality of the Rietveld fitting was evaluated by the weighted profile factor R_wp and the expected factor R_exp. Figure 7 shows the refinement results of samples MF0 and MF1 after curing for 28 days. The calculated patterns of these two samples showed comparatively good agreement with the measured patterns. The refinements of samples MF0 and MF1 after curing different days achieved R_wp values in the range of 7.35% to 7.55% and 7.70% to 7.81%, and R_exp values in the range of 5.85% to 5.90% and 6.01% to 6.12%, respectively. It is indicated that the refinement results for these samples were acceptable and reasonable.

• Pure MSWCM Paste

As shown in Table 6, the pure MSWCM paste without MSWI fly ash (sample MF0) contained C₂S (17.85–20.55%) and calcite (12.59–14.71%). The main hydration products of sample MF0 were C-S-H (Ca₅Si₆O₁₆(OH)·4H₂O), hydroxyapatite (Ca₅(PO₄)₃(OH)) and ettringite (3CaO·Al₂O₃·3CaSO₄·32H₂O).

The hydration reaction of C₂S and C₃S in the raw material produced calcium silicate hydrate (C-S-H) and calcium hydroxide (Equations (5) and (6)). The hydration reaction of C₂S is much slower than that of C₃S. Therefore, after 28 days of curing, there was still unreacted C₂S. The content of C-S-H and hydroxyapatite at different curing times was in the range of 19.8–24.7% and 11.29–19.87%, respectively. It can be seen from the results that a large amount of C-S-H and hydroxyapatite have been formed when the curing time is 3 days. This is very beneficial to the improvement of compressive strength and the solidification of MSWI fly ash. The generated alkaline activator Ca(OH)₂ can destroy and dissolve the silica vitreous structure and increase the amount of Ca²⁺, Al³⁺ and SiO₄²⁻. This may accelerate the hydration reaction, resulting in higher early compressive strength. With the continuous accumulation of various ions in the liquid phase, the solution reaches the supersaturated state of new hydration products, thus generating new hydration products. For example, Ca(OH)₂ further reacts with Al₂O₃ and CaSO₄ to form ettringite and a small amount of C-A-S-H (3CaO·Al₂O₃·2SiO₂·2H₂O) (Equations (7) and (8)). The generated C-A-S-H content was 2.35–4.71%. Compared with the C-A-S-H, the content of generated ettringite was relatively large, and with the prolonging of the curing time the content increased significantly. Its content were 4.15%, 17.9% and 16.9% at a curing time of 3, 7 and 28 days, respectively. It indicated that the hydration process was constantly evolving.

In addition, PO₄³⁻ produced from phosphoric acid sludge (PAS) reacted with Ca²⁺ in the aqueous solution to form a new hydration product, hydroxyapatite (Ca₅(PO₄)₃(OH)) (Equation (9)). The content of hydroxyapatite was 19.87%, 13.40% and 11.29% at a curing time of 3, 7 and 28 days, respectively. The possible hydration reactions that occurred in the pure MSWCM paste samples were listed as follows:

$${2\mathrm{C}}_3{\mathrm{S}+6\mathrm{H}}_2\mathrm{O} \to { 3\mathrm{C}-\mathrm{S}-\mathrm{H}+3\mathrm{Ca}\left(\mathrm{OH}\right)}_2$$

(5)

$${2\mathrm{C}}_2{\mathrm{S}+4\mathrm{H}}_2\mathrm{O} \to { 3\mathrm{C}-\mathrm{S}-\mathrm{H}+\mathrm{Ca}\left(\mathrm{OH}\right)}_2$$

(6)

$${3\mathrm{Ca}\left(\mathrm{OH}\right)}_2{+\mathrm{Al}}_2{\mathrm{O}}_3{+3\mathrm{CaSO}}_4·{\mathrm{H}}_2{\mathrm{O}+23\mathrm{H}}_2\mathrm{O} \to 3\mathrm{CaO}·{\mathrm{Al}}_2{\mathrm{O}}_3·{3\mathrm{CaSO}}_4·{32\mathrm{H}}_2\mathrm{O}$$

(7)

$${\mathrm{Ca}\left(\mathrm{OH}\right)}_2{+\mathrm{Al}}_2{\mathrm{O}}_{3 }{+\mathrm{SiO}}_2{+\mathrm{H}}_2\mathrm{O} \to \mathrm{CaO}·{\mathrm{SiO}}_2·{\mathrm{Al}}_2{\mathrm{O}}_3·{(n+1)\mathrm{H}}_2\mathrm{O} (\mathrm{C}-\mathrm{A}-\mathrm{S}-\mathrm{H})$$

(8)

$${5\mathrm{Ca}}^{2+}{+4\mathrm{OH}}^{-}{+3\mathrm{HPO}}_4^{2-} \to { \mathrm{Ca}}_5{\left({\mathrm{PO}}_4\right)}_3\left(\mathrm{OH}\right){+\mathrm{H}}_2\mathrm{O}$$

(9)

Some heavy metal oxides were detected at very low levels, such as PbO (0.17–0.3%) and PbO₂ (0.43–0.52%). And, after curing for 28 days, a small amount of mayenite (12CaO·7Al₂O₃) (0.55%) appeared.

Table 6. Quantitative results of sample MF0 using the Rietveld method.

Minerals	Chemical Formula	Rietveld Quantitative Results
		Contents/Mass %
		3 d	7 d	28 d
		Rwp = 7.55% Rexp = 5.88%	Rwp = 7.35% Rexp = 5.85%	Rwp = 7.36% Rexp = 5.90%
Calcite	CaCO3	14.70 (1) a	13.78 (1)	12.59 (1)
Ettringite	3CaO·Al2O3·3CaSO4·32H2O	4.15 (1)	17.90 (1)	16.90 (1)
C2S	2CaO·SiO2	20.10 (1)	20.55 (1)	17.85 (1)
C-S-H	Ca5Si6O16(OH)·4H2O	24.70 (2)	17.80 (2)	19.80 (2)
Gypsum	CaSO4·2H2O	4.08 (1)	3.09 (1)	4.23 (1)
Hydroxyapatite	Ca5(PO4)3(OH)	19.87 (1)	13.40 (1)	11.29 (1)
Portlandite	Ca(OH)2	0	0	0
Merwinite	12CaO·7Al2O3	9.79 (1)	9.37 (1)	11.44 (1)
Mayenite	Ca12Al14O33	0	0.39 (2)	0.55 (2)
PbO	-	0.30 (1)	0.16 (1)	0.21 (1)
PbO2	-	0	0.48 (1)	0.47 (1)
C-A-S-H	3CaO·Al2O3·2SiO2·2H2O	2.35 (1)	3.07 (1)	4.71 (1)
sum		100.04	100.00	100.04

^a Standard deviation.

Table 6. Quantitative results of sample MF0 using the Rietveld method.

Minerals	Chemical Formula	Rietveld Quantitative Results
		Contents/Mass %
		3 d	7 d	28 d
		Rwp = 7.55% Rexp = 5.88%	Rwp = 7.35% Rexp = 5.85%	Rwp = 7.36% Rexp = 5.90%
Calcite	CaCO3	14.70 (1) a	13.78 (1)	12.59 (1)
Ettringite	3CaO·Al2O3·3CaSO4·32H2O	4.15 (1)	17.90 (1)	16.90 (1)
C2S	2CaO·SiO2	20.10 (1)	20.55 (1)	17.85 (1)
C-S-H	Ca5Si6O16(OH)·4H2O	24.70 (2)	17.80 (2)	19.80 (2)
Gypsum	CaSO4·2H2O	4.08 (1)	3.09 (1)	4.23 (1)
Hydroxyapatite	Ca5(PO4)3(OH)	19.87 (1)	13.40 (1)	11.29 (1)
Portlandite	Ca(OH)2	0	0	0
Merwinite	12CaO·7Al2O3	9.79 (1)	9.37 (1)	11.44 (1)
Mayenite	Ca12Al14O33	0	0.39 (2)	0.55 (2)
PbO	-	0.30 (1)	0.16 (1)	0.21 (1)
PbO2	-	0	0.48 (1)	0.47 (1)
C-A-S-H	3CaO·Al2O3·2SiO2·2H2O	2.35 (1)	3.07 (1)	4.71 (1)
sum		100.04	100.00	100.04

^a Standard deviation.

• MSWCM Paste with MSWI Fly Ash

The mineral phase quantitative analysis of MSWCM paste with MSWI fly ash (sample MF1) under different curing times using the Rietveld method is listed in Table 7. It can be seen from the results that the phase composition is relatively complex. The dominant mineral phases were C-S-H, halite, sylvite, Friedel’s salt, K₃PO₄, hydroxyapatite, AlPO₄, calcite and gypsum, while the minor mineral phases were portlandite, periclase, mayenite, ettringite, grossite, katoite, PbO₂, CaCl₂, PbHPO₄ and lead phosphate (Table 7).

The main hydration products of sample MF1 were C-S-H (Ca₅Si₆O₁₆(OH)·4H₂O) and Friedel’s salt (3CaO·Al₂O₃·CaCl₂·10H₂O). The content of C-S-H and Friedel’s salt increased significantly with increases in curing time and their contents were 21.8%, 26.2% and 28.0% and 8.53%, 12.68% and 16.17% after the curing time of 3, 7 and 28 days, respectively. This result is consistent with the intensity variation of the diffraction peaks (Figure 6). At the same time, the amount of halite (NaCl) and sylvite (KCl) is decreasing, further illustrating the effect of Friedel’s salt on the solidification of the chloride ions in the MSWI fly ash. C-S-H has been proved to immobilize heavy metals (such as Zn, Cd and Pb) effectively [33]. With increases in curing time, the content of C-S-H increased, the stabilization efficiency should be better. This is consistent with the leaching results shown above (Figure 5). Additionally, the more C-S-H generated, the more beneficial it is to improve the compressive strength. This is consistent with the growth trend in Figure 3. In addition, there was a small amount of C-A-S-H (3CaO·Al₂O₃·2SiO₂·2H₂O) generated, the content of which is 2.54–4.22% at different curing times.

Halite (NaCl) and sylvite (KCl) were the main phases in the MSWI fly ash (Figure 1). NaCl can react with calcium aluminate hydrate to form Friedel’s salt (3CaO·Al₂O₃·CaCl₂·10H₂O) (Equation (10)) and potassium ions may form potassium phosphate (K₃PO₄) (Equation (11)). The content of halite and sylvite decreased with increases in curing days (Table 7). This suggested that some chloride ions in MSWI fly ash could be encapsulated in MSWCM paste by a chemical bonding effect.

A small amount of PbHPO₄ (0.51–0.96%) and lead phosphate Pb₃(PO₄)₂ (0.14–0.51%) were detected. Irene et al. and Ren et al. [35,47] found that the presence of phosphate was beneficial for the formation of insoluble lead phosphate salts (Equations (12) and (13)). Compared with other heavy metal ions, such as Zn, Cr and As, the phosphate ions reacted preferentially with Pb [48]. Due to the very low solubility of lead phosphate salts, Pb can be stably solidified in MSWCM paste and have a good curing effect. The quantitative change of the main hydration products is shown in Figure 8.

$${\mathrm{Al}}_2{\mathrm{O}}_3{+4\mathrm{CaO}+2\mathrm{NaCl}+11\mathrm{H}}_2\mathrm{O} \to 3\mathrm{CaO}·{\mathrm{Al}}_2{\mathrm{O}}_3·{\mathrm{CaCl}}_2·{10\mathrm{H}}_2\mathrm{O}+\mathrm{NaOH}$$

(10)

$${\mathrm{PO}4}^{3-}{+3\mathrm{OH}}^{-}{+3\mathrm{K}}^{+}\to { \mathrm{K}}_3\left({\mathrm{PO}}_4\right){+3\mathrm{H}}_2\mathrm{O}$$

(11)

$${2\mathrm{PO}4}^{3-}{+3\mathrm{Pb}}^{2+}\to { \mathrm{Pb}}_3{\left({\mathrm{PO}}_4\right)}_2$$

(12)

$${\mathrm{HPO}4}^{2-}{+\mathrm{Pb}}^{2+}\to { \mathrm{PbHPO}}_4$$

(13)

Table 7. Quantitative results of sample MF1 using the Rietveld method.

Minerals	Chemical Formula	Rietveld Quantitative Results
		Contents/Mass %
		3 d	7 d	28 d
		Rwp = 7.81% Rexp = 6.07%	Rwp = 7.77% Rexp = 6.12%	Rwp = 7.70% Rexp = 6.12%
Portlandite	Ca(OH)2	0.96 (2) a	0.17 (1)	0
C-S-H	Ca5Si6O16(OH)·4H2O	21.8 (1)	26.2 (1)	28.0 (1)
Periclase	MgO	0.49 (2)	2.32 (2)	1.43 (3)
Halite	NaCl	11.22 (3)	12.27 (3)	6.76 (3)
Sylvite	KCl	4.57 (2)	5.05 (2)	1.62 (2)
Mayenite	Ca12Al14O33	0.86 (4)	0	0
Gypsum	CaSO4·2H2O	5.52 (4)	5.17 (4)	5.79 (4)
Ettringite	3CaO·Al2O3·3CaSO4·32H2O	0	2.16 (1)	0.49 (1)
Friedel‘s salt	3CaO·Al2O3·CaCl2·10H2O	8.53 (5)	12.68 (5)	16.17 (6)
Grossite	CaAl4O7	1.36 (3)	2.09 (5)	1.08 (3)
Calcite	CaCO3	6.57 (5)	5.22 (5)	9.84 (5)
K3PO4	-	8.56 (3)	6.54 (3)	8.62 (3)
PbO2	-	0.68 (1)	0.27 (1)	0.41 (1)
AlPO4	-	7.07 (1)	7.57 (1)	6.71 (1)
Hydroxyapatite	Ca5(PO4)3(OH)	7.95 (5)	6.68 (45)	8.2 (6)
CaCl2	-	0.28 (1)	0.52 (1)	0.39 (1)
PbHPO4	-	0.96 (1)	0.73 (1)	0.51 (1)
Pb3(PO4)2	-	0.19 (1)	0.14 (1)	0.51 (1)
C-A-S-H	3CaO·Al2O3·2SiO2·2H2O	2.54 (5)	4.22 (4)	3.44 (4)
Carnallite	KCl·MgCl2·6H2O	9.87 (6)	-	-
Sum		100.0	100.0	100.0

^a Standard deviation.

Table 7. Quantitative results of sample MF1 using the Rietveld method.

Minerals	Chemical Formula	Rietveld Quantitative Results
		Contents/Mass %
		3 d	7 d	28 d
		Rwp = 7.81% Rexp = 6.07%	Rwp = 7.77% Rexp = 6.12%	Rwp = 7.70% Rexp = 6.12%
Portlandite	Ca(OH)2	0.96 (2) a	0.17 (1)	0
C-S-H	Ca5Si6O16(OH)·4H2O	21.8 (1)	26.2 (1)	28.0 (1)
Periclase	MgO	0.49 (2)	2.32 (2)	1.43 (3)
Halite	NaCl	11.22 (3)	12.27 (3)	6.76 (3)
Sylvite	KCl	4.57 (2)	5.05 (2)	1.62 (2)
Mayenite	Ca12Al14O33	0.86 (4)	0	0
Gypsum	CaSO4·2H2O	5.52 (4)	5.17 (4)	5.79 (4)
Ettringite	3CaO·Al2O3·3CaSO4·32H2O	0	2.16 (1)	0.49 (1)
Friedel‘s salt	3CaO·Al2O3·CaCl2·10H2O	8.53 (5)	12.68 (5)	16.17 (6)
Grossite	CaAl4O7	1.36 (3)	2.09 (5)	1.08 (3)
Calcite	CaCO3	6.57 (5)	5.22 (5)	9.84 (5)
K3PO4	-	8.56 (3)	6.54 (3)	8.62 (3)
PbO2	-	0.68 (1)	0.27 (1)	0.41 (1)
AlPO4	-	7.07 (1)	7.57 (1)	6.71 (1)
Hydroxyapatite	Ca5(PO4)3(OH)	7.95 (5)	6.68 (45)	8.2 (6)
CaCl2	-	0.28 (1)	0.52 (1)	0.39 (1)
PbHPO4	-	0.96 (1)	0.73 (1)	0.51 (1)
Pb3(PO4)2	-	0.19 (1)	0.14 (1)	0.51 (1)
C-A-S-H	3CaO·Al2O3·2SiO2·2H2O	2.54 (5)	4.22 (4)	3.44 (4)
Carnallite	KCl·MgCl2·6H2O	9.87 (6)	-	-
Sum		100.0	100.0	100.0

^a Standard deviation.

4. Conclusions

This study proposed the preparation of multiple solid waste-based cementitious material (MSWCM) from blast furnace slag (BFS), steel slag (SS), desulfurization ash (DA) and phosphoric acid sludge (PAS). The Rietveld method was applied for the qualitative and quantitative analysis of the hydration products of pure MSWCM pastes and MSWCM pastes with MSWI fly ash. The immobilization mechanism of MSWI fly ash by MSWCM binders was explored in depth. The main conclusions can be drawn as follows:

(a)

The optimal mass ratios of BFS:SS:DA:PAS for MSWCM preparation were 36:32:12:20 (w.t.%). The compressive strength of MSWCM with 80% MSWI fly ash content was 2.9 MPa during the 28-day curing period, which met the standard requirements. The leachability of As, Pb, Cr and Zn were far below Toxicity Characteristic Leaching Procedure (TCLP) standard requirement and the threshold value specified in Standard GB16889-2008.

(b)

The XRD analysis showed that the phase composition of the hydrated product was relatively complex. The most obvious diffraction peak in the pure MSWCM pastes was calcium carbonate and the main hydration products were hydroxyapatite, ettringite and calcium silicate hydrate (C-S-H and C-A-S-H). In contrast, strong peaks for chlorine salts such as halite and sylvite were observed in the mixed sample of 70% MSWCM and 30% MSWI fly ash. Additionally, small amounts of lead phosphate salts (PbHPO₄ and Pb₃(PO₄)₂) were formed.

(c)

The refinements of the samples MF0 and MF1 got low R_wp and R_exp values, which indicated that the refinement results were acceptable and reasonable. Most of the hydration reactions had started at the initial stage of curing and reacted quickly to form a large amount of hydration products. For the pure MSWCM paste, the main hydration products were C-S-H, hydroxyapatite and ettringite. With increased curing time, the content of ettringite increased obviously. For the MSWCM paste with MSWI fly ash, the content of C-S-H and Friedel’s salt increased significantly with increased curing time. During a curing period of 3 days to 28 days, the content of C-S-H and Friedel’s salt increased by 22% and 90%, respectively. Phosphate ions reacted with Pb to form a small amount of PbHPO₄ and Pb₃(PO₄)₂. It was further confirmed that phosphate had a good curing effect on lead ions. The hydration process could be deduced and evidenced by the changes in the content of the hydration products.