Micropore Structure Evolution and Macro-Micro Quantitative Analysis of Dredged Sludge Solidified with Ground Granulated Blast Furnace Slag, Carbide Slag, and Titanium Gypsum

Abstract

Revealing the evolution of micropore structure in industrial by-product solidified sludge is essential for elucidating strength development mechanisms and promoting the engineering utilization of industrial wastes. In this study, a series of tests, including unconfined compressive strength (UCS), low-field nuclear magnetic resonance, direct shear, and scanning electron microscopy coupled with energy-dispersive spectroscopy, were conducted on granulated blast furnace slag–carbide slag–titanium gypsum (GCT)-solidified sludge (GSDS) and cement-solidified sludge (CSDS). The results demonstrate that GSDS exhibits significantly superior compressive strength, deformation resistance, and pore-filling capacity compared with CSDS. With increasing curing age, both materials show logarithmic increases in UCS and mesopore volume fraction, accompanied by power-law decreases in total pore volume and the most probable pore size. On this basis, quantitative relationships between micropore characteristics and macroscopic mechanical properties are established for both solidified sludges. Microscopic analyses reveal that strength development in GSDS is primarily attributed to the formation of abundant C-(A)-S-H gels and expansive ettringite crystals, which effectively cement soil particles and refine interparticle pores. The synergistic solidification mechanism of GCT, involving ion exchange, cementitious bonding, and pore filling, promotes particle aggregation, enhances interparticle bonding, and refines pore structure, thereby markedly improving structural integrity and macroscopic strength in GSDS.

1. Introduction

Dredged sludge (DS) refers to the water–sediment mixture generated during dredging activities in rivers, lakes, harbors, and other aquatic environments, with its annual production steadily increasing [1,2]. This DS generally exhibits high water content and a fluid-like consistency, containing organic matter, microorganisms, and potentially hazardous substances such as heavy metals and various pollutants [3,4,5]. In recent years, its management has posed growing challenges due to difficulties in transportation and disposal. Moreover, prolonged storage without appropriate treatment may result in secondary environmental contamination, posing serious risks to both ecosystems and human health [6,7]. Therefore, the effective treatment and stabilization of DS are critical for both environmental protection and resource reutilization.

In response to these challenges, chemical solidification has emerged as a widely studied and increasingly adopted technique in both academic research and engineering practice. [8,9,10,11]. This technique involves the introduction of binders or reactive agents into DS, initiating a series of physicochemical reactions that enhance its mechanical strength and structural integrity, while effectively reducing the migration and release of hazardous substances. Consequently, chemical solidification is regarded as a key strategy for promoting sustainable engineering practices [12]. However, conventional binders such as ordinary Portland cement (OPC) and lime are linked to intensive energy consumption, elevated production costs, and considerable CO₂ emissions [6,13]. Specifically, producing 1 ton of OPC consumes approximately 1.16 tons of limestone and emits about 0.95 tons of CO₂, leading to substantial environmental burdens [14,15]. In response, growing efforts have been directed toward developing environmentally friendly solidification materials for DS treatment by partially substituting OPC with industrial by-products [16,17,18,19,20,21]. Among these, ground granulated blast furnace slag (GGBS), generated during iron production, has been systematically investigated for its applicability in geotechnical engineering through a series of comprehensive physical, mechanical, mineralogical, and durability assessments [22]. Moreover, owing to its high pozzolanic potential, arising from the abundance of reactive SiO₂ and Al₂O₃, GGBS can be effectively activated by alkaline and sulfate activators to yield substantial amounts of hydrated cementitious products [23,24]. Regarding potential activators, calcium carbide slag (CS), produced during acetylene gas generation from calcium carbide, has attracted significant attention due to its large availability and high Ca(OH)₂ content, making it a promising alkaline activator [25,26]. Meanwhile, titanium gypsum (TG), a powdery by-product derived from the sulfuric acid leaching of ilmenite during titanium dioxide production, is predominantly disposed of by landfilling, leading to extensive land occupation and considerable environmental impacts [27,28]. Relevant data indicate that TG production in China exceeds 20 million t annually, with cumulative stockpiles surpassing 130 million t [29]. Composed primarily of calcium sulfate dihydrate (CaSO₄·2H₂O), TG can serve as an effective sulfate activator, promoting the formation of expansive ettringite (AFt) under alkaline conditions, thereby significantly enhancing the microstructural densification of the solidified matrix [30]. Therefore, to facilitate the resource utilization of CS and TG, this study applies their synergistic activation of GGBS to develop a solidification material entirely derived from industrial by-products for treating DS with high water content.

In recent years, extensive research on solidified dredged sludge (SDS) has focused on optimizing mix designs incorporating various industrial by-products [31,32], evaluating mechanical properties [33,34], and elucidating the underlying solidification mechanisms [35,36]. These studies have demonstrated that the synergistic activation of multiple industrial by-products can be effectively leveraged to develop novel, environmentally sustainable binders capable of solidifying DS with elevated water content. However, investigations into the strength development mechanisms of DS solidified with industrial by-products have predominantly relied on analytical techniques such as scanning electron microscopy (SEM), X-ray diffraction (XRD), and Fourier transform infrared spectroscopy (FTIR) [1,9,31,37,38,39,40,41,42,43]. Although these techniques provide valuable insights into microstructural morphology and reaction products, they exhibit notable limitations in quantitatively characterizing the evolution of micropore structure. During solidification, the incorporation of curing agents into dredged sludge with high water content and large initial pore ratios induces the continuous generation of cementitious and expansive hydration products [44]. These products progressively bond soil particles and fill interparticle pores, resulting in gradual densification of the internal structure [45]. Accordingly, the strength development of solidified sludge is intrinsically associated with the evolution of its micropore structure. Nevertheless, studies that interpret strength development from the perspective of micropore structure evolution remain scarce. Furthermore, the correlation between macroscopic mechanical properties and micropore parameters has not yet been sufficiently elucidated, particularly for sludge solidified entirely with industrial by-products. To overcome the limitations of traditional microstructural characterization techniques, nuclear magnetic resonance (NMR) has recently garnered considerable attention in geotechnical engineering due to its non-destructive nature, rapid data acquisition, and capability for continuous monitoring [46,47]. Given that the transverse relaxation time (T₂) distributions furnish detailed characterization of pore size and associated pore water content [48,49], NMR has been extensively utilized to analyze geomaterial microstructures, offering significant insights into pore structure characteristics [50], freeze–thaw behavior [51], and soil–water interaction mechanisms [51,52]. Accordingly, the NMR technique was employed to investigate the temporal evolution of microstructural development in SDS by analyzing the progression of T₂ distribution curves throughout the curing process.

In summary, this study employs a novel curing agent composed of GGBS, CS, and TG (hereafter referred to as GCT) for the solidification of high water-content DS (GSDS), with conventional cement-solidified dredged sludge (CSDS) serving as a reference. On this basis, a series of unconfined compressive strength (UCS) tests, direct shear tests, and low-field nuclear magnetic resonance (LF-NMR) analyses were conducted on both GSDS and CSDS to systematically investigate the evolution of macroscopic mechanical properties and micropore structures under varying curing ages and curing agent contents. In addition, scanning electron microscopy coupled with energy-dispersive spectroscopy (SEM-EDS) was employed to elucidate the microstructural differences between GSDS and CSDS and to clarify the mechanisms governing strength development in GSDS. The results indicate that GCT exhibits superior solidification performance for high water-content DS compared with conventional cement. Moreover, quantitative relationships exist between the macroscopic mechanical properties and the micropore structure parameters of SDS.

2. Materials and Methods

2.1. Materials

The DS utilized in this study originated from the Yushan Lake dredging project located in Ma’anshan, China. The dredging site and morphological characteristics of the DS are illustrated in Figure 1, with its key physical properties detailed in

Table 1. Parameters of the test sludge.

Water Content (%)	Liquid Limit, LL (%)	Plastic Limit, LP (%)	Plasticity Index, PI	pH	Organic Content (%)	Specific Gravity, Gs	Void Ratio, e
84.20	36.46	18.88	17.58	7.13	6.18	2.68	1.672

. The natural water content was measured as 84.2%, which exceeds twice the liquid limit. In accordance with ASTM D2974-14 [53] and ASTM D4972-13 [54], the organic matter content and initial pH were determined to be 6.18% and 7.13, respectively. The particle size distribution curve, presented in Figure 2, indicates that particles finer than 0.075 mm (fine fraction) comprise approximately 88% of the total. The chemical composition presented in

Table 2. Chemical composition of raw materials/%.

	Component	SiO2	Al2O3	CaO	MgO	Fe2O3	SO3	P2O5	Other Components
Material
Dredged sludge		61.30	18.40	4.46	2.45	7.35	0.19	0.23	5.62
Ground granulated blast furnace slag		33.06	15.04	39.29	9.96	/	1.90	/	0.75
Calcium carbide slag		2.60	1.41	69.68	0.17	0.42	0.85	/	24.87
Titanium gypsum		1.53	1.00	36.54	0.10	12.40	44.85	/	3.58
Ordinary Portland cement		15.24	1.59	71.25	6.60	0.32	3.69	0.02	1.29

Note: “/” indicates not detected.

, as determined by X-ray fluorescence (XRF) analysis, identifies SiO₂ and Al₂O₃ as the predominant components.

The industrial by-product-based binder employed in this study consisted of GGBS, CS, and TG. GGBS, supplied by Henan Borun Foundry Materials Co., Ltd. (Zhengzhou, China), appears as a grayish-white powder possessing a specific surface area above 400 m²/kg. It is primarily composed of reactive SiO₂, Al₂O₃, and CaO, and is characterized by a loss on ignition of 0.8%, a glassy phase content of 99%, and a 28-day activity index of 95.5%, which collectively reflect its high reactivity and pozzolanic potential. CS, derived from an acetylene plant in Henan (Xinxiang, China), consists mainly of Ca(OH)₂, offering a strongly alkaline environment necessary for activating the ternary system. TG, sourced from Jiangsu Taibai Group Co., Ltd. (Zhenjiang, China), is a reddish-yellow powdery by-product generated from the sulfuric acid leaching of ilmenite during titanium dioxide production, primarily composed of calcium sulfate dihydrate (CaSO₄·2H₂O) and functioning as a sulfate activator. Additionally, OPC (#42.5) was utilized as a reference binder for comparison with the GCT system. The particle size distribution and chemical compositions of all materials are presented in Figure 2 and Table 2. It should be noted that, owing to variations in the types of industrial by-products and the intrinsic physical properties of dredged sludge, the optimal mixing proportions of industrial wastes may differ accordingly [55]. Therefore, mix-proportion optimization tests were first conducted for GGBS, CS, and TG, tailored to the tested sludge in this study. The optimal mass ratio of GGBS, CS, and TG was identified as 56.3:18.7:25 and was subsequently applied throughout all subsequent experiments.

2.2. Experimental Program

The experimental variables and test items are summarized in

Table 3. Experimental variables and test items.

Type of Curing Agent	Curing Agent Content (α)/(kg/m3)	Curing Age (t)/d	Test Items
			I	II	III
			UCS Test (Direct Shear Test)	LF-NMR	SEM (-EDS)
GCT/Cement	100	3	√	√
		7	√	√	√
		14	√	√
		28	√	√
GCT/Cement	150	3	√	√	√
		7	√	√	√
		14	√	√	√
		28	√	√	√
GCT/Cement	200	3	√	√
		7	√	√	√
		14	√	√
		28	√	√

Note: The content in parentheses refers exclusively to tests conducted for the GCT system.

and categorized into three investigative series. Series I investigates the macroscopic strength and deformation characteristics of GSDS and CSDS under varying curing agent content and curing ages, aiming to compare their performance and validate the solidification efficacy of GCT. Series II focuses on the temporal evolution of micropore structure at different curing agent contents and establishes quantitative correlations between micropore structure characteristics and macroscopic strength parameters. Series III examines the mineralogical composition and microstructural morphology of GSDS and CSDS under different curing conditions, thereby elucidating the microcosmic mechanism underlying the mechanical differences between GSDS and CSDS, as well as the strength development in GSDS.

2.3. Specimen Preparation and Testing Methods

(1) Sample preparation

Initially, the dried DS was manually crushed and passed through a 2 mm sieve, followed by the addition of water to adjust the water content to 100%. Subsequently, the pre-blended GCT powder was incorporated into the resulting slurry at the designated dosage and thoroughly homogenized to ensure uniform distribution. The prepared mixture was then placed into cylindrical PVC split molds of dimensions ϕ 61.8 mm × 20 mm for direct shear tests and ϕ 50 mm × 100 mm for UCS tests. Mechanical vibration was applied to eliminate entrapped air. Following molding, specimens were transferred to a curing chamber (20 ± 2 °C, relative humidity ≥ 95%), then demolded after 24 h, sealed in plastic film, and continuously cured until testing (Figure 3a).

(2) UCS tests and direct shear tests

Specimens cured to the designated age were subjected to UCS and direct shear testing (Figure 3b). UCS testing conformed to ASTM D4219-08 [56], employing a constant displacement rate of 1.0 mm/min until peak strength or 10% axial strain was attained. Direct shear tests were carried out under rapid shearing at 0.8 mm/min, applying normal stresses of 50, 100, 150, and 200 kPa. Cohesion (c) and internal friction angle (φ) were derived via linear regression analysis in accordance with GB/T 50123-2019 [57]. To verify reproducibility, three replicates were tested for each condition, with mean values reported.

(3) LF-NMR tests

LF-NMR tests were conducted on cylindrical specimens (ϕ 50 mm × 100 mm) using a MicroMR12-150 NMR analyzer (Niumag, Suzhou, China), which accommodates a maximum sample size of ϕ 60 mm × 100 mm (Figure 3c). Prior to testing, the free induction decay (FID) sequence was employed to automatically identify the center frequency and optimize the hard-pulse width for initial system calibration. Subsequently, T₂ spectra were acquired using the Carr–Purcell–Meiboom–Gill (CPMG) pulse sequence. The optimized CPMG parameters were as follows: center frequency of 12 MHz, echo spacing of 0.1 ms, 1000 echoes, and a waiting time of 1500 ms. After each measurement, the specimens were hermetically sealed and returned to the curing chamber for continued curing. To ensure reliability and repeatability, three parallel specimens were tested for each condition, and the reported results represent the average values.

(4) SEM-EDS tests

Following UCS testing, SEM-EDS was employed to characterize the hydration products and microstructural features of representative specimens (Figure 3d). Specimen cores were first immersed in anhydrous ethanol for hydration arrest and microstructure preservation, then subjected to liquid nitrogen freeze-drying. After vacuum sublimation for 48 h, dried fragments no larger than 7 mm were selected for SEM examination. Imaging was conducted utilizing a Tescan Mira3 scanning electron microscope (Tescan, Shanghai, China), with sample surfaces gold-coated prior to observation. The elemental composition (e.g., Ca, Si, Al, and S) of the identified hydration products was subsequently analyzed using an EDS detector integrated within the SEM system in point-scan mode. EDS spectra were collected at an accelerating voltage of 15 kV and a beam current of 62 μA over an energy range of 0–20 keV.

3. Results and Discussions

3.1. Unconfined Compressive Strength

Figure 4 illustrates the variation in UCS with different curing agent contents for CSDS and GSDS. UCS notably increases with rising curing agent content in both materials. At the same curing age, the UCS values for curing agent contents of 150 kg/m³ and 200 kg/m³ are approximately 2~3 times and 3~5 times greater than those at 100 kg/m³, respectively. Notably, under identical curing conditions, GSDS exhibits a markedly greater UCS compared to CSDS. To further quantitatively assess the strength enhancement of GSDS over CSDS, the UCS ratio (UCSr) is defined as follows:

$${\mathrm{UCSr} = \mathrm{q}}_{\mathrm{u}\text{-}\mathrm{GSDS}}{/\mathrm{q}}_{\mathrm{u}\text{-}\mathrm{CSDS}}$$

(1)

where q_u-GSDS and q_u-CSDS denote the UCS of GSDS and CSDS, respectively, in kPa. The variation in UCSr with curing agent content is also presented in Figure 4. It can be observed that the UCS of GSDS was approximately 5~6 times that of CSDS under comparable conditions. Moreover, at a GCT content of 200 kg/m³, the UCS of GSDS reached nearly 3.0 MPa after 28 days of curing, demonstrating the effectiveness of GCT in solidifying DS with high water content.

Figure 5 illustrates the evolution of UCS over curing time for both CSDS and GSDS. The results indicate that UCS increases logarithmically with curing age in both materials. Moreover, the rate of UCS development becomes more pronounced with higher curing agent content. Notably, compared with CSDS, GSDS demonstrates a substantially greater strength gain within the initial 7 days of curing, reaching approximately 70% of the 28-day strength. This indicates the significant early-age strength performance provided by the GCT curing agent.

In practical engineering, the deformation modulus (E₅₀), defined as the secant modulus at 50% of the failure strain, is extensively employed to characterize the deformation behavior of SDS [58]. Figure 6 illustrates the correlation between E₅₀ and UCS for GSDS and CSDS, where q_u denotes the UCS of SDS in kPa. In both materials, E₅₀ demonstrates a strong positive linear relationship with q_u, represented by E₅₀ = 95.8 q_u for GSDS and E₅₀ = 61.2 q_u for CSDS, respectively. This indicates that, compared to CSDS, GSDS generates a greater quantity of expansive AFt, resulting in a denser microstructure and significantly improved deformation resistance. Additionally, Ding et al. [59] reported E₅₀ = 55.2 q_u for SDS solidified with cement and phosphogypsum, which is comparable to that of CSDS but considerably lower than that of GSDS. This difference indicates a stronger bonding and filling effect of hydration products in GSDS, resulting in a denser internal structure and a more pronounced resistance to deformation.

3.2. Pore Structure Analysis

3.2.1. T₂ Curves

Figure 7 displays the T₂ distribution curves of CSDS and GSDS at curing agent contents of 100 kg/m³, 150 kg/m³, and 200 kg/m³ across various curing ages, where T₂ represents the transverse relaxation time (in ms). It is evident that, for both CSDS and GSDS, a distinct double-peak structure is observed in the T₂ curves at all curing ages. Moreover, the curves progressively shift towards the lower left with increasing curing agent content and curing age, accompanied by a gradual decrease in both peak intensity and integral area. This trend indicates a continuous decrease in total pore volume and an increasing proportion of smaller pores within SDS, along with enhanced transformation of pore water into chemically bound water. Notably, compared to CSDS, the integral area of the T₂ curves for GSDS decreases significantly within the first 7 days of curing, reflecting rapid pore filling among aggregated particles and significant microstructural evolution during the early curing stage. However, as the curing period extends from 7 to 28 days, morphological changes in the T₂ curves of GSDS become less pronounced, implying that hydration reactions are particularly intense at the early stage and gradually decelerate thereafter. This trend closely corresponds to the rapid early strength development of GSDS, suggesting a strong correlation between micropore structure evolution and mechanical strength development. Moreover, under identical curing ages and curing agent contents, both the integral area (i.e., total pore volume) and the T₂ value corresponding to the main peak (T_2-main-peak, i.e., the most probable pore size) of the T₂ curves for GSDS are significantly lower than those of CSDS, indicating a denser structure and a finer pore size distribution in GSDS.

Figure 8 further illustrates the evolution of the T₂ curve integral area and the T_{2-main peak} for both CSDS and GSDS with curing age, along with the corresponding ratio of GSDS to CSDS. For both materials, total pore volume and the most probable pore size all exhibit a power-law decline with increasing curing age. However, under identical curing age and curing agent content, both the total pore volume and the most probable pore size of GSDS specimens range from approximately 60% to 80% of those observed in CSDS, highlighting a more efficient macropore-filling capability in GSDS.

3.2.2. Pore Size Distribution

The relationship between transverse relaxation time T₂ and pore morphology is commonly characterized using the volume V and surface area S of the pore, as shown below [60]:

$${\displaystyle \frac{1}{T_2}}={\rho }_2{\displaystyle \frac{S}{V}}$$

(2)

where ${\rho }_2$ denotes the surface relaxivity, a critical parameter for characterizing the pore structure of cement-based materials using LF-NMR. In this study, ρ₂ was set to 0.012 μm/ms [61]. Assuming cylindrical pore geometry, an approximate linear correlation between the equivalent pore diameter (d) and T₂ was derived [48]:

$$d=4{\rho }_2{T}_2$$

(3)

Accordingly, applying Equation (3) enables transformation of T₂ distribution curves into pore size distribution curves.

Figure 9 presents the variation in pore size distribution curves for CSDS and GSDS with curing age at different curing agent contents. The pore size distributions closely align with the T₂ spectrum for both materials, exhibiting a progressive downward and leftward shift with increasing curing age and curing agent content. Specifically, with 100 kg/m³ of curing agent, the most probable pore size decreases from 0.075 μm at 3 days to 0.046 μm at 28 days for GSDS, and 0.102 μm to 0.068 μm for CSDS. Furthermore, at 14 days, increasing the GCT content from 100 kg/m³ to 200 kg/m³ reduces the most probable pore size from 0.055 μm to 0.029 μm for GSDS, and 0.080 μm to 0.043 μm for CSDS, respectively. These results suggest that the enhanced generation and development of cementitious hydration products significantly facilitate the filling of larger pores, thereby contributing to the refinement and optimization of the pore structure.

To further investigate the evolution of the micropore structure in SDS, the IUPAC pore classification system was employed to categorize pores into micropores (<2 nm), mesopores (2 nm–50 nm) and macropores (>50 nm) [62,63]. The variation in volumetric fractions of these pore types in GSDS and CSDS across curing ages is depicted in Figure 10. Notably, the volumetric fraction of micropore constitutes a minor proportion, ranging from 3% to 5%, and demonstrates little sensitivity to variations in curing age and curing agent content, whereas mesopores and macropores comprise the dominant pore volume. Moreover, the volumetric fraction of mesopores increases progressively in response to extended curing age and higher curing agent content, with a concomitant reduction in macropores. Figure 11a illustrates the correlation between mesopore volume fraction and the integral area of the T₂ curves, revealing a strong negative linear correlation for both CSDS and GSDS. This indicates that the reduction in overall pore volume within SDS primarily results from the conversion of macropores into mesopores. Furthermore, Figure 11b further presents the evolution of mesopore volume fraction over curing age, demonstrating a logarithmic growth pattern for both SDS, which closely aligns with the trend observed in UCS development.

Additionally, compared with CSDS, GSDS exhibits a significantly higher mesopore volume fraction, particularly at early curing stages. Specifically, at a curing agent content of 100 kg/m³, the mesopore volume fraction in GSDS is more than 1.5 times that of CSDS and even exceeds twice that of CSDS at a curing age of 7 days. This difference in pore structure between GSDS and CSDS mainly results from the continuous generation and accumulation of expansive AFt in GSDS, which effectively fills macropores and enhances the interparticle contact among aggregates, thereby improving overall structural integrity and facilitating strength development.

3.2.3. Relationship Between UCS and Micropore Structure

Figure 12 and Figure 13 illustrate the relationships between UCS and the integral area of T₂ curves, as well as the mesopore volume fraction, respectively. It is observed that UCS exhibits a power-law decline with increasing total pore volume for both CSDS and GSDS. With respect to mesopore volume fraction, UCS in GSDS shows a power-law increase, while a linear relationship is observed in CSDS. These results underscore that the macroscopic mechanical properties of SDS are intrinsically associated with the evolution of its micropore structure characteristics, wherein the cementation and pore-filling actions of hydration products refine the pore network and contribute to a significant enhancement in overall strength.

Overall, these findings demonstrate that GCT exhibits significant advantages over conventional cement binders in enhancing the compressive strength and filling interparticle pores in DS with high water content. Furthermore, the macroscopic mechanical properties of SDS exhibit a strong dependence on the progressive development of its microscopic pore structure.

3.3. Direct Shear Parameters of GSDS

Figure 14 shows the evolution of shear strength parameters, cohesion (c), and internal friction angle (φ) in GSDS over curing age at different GCT contents. As shown, both c and φ exhibit logarithmic increases with prolonged curing ages, which correspond with the UCS development over curing ages. Notably, the increment in cohesion with increasing GCT content and curing age is substantially more pronounced than that of the internal friction angle. Specifically, when the GCT content increases from 100 kg/m³ to 200 kg/m³, the cohesion of GSDS cured for 28 days rises from 159.40 kPa to 367.00 kPa, representing a 130% increase, while the internal friction angle increases from 29.72° to 53.02°, corresponding to a 78% increase. This behavior is attributable to the fundamental properties of GSDS as a cementitious composite structure, where strength is primarily derived from the binding and pore-filling actions of hydration products that contribute to the creation of a dense and continuous bonding network. Furthermore, when GCT content exceeds 150 kg/m³, the internal friction angle of GSDS cured for 28 days surpasses 39°, reaching the frictional characteristics of typical coarse-grained soils. This observation indicates that the substantial generation of cementitious materials under high GCT content effectively encapsulates individual soil particles into larger aggregates, thereby enhancing interparticle mechanical interlocking and progressively increasing its contribution to shear strength.

Figure 15 depicts the correlation between UCS and the shear strength parameters of GSDS. Cohesion demonstrates a strong linear relationship with UCS, supported by a high determination coefficient (R²) of 0.978, while the internal friction angle exhibits a nonlinear power-law correlation (R² = 0.940). These findings indicate that the shear strength characteristics of GSDS can be accurately inferred from UCS values using the proposed empirical formulations. This establishes a practical and efficient approach for evaluating shear strength parameters in engineering applications where direct shear testing may be impractical or time-consuming. Regarding the correlation between shear strength parameters and micropore structure, Figure 16 and Figure 17 present the relationships between c and φ with the integral area of T₂ curves and mesopore volume fractions, respectively. It can be seen that both c and φ exhibit a power-law decline with increasing total pore volume. Moreover, c shows a power-law relationship with mesopore volume fraction, while demonstrating a linear correlation.

3.4. Mechanism Analysis

3.4.1. SEM-EDS Analysis

Figure 18 presents the EDS spectra of representative regions in the microscopic morphology of GSDS. As no distinct characteristic peaks were detected at energies above 10 keV, the spectra were therefore analyzed within the energy range of 0–10 keV. Region P1 reveals a flocculent gel-like substance primarily composed of O, Ca, Si, and Al. The measured Ca/Si ratio of 1.26 indicates the formation of C-S-H gels [38]. Furthermore, the relatively high Al content, along with a Ca/(Al + Si) ratio of 0.93, implies that Al partially substitutes for Si within the C-S-H structure, leading to the generation of C-A-S-H gels [64,65]. Region P2 is characterized by the presence of needle-like crystalline products. The elemental composition in this region exhibits a mass ratio of m(Al):m(S):m(Ca) = 1:1.53:4.75, which closely corresponds to the theoretical composition of 3CaO·Al₂O₃·3CaSO₄·32H₂O, confirming the identification of the crystalline phase as AFt [66].

Figure 19 presents the microstructural morphology of CSDS and GSDS at different curing ages. As the curing period progresses, both materials demonstrate a substantial increase in the generation of cementitious products, consequently strengthening their interaction with surrounding soil particles and enhancing overall interparticle bonding. In CSDS, soil particle surfaces are encapsulated by flocculent C-(A)-S-H gels, which further aggregate to form clustered soil aggregates. Notably, acicular AFt crystals are detected in only limited quantities within CSDS, resulting in a large number of inter-aggregate pores remaining unoccupied. Consequently, the strength development and the progressive refinement of pore structure in CSDS primarily rely on the envelopment and aggregation effects of C-(A)-S-H gels. In contrast, GSDS demonstrates a distinctly denser microstructure, with soil particle surfaces extensively covered by flocculent and lamellar C-(A)-S-H gels and simultaneously exhibiting a widespread distribution of needle- and rod-shaped AFt crystals. The expansive filling effect of these AFt crystals markedly reduces inter-aggregate pore volume, promoting denser packing of interconnected soil particles and refining the internal microstructure. Therefore, GSDS displays a significantly denser pore structure compared to CSDS. Furthermore, the extensive intergrowth of AFt crystals with C-(A)-S-H gels facilitates the development of a dense three-dimensional network that robustly connects and consolidates soil aggregates. These interconnected structures provide mechanical support, enabling GSDS to achieve higher strength at early curing stages.

Figure 20 presents the microstructure of GSDS at varying curing ages and GCT contents. In conjunction with Figure 19c,d, the microstructural evolution of GSDS can be further analyzed. As illustrated in Figure 19c,d and Figure 20a,b, a substantial number of fine, needle-like AFt crystals are already observed at the 3-day curing age. With prolonged curing, these AFt crystals continue to grow and undergo a morphological transformation from slender, needle-like forms to thicker, columnar structures, thereby enhancing their pore-filling capacity. A comparison of Figure 19c and Figure 20c,d reveals a marked enhancement in the generation of both C-(A)-S-H gels and AFt crystals within the GSDS with increasing GCT content, accompanied by the development of a more continuous interparticle bonding network. These observations suggest that the primary characteristic of the GCT binder is to promote the formation of AFt crystals, which enhances pore filling within soil aggregates and reinforces interparticle bonding in conjunction with C-(A)-S-H, thereby substantially improving the overall structural integrity of GSDS.

3.4.2. Strength Development Mechanism of GSDS

Based on the preceding analysis, the strength development mechanism of DS solidified with GGBS-CS-TG is illustrated in Figure 21 and described by Equations (4)–(10) [44]. Prior to GCT binder addition, sludge particles are coated with a thick adsorbed water film that induces large interparticle pores and weakens structural cohesion, resulting in flowable, plastic behavior. Upon dissolution of GCT components in pore water, a series of physicochemical reactions is triggered among GGBS, CS, TG, water, and soil particles. The solidification process proceeds through three primary mechanisms: ion exchange, cementitious reinforcement, and pore filling. Initially, CS and TG dissolve, releasing Ca²⁺, OH⁻, and SO4²⁻ ions into the pore solution. Released Ca²⁺ then induces ion exchange with surface-absorbed Na⁺ and K⁺, resulting in double-layer compression and enhanced flocculation through reduced interparticle repulsion [67]. Simultaneously, under alkaline conditions, reactive components in GGBS are activated, releasing Ca²⁺, Si⁴⁺, and Al³⁺ ions. The liberated Si⁴⁺ and Al³⁺ subsequently react with OH⁻ and H₂O to form tetrahedral [H₃SiO₄]⁻ and [H₃AlO₄]²⁻, as well as octahedral [Al(OH)₆]³⁻. Thereafter, the tetrahedral [H₃SiO₄]⁻ and [H₃AlO₄]²⁻ then combine with Ca²⁺ to form a C-(A)-S-H gels, whereas the octahedral [Al(OH)₆]³⁻ react with Ca²⁺, SO₄²⁻, and H₂O to generate AFt, as described in Equations (6)–(8) [44]. Additionally, Ca²⁺ and OH⁻ ions in the pore solution carbonate with CO₂, leading to CaCO₃ precipitation. [68]. The ion exchange facilitated by Ca²⁺ effectively compresses the electrical double layer, thereby promoting particle aggregation. The resulting C-(A)-S-H gels and CaCO₃ function as binding matrices that tightly encapsulate and interconnect soil particles, resulting in an interwoven microstructure [68]. Meanwhile, the generation of AFt effectively occupies larger interparticle pores within the flocculated aggregates, significantly reducing porosity and reinforcing interparticle bonding [42]. Concurrently, the intergrowth synergy between C-(A)-S-H gels and AFt further promotes the establishment of a more compact and stable matrix structure. Moreover, the substantial consumption of free water by C-(A)-S-H gels and AFt formation enhances the efficacy of the GCT binder, particularly in the solidification of DS with elevated water content. In summary, these hydration processes effectively enhance particle aggregation, interparticle bonding, and pore filling within soil aggregates, thereby promoting the densification of the DS microstructure into a compact, interconnected three-dimensional network and substantially improving the mechanical strength of GSDS.

$${\mathrm{SiO}}_2+{\mathrm{OH}}^{-}+{\mathrm{H}}_2\mathrm{O}\to {{[\mathrm{H}}_3{\mathrm{SiO}}_4]}^{-}$$

(4)

$${\mathrm{Al}}_2{\mathrm{O}}_3{+\mathrm{OH}}^{-}{+\mathrm{H}}_2\mathrm{O}\to {[\mathrm{Al}{\left(\mathrm{OH}\right)}_4]}^{-}, [\mathrm{Al}{\left(\mathrm{OH}\right)}_6{]}^{3^{-}}$$

(5)

$${\mathrm{Ca}}^{2+}{+2\mathrm{OH}}^{-}\to \mathrm{Ca}{\left(\mathrm{OH}\right)}_2$$

(6)

$$x{\mathrm{Ca}}^{2+}+y{{[\mathrm{H}}_3{\mathrm{SiO}}_4]}^{-}+\left(z-x-y\right){\mathrm{H}}_2\mathrm{O}+\left(2x-y\right){\mathrm{OH}}^{-}\to \mathrm{C}x-\mathrm{S}y-\mathrm{H}z (\mathrm{C}\text{-}\mathrm{S}\text{-}\mathrm{H} \mathrm{gel})$$

(7)

$$\mathrm{C}\text{-}\mathrm{S}\text{-}\mathrm{H} \mathrm{gel}+[\mathrm{Al}{\left(\mathrm{OH}\right)}_4{]}^{-}\to \mathrm{C}\text{-}\mathrm{A}\text{-}\mathrm{S}\text{-}\mathrm{H} \mathrm{gel}$$

(8)

$${6\mathrm{Ca}}^{2+}{{+3\mathrm{SO}}_4}^{2-}+2 [\mathrm{Al}{\left(\mathrm{OH}\right)}_6{]}^{3-}{+26\mathrm{H}}_2\mathrm{O}\to 3\mathrm{CaO}·{\mathrm{Al}}_2{\mathrm{O}}_3·{3\mathrm{CaSO}}_4·{32\mathrm{H}}_2\mathrm{O}$$

(9)

$$\mathrm{Ca}{\left(\mathrm{OH}\right)}_2{+\mathrm{CO}}_2\to {\mathrm{CaCO}}_3{+\mathrm{H}}_2\mathrm{O}$$

(10)

4. Conclusions

In this study, a series of UCS tests, LF-NNR analyses, direct shear tests, and SEM-EDS observations were conducted to investigate the strength development, micropore structure evolution, hydration products, and microstructural morphology of DS solidified with both cement and an entirely industrial by-product-based curing agent (GCT), composed of GGBS, CS, and TG. Based on these investigations, the applicability and solidification advantages of GCT for high water-content DS were demonstrated, quantitative correlations linking macroscopic mechanical behavior and microscopic pore characteristics were identified, and the microstructural differences between GSDS and CSDS, as well as the mechanisms governing strength development in GSDS, were systematically analyzed. The following conclusions were drawn:

• For both CSDS and GSDS, UCS and mesopore volume fraction exhibit logarithmic growth with increasing curing age. In contrast, the total pore volume and the most probable pore size decrease following a power-law relationship. The decline in pore volume primarily results from the transformation of macropores into mesopores, while the micropore volume fraction remains relatively stable at approximately 3~5%, showing minimal variation with changes in curing age and curing agent content. Compared to CSDS, the UCS of GSDS is approximately 5~6 times higher under comparable conditions, along with significantly enhanced deformation resistance. In addition, GSDS exhibits significantly greater early strength development and a more rapid decrease in pore volume within the first 7 days, with its 7-day UCS reaching approximately 70% of the 28-day value. Furthermore, GSDS presents a markedly higher mesopore volume fraction than CSDS, demonstrating superior structural integrity and overall strength development.
• Both CSDS and GSDS exhibit a power-law decline in UCS with increasing total pore volume. Regarding mesopore volume fraction, UCS in GSDS increases following a power-law trend, while a linear increase is observed in CSDS. Additionally, in GSDS, both c and φ increase logarithmically with curing age. A strong linear correlation exists between c and UCS, whereas φ exhibits a power-law increase with UCS. Furthermore, both c and φ decrease with increasing total pore volume according to power-law relationships. As mesopore volume fraction increases, c increases in a power-law manner, whereas φ increases linearly. Macroscopic mechanical performance is strongly correlated with the evolution of microscopic pore structure.
• In CSDS, soil particle surfaces are encapsulated by flocculent C-(A)-S-H gels, while substantial inter-aggregate pores remain unfilled. Strength development and pore structure refinement in CSDS primarily result from the envelopment and aggregation effects of C-(A)-S-H gels. In GSDS, GGBS is effectively activated under the combined action of CS and TG, leading to the formation of C-(A)-S-H gels and abundant expansive AFt crystals that immobilize substantial amounts of free water. The synergistic solidification mechanism in GSDS involves ion exchange, cementitious bonding, and pore filling, which effectively enhance particle aggregation, strengthen interparticle bonding, and refine the pore structure, thereby significantly improving the structural integrity and macroscopic strength of GSDS.