Flocculation Performance and Interfacial Adsorption Mechanism of Aluminum Hydroxide–Polyacrylamide in Coal Slime Water Treatment

Abstract

Effective treatment of coal slime water is essential for sustainable coal preparation plant operation but hindered by the stable suspension of fine, negatively charged particles. To address this, a novel star-shaped inorganic–organic hybrid polymer (aluminum hydroxide-polyacrylamide, Al-PAM) was synthesized via in situ polymerization. Its performance was systematically compared with well-established coagulants/flocculants—polyaluminum chloride (PAC), non-ionic polyacrylamide (NPAM), and their binary combination through settling tests and quartz crystal microbalance with dissipation monitoring (QCM-D). The results showed a positive correlation between the molecular weight of Al-PAM and its flocculation efficiency. The optimal variant, Al-PAM-442, achieved an exceptionally high initial settling rate (50.4 m/h) and low supernatant turbidity (45.77 NTU) at an ultralow dosage of 6 mg/L. QCM-D analysis elucidated the mechanism: Al-PAM forms a thick, soft, and irreversibly adsorbed hydrated layer on silica, enabling strong electrostatic anchoring and effective polymer bridging. In contrast, PAC adsorption was reversible, while NPAM formed a thin, compact film with poor bridging capacity. Although the combined PAC/NPAM system showed synergistic performance, it required a significantly higher dosage (70 mg/L). This study demonstrates that the star-shaped Al-PAM architecture successfully integrates charge neutralization and bridging into a single molecule, offering a highly efficient and practical solution for industrial coal slurry dewatering.

1. Introduction

The efficient treatment of coal slime water, a highly stable colloidal suspension generated in large volumes during wet coal preparation, is paramount for ensuring water recycling, minimizing environmental footprint, and maintaining the economic viability of coal processing plants [1]. The key difficulty lies in its composition: a high ash content (25–75%) and a predominance of fine particles (50–85% smaller than 0.045 mm) [2,3,4,5]. These ultrafine particles are primarily negatively charged clay minerals like kaolinite and montmorillonite, which hinder aggregation and sustain suspension stability [6]. Beyond operational challenges, improper disposal of these slimes poses serious environmental risks, such as soil contamination and heavy metal mobilization [7]. Thus, rather than being a terminal waste, coal slime sludge can be regarded as a potential resource. Efficient dewatering is key to enabling its conversion into a fuel for co-combustion, advancing waste-to-energy recovery and circular economy principles within the industry. Therefore, developing highly effective, cost-efficient, and environmentally sound treatment technologies remains a critical industrial and research priority.

Among various physicochemical methods, flocculation using synthetic polymers, particularly polyacrylamide (PAM) and its derivatives, is the most prevalent industrial practice due to its effectiveness and scalability [8]. The performance of these flocculants hinges on their ability to destabilize particles through mechanisms like charge neutralization and polymer bridging. For instance, cationic PAM (CPAM) has demonstrated superior efficiency over non-ionic (NPAM) and anionic (APAM) variants in treating clay-rich slimes, primarily by effectively neutralizing the negative surface charge of minerals. Despite this, conventional PAMs exhibit limitations, including sensitivity to water chemistry, inefficient adsorption on specific mineral faces, and the risk of overdosing which can lead to re-stabilization [9]. These shortcomings have spurred the development of advanced flocculants, with a clear trend towards designing hybrid or grafted copolymers that integrate multiple functional groups to synergize different flocculation mechanisms.

A promising recent innovation is the inorganic–organic hybrid flocculant aluminum hydroxide–polyacrylamide (Al-PAM). Characterized by a unique star-shaped architecture—with PAM chains radiating from a central Al(OH)₃ core, as revealed by AFM [10]—this structure endows Al-PAM with dual functionality. The cationic core provides sites for electrostatic charge neutralization, while the extended PAM arms enable effective inter-particle bridging (see the schematic model proposed by Alagha et al. [11] and a conceptual illustration in the Graphical Abstract of this work). The more open conformation of such branched polymers is theoretically favorable for capturing fine particles [12]. Preliminary applications, notably in treating oil sands tailings (a complex wastewater containing 30–35% negatively charged clays [13]), demonstrate its effectiveness [14]. Studies show Al-PAM achieves rapid flocculation and yields clearer supernatant (e.g., 68 NTU vs. 470 NTU with a very effective commercial polymer–Magnafloc 1011) [15], also serving as a potent filtration aid [16]. Despite its proven efficacy in clay-rich suspensions like oil sands tailings, a significant knowledge gap exists regarding its application to coal slime water. More critically, current research has focused on synthesizing novel structures and evaluating their macro-scale performance, leaving a fundamental lack of understanding about how the star-shaped structure governs dynamic interfacial behavior at the solid–liquid interface. Key questions concerning its adsorption kinetics, adsorbed-layer conformation, and core-arm synergy remain unanswered, creating a disconnect between its structure and flocculation performance.

This lack of fundamental insight presents a major bottleneck for the rational design of next-generation flocculants. As noted in recent reviews, theoretical understanding of flocculation mechanisms lags behind applied process development, and advanced tools are needed to probe these interfacial phenomena [17,18]. Quartz crystal microbalance with dissipation monitoring (QCM-D) is a powerful technique that addresses this need by providing real-time, in situ measurements of adsorbed mass and, crucially, the viscoelastic properties (softness/stiffness) of the adsorbed polymer layer—a direct indicator of its conformation and bridging potential [19,20,21].

To bridge the critical gap between macroscopic performance and nanoscale mechanisms for Al-PAM, this study tests the hypothesis that its superior efficacy originates from a unique adsorbed layer structure. We synthesized a series of Al-PAM polymers with varying molecular weight (MW) to conduct a comparative investigation with three main aims: (1) to evaluate the coagulation–flocculation and sedimentation performance of Al-PAM in coal slime water against benchmark reagents (polyaluminum chloride (PAC), NPAM, and their combination); (2) to employ QCM-D for in situ characterization and comparison of the adsorption kinetics and viscoelastic properties of the layers formed by these agents on a model coal slime surface; and (3) to establish a coherent mechanism linking Al-PAM’s molecular structure and interfacial behavior to its macroscopic coagulation–flocculation and sedimentation efficiency. It is anticipated that this work will elucidate the fundamental interfacial mechanisms governing Al-PAM’s effectiveness, thereby providing a theoretical basis and a molecular design strategy for developing high-efficiency flocculants for coal slime water treatment.

2. Materials and Methods

2.1. Preparation of Coal Slime Water

The experimental sample was prepared using raw coal provided by Zhengtong Coal Preparation Plant, Boxuan Company, Northwest Mining Co., Ltd. (Xi’an, China) Raw coal (with an ash content of 17.07%) and gangue from the same plant (with an ash content of 85.69%). Both materials were crushed to below 1 mm using a jaw crusher (KER-MFEP 100 × 60, Zhenjiang Kerui Sample Preparation Equipment Co., Ltd., Zhenjiang, China) and a roll crusher (KER-MF 200 × 125, Zhenjiang Kerui Sample Preparation Equipment Co., Ltd., Zhenjiang, China), respectively. With reference to the typical ash content (42.5%) of the coal slime at Zhengtong Coal Preparation Plant, the crushed raw coal and gangue were blended in a 2:1 ratio to produce a sample with an ash content of 40.43% (determined via proximate analysis, see

Table 1. Proximate analysis results of the coal slime samples.

Proximate Analysis Item 1	Mad	Aad	Vad	FCad
Mass Percentage/%	1.47	40.43	20.95	37.15

¹ Proximate analysis parameters (on an air-dried basis) are abbreviated as follows: M_ad, Moisture content (air-dried basis); A_ad, Ash content (air-dried basis); V_ad, Volatile matter (air-dried basis); FC_ad, Fixed carbon (air-dried basis).

). This prepared sample was used to simulate the plant’s coal slime water and prepare a 30 g/L suspension for settling tests.

2.2. Synthesis of Al(OH)₃-polyacrylamide

2.2.1. Materials

Acrylamide (AR, 99.0%), ammonium carbonate (AR, 30% NH₃ basis), anhydrous aluminum chloride (99%), sodium bisulfite (99.99% metals basis), and ammonium persulfate (AR, ≥98%) for the synthesis of the inorganic–organic hybrid polymer Al-PAM were purchased from Shanghai Aladdin Biochemical Technology Co., Ltd (Shanghai, China). All organic solvents were supplied by Tianjin Fuyu Fine Chemical Co., Ltd (Tianjin, China). The NPAM (5 × 10⁴ Da) and sodium dodecyl sulfate (SDS, ≥98.0%) were acquired from Shanghai Macklin Biochemical Technology Co., Ltd. (Shanghai, China). PAC (analytical reagent grade) was obtained from Tianjin Damao Chemical Reagent Factory (Tianjin, China).

2.2.2. Synthesis of Al(OH)₃ Colloid

The Al(OH)₃ colloid was prepared at room temperature by titrating 36 g of a 0.1 M (NH₄)₂CO₃ solution (0.5 g/min) into 25 g of a 0.1 M AlCl₃ solution with stirring (500 rpm), followed by continued stirring at 300 rpm for 1 h. The obtained Al(OH)₃ colloid, exhibiting a clear Tyndall effect, was used fresh on the same day for Al-PAM synthesis.

2.2.3. Synthesis of Al-PAM

A mixture of 25.5 mL of the freshly prepared Al(OH)₃ colloid suspension and 4.5 g of acrylamide were added into a 100 mL three-neck flask. Under a N₂ purge (to remove oxygen) and light protection, the mixture was stirred (250 rpm) and heated to 40 °C in a water bath for 30 min. Subsequently, 2 mL of a mixed initiator solution containing (NH₄)₂S₂O₈ (1 g/L) and NaHSO₃ (0.5 g/L) was added dropwise to the reaction flask using a micro peristaltic pump at a rate of 0.13 g/min. After 10 min, the N₂ purging and stirring were ceased, the flask was sealed, and the polymerization was allowed to proceed at 40 °C for 6 h, yielding a transparent polymer gel.

2.2.4. Purification and Drying of Al-PAM

The polymer gel was purified via a dissolution-reprecipitation process to remove unreacted monomers and initiators. The prepared polymer gel was diluted with ultrapure water to 10 wt.% and stirred at 300 rpm until the polymer dissolved, forming a colloidal solution. This solution was then precipitated into a 6:4 (v/v) ethanol/water mixture under continuous stirring. The resulting precipitate was isolated, repeatedly washed with the same ethanol/water mixture, and further rinsed with acetone to eliminate impurities such as unreacted acrylamide and aluminum hydroxide. Finally, the purified solid was dried under vacuum at 55 °C for 8 h to constant weight.

Motivated by the reported positive correlation between the performance of a flocculant and its MW [22], we synthesized Al-PAM polymers with varied MWs by adjusting the concentration of the oxidation/reduction initiator system, specifically to investigate their effect on coal slime water treatment. The MW of the polymers were determined using the viscometry method [23], and the pH of each Al-PAM sample was measured in its 1000 ppm aqueous solution, with the results summarized in

Table 2. Synthesis and property parameters of Al-PAM.

Polymer	(NH4)2S2O8 (g/L)	NaHSO3 (g/L)	Intrinsic Viscosity (mL/g)	Relative Molecular Weight (104 Da)	pH
Al-PAM-264	1.0	0.5	862.94	264	6.87
Al-PAM-370	0.5	0.25	1131.97	370	6.76
Al-PAM-442	0.125	0.0625	1305.08	442	6.56

. Different samples are distinguished by the suffix Al-PAM-X, where X denotes the MW (in units of 10⁴ Da).

2.3. Settling Test

Settling tests were conducted on 30 g/L coal slime suspensions. With stock solutions of PAC at 5000 ppm, and Al-PAM-x/NPAM at 1000 ppm, the settling test procedure was: stirring the coal slime suspension (700 rpm) using a magnetic stirrer (MS-H280-Pro, Dalong Xingchuang Experimental Instrument Co., Ltd., Beijing, China) for 5 min, transferred the suspension to a graduated 500 mL settling cylinder and inverted five times, adding a specified dosage of polymer and inverting another five times (this step was repeated if two chemicals were added), then initiating timed settling. The position of the mud-line was recorded over time. The settling curve was obtained by plotting the normalized mud-line height (h/H, where h is the mud-line height at settling time t, and H is the initial height of the coal slime water in the cylinder) against the settling time. The initial slope of this curve was taken as the initial settling rate (ISR). After 10 min, the sediment thickness was recorded, and the turbidity of 30 mL supernatant extracted from a position 10 cm below the water surface was measured with a WZS-188 turbidimeter (WZS-188, Shanghai Leici Instrument Co., Ltd., Shanghai, China).

2.4. Zeta Potential (ζ) Measurement

Zeta potential measurements were conducted using a Malvern ZEN3690 analyzer (Malvern Panalytical Ltd., Malvern, UK) to assess the colloidal stability of the system. Directly after the 10 min settling test, a sample of the supernatant was collected and analyzed without further dilution. Each measurement was performed at 25 °C and repeated in triplicate.

2.5. Measurements of Polymer Adsorption by QCM-D

The adsorption behavior of the polymers was investigated using a quartz crystal microbalance with dissipation monitoring (QCM-D; Q-Sense Explorer, Biolin Scientific, Gothenburg, Sweden). The measurement principle is based on the piezoelectric effect of the quartz crystal sensor. Mass adsorption on the crystal surface leads to a shift in its resonant frequency (f) [17]. Generally, a larger negative frequency shift (∆f) indicates a higher adsorbed mass. For a thin, rigid, and uniform adsorbed layer, the frequency shift is proportional to the mass change (∆m), following a linear relationship [24]. However, this linear relationship between ∆f and ∆m no longer holds for soft, viscoelastic adsorbed layers in a liquid environment. Therefore, the dissipation factor (D) is introduced to characterize the physical properties of the adsorbed layer. D is defined as the ratio of the energy dissipated to the energy stored during one oscillation cycle. The dissipation shift (ΔD) depends on the viscoelasticity of the adsorbed layer; a larger ΔD indicates a looser and softer structure [25]. The QCM-D measurements were performed using AT-cut quartz crystal sensors (14 mm diameter, 5 MHz fundamental shear oscillation frequency) coated with silicon dioxide (SiO₂) to simulate the silica surface of coal slime. All experiments were conducted at 20.0 ± 0.2 °C.

• Sensor Preparation: Prior to each measurement, the sensor was cleaned by sequential immersion in a 2 wt.% SDS solution for 30 min, thorough rinsing with ultrapure water, drying under a nitrogen stream, and finally a 20 min UV-ozone treatment.
• Solution Preparation: The concentrations of PAC were fixed at 5000 ppm for QCM-D experiments. Considering that a highly viscous or dense medium on the sensor chip can lead to excessive damping and compromise measurement accuracy, the concentrations of NPAM and Al-PAM were set to a lower level of 100 ppm. Ultrapure water was used as the background solution.
• Measurement Protocol: The background solution was first flowed through the system at 0.15 mL/min until stable baselines for frequency (f) and dissipation (D) were established. Polymer solutions and subsequent rinsing water were then introduced at the same flow rate. Frequency and dissipation shifts were monitored in real-time at 5, 15, 25, 35, 45, 55, and 65 MHz, respectively.
• Data Analysis: The frequency and dissipation shifts from the 3rd, 5th, and 7th overtones were primarily utilized for data analysis due to their optimal sensitivity and stability. These multi-overtone datasets were fitted using the Viscoelastic model in the Dfind software (version 2.2) to determine the adsorbed mass and viscoelastic properties. For consistency in reporting, all Δf and ΔD values presented in this study correspond to the changes measured at the 3rd overtone.

3. Results and Discussion

3.1. Characterization of Coal Slime Water and Verification of Synthesized Al-PAM

Characterization of the raw coal slime (

Table 3. Size fraction and ash distribution of the coal slime samples.

Size Fraction 1/mm	Yield /%	Ash Content /%	Cumulative Yield/%	Cumulative Ash Content/%
+0.125	7.14	46.14	7.14	46.14
−0.125 + 0.074	18.47	39.77	25.61	41.54
−0.074 + 0.045	11.90	39.99	37.51	41.05
−0.045	62.49	38.28	100.00	39.32

¹ Note: The “+” and “−” signs denote oversize and undersize fractions relative to the stated sieve sizes, respectively.

) revealed a high content of fine particles, with a particle size fraction below 0.045 mm, constituting 62.49% of the total. This fine size distribution presents a significant challenge for solid–liquid separation due to the high specific surface area and colloidal stability of the suspension.

The mineral composition of the coal slime was determined using an X-ray diffractometer (D8 ADVANCE, Bruker AXS GMBH, Karlsruhe, Germany), as shown in Figure 1. The main mineral components of the coal slime are quartz and kaolinite. The strongest peak at 26.64° corresponds to the quartz structure, while the strongest peaks at 12.23° and 24.73° are attributed to the kaolinite structure. Given that quartz (SiO₂) is a primary component and that the siloxane basal surfaces of kaolinite are also siloxane- or silanol-dominated, a silica sensor was therefore employed in QCM-D measurements. This provides a representative and well-defined model surface of coal slime particles, enabling the systematic study of flocculant adsorption mechanisms under controlled conditions.

Concurrently, the successful synthesis of the target inorganic–organic hybrid flocculant, Al-PAM, was confirmed. FT-IR spectroscopy (performed using a Fourier transform infrared spectrometer, Nicolet iN10, Thermo Fisher Scientific Inc., Waltham, MA, USA) of Al-PAM-442 (Figure 2) revealed distinct absorption bands at 481 and 765 cm⁻¹, corresponding to Al-O stretching vibrations, and a band at 1086 cm⁻¹ associated with hydroxyl bending. These features, absent in pure PAM, provide direct evidence for the incorporation of Al(OH)₃ colloids into the polymer matrix. Furthermore, the hybrid polymer exhibited a positive zeta potential of +0.62 mV in aqueous solution, confirming its cationic nature. This net positive charge is a critical design outcome, equipping Al-PAM with the inherent capability for electrostatic attraction with the negatively charged mineral surfaces in the coal slime, thereby integrating the charge-neutralization function directly into its molecular structure.

3.2. Settling Performance of Al-PAM with Different Molecular Weights

3.2.1. Dosage Effect of Al-PAM-264

Al-PAM was synthesized according to a previously established method, and its MW was determined to be 2.64 × 10⁶ Da [26]. Its flocculation performance for coal slurry water was first evaluated by assessing the settling behavior across a dosage range of 6–26 mg/L (Figure 3). Visual observation after 10 min of settling (Figure 3a) revealed a distinct trend in supernatant clarity. Turbidity initially decreased with increasing Al-PAM dosage, reaching a minimum of 156.1 NTU at 10 mg/L, before increasing again at higher dosages. The optimal dosage of 10 mg/L effectively reduced the system’s zeta potential from −26.87 mV to −4.59 mV. This demonstrates that for Al-PAM, in addition to the conventional adsorption and bridging, charge neutralization serves as a crucial flocculation mechanism owing to its cationic inorganic–organic hybrid structure.

The sedimentation kinetics were further investigated by tracking the mud line height over time (Figure 3b). For any given dosage, the mud line descended rapidly during the first 25 s before plateauing, signifying the transition from initial floc formation to a stable compression phase. Increasing Al-PAM-264 dosage from 6 to 10 mg/L thickened the sediment layer, but a further increase to 14 mg/L slightly reduced it. Beyond this, additional dosage had little impact on sediment thickness.

After 10 min of settling, when the system had stabilized, key parameters including the ISR, final sediment thickness, and supernatant turbidity were measured for different Al-PAM-264 dosages (Figure 3c). The results demonstrate a non-linear dose–response relationship: increasing the dosage from 6 to 18 mg/L progressively enhanced the ISR, which peaked at 40.68 m/h, and reduced the supernatant turbidity to a minimum of 156.1 NTU at 10 mg/L. However, beyond this optimal range, a dosage of 22 mg/L caused the ISR to decline, the turbidity to rise sharply, and the sediment layer to thicken. This deterioration in performance indicates overdosing. The concurrent increase in turbidity and sediment layer thickening at 22 mg/L, despite the zeta potential remaining negative (−3.03 mV), suggests a mechanism distinct from simple charge reversal. Instead, it is consistent with polymer-induced restabilization. We propose that beyond the optimal dosage, particle surfaces become saturated with adsorbed polymer. This surface saturation likely reduces the availability of functional sites for inter-particle bridging and may introduce steric repulsion, leading to the formation of weaker, more hydrated flocs that entrap water, are susceptible to breakage, and release fine particles—thereby increasing the supernatant turbidity. This delineates a precise optimal dosage window and underscores that exceeding it compromises the goal of low chemical consumption without sacrificing performance. The detailed interfacial mechanism of this overdosing phenomenon is investigated via subsequent QCM-D study. Collectively, Figure 3a demonstrates that excessive Al-PAM dosage leads to polymer-induced restabilization, resulting in a turbid supernatant with a diffuse mud line and inferior settling performance.

3.2.2. Impact of Molecular Weight

To investigate the effect of MW, Al-PAM-370 and Al-PAM-442 were synthesized. As shown in Figure 4a–c, the treatment performance improved significantly with increasing MW. The ISR increased consistently, with Al-PAM-442 achieving a maximum of 50.4 m/h, substantially surpassing Al-PAM-370 (41.4 m/h) and Al-PAM-264 (40.68 m/h). Critically, this enhanced settling efficiency did not compromise supernatant clarity. The turbidity of the clarified layer after 10 min of settling (Figure 4d) for all Al-PAM samples first decreased and then increased with increasing dosage. Al-PAM-370 reached its minimum supernatant turbidity of 107.6 NTU at 10 mg/L, already lower than the 156.1 NTU achieved by Al-PAM-264 at its optimal dose. Remarkably, Al-PAM-442 achieved its lowest turbidity of 45.77 NTU at a dose of only 6 mg/L, demonstrating a far more efficient capture of fine particles.

The superior performance of higher MW Al-PAM stems from the synergistic interplay of its extended polymer chains and unique hybrid architecture. The longer chains significantly enhance the “adsorption and bridging” capability, forming larger, denser flocs for rapid settling while effectively capturing fine particles to lower turbidity. This effect is further amplified by the star-like architecture of Al-PAM, where the inorganic Al(OH)₃ core with radiating PAM arms provides a larger spatial footprint and more anchoring sites for efficient multi-point particle capture [27]. Crucially, the charge-neutralizing function of the positively charged Al(OH)₃ core is quantitatively substantiated by zeta potential measurements: the potential of the coal slime water shifted from −26.87 mV to approximately −3 mV upon treatment, effectively destabilizing the suspension and promoting initial aggregation. Collectively, the robust bridging of long chains, the multi-point contact enabled by the star-like geometry, and the electrostatic attraction via charge neutralization work in concert to achieve the highest flocculation efficiency.

Because a nonlinear dependence of the key flocculation-settling parameters on dosage was observed for all three Al-PAM polymers, the comprehensive settling index (CSI, Equation (1)) was employed to evaluate this relationship quantitatively and determine the optimal dosage, as plotted in Figure 4e.

$$CSI={\displaystyle \frac{v\times \left(1-\raisebox{1ex}{$h$}\!\left/ \!\raisebox{-1ex}{$H$}\right.\right)\times 100\%}{T_s}}$$

(1)

where $v$ is the settling rate and $T_s$ is the turbidity of the supernatant.

As shown in Figure 4e, this trend was consistent across all polymers, with optimal performance at 10 mg/L. Given its superior performance, Al-PAM-442 at this dosage was selected for comparative analysis against other agents for coal slime water treatment.

3.3. Settling Performance Comparison of Coagulants and Flocculants

3.3.1. Settling Performance of PAC and NPAM

The flocculation performance of the synthesized inorganic–organic hybrid Al-PAM was critically evaluated by benchmarking it against two key agents: PAC (a coagulant, also a precursor for Al-PAM synthesis) and NPAM (a non-ionic flocculant). This approach allowed for a direct assessment of the enhancements afforded by the hybrid structure. All experiments were conducted with 30 g/L of coal slime water.

Settling performance of PAC. The effect of the coagulant PAC was investigated using a 0.5 wt.% stock solution. Since a clear mud line was difficult to distinguish with PAC alone, supernatant turbidity served as the primary evaluation metric. As shown in Figure 5a, turbidity decreased and then increased with PAC dosage, reaching a minimum of 151.3 NTU at 60 mg/L. This trend is explained by the dual role of PAC hydrolysis products (polymeric hydroxyl complexes) [28], which neutralize surface charge and promote particle aggregation via adsorption and bridging. Zeta potential measurements verified the charge neutralization mechanism: the potential decreased from −26.87 mV (untreated) to −19.67 mV at 20 mg/L, indicating partial charge neutralization and high residual turbidity. At the optimal 60 mg/L dosage, the potential reached −5.99 mV, suggesting near-complete charge neutralization and minimized electrostatic repulsion, resulting in the lowest turbidity. Further dosage increase to 120 mg/L caused charge reversal to +1.43 mV, re-stabilizing the suspension through electrostatic repulsion between now positively charged particles and sharply increasing turbidity. This classic over-neutralization and charge reversal phenomenon defines the optimal dosage for charge-neutralizing coagulants. Consequently, any dosage beyond this optimum is counterproductive, wasting chemicals while actively impairing clarification.

Settling performance of NPAM. A non-ionic PAM (NPAM-500) with molecular-weight of 5.0 × 106 Da was examined for comparison with Al-PAM-442 (4.42 × 106 Da), using a 0.1 wt.% stock solution. Figure 5b shows photographs of coal slime water treated with 6, 14, and 22 mg/L NPAM-500 after 10 min of settling. The supernatant remained highly turbid with no clear mud line formation, indicating inadequate flocculation. Consequently, quantitative turbidity analysis was not performed, underscoring the limited efficacy of NPAM alone under these conditions.

3.3.2. Binary PAC/NPAM System versus Al-PAM-442

Based on the above finding that neither PAC nor NPAM alone could compete with the performance of Al-PAM, the settling efficacy of a combined PAC/NPAM system was investigated to determine if it could surpass the synthesized inorganic–organic hybrid polymer. According to prior single-factor tests (Section 3.3.1), a fixed optimal PAC dosage of 60 mg/L was used in combination with varying dosages of NPAM-500.

The results presented in Figure 6 demonstrate a significant enhancement when NPAM-500 is used in conjunction with PAC. Compared to NPAM alone (Figure 5b), the combined system produced a clear mud line, with supernatant turbidity decreasing progressively as the NPAM-500 dosage increased (Figure 6a,b). This superior settling performance, which was particularly evident at NPAM-500 dosages from 10 to 26 mg/L, underscores the powerful synergy effect between the two agents. The mechanism aligns with classical coagulation–flocculation theory: the high charge density of PAC effectively neutralizes and destabilizes fine particles, creating ideal conditions for the long chains of NPAM-500 to bridge them into large, settleable flocs, thereby yielding a clearer supernatant. Notably, at an NPAM-500 dosage of 22 mg/L (with 60 mg/L PAC), the system achieved an excellent turbidity of 12.39 NTU. In terms of settling rate, the ISR of the PAC/NPAM system also increased with NPAM-500 dosage, reaching 52.92 m/h (Figure 6c). However, a critical comparison with Al-PAM-442 reveals a distinct advantage in dosage efficiency for the hybrid flocculant. Despite its lower MW, Al-PAM-442 alone yielded a higher ISR than the PAC/NPAM combination only when overdosed at 26 mg/L. More importantly, Al-PAM-442 achieved a comparable supernatant turbidity (45.77 NTU) at a mere 6 mg/L, whereas the PAC/NPAM system required a total dosage of 70 mg/L to reach a similar clarity. In addition, the CSI for Al-PAM-442 at 6 mg/L (140.8) was also higher than that of the PAC/NPAM system at 70 mg/L (116.72).

Therefore, while the combination of PAC and NPAM-500 can achieve exceptionally low supernatant turbidity due to the powerful synergy of charge neutralization and bridging, this comes at the cost of a significantly higher total reagent consumption and a more complex dual-dosing process. In contrast, the star-shaped hybrid structure of Al-PAM-442 integrates both functions into a single molecule, enabling effective flocculation at a drastically lower dosage. This translates to simplified operation, reduced costs, and strong potential for industrial application in coal slime water treatment, even if the ultimate clarity is slightly less than the optimal combined system.

3.4. Interfacial Adsorption Mechanisms Probed by QCM-D

The adsorption of polymer agents on coal slime surfaces governs flocculation efficiency, where agent–particle interactions and the conformational evolution of the adsorption layer drive particle aggregation and sedimentation. QCM-D enables real-time, in situ tracking of adsorption, providing both quantitative mass uptake data and insight into conformational and viscoelastic layer changes. Therefore, to investigate the adsorption kinetics of the organic–inorganic hybrid polymer and conventional polymers, QCM-D was employed on a silica sensor to simulate the dominant surfaces of coal slime particles.

3.4.1. Influence of Molecular Weight on Al-PAM Adsorption

The adsorption kinetics of Al-PAM with different MWs on silica were first studied and the results are shown in Figure 7. After establishing a stable baseline using ultrapure water as the background solution, a 100 ppm polymer solution was introduced. As observed in Figure 7a,b, ∆f decreased and ∆D increased rapidly until stabilization, after which washing with ultrapure water was performed to examine desorption. The results indicate that higher MW Al-PAM produced a larger shift and a longer adsorption equilibrium time. Concurrently, the greater ∆D change with increasing MW suggests the formation of a softer, more elastic adsorbed layer. This can be attributed to the more extended structure and enhanced interchain interactions (e.g., entanglement and sliding between chains) characteristic of higher MW star-shaped Al-PAM, which increase energy dissipation during segmental rearrangement and prolong the time needed to attain an equilibrium conformation. After washing, neither ∆f nor ∆D changed significantly, confirming that Al-PAM adsorption on silica is essentially irreversible.

To analyze the interfacial adsorption behavior of the polymers on the silica surface, ∆D versus -∆f curves were plotted (Figure 7c). The instantaneous slope (∆D/-∆f) reflects the degree of structural change induced per unit mass adsorbed. The higher instantaneous slope for Al-PAM-442 indicates that, per unit mass, it forms a thicker, more hydrated, and viscoelastic layer than lower-MW variants. This extended, hydrated conformation underpins its “strong adsorption bridging” in flocculation, enabling longer-range particle adsorption and larger floc formation, which directly contribute to the higher observed ISR. The linear ∆D-∆f relationship (Figure 7c) indicates that the adsorption process followed a consistent, self-similar growth pattern, with the viscoelastic properties of the adsorbed layer scaling proportionally with its mass. As shown in Figure 7d, higher MW also led to greater final adsorbed amount, due to more binding sites, stronger interfacial affinity, and entanglement-induced desorption resistance. The resulting irreversibly anchored, thick hydrated layer correlates with excellent floc shear stability. Furthermore, the high adsorption capacity and efficiency of Al-PAM-442 at low dosage (6 mg/L) reveals the microscopic origin of its superior dose efficiency (CSI).

In summary, MW governs interfacial behavior by tuning hydration, conformation, and anchoring strength. The thick, soft, and stable adsorbed layer formed by high-MW Al-PAM translates directly into the integrated macroscopic advantages of fast settling, robust shear resistance, and low chemical consumption.

3.4.2. Adsorption of PAC, NPAM, and Their Combination

The adsorption kinetics of PAC, NPAM, and their sequential combination were also examined to provide a comparative basis for understanding the superiority of Al-PAM.

The adsorption results of PAC on silica sensor are shown in Figure 8a–c. Upon introducing the PAC solution into the background ultrapure water, ∆f dropped rapidly to approximately −14 Hz, indicating fast initial adsorption of a considerable amount of PAC onto the SiO₂ surface. This rapid uptake is attributed to the electrostatic attraction between the positively charged PAC hydrolysates and the negatively charged silica surface in aqueous phase. Concurrently, ∆D increased to around 4, suggesting the formation of a relatively soft adsorption layer. This softness likely results from the highly hydrated nature of the adsorbed PAC hydrolysates, leading to a loose, water-rich interfacial structure. When the system was rinsed with ultrapure water, ∆f recovered quickly, demonstrating significant desorption of PAC. The adsorption process is therefore largely reversible. This reversibility indicates that the adsorption is weak and readily reversible, consistent with a predominantly electrostatic interaction that can be easily disrupted by changing the interfacial environment, as opposed to strong specific adsorption [29]. The ∆D-∆f plot (Figure 8c) reveals two distinct adsorption regimes. In Stage I, ∆D rose steeply with decreasing ∆f, which reflects ample freedom for PAC hydrolysates to adsorb and reorganize at the solid–liquid interface. As adsorption proceeded into Stage II, the available surface sites became increasingly occupied, and further interfacial rearrangement occurred under more confined conditions, resulting in a slower increase in ∆D and hence a lower slope in the ∆D-∆f trajectory. The mass uptake of PAC over time, derived from QCM-D fitting, is presented in Figure 8c. The mass increased rapidly to about 500 ng/cm⁻² and then grew slowly to around 650 ng/cm⁻² by 15 min. After rinsing with ultrapure water, the adsorbed mass dropped sharply to approximately 40 ng/cm⁻², confirming the reversible nature of the adsorption. The reversible electrostatic adsorption and the resulting soft, hydrated layer of PAC hydrolysates on particle surfaces account for the formation of only small, weak flocs, which leads to the poor solid–liquid separation (indistinct interface) observed during the settling test.

The adsorption results of NPAM-500 on a silica sensor are shown in Figure 8d–f. A rapid Δf decrease to −7.5 Hz and a concurrent ∆D increase to ~0.5 indicate fast uptake and the formation of a compact layer. This structure results from the linear chains of NPAM adsorbing in a flat conformation, driven primarily by likely hydrogen bonding between the amide groups (-CONH₂) of NPAM and the silanol groups (-SiOH) on the silica surface, potentially supplemented by dipole–dipole interactions. This mechanism, widely reported for polyacrylamide–silica systems [11], is consistent with the observed irreversible adsorption under our experimental conditions. This efficient surface coverage yields a film that is thin and strongly coupled to the sensor, as indicated by the substantial ∆f decrease accompanied by a minimal increase in ∆D. Rinsing with ultrapure water caused no significant change in ∆f or ∆D, demonstrating that the adsorption was largely irreversible under the experimental conditions. A comparison between Figure 8f and Figure 7d shows that the equilibrium adsorbed mass of NPAM-500 is only comparable to that of the lower-MW Al-PAM-264. This similarity in mass uptake contrasts with their distinct adsorbed-layer structures: Al-PAM-264 forms a thicker, looser layer, whereas NPAM-500 yields a thin, compact film. The difference arises from their adsorption mechanisms. For Al-PAM, the positively charged aluminum hydroxide cores provide strong electrostatic attraction to the negatively charged silica, while inter-molecular repulsion and steric effects promote interfacial rearrangement that accommodates more material. In contrast, linear NPAM-500 chains adsorb in a flat, high-coverage conformation that saturates the surface quickly but offers little capacity for extended layer growth. This highlights the structural advantage of the star-like Al-PAM architecture, which not only initiates adsorption via its charged core but also, through optimized spatial arrangement, enhances the binding capacity and creates a more open, hydrated interface. Consequently, the thin, compact layer of NPAM-500 lacks the extended bridging conformation required for effective inter-particle flocculation, explaining its poor performance in settling coal slime water.

The sequential adsorption kinetic results of PAC and NPAM-500 on silica are shown in Figure 8g–i. Pre-adsorption of PAC (0.5 wt.%) caused a ∆f shift to −14 Hz. Subsequent injection of NPAM-500 drove a further sharp decrease to −32 Hz, indicating substantial adsorption onto the PAC-modified surface. This secondary adsorption process likely involves hydrogen bonding between the amide groups of NPAM and the hydroxyl groups of the PAC hydrolysates or the exposed silanol groups on silica. Rinsing with ultrapure water removed loosely bound species (shown in Figure 8h), leading to a more compact layer configuration. The adsorbed mass (see Figure 8i) reached ~1200 ng/cm² for the PAC+NPAM-500 system, nearly double that of Al-PAM-442 (~620 ng/cm²). This microscopic synergy directly explains the superior settling performance observed in Figure 6: the PAC-primed interface enables extensive NPAM-500 uptake, facilitating effective particle bridging and leading to a clear mud line and low supernatant turbidity. However, this two-step mechanism requires a high total dosage (e.g., 70 mg/L PAC+NPAM). In contrast, the star-shaped Al-PAM-442 achieves better clarification (e.g., 45.77 NTU at 6 mg/L) and a higher CSI through its integrated molecular design, which combines electrostatic anchoring and polymeric bridging in a single step. This fundamental difference underscores Al-PAM’s superior dose efficiency and operational simplicity over the binary PAC/NPAM system.

4. Conclusions

This study successfully synthesized and evaluated a series of star-shaped inorganic–organic hybrid flocculants, Al-PAM, for coal slime water treatment. The results demonstrate that Al-PAM, particularly the high MW Al-PAM-442, delivers superior coagulation-flocculation performance compared to conventional PAC, NPAM, and their binary combination. At an ultra-low dosage of 6 mg/L, Al-PAM-442 achieved a high ISR of 50.4 m/h and reduced supernatant turbidity to 45.77 NTU, demonstrating exceptional dose efficiency. The key to its performance lies in its unique integrated molecular architecture. The cationic aluminum hydroxide core enables effective electrostatic anchoring and charge neutralization, while the radiating PAM chains facilitate particle bridging. QCM-D analysis provided crucial interfacial validation, revealing that Al-PAM adsorbs to form a thick, highly hydrated layer, in contrast to the reversible, hydrated layer from PAC and the thin, compact film from NPAM. Importantly, the study clarifies that exceeding the optimal dosage degrades performance via distinct pathways: charge reversal for PAC and steric hindrance due to polymer surface saturation for Al-PAM, underscoring the precision of the identified optimal window. Although the sequential PAC/NPAM system achieves high clarity through a two-step synergy, it requires a significantly higher total dosage. In summary, the star-shaped Al-PAM integrates dual functions into a single molecule, which translates into the integrated macroscopic advantages of rapid settling, high clarity, and low chemical consumption within a defined optimal range, offering an efficient and mechanistic rationale for advanced coal slime water treatment.