Hydrothermal Modification of Coal Gangue for Cd²⁺ and Pb²⁺ Adsorption: Modelling and Optimization of Process Parameters by Response Surface Methodology

Abstract

Hydrothermal treatment has been demonstrated as a highly effective approach to enhance the adsorption properties of coal gangue (CG) for Cd²⁺ and Pb²⁺. This study evaluated the influence of NaOH concentration (X₁), hexadecyltrimethylammonium bromide dosage (X₂) and hydrothermal duration (X₃) on the modified CG’s composition and its adsorption capacities for Cd²⁺ (q_e,Cd) and Pb²⁺ (q_e,Pb). Response Surface Methodology (RSM) was utilized to explore the interactions among these variables, while XRD, SEM-EDX, FTIR and XPS techniques were employed to elucidate the adsorption mechanisms. The results reveal that hydrothermal treatment facilitated the formation of zeolite materials in the modified CG samples. Optimal conditions for Cd²⁺ adsorption (q_e,Cd = 58.4 mg/g) were determined as X₁ = 2.9 mol/L, X₂ = 1 g and X₃ = 16.8 h. For Pb²⁺ adsorption (q_e,Pb = 233.6 mg/g), the optimal conditions were X₁ = 2.4 mol/L, X₂ = 0.57 g and X₃ = 20.7 h. RSM analysis indicated that X₃ significantly influences q_e,Cd with the appropriate range of 14–18 h, while all three factors (X₁, X₂ and X₃) notably affect q_e,Pb, with suitable ranges identified as follows: X₁ = 1.8–3 mol/L, X₂ = 0.4–0.8 g and X₃ = 16–24 h. The primary adsorption mechanism for Cd²⁺ and Pb²⁺ was determined to be ion exchange, with additional Pb²⁺ removal achieved through the precipitation of cerussite (PbCO₃).

1. Introduction

Over the past decades, heavy metal ions released from industrial activities such as mining operations, electroplating, ceramics production, alkaline battery manufacturing and pigment processing have emerged as a significant environmental concern due to their toxicity and persistence [1]. Various techniques have been developed to remove heavy metal ions from aqueous solutions, including solvent extraction, microbiological degradation, reverse osmosis, electrodialysis, and adsorption. Among these, adsorption stands out as one of the most effective approaches. For large-scale applications, it is crucial to select an adsorbent that balances cost-effectiveness with removal efficiency [2,3,4].

CG is a kind of hazardous solid waste generated during coal mining and washing processes, typically accounts for 15–20% of the total raw coal output [5]. It is estimated that the accumulated CG in China has exceeded 700 million tons, with an annual production of approximately 0.9 million tons in recent years [6]. Given its primary chemical constituents, Al₂O₃ and SiO₂, as well as its predominant mineral composition of kaolin, coal gangue has emerged as a promising raw material for preparing heavy metal adsorbents [7]. Two main approaches for utilizing CG in this context include modifying it to produce CG-based adsorbents and converting it into CG-based zeolite materials through zeolitization.

Various modification techniques have been investigated to enhance the adsorption performance of CG, including calcination [8,9,10], acid washing [9,10], alkali washing [11,12], hydrothermal treatment [13,14] and intercalation [15,16]. Qiu et al. [8] developed a novel calcination method by incorporating sodium tetraborate (Na₂B₄O₇·10H₂O), which increased the adsorption capacity for Mn²⁺ from 3.4 to 24.4 mg/g. Additionally, they employed hydrothermal treatment to modify CG, achieving an improved Cd²⁺ adsorption capacity of 183.7 mg/g [13]. Gao et al. [11] found that HCl modification was more effective than KOH modification for enhancing the adsorption capacity of CG for U⁶⁺. Jabłońska et al. [9] compared the structural and surface properties of natural CG with CG modified using H₂NO₃, H₂O₂ and calcination. Furthermore, modified CG-SH was synthesized through surface grafting of CG with (3-mercaptopropyl) trimethoxysilane [15], resulting in maximum adsorption capacities of 332.8, 110.4 and 179.2 mg/g for Pb²⁺, Cd²⁺ and Hg²⁺, respectively.

Alkaline fusion-assisted hydrothermal crystallization is the most widely used method for synthesizing zeolites from CG [17]. Researchers have successfully produced various zeolites from CG, including NaX [2,18,19], NaY [6] and NaA [19]. Lu et al. [18] synthesized NaX zeolite from CG, achieving adsorption capacities of 45.05 mg/g for Cu²⁺ and 44.53 mg/g for Co²⁺. Ge et al. [2] reported an impressive Pb²⁺ adsorption capacity of 457 mg/g under optimal conditions using NaX zeolite derived from CG. Bu et al. [6] produced NaY zeolite from CG and identified ion exchange as the primary mechanism for Pb²⁺ adsorption. Jin et al. [20] developed a cost-effective method for synthesizing NaA zeolite from CG by eliminating the high-temperature alkali fusion process, which typically requires significant amounts of NaOH. Acid leaching is critical for zeolite synthesis when CG has high iron and quartz content [21]. Additionally, Li et al. [22] prepared a zeolite-activated carbon composite from CG through CO₂ activation followed by hydrothermal synthesis.

Compared to CG-based zeolite materials, modified CG generally exhibits lower adsorption efficiency but significantly reduced processing costs, making it more feasible for industrial-scale applications. In our prior research, we evaluated the impacts of three modification techniques (i.e., acid washing, alkali washing and hydrothermal treatment) on adsorption performance, identifying hydrothermal modification as the most effective approach for enhancing the adsorption of Cd²⁺ and Pb²⁺ under controlled conditions [14]. This study focuses on optimizing key parameters of the hydrothermal process, including NaOH concentration, HTAB dosage and treatment duration, with particular emphasis on their interactive effects. Using RSM, we systematically investigated these interactions, while advanced characterization techniques such as XRD, SEM-EDX, FTIR and XPS techniques were used to elucidate the adsorption mechanisms. The results of this study can provide valuable guidance for the production of low-cost CG-based adsorbents for heavy metal ions.

2. Materials and Methods

2.1. Materials

The CG utilized in this study was obtained from the Yangshita mine in Inner Mongolia, hereafter referred to as YST. Its chemical composition was analyzed using X-ray fluorescence spectroscopy (XRF-1800, Shimadzu, Kyoto, Japan). According to the results presented in

Table 1. Main chemical composition of the CG sample.

Sample	SiO2	Al2O3	Fe2O3	K2O	Na2O	CaO	MgO	SO3	P2O5	Others	SiO2/Al2O3
YST	64.64	21.43	3.56	3.59	0.87	2.51	0.92	0.30	0.18	2.00	5.13

, the molar ratio of SiO₂ to Al₂O₃ was determined to be 5.13. Proximate analysis was conducted in accordance with the Chinese national standard GB/T 212-2008 [23]. The contents of moisture, ash, volatile matter and fixed carbon were found to be 2.39 wt.%, 84.12 wt.%, 8.22 wt.% and 5.27 wt.%, respectively.

The initial CG sample was first ground and passed through a sieve to collect particles under 180 μm, which were then oven-dried at 105 °C over 24 h. The dried material obtained was thereafter employed in the modification procedure. All chemical reagents, namely HCl, NaOH, hexadecyltrimethylammonium bromide (HTAB), CdCl₂ and Pb(NO₃)₂, were supplied by Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). Among these, NaOH, CdCl₂ and Pb(NO₃)₂ were of analytical grade. HCl (36–38%) and HTAB (99%) were utilized without further purification.

2.2. Hydrothermal Modification of CG

The hydrothermal modification of CG was carried out based on the method reported by Qiu et al. [13]. Typically, 10 g of CG was mixed with 100 mL of 2 M NaOH solution and 1 g of HTAB in a 200 mL three-necked round-bottom flask fitted with a reflux condenser. The reaction was performed under continuous stirring at 100 °C for 24 h. After the reaction, the colloidal suspension obtained was vacuum-filtered, rinsed thoroughly with distilled water, and the collected solid was dried at 105 °C for 12 h to produce the hydrothermally modified CG. To identify optimal modification conditions, single-factor tests were conducted by varying key parameters as listed in

Table 2. Reaction condition of hydrothermal modification of CG.

Run No.	NaOH Concentration (mol/L)	HTAB Dosage (g)	Hydrothermal Duration (h)
effect of NaOH concentration
1	1	1	18
2	2
3	3
4	4
effect of HTAB dosage
5	2	0.25	18
6		0.5
7		1
8		2
effect of hydrothermal duration
9	2	1	12
10			18
11			24
12			30

. These included NaOH concentration (1–4 M), HTAB mass (0.25–2 g) and treatment time (12–30 h), while keeping other experimental settings unchanged.

In an earlier study, CG was first calcined at 600 °C for 5 h in a muffle furnace prior to hydrothermal treatment [14]. However, this high-temperature calcination step is energy-intensive and contributes to increased processing costs. Therefore, in the present study, we also explore the impact of calcination pretreatment on the adsorption properties of the modified CG. Calcined CG and uncalcined CG underwent hydrothermal treatment under specified conditions: a NaOH concentration of 2 mol/L, an HTAB dosage of 1 g and a hydrothermal duration of 24 h. The resulting modified CG samples were designated as CG-C-HY and CG-HY, respectively.

2.3. Box–Behnken Method

The Box–Behnken design, a RSM approach, was implemented via Design Expert 13 software [24,25] to formulate the experimental scheme and construct a predictive model. This design aimed to optimize the process parameters and identify the conditions for maximum adsorption capacity. The independent variables investigated were NaOH concentration (X₁), HTAB dosage (X₂) and hydrothermal treatment time (X₃). The corresponding adsorption capacities for Cd²⁺ and Pb²⁺ served as the response variables. As shown in

Table 3. Experimental design in the Box–Behnken method.

Run No.	X1 (mol/L)	X2 (g)	X3 (h)	qe,Cd	qe,Pb
1	3	1	18	55.5	190.0
2	2	0.5	18	55.1	220.2
3	2	0.5	18	52.0	224.7
4	3	0.25	18	52.2	207.2
5	3	0.5	12	52.1	177.7
6	2	1	24	45.6	190.0
7	2	0.5	18	52.5	237.9
8	1	1	18	54.2	137.8
9	1	0.5	12	52.1	148.9
10	2	0.5	18	50.4	229.9
11	1	0.5	24	42.5	180.4
12	2	1	12	52.1	156.7
13	2	0.25	12	50.2	170.5
14	1	0.25	18	55.1	160.6
15	3	0.5	24	45.0	219.7
16	2	0.25	24	45.1	207.2
17	2	0.5	18	53.1	244.4

, each factor was tested across a range established in prior single-factor experiments, with all other conditions held constant. It is noteworthy that the values presented in Table 3 are the actual values used in the experiments.

2.4. Adsorption Capacity Test of Cd²⁺ and Pb²⁺

Stock solutions of Cd²⁺ and Pb²⁺ (100 mg/L) were prepared by dissolving CdCl₂·2.5H₂O and Pb(NO₃)₂ in distilled water, respectively. The initial pH values of these solutions were approximately 6 for Cd²⁺ solution and 5 for Pb²⁺ solution. In the batch adsorption tests, 0.1 g of the modified CG was introduced into 50 mL of the Cd²⁺ solution, whereas a smaller mass of 0.02 g was used for the Pb²⁺ solution at the same volume. These mixtures were then agitated continuously for 1800 min at room temperature on a horizontal shaker (HY-2, Supo, Shaoxing, China). This extended duration was chosen to ensure the adsorption process reached equilibrium [14]. Subsequently, the samples were filtered using a 0.45 μm membrane, and the residual metal ion concentrations in the filtrate were analyzed with a UV-visible spectrophotometer (UV-1100, Mapada, Shanghai, China) [26]. The testing procedure is briefly outlined in the Supporting Material. The adsorption capacity was finally calculated using Equation (1) [14]:

$$q_e={\displaystyle \frac{\left(c_0-c_e\right)\times V}{m}}$$

(1)

where q_e stands for the adsorption capacity in mg/g. The terms c₀ and c_e signify the initial and equilibrium concentrations of the heavy metal ions, respectively (mg/L). V is the volume of the solution (L), and m is the mass of the adsorbent (g). To ensure reliability, all adsorption tests were performed in triplicate, and the reported results are based on the mean values. The adsorption capacities for Cd²⁺ and Pb²⁺ are specifically denoted as q_e,Cd and q_e,Pb, respectively. Statistical analysis of the adsorption capacity data was performed using Tukey’s test in OriginPro 2017 software. The significance of differences was determined based on the calculated p-values and indicated with asterisk-based significance codes: one asterisk (*) denotes a significant difference (p < 0.05), two asterisks (**) represent a highly significant difference (p < 0.01), three asterisks (***) indicate an extremely significant difference (p < 0.001) and “ns” denotes no significant difference (p ≥ 0.05).

2.5. Characterization of Modified CG

The crystalline phases present in the modified CG samples were identified by X-ray diffraction (XRD) on a Bruker D8 diffractometer (Billerica, MA, USA) with Cu Kα radiation (λ = 0.15418 nm), operating at 40 kV and 100 mA. Diffraction patterns were recorded across a 2θ interval of 5–60° at a scanning speed of 10°/min. Sample morphology was assessed using a ZEISS Gemini SEM 300 scanning electron microscope (SEM), Oberkochen, Germany.

To elucidate the adsorption mechanism, the samples were characterized both before and after the adsorption process using EDX, XPS and FTIR. Elemental distribution mapping of heavy metals was conducted with a Bruker XFlash Detector 5010 energy dispersive spectrometer (Ettlingen, Germany). XPS analysis was performed on a Thermo Escalab 250XI instrument (Waltham, MA, USA) employing an Al-Kα X-ray source (hv = 1486.6 eV), with survey scans covering binding energies from 0 to 1300 eV. FTIR spectra were collected using a Nicolet 6700 spectrometer (Waltham, MA, USA), scanning the mid-IR region from 400 to 4000 cm⁻¹.

3. Results

3.1. Effect of Calcination Pretreatment on CG Modification

Figure 1 presents the XRD spectra and adsorption capacities of the modified CG samples. In Figure 1a, it can be observed that NaP zeolite was formed after calcination and hydrothermal treatment. NaP zeolite could also be synthesized when the raw CG underwent only hydrothermal treatment; however, the diffraction intensity was notably lower. Figure 1b shows that CG-C-HY exhibited better adsorption performance than CG-HY, though the difference between the two was relatively modest, with q_e,Cd and q_e,Pb values of approximately 10 and 40 mg/g, respectively. Based on these results, the high-energy-consuming calcination process was excluded from the hydrothermal modification of CG samples. No CG samples underwent high-temperature calcination prior to the hydrothermal modification process described in the following sections.

3.2. Effect of Different Parameters on CG Modification

3.2.1. Effect of NaOH Concentration on CG Modification

The impact of NaOH concentration on both the crystalline phase evolution and the adsorption performance of the modified CG is presented in Figure 2. As shown in Figure 2a, the XRD pattern of the sample treated with 1 mol/L NaOH is predominantly characterized by quartz. This indicates that the alkalinity at this level was inadequate to sufficiently dissolve the Si and Al components in CG for zeolite crystallization. In contrast, with the NaOH concentration elevated to 2 mol/L and beyond, a distinct zeolite phase identified as Na₈(Al₆Si₆O₂₄)(NO₂)₂·3H₂O (labeled Z₁) emerges. The diffraction peaks for this phase intensify with further increases in NaOH concentration. Correspondingly, Figure 2b demonstrates a consistent enhancement in q_e,Pb with rising NaOH concentration. This trend is closely linked to the increased yield of the zeolite phase and the consequent creation of more active adsorption sites. Statistical analysis revealed that NaOH concentration had a significant effect on q_e,Pb. In contrast, a NaOH concentration of up to 4 mol/L was required to exert a significant effect on q_e,Cd.

3.2.2. Effect of HTAB Dosage on CG Modification

Figure 3 shows the influence of HTAB dosage on the characteristics of the modified CG products. As shown in Figure 3a, the formation of two distinct zeolitic phases, denoted as Z₁ and Z₂, was observed within the HTAB dosage range of 0.25 to 1 g. These phases were identified as Na₈(Al₆Si₆O₂₄)(NO₂)₂∙3H₂O and K₅Ca₂(Al₉Si₂₃O₆₄)∙24H₂O, respectively. NaP zeolite was identified when 2 g of HTAB was used. HTAB is commonly employed as a modifier for coal-based solid wastes in previous studies. It is hypothesized that HTAB functions as an organic template, promoting the formation of crystalline NaP zeolite. Figure 3b shows that, with increasing HTAB dosage, q_e,Pb initially increased and then decreased, reaching a maximum value of 213.4 mg/g at 0.5 g. In contrast, q_e,Cd exhibited a peak value of 62.4 mg/g at 2 g. The observed disparity in adsorption performance may be attributed to distinct removal mechanisms for Cd²⁺ and Pb²⁺. Specifically, the formation of NaP zeolite during modification primarily enhances Cd²⁺ uptake, while the treatment with 0.5 g of HTAB favors the adsorption of Pb²⁺. Statistical analysis indicated that a HTAB dosage of 2 g was required to significantly affect q_e,Cd, implying that enhancing q_e,Cd may necessitate an increase in HTAB usage. Furthermore, HTAB dosage exhibited a moderate yet discernible influence on q_e,Pb across the investigated range.

3.2.3. Effect of Hydrothermal Treatment Duration on CG Modification

Figure 4 illustrates the effects of hydrothermal treatment duration on the XRD pattern and adsorption capacity of the modified CG samples. As shown in Figure 4a, a zeolite material Na₈(Al₆Si₆O₂₄)(NO₂)₂∙3H₂O (denoted as Z₁) is formed after 12 h of hydrothermal treatment, with diffraction intensity increasing as the treatment duration is extended. This is because a longer processing time facilitates the polymerization of silicate and aluminate ions, promoting the formation of zeolite with higher crystallinity. However, a different zeolite phase K₅Ca₂(Al₉Si₂₃O₆₄)∙24H₂O (denoted as Z₂) is identified after 24 h of treatment. This may result from the complex phase transitions occurring during the zeolite formation from coal-based solid wastes. During crystallization, the aluminosilicate gel first gives rise to metastable intermediates, which then gradually transform into more stable crystalline phases over time. At a hydrothermal treatment duration of 18 h, both q_e,Cd and q_e,Pb reached their maximum values of 55.5 and 191.1 mg/g, respectively. This suggests that excessively long processing times may be detrimental, and an optimal treatment duration is required to achieve the best adsorption performance. Statistical analysis revealed that the duration of hydrothermal treatment influenced both q_e,Cd and q_e,Pb, with a more pronounced effect on q_e,Pb.

3.3. Optimization of Modification Conditions by RSM

3.3.1. RSM Model Fitting

The adsorption data from the 17 trials designed by the Box–Behnken method are summarized in Table 3. To model the responses, second-order polynomial equations were fitted to the experimental data for q_e,Cd and q_e,Pb using RSM. The resulting quadratic models are given in Equations (2) and (3).

q_e,Cd = 24.98 − 5.63X₁ − 9.34X₂ + 4.54X₃ + 2.21X₁X₂ + 0.108X₁X₃ − 0.063X₂X₃ + 0.636X₁² + 6.12X₂² − 0.148X₃²

(2)

q_e,Pb = −216.62 + 123.03X₁ + 231.13X₂ + 23.61X₃ + 6.77X₁X₂ + 0.438X₁X₃ − 0.419X₂X₃ − 28.46X₁² − 208.63X₂² − 0.591X₃²

(3)

3.3.2. Analysis of Variance (ANOVA) Tests

The significance and adequacy of the developed models were evaluated through ANOVA. The results are shown in Tables S1 and S2 for q_e,Cd and q_e,Pb, respectively. The results demonstrate that both models are statistically significant at a 95% confidence level. This conclusion is supported by high F-values (9.64 for q_e,Cd and 26.27 for q_e,Pb) and notably low p-values (0.0034 for q_e,Cd and 0.0001 for q_e,Pb). In this analysis, the p-value indicates the probability of error, with values below 0.05 denoting statistical significance, and those under 0.0001 reflecting a particularly strong effect. For q_e,Cd, the significant terms are X₃ and X₃². For q_e,Pb, X₁, X₂, X₃, X₁², X₂² and X₃² are identified as significant. Among these, hydrothermal treatment duration (X₃) and NaOH concentration (X₁) emerge as the most influential factors for Cd²⁺ and Pb²⁺ adsorption, respectively, as confirmed by their substantial F-values (36.55 and 48.98) and minimal p-values (≤0.0005).

3.3.3. Variables Interactions Analysis

Figure 5 illustrates the combined effects of NaOH concentration, HTAB dosage and hydrothermal treatment duration on the adsorption capacity for Cd²⁺. The RSM plot in Figure 5a appears relatively flat, suggesting that both NaOH concentration and HTAB dosage have a minimal impact on q_e,Cd. However, Figure 5b,c demonstrate that the hydrothermal treatment duration has a significant effect on the Cd²⁺ adsorption capacity. For instance, when the HTAB dosage is 0.625 g and the NaOH concentration is 2 mol/L, q_e,Cd initially increases and then decreases as the treatment duration extends from 12 to 24 h, reaching a maximum of 53.2 mg/g at 16 h. To reduce the cost of preparing modified CG for Cd²⁺ adsorption, a lower HTAB dosage and NaOH concentration can be chosen while maintaining the treatment duration within the range of 14 to 18 h.

Figure 6 illustrates the combined effects of the variables on the adsorption capacity for Pb²⁺. All three RSM plots exhibit a saddle-shaped profile, suggesting that each of the three factors significantly impacts the Pb²⁺ adsorption capacity. For example, at an HTAB dosage of 0.625 g and a hydrothermal treatment duration of 18 h, the q_e,Pb reached a maximum of 235.5 mg/g when the NaOH concentration was 2.4 mol/L. The adsorption capacity initially increased, then decreased as the NaOH concentration varied from 1 to 3 mol/L. To optimize the adsorption performance of the modified CG for Pb²⁺, all three parameters must be adjusted within specific ranges: NaOH concentration between 1.8 and 3 mol/L, HTAB dosage between 0.4 and 0.8 g, and hydrothermal treatment time between 16 and 24 h.

3.3.4. Validation

The optimization results indicated that the maximum q_e,Cd value (55.9 mg/g) would be achieved under the following conditions: 2.9 mol/L NaOH, 1 g HTAB and a hydrothermal treatment time of 16.8 h. For q_e,Pb, the model predicted an optimal capacity of 240.2 mg/g with parameters set at 2.4 mol/L NaOH, 0.57 g HTAB and a duration of 20.7 h. To validate these predictions, verification experiments were performed at the specified optimum conditions, yielding measured capacities of 58.4 mg/g for Cd²⁺ and 233.6 mg/g for Pb²⁺. The strong consistency between the experimental and predicted values successfully corroborates the reliability of the regression models.

3.4. Characterization of Modified CG Before and After Adsorption

To further elucidate the adsorption mechanisms, the modified CG samples prepared under optimized conditions were characterized both before and after adsorption using XRD, SEM-EDS, FTIR and XPS techniques. The samples were labeled as CG-HY-1 and CG-HY-2 for the modified CG samples prior to adsorption for Cd²⁺ and Pb²⁺, respectively, and as CG-HY-1 (Cd) and CG-HY-2 (Pb) after adsorption.

3.4.1. XRD

Figure 7 illustrates that minor alterations can be observed in the XRD patterns after the adsorption of Cd²⁺ and Pb²⁺ ions. Specifically, the peaks corresponding to the lattice planes of the zeolite material Na₈(Al₆Si₆O₂₄)(NO₂)₂∙3H₂O show a decrease post-adsorption. This reduction is likely due to the introduction of hydrated Cd²⁺ and Pb²⁺ into the zeolite pore structure, which may have disrupted the crystal lattice to some extent, leading to a decline in the overall crystallinity of the material. Additionally, cerussite (PbCO₃) was identified in the CG-HY-2 (Pb) sample, indicating that Pb²⁺ could also be retained in the modified CG through precipitation.

3.4.2. SEM-EDX

Figure 8 shows the SEM images and EDX spectra of modified CG samples after the adsorption of Cd²⁺ and Pb²⁺. The SEM images reveal the presence of aggregated small particles in these samples. Given that the CG samples did not undergo calcination pretreatment, it is hypothesized that the NaOH solution only dissolved a limited amount of Si⁴⁺ and Al³⁺, leading to the formation of crystal nuclei of zeolite materials. As shown in Figure 8a,c, octahedral granules are observed adhering to the surfaces of CG particles, which are inferred to be well-crystallized zeolite structures. However, these granules are no longer visible after the adsorption of Cd²⁺ and Pb²⁺ (Figure 8b,d), potentially due to the occupation of zeolite pore channels by Cd²⁺ and Pb²⁺ ions, disrupting the crystal lattice structure. The EDX analysis further confirms adsorption, as evidenced by the increase in Cd and Pb content from 0 and 0.07 wt.% to 1.18 wt.% and 1.45 wt.%, respectively. Simultaneously, a reduction in Na content was observed, which is attributed to an ion-exchange mechanism, where Na⁺ ions in the zeolite framework are replaced by Cd²⁺ and Pb²⁺ ions.

3.4.3. FTIR

The FTIR spectra of the modified CG samples, both before and after metal ion adsorption, are displayed in Figure 9. A prominent absorption band at 3423 cm⁻¹ is assigned to the O–H stretching vibration. The feature near 1640 cm⁻¹ is associated with H–O–H bending. The presence of peaks at approximately 2920 and 2855 cm⁻¹, which correspond to the asymmetric and symmetric C–H stretching vibrations, may be attributed to the successful incorporation of the hexadecyltrimethylammonium surfactant into the material framework, as well as to the inherent organic matter present in the CG. The strong absorption at 1000 cm⁻¹ originates from the asymmetric stretching of Si–O–Si/Al bonds, and the peak at 460 cm⁻¹ is characteristic of the D4R ring vibrations in zeolitic structures. Following the adsorption of Cd²⁺ and Pb²⁺, the core spectral features of the adsorbent remained largely intact, indicating that the adsorption process did not induce major changes to its fundamental chemical structure.

3.4.4. XPS

Figure 10 displays the results of XPS analysis conducted on the modified CG samples both before and after the adsorption of Cd²⁺ and Pb²⁺ ions. As illustrated in Figure 10a,c, the full spectra confirm that the binding energies of Na 1s, O 1s, Si 2p and Al 2p in CG-HY-1 and CG-HY-2 align well with previously reported values in the literature [6,27]. The adsorption of Cd²⁺ onto CG-HY-1 is validated by the two newly observed peaks at 404.9 eV and 411.7 eV, corresponding to Cd 3d_5/2 and Cd 3d_3/2, respectively, as shown in Figure 10b. Likewise, Figure 10d demonstrates the adsorption of Pb²⁺ onto CG-HY-2, evidenced by peaks at 138.4 eV and 143.3 eV, characteristic of Pb 4f_7/2 and Pb 4f_5/2, respectively. A marked reduction in the Na 1s peak intensity in CG-HY-2 (Pb) compared to that in CG-HY-2 indicates that ion exchange occurs between Pb²⁺ and Na⁺ ions during the adsorption process. These findings from XPS analysis are consistent with those obtained through SEM-EDX.

4. Discussion

4.1. Synthesis Strategies for Low-Cost Hydrothermally Modified CG Adsorbents

This study employed a hydrothermal method to modify CG for the preparation of an adsorbent targeting Cd²⁺ and Pb²⁺. Although pre-calcination prior to hydrothermal modification improved the adsorption capacity of the modified CG (CG-C-HY), the difference compared to the non-calcined sample (CG-HY) was not statistically significant. Therefore, the high-cost, high-temperature calcination step can be omitted to reduce production expenses. Furthermore, for the hydrothermal modification of CG, hydrothermal duration (X₃) was identified as a critical factor affecting the adsorption capacity for Cd²⁺, while NaOH concentration (X₁), HTAB dosage (X₂) and hydrothermal duration (X₃) all significantly influenced the equilibrium adsorption capacity for Pb²⁺. Given the high cost of HTAB, optimizing NaOH concentration and treatment duration to minimize HTAB usage presents a cost-effective strategy for adsorbent synthesis. By constraining q_e,Pb at 200 mg/g and analyzing the interplay among these three variables using Equation (3), the results are presented in Figure 11. The analysis indicates that when the hydrothermal duration is less than 20 h, the required HTAB dosage increases with prolonged treatment time, leading to higher modification costs. In contrast, when the duration exceeds 20 h, the HTAB dosage decreases. Similarly, at a fixed treatment duration, the HTAB dosage initially increases and then decreases with increasing NaOH concentration. Overall, the minimum HTAB dosage of 0.68 g is achieved at a NaOH concentration of 3 mol/L and a treatment duration of 14 h. Since the price of industrial-grade HTAB is five times or even higher than that of industrial-grade NaOH, and the hydrothermal process is conducted at a low temperature of only 100 °C (resulting in low energy consumption and cost), reducing the dosage of HTAB can significantly enhance the industrial scale-up potential of this modification process.

4.2. Adsorption Mechanism of Hydrothermally Modified CG for Cd²⁺ and Pb²⁺

Characterization of the hydrothermally modified CG samples before and after metal uptake points to ion exchange as the dominant mechanism for the removal of Cd²⁺ and Pb²⁺. In this process, Cd²⁺ and Pb²⁺ ions migrate into the zeolite pores and replace the Na⁺ ions that balance the framework’s negative charge. This exchange is evidenced by a considerable decrease in Na⁺ content post-adsorption, along with a concomitant reduction in the intensity of the characteristic zeolite XRD peaks. Such findings are consistent with reports in the literature [6]. Bu et al. [6] attributed a shift in the I(2 2 0)/I(3 1 1) intensity ratio of NaY zeolite after Pb²⁺ adsorption to the displacement of Na⁺ ions from supercages by Pb²⁺. Furthermore, precipitation contributes to the removal of Pb²⁺. The identification of cerussite (PbCO₃) in the CG-HY-2 (Pb) sample suggests that carbonate ions released from the CG component reacted with Pb²⁺ to form an insoluble precipitate. A parallel mechanism was reported by Lyu et al. [28], who identified the formation of PbCO₃ and Pb₃(CO₃)₂(OH)₂ during Pb²⁺ adsorption on modified red mud and concluded that the removal involved both ion exchange and precipitation. Figure 12 summarizes the adsorption mechanism of Cd²⁺ and Pb²⁺ by modified CG.

The adsorption capacity results in this study further indicate that the uptake of Pb²⁺ by hydrothermally modified CG samples is approximately three times higher than that of Cd²⁺. This difference can be attributed to two main factors. First, a portion of Pb²⁺ is removed through precipitation. Second, the relative atomic mass of Pb (207.2) is nearly twice that of Cd (112.4), which also contributes to the higher measured adsorption capacity for Pb²⁺ in mass-based units. Moreover, the hydrated ionic radius of Cd²⁺ (0.43 nm) is larger than that of Pb²⁺ (0.40 nm), suggesting that Cd²⁺ may have limited access to certain exchange sites within the zeolite structure compared to Pb²⁺. It is therefore inferred that the number of exchangeable sites available for Cd²⁺ in hydrothermally modified coal gangue is fewer than those for Pb²⁺.

4.3. Adsorption Capacity of Hydrothermally Modified CG for Cd²⁺ and Pb²⁺

Table 4. Adsorption capacity of different adsorbents used for removal of Cd²⁺ and Pb²⁺.

Adsorbent	Adsorption Capacities (mg/g)
	Cd2+	Pb2+	Reference
Hydrothermally Modified CG	58.4	233.6	This Work
Raw CG of Yangshita Mine	7.7	26.8	[14]
Hydrothermally Modified CG of Yangshita Mine	72.7	247.5	[14]
NaX Zeolite Parepared from Yangshita CG	187.5	427.6	[29]
Mercapto Modified CG	110.4	332.8	[15]
Coal-analcime Composite from CG	54.1	/	[30]
Low-temperature KOH Modified CG	70.7	/	[12]
Hydrothermally Modified Fly Ash	183.7	/	[13]
Natural Zeolite Treated with NaCl	/	91.2	[31]
Low-temperature NaOH Modified Fly Ash	/	126.6	[32]
NaA Zeolite Prepared from Kaolin	/	213.4	[33]

compares the adsorption capacities of hydrothermally modified CG synthesized in this study with those of adsorbents reported in the literature for Cd²⁺ and Pb²⁺. Although the adsorption conditions differ between this study and the literature, the data in the table remain reasonably comparable. In our previous study, we found that the raw CG from Yangshita Mine exhibited q_e,Cd and q_e,Pb values of only 7.7 and 26.8 mg/g, respectively. After hydrothermal modification, these values significantly increased to 72.7 and 247.5 mg/g [14]. In this study, CG was not subjected to high-temperature calcination (600 °C) pretreatment. Under the optimized modification conditions, q_e,Cd and q_e,Pb reached 58.4 and 233.6 mg/g, respectively. The zeolitization of CG is an effective method for enhancing adsorption capacity. In our previous study, NaX-type zeolite synthesized through alkali fusion-hydrothermal treatment achieved q_e,Cd and q_e,Pb values of 187.5 and 427.6 mg/g, respectively [29]. Shang et al. [15] prepared mercapto-modified coal gangue with q_e,Cd and q_e,Pb values of 110.4 and 332.8 mg/g, respectively, which are notably higher than those of the hydrothermally modified CG prepared in this study. However, their method involves a complex procedure requiring multiple reagents including H₂SO₄, KMnO₄, N₂H₄·H₂O, MPTMS (3-mercaptopropyl trimethoxysilane) and methanol, through steps of acid washing, oxidation and sulfurization. Obviously, although both zeolite synthesis and mercapto modification can enhance q_e,Cd and q_e,Pb, they are associated with drawbacks such as multi-step processes and high costs.

Regarding the enhancement of q_e,Cd of CG, Jin et al. [30] prepared a low-cost analcime-coal composite adsorbent from CG, achieving a q_e,Cd of 54.1 mg/g, which is comparable to the hydrothermally modified CG in this study. Their method eliminates the need for high-temperature alkali fusion, thereby reducing synthesis costs and energy consumption, demonstrating significant potential for industrial-scale application. Zhao et al. [12] modified CG via a low-temperature alkali roasting approach, increasing q_e,Cd to 70.7 mg/g. However, the mass ratio of KOH to CG employed in their process was 3:2, indicating substantial alkali consumption, which limits its industrial feasibility. It should be noted that Qiu et al. [13] achieved a q_e,Cd of 183.7 mg/g through hydrothermal treatment of circulating fluidized bed fly ash from a coal gangue-fired power plant, approaching the adsorption capacity of NaX zeolite synthesized from Yangshita CG [29] and significantly exceeding the performance of the hydrothermally modified CG in this study. They also identified the zeolitic phases formed in the material as the active components responsible for adsorption. This enhancement could be primarily attributed to the transformation of inert mineral phases in CG into active phases during high-temperature combustion, which facilitates zeolite formation under subsequent hydrothermal conditions.

4.4. Future Perspectives

Cadmium ions (Cd²⁺) represent a pollutant of primary concern in CG from the Shendong mining area, China. When CG is used in backfilling and reclamation projects, Cd²⁺ can be leached by rainfall, posing a potential risk of groundwater contamination [29]. A promising mitigation strategy involves blending the CG with a CG-based adsorbent within the backfill material. The incorporated adsorbent can immobilize Cd²⁺, thereby reducing the toxicity and environmental mobility of this heavy metal. While this study employed a low-cost hydrothermal modification method to synthesize a CG-based adsorbent, its adsorption capacity for Cd²⁺ remains limited. To enhance the Cd²⁺ uptake performance of modified CG, future research could explore the following directions: (1) Utilizing mesoporous templates such as poly dimethyl diallyl ammonium chloride during synthesis to generate zeolitic structures with larger pore sizes [34]. This approach aims to create more accessible ion-exchange sites for Cd²⁺, facilitating its diffusion and capture. (2) Incorporating other materials such as biochar [27] or iron-manganese oxides [35] during the modification process. These components can enhance the overall adsorption capacity and promote co-precipitation mechanisms for Cd²⁺ removal. (3) Functionalizing the CG surface with sulfide ions (S²⁻) [36]. Given the exceptionally low solubility product (K_sp) of CdS, the introduction of S²⁻ provides specific sites for the effective precipitation and immobilization of Cd²⁺.

5. Conclusions

This study investigated the modification of CG via hydrothermal treatment and evaluated its adsorption performance for Cd²⁺ and Pb²⁺. The modification parameters, including NaOH concentration (X₁), hexadecyltrimethylammonium bromide dosage (X₂) and hydrothermal duration (X₃), were optimized using RSM method. Various characterization techniques were employed to elucidate the adsorption mechanisms. The main findings are summarized as follows:

(1)

The optimal modification conditions for Cd²⁺ adsorption (q_e,Cd = 58.4 mg/g) were determined as X₁ = 2.9 mol/L, X₂ = 1 g and X₃ = 16.8 h. For Pb²⁺ adsorption (q_e,Pb = 233.6 mg/g), the optimal parameters were X₁ = 2.4 mol/L, X₂ = 0.57 g and X₃ = 20.7 h.

(2)

Hydrothermal duration was identified as a critical factor influencing Cd²⁺ adsorption, with a suitable range of 14–18 h. For Pb²⁺ adsorption, all three parameters—X₁, X₂, X₃—played significant roles, with optimal ranges of X₁ = 1.8–3 mol/L, X₂ = 0.4–0.8 g, and X₃ = 16–24 h.

(3)

Hydrothermal modification promoted the formation of zeolite structures in the modified CG, enhancing its adsorption properties. The primary adsorption mechanism for Cd²⁺ and Pb²⁺ involved ion exchange, with additional Pb²⁺ removal attributed to the precipitation of cerussite (PbCO₃).