Efficient Adsorption of Pollutants from Aqueous Solutions by Hydrochar-Based Hierarchical Porous Carbons

Abstract

Three-dimensional hierarchical porous carbons (HPCs) created through hydrothermal carbonization and the subsequent chemical activation of miscanthus were tested as adsorbents of Pb²⁺ and methylene blue from the aqueous solution. The HPC pore structure was customized using various hydrochar precursors obtained through a longer reaction time and by adding acetic acid. HPC obtained from hydrochar derived from acetic acid’s addition exhibited the highest specific surface area due to a larger micropore volume. This adsorbent proved to be the most efficient in removing lead from aqueous solutions. The Langmuir isotherm best described the lead adsorption process onto HPC with q_m = 155.6 mg g⁻¹ and the pseudo-second-order kinetic model. HPC obtained from hydrochar produced with a longer reaction time exhibited improved methylene blue adsorption properties. The pseudo-second-order kinetic model and the Freundlich isotherm best described the experimental data. The theoretical maximum adsorption capacity for methylene blue was 316.0 mg g⁻¹. The type of hydrochar significantly impacted the yield and physical structure of HPCs, while having a lesser effect on the composition of surface functional groups. The results revealed the binding mechanism of each pollutant, highlighting the importance of biomass pretreatment on the structure of the resulting HPC and its effectiveness in water purification.

1. Introduction

The urgent need to address environmental pollution has led to the investigation of renewable, cost-effective, and eco-friendly materials with a high adsorption potential and adjustable surface chemistry. In recent times, there has been a significant interest in the development of three-dimensional biomass-based hierarchical porous carbons (3D HPCs) due to their large surface area, favorable pore architecture, and abundance of functional groups [1,2,3]. Researchers have been primarily focusing on producing porous carbon (PC) materials from agro-industrial resources for several reasons. Firstly, the large-scale production of activated carbons requires a significant amount of inexpensive material with minimal transportation costs. Secondly, the global safe recycling movement and the reduction of water, air, and soil pollution necessitate the cessation of harmful substances emission caused by poor waste management. Thirdly, PCs can be controllably produced with one-dimensional, two-dimensional, and three-dimensional structures by inheriting the bio-template structure of the biomass. Lastly, N-containing precursors and minerals in plant biomass allow for the preparation of activated carbons without expensive activators [4,5,6].

The improvement of PCs’ functionality is focused on three main aspects: expanding the specific surface areas (SSAs), tailoring the pore size, and modifying the surface chemistry. Increasing the dimensionality can enlarge the active surface area, which can enhance the adsorption abilities of PC. The biomass precursor is a crucial factor, as it affects the final carbon yield and structural characteristics. Additionally, the choice of synthesis method plays a significant role in optimizing the microstructure, SSA, and porosity, which are necessary for achieving the high adsorption performance of water pollutants [7,8,9,10].

In this study, high-performance HPCs are produced through hydrothermal carbonization (HTC) and subsequent chemical activation of the feedstock [11,12,13]. HTC is a process of biomass transformation to carbon-rich material called hydrochar (HC), using water as the medium in a closed autoclave at moderate temperatures (180–250 °C) and autogenic pressure. HTC produces solid HC, a carbon-rich material, along with some HTC process water (PW) and a negligible amount of gas [14]. HCs have a unique microsphere structure, with a hydrophobic core and hydrophilic shell rich in reactive oxygenated functional groups (OFGs) like carbonyl and hydroxyl/phenolic. They are energy-dense solid fuels that show promise for applications as adsorbents and precursors for engineered/modified PC production [15,16,17,18]. The second step in preparing high-performance HPCs is chemical activation, where the HC and the activator are mixed to ensure full contact. The molten mixture is then converted into HPC through a complex chemical reaction under high temperatures. Potassium hydroxide (KOH), the most common alkaline activator, promotes the formation of micropores with a large SSA [19,20,21].

It should be emphasized that the pre-carbonization of biomass using HTC has a significant impact on the morphology and porous structure of the produced HPC. This is especially true for the formation mechanism of the 3D hierarchical pore structure. Studies have suggested that the reactive OFGs in HCs play a crucial role in fabricating the 3D honeycomb-like porous architecture [22]. It has been found that the direct KOH activation of some biomasses can lead to complete burnout or the formation of PCs with an irregular structure, inferior pore size distribution, and low SSA [22,23]. On the other hand, hydrothermally carbonized biomass has been successfully converted into HPCs [7]. When HC is mixed with KOH and heated, OFGs react with KOH, producing a molten mixture of corresponding potassium salts and HC. As the temperature increases, these salts and HC begin to pyrolyze, releasing CO, CO₂, H₂O, and other gases. This causes the molten mixture to expand, similar to bread leavening, which generates a porous structure like a soft template. The continuous pore structure forms carbon walls through the pyrolysis of HC, while KOH etches the carbon walls, creating numerous micropores. Additionally, the literature suggests that HCs obtained at 180 °C contain the highest amount of reactive OFGs [24]. The utilization of acetic acid as a modifier in HTC is an effective approach for enhancing the performance of HPCs [12]. The resulting 3D HPCs exhibit excellent mechanical strength, thermal stability, and adsorption properties, making them suitable for use in a wide range of other industrial applications (i.e., CO₂ capture, lithium batteries, supecapacitor electrodes, electrocatalytic oxygen reduction, etc.).

This study aimed to obtain different 3D HPCs through a two-step thermochemical conversion of miscanthus (MIS). Although biomass has a limited impact on overall energy production, energy crops have experienced rapid development and expansion. Among these, Miscanthus × giganteus Greef et Deu, a sterile hybrid genotype, has been found to exhibit remarkably robust growth. Utilizing this biomass source is eco-friendly and efficient, relying on continuous production from renewable crops [25]. A thorough analysis of the structure, morphology, and physicochemical properties of miscanthus, HCs, and 3D HPCs materials will help optimize the two-step conversion process and provide insight into the chemical changes happening during treatment. For the first time, this study evaluated the adsorption abilities of miscanthus-based 3D HPCs for removing Pb²⁺ and methylene blue (MB) from wastewater, studying their adsorption kinetics and equilibrium to select the best materials for practical use in wastewater treatment.

2. Materials and Methods

Miscanthus × giganteus Greef et Deu was harvested from the experimental field of Institute INEP, located in Zemun, Serbia. The collected miscanthus plants were air-dried and then ground to homogenize them. Fractions below 0.5 mm were used for further experiments. All chemicals of analytical grade, such as acetic acid (CH₃COOH), potassium hydroxide (KOH), hydrochloric acid (HCl), lead nitrate (Pb(NO₃)₂), and methylene blue, were purchased from Sigma-Aldrich, Inc. (Merck KGaA, Darmstadt, Germany).

2.1. Preparation of the HCM and C-HCM

The hydrochars were prepared using a 250 mL high-pressure laboratory autoclave (Carl Roth, Karlsruhe, Germany, Model II) with a temperature controller and thermocouple. To ensure a uniform mixing of the suspension during carbonization, each reaction load consisted of 10 g of biomass and aqueous phase in a 1:15 ratio. The suspensions were heated in an HTC autoclave from room temperature to 180 °C (at a rate of 4 °C min⁻¹) while being stirred at 150 rpm to prevent overheating. Biomass and distilled water were mixed and heated for 12 h to prepare HCM. C-HCM was prepared by mixing biomass with a mixture of distilled water and acetic acid (at 70/30 vol% ratio) and being carbonized for 2 h. After the solid and liquid products were separated through filtration, the resulting miscanthus hydrochars (HCM and C–HCM) were thoroughly washed with ultrapure water five times to neutral pH, and dried at 105 °C until a constant mass of hydrochar was achieved, before storage and further experiments. All experiments were repeated three times.

2.2. Preparation of the HPCM and C-HPCM

An amount of 5 g of the obtained HCM and C-HCM was mixed with KOH (an activating agent) at a ratio of 1:1 (1 g of KOH per 1 g of HC) in 60 mL of distilled water. The suspensions were oscillated for an hour (220 rpm) until they became homogeneous, then dried in an oven at 80 °C. Next, the mixtures were transferred to porcelain boats and heated in a furnace (Nabertherm 30–3000 °C, Lilienthal, Germany) at 800 °C (heating rate 15 °C min⁻¹) for 1 h under a constant Ar flow. After cooling down to room temperature, the resulting HPCs were washed with diluted HCl, and then thoroughly cleaned with boiled distilled water until the pH was neutral. Finally, the products were dried and labeled as HPCM and C-HPCM.

2.3. Characterization Methods

An elemental analysis of the samples was conducted using a Vario EL III CHNS Analyzer in three repetitions. The content of oxygen was calculated by subtracting the C, H, N, S, and ash content from 100%. The textural properties of all materials were determined based on N₂ adsorption–desorption isotherms obtained at −196 °C using a Sorptomatic 1990 Thermo Finnigan (Thermo Fisher Scientific, Waltham, MA, USA) apparatus. Before measurement, the HCM and C-HCM samples were degassed at the preparation site for 4 h at room temperature under a vacuum, and then for another 24 h at 80 °C under a vacuum pressure better than 0.3 Pa. HPCM and C–HPCM samples were prepared similarly but degassed for 24 h at 200 °C at the same residual pressure. The N₂ isotherms were analyzed using ADP 5.13 Thermo Electron Software. The specific surface areas (SSA_BET) of the samples were calculated using the Brunauer–Emmett–Teller (BET) equation [26] from the part of the adsorption isotherms selected according to the Rouquerol criteria [27]. The total pore volume (V_tot) of the samples was determined using the Gurevitsch rule [26] at p/p₀ = 0.98 (where p and p₀ denote the equilibrium and saturation nitrogen pressures at the adsorption temperature). The Dubinin–Radushkevich method [28] was used to determine the micropore volume (V_micro-DR). The mesopore volume (V_meso) was obtained from the adsorption branch of the isotherm for the p/p₀ region corresponding to the mesopores. The infrared (IR) spectra of the samples were obtained using a Thermo Scientific Nicolet iS50 FT-IR spectrometer in transmission mode. KBr pastilles were made for each sample by combining 0.8 mg of the sample and 80 mg of KBr. The spectra were recorded at a resolution of 4 cm⁻¹ and over a spectral range of 4000 to 400 cm⁻¹ by performing 32 scans. Scanning electron microscopy with energy-dispersive spectrometry (SEM–EDS analysis) was conducted using a JEOL JSM-7001F field-emission scanning electron microscope from Tokyo, Japan. This was coupled with an Oxford Instruments Xplore 15 energy-dispersive X–ray spectrometer from Abingdon, UK, operating in high vacuum mode (approximately 10⁻⁴ Pa). The samples were coated with a 15 nm thick layer of electrically conductive gold. For SEM imaging, the experiment used an accelerating voltage of 30 kV, a probe current of approximately 0.5 nA, and a working distance of 6 mm.

2.4. Water Decontamination Assays

The initial adsorption efficiency assessment of HCM, C–HCM, HPCM, and C–HPCM was determined based on their experimental adsorption capacities for Pb²⁺ and MB from aqueous solutions in a batch system. To do this, 20 mg of the char particles was dispersed in 30 mL of 200 mg L⁻¹ Pb²⁺ or MB solution and shaken (at 220 rpm) for 24 h at 25 °C. After shaking, the chars were separated by filtration. A PinAAcle 900T Atomic Absorption Spectrometer (Perkin Elmer, Waltham, MA, USA) was used to analyze the residual solution for Pb²⁺ concentrations. The UV–VIS Spectrophotometer SPEKOL 1300 (Analytik Jena GmbH, Jena, Germany) was used to analyze MB concentrations in the residual solution at 664 nm.

The equilibrium adsorption capacity of the obtained carbonaceous materials for Pb²⁺ and MB was calculated by Equation (1):

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

(1)

where q_e represents the mass of the adsorbed pollutant (mg g⁻¹), while C_o and C_e denote the initial and final concentrations of contaminants (mg L⁻¹), respectively. m and V represent the weight of the adsorbent (g) and the volume of the solution (L), respectively.

The following study involved a comprehensive analysis of the most effective adsorbents for different pollutants, including an examination of the impact of the initial pH of the solution, various adsorption kinetics, and adsorption equilibrium models.

The Pb²⁺ and MB solution, both at a concentration of 200 mg L⁻¹, were adjusted to appropriate pH values using HCl or NaOH to study the impact of the initial pH on the adsorption process. The suspensions were shaken at 25 °C and 220 rpm for 24 h.

Kinetic assays were conducted using 200 mg L⁻¹ Pb²⁺ or MB solutions at 25 °C, and samples were collected at appropriate time intervals. Non-linear kinetic models, including the Lagergren pseudo-first-order (PFO) model Equation (2) [29], the pseudo-second-order (PSO) model Equation (3) [30], and the Weber–Morris intra-particle diffusion model Equation (4) [31], were applied to the experimental data.

$$q_t=q_e(1-e^{-k_{1}t})$$

(2)

$$q_t={\displaystyle \frac{{q_e}^2{k}_{2}t}{1+q_e{k}_{2}t}}$$

(3)

$$q_t=K_{diff}{t}^{0.5}+C$$

(4)

The symbols in Equations (2)–(4), q_e and q_t (mg g⁻¹), represent the observed adsorption capacities at equilibrium and at time t (min), while k₁ (min⁻¹) and k₂ (g mg⁻¹ min⁻¹) are the pseudo-first-order and pseudo-second-order rate constants, respectively. K_diff (mg g⁻¹ min^−0.5) represents the intra-particle diffusion rate constant, and C (mg g⁻¹) is the intercept.

The adsorption equilibrium and maximum adsorption capacities of the selected HPCs were determined using Pb²⁺ solutions with initial concentrations ranging from 20 to 200 mg L⁻¹ and MB solutions with initial concentrations ranging from 50 to 500 mg L⁻¹ at 25 °C for 24 h. The Langmuir [32], Freundlich [33], and Sips [34] isotherms were used to determine the most accurate adsorption model based on the experimental data. The non-linear forms of these isotherm models are expressed by Equations (5)–(7), respectively.

$$q_e={\displaystyle \frac{q_m{K}_L{C}_e}{1+K_L{C}_e}}$$

(5)

$$q_e=K_F{C_e}^{1/n}$$

(6)

$$q_e={\displaystyle \frac{q_m{K}_s{C}_e^{n_s}}{1+K_s{C}_e^{n_s}}}$$

(7)

where q_e represents the adsorption capacity at equilibrium (mg g⁻¹), q_m represents the maximum quantity of adsorbed pollutant (mg g⁻¹), C_e represents the equilibrium concentration of the adsorbate (mg L⁻¹), K_L (L mg⁻¹) is the Langmuir empirical constant related to energy of adsorption, K_F (mg g⁻¹) represents the Freundlich constant that refers to the adsorption capacity, and K_S represents the Sips equilibrium constant (L mg⁻¹), while n and n_s are the model exponents that indicate adsorption intensity.

Origin 9.0 software and non-linear fitting methods were used to investigate the adsorption kinetics and isotherm models. All experiments were carried out three times to ensure accuracy.

3. Results

3.1. Characterization of the Carbonaceous Materials

Table 1. Elemental analysis of the samples.

	(wt %)
Sample	Yield	Ash	Elemental Composition *
			C	H	O **	N
HCM	49.2 ± 0.6	1.49 ± 0.01	65.4 ± 0.04	5.3 ± 0.04	27.4 ± 0.04	0.37 ± 0.04
C-HCM	46.6 ± 0.7	2.05 ± 0.02	52.6 ± 0.03	7.1 ± 0.04	38.0 ± 0.02	0.22 ± 0.03
HPCM	31.7 ± 0.3	2.54 ± 0.02	81.2 ± 0.07	1.9 ± 0.01	13.6 ± 0.01	0.80 ± 0.01
C-HPCM	15.9 ± 0.4	4.62 ± 0.03	73.2 + 0.09	3.0 ± 0.05	18.7 ± 0.08	0.42 ± 0.02

Note(s): * Sulfur was not detected in the samples, ** Oxygen was assessed by the difference.

shows the elemental composition of the synthesized HCs and HPCs. The carbon content increased to 81.2% and 73.2% in HPCM and C–HPCM, respectively, compared to 65.4% and 53.0% in the HC precursors. Conversion at higher temperatures promoted the dehydration and decarboxylation of lignocellulosic components in the HCMs, resulting in a halving of the hydrogen and oxygen content, while the nitrogen content in the HPCs doubled. These findings align with previous research on the elemental analysis of biomass-based 3D HPC [35,36]. Furthermore, the deposition of impurities and water-soluble components during conversion, especially in the sample hydrothermally treated in the presence of acetic acid [12], led to reduced mass yields of the HPCM and C–HPCM to 31.7% and 15.9%, respectively, and increasing the ash content in the latter.

The SSA_BET and pore volumes for the HCs and HPCs are provided in

Table 2. The textural properties of the samples.

Sample	SSABET (m2 g−1)	Vmic-DR (cm3 g−1)	Vmeso (cm3 g−1)	Vtot (cm3 g−1)
HCM	7.2	0.003	0.021	0.031
C-HCM	4.3	0.002	0.021	0.029
HPCM	705	0.271	0.031	0.295
C-HPCM	1040	0.404	0.040	0.438

. Both hydrochars produced at 180 °C exhibit a relatively low SSA and pore volumes, regardless of the carbonization period and the addition of acetic acid. Specifically, the SSAs were 7.2 m² g⁻¹ and 4.3 m² g⁻¹, for HCM and C-HCM, respectively. These values are very similar to the previously published SSA for rapeseed, whitewood, and brown seaweed hydrochars obtained at similar HTC temperatures [8]. On the other hand, KOH activation and subsequent pyrolysis dramatically increased the SSA_BET and pore volume of the obtained HPCs. Specifically, HPCM showed a surface area of 705 m² g⁻¹ and a micropore volume of 0.271 cm³ g⁻¹, while C-HPCM showed a surface area of 1040 m² g⁻¹ and a micropore volume of 0.404 cm³ g⁻¹. The mesopore volume was 0.031 cm³ g⁻¹ and 0.040 cm³ g⁻¹ for HPCM and C–HPCM, respectively. When comparing the value of V_mic, it has been found that almost all the increase in SSA came from the increase in micropore volume. The microporous structure in the sample mainly resulted from the self-crosslinking reaction of the benzyl group between the benzene rings. The greater the degree of crosslinking between the hydrochar microspheres, the denser the crosslinked bridge structure, leading to the formation of more micropores [10]. The results confirmed that the addition of acetic acid during HTC promotes the formation of more regular HC microspheres, resulting in a higher micropore volume and a higher specific surface area for the material after activation (C–HPCM).

The FT-IR spectra of the analyzed samples are displayed in Figure 1., encompassing the relevant spectral regions (4000–400 cm⁻¹). The FT-IR results are comparable to those from the elemental analysis, as listed in Table 1. The levels of hydrogen and oxygen decreased during the activation process. The prominent bands of the MIS, HCM, and C-HCM are located similarly to those previously observed in FT-IR spectra of miscanthus and its hydrochars obtained at comparable temperatures [23]. The high-temperature transformation of HCM and C-HCM to corresponding HPCMs is indicated by the reduction in the usual carbohydrates peaks (C-OH at ~3340 cm⁻¹, and -C-H stretching vibrations at ~2900 cm⁻¹). Cellulose (C-O) stretching peaks (at 1113 cm⁻¹, 1060 cm⁻¹, and 1030 cm⁻¹), asymmetrical C-O-C stretching (at 1163 cm⁻¹), and C-C stretching (at 670 cm⁻¹)), and peaks of fatty esters groups (O-C=O at ~1700 cm⁻¹, 1274 cm⁻¹, and 1163 cm⁻¹) disappeared in HPCs. Lignin-related groups (at ~1600 cm⁻¹ (aromatic C=C), aromatic skeletal vibration with C-H plane deformation at ~1500 cm⁻¹ and 1457 cm⁻¹, at 1314 cm⁻¹ (syringyl), and 1274 cm⁻¹ (guaiacyl)), also disappeared in HPCs, as well as the amorphous absorption band at 899 cm⁻¹ associated with the C-O-C stretching of β-glucosidic linkages, and the aliphatic C-H deformation vibrational band (at 1370 cm⁻¹) of cellulose and hemicellulose present in hydrochars. The remaining IR bands at 1560 cm⁻¹ and 1070 cm⁻¹ in HPCM and C-HPCM correspond to the C=C vibration of the condensed aromatic rings, and the C-C-O and C-O-C stretching of non-cyclic anhydrides, which are typical for biochars [1,8,37].

SEM micrographs of HCs and 3D HPCMs are depicted in Figure 2 to provide clear morphological representations of the samples. The image of the HCM shows the typical pores and channels of a lignocellulosic template. However, hydrothermal conversion in the presence of acetic acid promotes the formation of spherical carbon particles, which are visible in the C–HCM micrograph. During the high-temperature conversion process with KOH, the lignocellulosic template shrank and became densely stacked, while releasing a significant amount of gases, resulting in the formation of gas bubbles. In this way, the 3D foam-like structure of the thin carbon skeleton was formed. The HPCM image shows randomly opened macropores of different sizes and large amounts of uneven micropores. The SEM image of C−HPCM depicts the thin carbon skeleton of the micropores, resembling foam-like honeycomb structures. These findings align with the textural properties of the materials under examination.

3.2. Heavy Metal and Dye Adsorption Study

3.2.1. Preliminary Tests

The above results show that HPCs are materials abundant in micropores and active adsorption sites. Furthermore, HPCs have a porous structure with interconnected pores, significantly enhancing their ability to function as an effective adsorbent with a high adsorption performance. Therefore, it is not surprising that in a preliminary experiment to test the adsorption abilities of the samples, the C−HPCM and HPCM samples were found to have the best potential for removing Pb²⁺ and MB from water solutions, respectively (refer to Figure 3). For more details on each sample’s adsorption process and mechanism, please see the relevant subsections below.

3.2.2. Batch Experiment on Pb²⁺ Removal Using C−HPCM

Figure 4a illustrates how the pH of a solution affects the absorption of Pb²⁺ using C−HPCM. In acidic solutions, lead primarily exists as Pb²⁺, and it forms a soluble monovalent hydroxyl complex (PbOH⁺) at pH 4.9 [17]. Notably, an insoluble Pb(OH)₂ precipitate was previously observed at pH 5.8 [1,15]. Therefore, all lead absorption experiments are conducted at pH = 5.5 to prevent potential precipitation. The results showed that C−HPCM can absorb more Pb²⁺ as the pH level increases from 2 (13.5 mg g⁻¹) to 5.5 (70.5 mg g⁻¹). At higher acidic pH levels, the surface of the C–HPCM becomes more negatively charged, allowing it to attract more Pb²⁺ and (PbOH⁺) cations. These findings are consistent with previously published research on biomass-based HPCs’ absorption capacity for lead [1]. Figure 4b,c display the fitting results of the adsorption process using the PFO kinetic model [29], PSO kinetic model [30], and Weber–Morris intraparticle diffusion model [31] applied to the experimental data.

Table 3. Kinetic parameters for Pb²⁺ adsorption onto C–HPCM.

qeq.exp (mg g−1)			70.5
Pseudo-First-Order Model		Pseudo-Second-Order Model
qeq.cal (mg g−1)	65.4	qeq.cal (mg g−1)	69.2
k1 (min−1)	0.09	k2 (g mg−1 min−1)	0.002
R2	0.900	R2	0.955
Weber–Morris diffusion Model
Kdiff1 (mg g−1 min−1/2)	1.64	Kdiff2 (mg g−1 min−1/2)	0.188
C1 (m g−1)	44.5	C2 (m g−1)	63.3
R2	0.854	R2	0.994

summarizes the obtained kinetic parameters.

The kinetic model using PSO demonstrated a higher correlation coefficient (R²) compared to the PFO model, with an R² exceeding 0.95. The experimental equilibrium adsorption capacity (q_eq.exp of 70.5 mg g⁻¹) was very close to the theoretical adsorption capacity (q_eq.cal of 69.2 mg g⁻¹) calculated by the PSO kinetic model. This suggests that the PSO kinetic model was more suitable for describing the adsorption process of Pb²⁺ by C–HPCM. This also agrees with the previously published results on the kinetics of lead adsorption on similar materials [38,39]. This model suggests that the chemisorption of Pb²⁺ ions on the surface of the adsorbent significantly contributes to the overall removal process [30]. To better understand the potential impact of diffusion on Pb²⁺ adsorption by C–HPC, the Weber–Morris intra-particle model was applied to obtained experimental data showing the two linearity zones (Figure 4c). This result suggests that the diffusion of Pb²⁺ from the solution onto C–HPCM was not solely due to the rate-controlling step, as other mechanisms are also involved [17,20]. The initial rapid adsorption of Pb²⁺ took place on the external surface of C–HPCM, followed by intraparticle diffusion, and the slow movement of Pb²⁺ into micropore active sites.

The Langmuir, Freundlich, and Sips isotherm models were used to analyze the impact of the initial Pb²⁺ concentration on the removal effectiveness of C–HPCM and to assess the equilibrium adsorption capacity of the pollutant based on the experimental data. Figure 5a displays the non-linear curves of the applied isothermal models. The fitting results are outlined in

Table 4. Fitting parameters of isothermal models for Pb²⁺ adsorption onto C–HPCM (at 25 °C).

Langmuir Model		Freundlich Model		Sips Model
qm (mg g−1)	155.6	KF (mg(1−1/n) L1/n g−1)	2.32	qm (mg g−1)	141.2
KL (L mg−1)	0.006	n	1.43	Ks (L mg−1)	0.009
				ns	1.12
R2	0.994	R2	0.988	R2	0.993

. The correlation coefficient of the Langmuir model (R² = 0.994) was higher than that of the Freundlich and Sips models. This result shows that the active adsorption sites on C–HPCM are homogeneous, and the Pb²⁺ is adsorbed at these sites in a monolayer [32]. The theoretical maximum adsorption capacity (q_m) of C–HPCM for Pb²⁺ was 155.6 mg g⁻¹, as determined by the Langmuir isotherm model. The q_m value of C–HPC for Pb²⁺ is significantly higher than that of most other biomass-based HPC adsorbents, such as activated cork (74.3 mg g⁻¹) [1], activated coconut shell (32.1 mg g⁻¹ and 92.4 mg g⁻¹) [40], and activated soybean cake (89.7 mg g⁻¹) [39].

The FT–IR analysis of C-HPCM conducted before and after Pb²⁺ adsorption revealed changes in the spectra between 1400 and 400 cm⁻¹ (Figure 5b). After adsorption, a strong and distinct peak emerged at 1384 cm⁻¹, attributed to the C-H deformation vibration of conjugated molecules. However, the peak at 943 cm⁻¹ (C-H out-of-plane deformation vibration) [41] diminished. Additionally, more prominent peaks at 1070 cm⁻¹ (C-C-O and C-O-C stretching vibration) and 789 cm⁻¹ (CH₃-metal groups due to CH₂ rocking vibration) [39], were observed on the C–HPCM surface after adsorption. These changes suggest that lead is likely binding due to cation-π interaction and complexation of Pb²⁺ ions by the functional groups [39,42].

3.2.3. Batch Experiment on MB Removal Using HPCM

The graph in Figure 6a shows how the pH of a solution affects the absorption of MB by HPCM. The results indicate that HPCM can absorb slightly more MB as the pH increases from 2.0 (233 mg g⁻¹) to 12.0 (268 mg g⁻¹). The highest absorption was observed at a pH of 7.0 (274.4 mg g⁻¹). When the pH of the solution ranged from 2.0 to 12.0, the functional groups on the surface of the HPCM became negatively charged, while the cationic dye methylene blue transformed into cationic MB⁺. This led to stronger electrostatic attraction between the negatively charged biochar surface and the cationic dye MB⁺ at higher pH values, resulting in greater adsorption capacity [19]. Nonetheless, the similar quantities of MB adsorbed at pH 2 and 12, despite the difference in value, suggest that electrostatic attraction was not the dominant adsorption mechanism of MB onto HPCM [9,36]. Figure 6b,c display the fitting results of the adsorption process using different kinetic models [29,30,31] applied to the experimental data.

Table 5. Kinetic parameters for MB adsorption onto HPCM.

qeq.exp (mg g−1)			274.4
Pseudo-First-Order Model		Pseudo-Second-Order Model
qeq.cal (mg g−1)	260.4	qeq.cal (mg g−1)	272.4
k1 (min−1)	0.063	k2 (g mg−1 min−1)	0.0004
R2	0.964	R2	0.996
Weber–Morris diffusion Model
Kdiff1 (mg g−1 min−1/2)	7.81	Kdiff2 (mg g−1 min−1/2)	0.149
C1 (m g−1)	158.1	C2 (m g−1)	268.7
R2	0.868	R2	0.997

summarizes the obtained kinetic parameters.

The PSO kinetic model demonstrated a higher correlation coefficient value than the PFO model, with R² values exceeding 0.99. The experimental equilibrium adsorption capacity (q_eq.exp of 274.4 mg g⁻¹) closely matched the theoretical adsorption capacity (q_eq.cal of 272.4 mg g⁻¹) calculated by the PSO kinetic model. Similar to the adsorption of Pb onto C–HPCM, these results suggest that the PSO kinetic model is more suitable for describing the adsorption process of MB. This finding is consistent with previously published results on the kinetics of MB adsorption on similar materials [8,19,36]. It supports chemisorption as the primary mechanism of the MB removal process using HPCM [30]. The experimental data were analyzed using the Weber–Morris intra-particle model, revealing two linearity zones (see Figure 6c), indicating that the reaction rate is not solely controlled by the diffusion of MB from the solution onto HPCM.

Figure 7a shows the non-linear curves of the applied isothermal models. The fitting results are listed in

Table 6. Fitting parameters of isothermal models for MB adsorption onto HPCM (at 25 °C).

Langmuir Model		Freundlich Model		Sips Model
qm (mg g−1)	316.0	KF (mg(1−1/n) L1/n g−1)	151.1	qm (mg g−1)	685.9
KL (L mg−1)	3.10	n	5.79	Ks (L mg−1)	0.0002
				ns	0.174
R2	0.836	R2	0.981	R2	0.977

. The correlation coefficient of the Freundlich model (R² = 0.981) was higher than that of the Langmuir and Sips models. The Freundlich isotherm describes adsorption/sorption onto heterogeneous adsorbate surfaces with different site energies. The exponent value of n > 1 indicates the potential for heterogeneous adsorption involving physical binding forces [43,44]. The results for HPCM adsorption demonstrate strong agreement with the previously studied mechanisms of potential heterogeneous MB adsorption associated with cation-π interaction, pore filling, and π-π stacking [1,8,43,44].

The FTIR analysis of HPCM conducted before and after adsorption confirms MB’s presumed binding mechanism, revealing changes in the spectra between 1400 cm⁻¹ and 400 cm⁻¹ (Figure 7b). After adsorption, a strong and distinct peak emerged at 1384 cm⁻¹ and 1324 cm⁻¹, attributed to C-H deformation vibration in conjugated molecules [25,41]. Additionally, decreases in the intensity of the peaks at 1560 cm⁻¹ (C=C vibration of the condensed aromatic rings) and 1070 cm⁻¹ (C-C-O and C-O-C stretching vibration), and the appearance of a peak at 883 cm⁻¹ (out-of-plane-deformation of aromatic C-H atoms) [41], were observed on the HPCM surface due to surface interaction with MB [45]. These changes indicate that MB likely binds to HPCM through cation-π interaction, π-π-stacking, and pore-filling mechanisms.

4. Discussion

In this study, a two-stage thermochemical process was used to create novel high-performance HPCs from miscanthus. The feedstock was first subjected to HTC pretreatment at 180 °C for 12 h, followed by alkaline pyrolysis activation, resulting in the production of HPCM. In the second procedure, acetic acid was added during the two-hour HTC at 180 °C, before alkali pyrolysis, producing C–HPCM. The elemental, BET, SEM, and FT–IR analysis showed that the differences in HTC pretreatment significantly impacted the yield and the physical structure of the HPCs, while having a lesser effect on the composition of surface functional groups. The addition of acetic acid in the pretreatment increased the specific surface area and micropore content of the obtained C–HPCM. Huang et al. reached a similar conclusion in their earlier research [12]. On the other hand, HPCM was characterized by a higher degree of condensation of carbohydrate molecules and a higher yield. FT−IR analysis of miscanthus HPCs to a certain extent confirmed previous general conclusions that HPC refers to pyrolytic carbons with minimal functional groups that primarily interact with other molecules through physical interactions [36,46]. Moreover, the heavy metal sorption capacity of biochars, except the temperature of carbonization, was highly related to the ash and mineral element content [47]. The results obtained in this study confirmed that the superior adsorption coefficient of C−HPCM for Pb²⁺ can be associated with these earlier findings. C–HCPM stands out for its higher specific surface area, micropore content, and ash content in comparison to the other samples obtained. Therefore, the efficient adsorption of Pb²⁺ by C–HPCM primary occurs through electrostatic interactions between positive metal species and the negative functional groups of C–HPCM, the ion exchange mechanism, and the sorbent’s pore filling [38,48,49]. The pseudo-second-order kinetic model and the Langmuir isotherm best described the adsorption process. The theoretical maximum adsorption capacity (q_m) of C−HPCM for Pb²⁺ was 155.6 mg g⁻¹, as determined by the Langmuir isotherm model. This is, on average, a higher lead removal efficiency compared to similar materials previously reported (see

Table 7. The adsorption coefficients of different biomass-based HPCs for Pb²⁺ and MB removal from aqueous solution at 25 °C.

Adsorbents	Contaminants	qm * (mg g−1)	SSABET (m2 g−1)	Reference
Activated cork	Pb2+	74.3	1853	[1]
Coconut shell	Pb2+
GCN816G	Pb2+	32.08	N/A	[40]
GCN1240	Pb2+	92.39	1150	[40]
Cork-based CAC@CDs-BPEI	Pb2+	231.48	1089	[39]
C−HPCM	Pb2+	155.6	1038	This study
Soybean cake	Pb2+	133.6	N/A	[38]
Rice husk	Pb2+	261.2	1040	[50]
Bamboo	MB	67.46	562	[19]
Activated cork	MB	887.7	1853	[1]
HPCM	MB	316.0	705	This study

Note(s): * Based on the theoretical maximum adsorption capacity determined by the Langmuir isotherm models.

).

On the other hand, HPCM achieved the highest adsorption coefficient for MB. The pseudo-second-order kinetic model and the Freundlich isotherm best described the adsorption process of MB removal. This indicates that the process of adsorption implies, to the greatest extent, chemical adsorption of MB on a heterogeneous surface. The theoretical maximum adsorption capacity (q_m) of HPCM for MB was 316.0 mg g⁻¹, as determined by the Langmuir isotherm model. Previous studies on the adsorption of MB using similar materials have confirmed that the main mechanisms include cation-π interaction between the MB and the surface, as well as pores filling [1,8,18,19]. However, the HPCM adsorbed MB somewhat more efficiently, due to its higher aromatic graphite-type structure and more effective π-π stacking which was confirmed by FT−IR analysis.

Compared to the other HPC adsorbents listed in Table 7, synthesized 3D HPCs made from miscanthus have shown favorable adsorption capacities for Pb²⁺ and MB. This is mainly due to their large specific surface area and pore volume, indicating that 3D HPC materials have great potential for treating different kinds of wastewater pollutants. This study also showed that optimizing the HTC pretreatment greatly influences the tailoring of the structure of the obtained HPC. The possibilities of obtaining active precursors of HPC for specific applications should be further investigated.