Dynamic Adsorption of Low-Concentration Formaldehyde Using One-Step Synthesized Amino-Functionalized Hyper-Cross-Linked Ionic Copolymers

Abstract

Formaldehyde (FA) emissions seriously influence the environment and human health, while traditional adsorbents are restricted by low capacity and poor selectivity. To address these limitations, amino-functional hyper-cross-linked copolymer ionic compounds (HPIL-Cl-Xs) were designed and synthesized through a one-step hyper-cross-linking and quaternization reaction involving benzimidazole, dichloro-p-xylene, and functional monomers. These polymers provide an ionic environment, active adsorption sites, and a microporous structure, offering abundant adsorption sites. The synthesis parameters were studied to optimize the preparation conditions. Under conditions of 8.6 ppm and WHSV of 54,000 h^–1, the equilibrium adsorption capacity of HPIL-Cl-Phe (phenylalanine) reached 11.3 mg/g with a partitioning coefficient (PC) of 0.44 mol·kg^–1·Pa^–1, surpassing that of conventional adsorbents. The impacts of the adsorption temperature, WHSV, and relative humidity on adsorption were explored, confirming the adaptability of HPIL-Cl-Xs to various environmental conditions. DFT calculations, XPS, and FT-IR confirmed the existence of hydrogen bond interactions and nucleophilic addition reactions. HPIL-Cl-Phe demonstrated an excellent cycling performance with stable adsorption over multiple cycles.

1. Introduction

Formaldehyde (FA), a common volatile organic compound (VOC), is widely present in indoor air and is also emitted by various industries such as pharmaceuticals, furniture manufacturing, and printing. This caused significant environmental concerns and poses health risks to humans. Traditional methods for FA removal include catalytic oxidation, , photocatalytic oxidation, , combustion, biotechnology, and adsorption. , Among these, adsorption is widely adopted due to its low energy consumption, operational simplicity, and broad applicability. Activated carbon , and molecular sieves , are commonly used adsorbents, exhibiting relatively good adsorption performance toward nonpolar or weakly polar conventional VOCs, such as benzene derivatives. However, their adsorption capacity and selectivity for FA remain comparatively low. Surface modification and property adjustments have shown limited effectiveness in improving FA adsorption. , Moreover, the actual dynamic adsorption capacity is significantly lower than the static capacity. MOFs (Metal–Organic Frameworks), POPs (Porous Organic Polymers), , and other new materials had pretty good adsorption capacities for common VOCs. However, their synthesis processes are complex, and their applications for FA removal remain limited. Most current FA adsorbents are primarily evaluated under high-concentration conditions, despite FA typically occurring in low-concentration environments. The performance of these materials under high-concentration settings may not accurately represent their efficacy in real-world, low-concentration scenarios. Therefore, it is crucial to develop novel adsorbents and assess their adsorption performance under conditions that reflect low FA concentrations and high flow velocities.

POPs possess a high specific surface area, tunable pore structures, and excellent physicochemical stability, making them suitable for applications such as gas storage, heterogeneous catalysis, organic photovoltaics, and separation. , POPs can be categorized into hyper-cross-linked polymers (HCPs), covalent organic frameworks (COFs), and conjugated microporous polymers (CMPs). , The highly cross-linked polymer chains in HCPs provide structural rigidity, preventing skeletal collapse and maintaining a permanent porous architecture. With outstanding thermal and chemical stability, low production costs, and wide applicability, HCPs are extensively utilized in adsorption, catalysis, and electronics. Their specific surface area and pore size distribution can be flexibly adjusted by varying the types and ratios of monomers and cross-linking agents. − Introducing dynamic ionic pairs enables the creation of tailored charge environments, enhancing the selective interaction of POPs with target compounds. , Functionalization with nitrogen- or oxygen-containing groups significantly improves formaldehyde (FA) adsorption performance, − where O^δ− acts as a hydrogen bond acceptor and C^δ+ serves as an electrophilic reagen, ^{, ,} Incorporating these groups into HCPs as active adsorption sites, along with the establishment of an ionic environment, enhances FA adsorption capacity. Functionalization can be achieved through either postmodification or presynthesis approaches. Postmodification involves chemical alterations after HCP formation, although the introduction of additional groups may limit pore accessibility and reduce structural flexibility. In contrast, presynthesis integrates functional molecules into HCPs during polymerization, enabling precise control over the content and spatial distribution of functional groups by adjusting the precursor structure, while maintaining a relatively simple synthetic procedure.

Hyper-cross-linked ionic polymers (HPILs) integrate the features of microporous polymers and ionic materials by embedding IL-like ionic sites into the polymer backbone. Phenylic ILs and imidazole-based compounds capable of undergoing Friedel–Crafts alkylation and quaternization reactions can act as monomers for HPILs. , The synthesized HPILs exhibit a well-developed meso-microporous structure with tunable IL content. However, anionic functionalization of IL monomers may reduce the reactivity during the hyper-cross-linking process. Postsynthesis modifications, such as anion exchange, can be incomplete due to steric hindrance effects. Moreover, these functionalization strategies often involve lengthy procedures and complex synthesis steps. Alternatively, phenylic compounds with targeted functional groups can be used as monomers in a one-step hyper-cross-linking reaction between benzimidazole and a cross-linking agent, yielding HPILs that combine an ionic polymer matrix with a high specific surface area and a microporous architecture. The high BET surface area and dominant microporosity of these HPILs enhance FA accessibility, while the embedded ionic pairs create a favorable ionic environment that promotes the adsorption process. Our previous study demonstrated that amino acid groups serve as effective active sites for FA adsorption. Our previous research has demonstrated that amino groups, such as those found in amino acids, act as effective active sites for FA adsorption, significantly improving its adsorption capacity. Therefore, the development of a novel FA adsorbent featuring a high specific surface area, a micromesoporous architecture, and integrated amino groups through a one-step hyper-cross-linking approach holds considerable significance.

Herein, we prepared the synthesis of a series of novel amino-functionalized hyper-cross-linked ionic copolymers (HPIL-Cl-Xs) that integrate both an ionic environment and amino functional groups, endowing them with strong hydrogen-bonding basicity and nucleophilic addition capability. These materials were fabricated via a direct one-step Friedel–Crafts alkylation reaction between benzimidazole (BMZ), dichloro-p-xylene (DCX), and amino-functionalized monomers (FM). The resulting HPIL-Cl-Xs possess a well-developed micromesoporous architecture, a distinct ionic environment, and tunable functional group density. We systematically investigated the influence of polymerization parameters and functional monomers on the composition, structural characteristics, and adsorption performance of the HPIL-Cl-Xs. Among the synthesized materials, HPIL-Cl-Phe exhibited significantly enhanced adsorption capacity compared to the nonfunctionalized HPIL-Cl and conventional adsorbents. The adsorption behavior under various operational conditions was also evaluated. To elucidate the underlying adsorption mechanism and assess the regeneration performance of HPIL-Cl-Phe toward formaldehyde (FA), comprehensive analyses were conducted by using XPS, FT-IR, and DFT calculations.

2. Experiment and Calculation

2.1. Chemicals

The chemicals used in this study are listed in Table S1, with their purity percentages provided by the suppliers as mass fractions. All chemicals were employed as received without further purification. Distilled water was used after being freshly distilled in a glass apparatus.

2.2. Synthesis of HPIL-Cl-Xs

The specific synthesis route of HPIL-Cl-Xs is shown in Figure . A defined mole ratio of BMZ, DCX, and functional monomers (such as PA, DCA, Phe, and Trp) was introduced into a Schlenk flask, followed by the addition of an appropriate amount of DCE to dissolve the reactants. The reaction was conducted at 80 °C for 24 h under a nitrogen atmosphere by using anhydrous iron­(III) chloride as the catalyst. Upon completion, the resulting product was filtered, washed, and dried. Synthesis details can be found in the Supporting Information. The obtained HPIL-Cl-Xs are HPIL-Cl-Phe, HPIL-Cl-DCA (2,5-Dichloroaniline), HPIL-Cl-PA (Phenylamine), and HPIL-Cl-Trp (Tryptophan).

1.

Synthesis routes of HPIL-Cl-Xs (X means different functional monomers, such as PA, Phe, Trp, and DCA).

2.3. Characterizations

The composition, structure, and physical properties of HPIL-Cl-Xs were investigated by elemental analysis (EA), Fourier transform infrared (FT-IR), N₂ adsorption and desorption experiment, Scanning Electron Microscopy (SEM), and Thermogravimetric analysis (TGA). The adsorption mechanism was characterized by FT-IR and X-ray Photoelectron Spectroscopy (XPS), and the specific test details are shown in the Supporting Materials.

2.4. Adsorption Experiments and Evaluation

The adsorption capacities of HPIL-Cl-Xs were evaluated by using a self-assembled microtubular reactor, as shown in Figure S1. The system consisted of a gas generation unit and a detection apparatus. The experimental procedure was as follows: 100 mg of HPIL-Cl-X was placed in a quartz tube with a 6 mm inner diameter through which FA gas was passed. The FA concentration was controlled by adjusting both the gas generation temperature and the flow rate of the dilution gas. Weight Hourly Space Velocity (WHSV) was varied by adjusting the dilution gas flow, while the relative humidity (RH) was controlled by bypassing water. FA concentration was measured using the MBTH spectrophotometry method, with detailed information provided in the Supporting Information. The adsorption capacities of HPIL-Cl-Xs were determined using eq .

$$Q=\frac{q\times {10}^{-6}\times {\int }_0^t(C_0-C_t)\mathrm{d}t}{m}$$ 1

The partitioning of FA between the gas and adsorbent (solid) phases can be expressed as the partitioning coefficient (PC) (eq ), which can be used to evaluate the affinity of FA for adsorbent sites at the partial pressure of free FA (molecular weight [MWt.] = 30.03 g/mol) in the gas phase as P (Pa).

$$\mathrm{PC}=\frac{Q}{{\mathrm{MW}}_{\mathrm{FA}}\times P}$$ 2

where Q represents the FA adsorption capacity (mg/g); Q _e is the equilibrium adsorption capacity (100%C_t /C ₀), in mg/g; Q _p is the penetrating adsorption capacity (1%C_t /C ₀), in mg/g; q denotes the inlet airflow, in mL/min; C ₀ and C_t are the inlet and outlet FA concentrations, in mg·m^–3; t is the adsorption time, in minutes; and m is the mass of the HPIL-Cl-X adsorbent (g); P is the partial pressure of FA, Pa.

2.5. DFT Calculations

The representative fragment was selected as a simplified HPIL-Cl-Phe, as shown in Figure S2. Geometry optimization and frequency analysis were performed at the B3LYP/6-31+g­(d,p) level on the ORCA software. , Due to abundant aromatic rings in model compounds, DFT-D3 dispersion corrections were also taken into consideration. Imaginary frequency was not found after geometry optimization and frequency analysis. Then visualization of noncovalent interactions was carried out based on wave function analysis, including ESP (electrostatic potential) isosurface and RDG (reduced density gradient) function. To be specific, the .gbw file obtained from ORCA was utilized, while several softwares (Multiwfn and VMD) were used in the subsequent analysis analysis ^,

Through an ESP isosurface analysis, potential adsorption sites can be identified. RDG analysis further elucidates these interaction sites, allowing for the evaluation of interaction types and strengths based on the color and intensity of the RDG surfaces. Two adsorption systems were created based on possible adsorption sites (−NH– in the BMZ ring and amino acid group in Phe), which was confirmed by the ESP isosurface. Based on visualization of noncovalent interactions in two adsorption systems, the strength of hydrogen bonds could be quantified by the core–valence bifurcation index (CVB) index. Electron localization function (ELF) was utilized to describe the localization of electrons. CVB can be obtained based on ELF and used to evaluate the strength of hydrogen bonds. CVB index can be calculated by eq .

$$\mathrm{CVB}\mathrm{index}=\mathrm{ELF}(\mathrm{C‐V})-\mathrm{ELF}(\mathrm{DH‐A})$$ 3

ELF­(C-V) denotes the ELF value at the bifurcation point between the core basin and the valence basin, while ELF­(DH-A) represents the ELF value at the bifurcation point between the valence basin of the hydrogen bond donor (V­(D-H)) and that of the acceptor (V­(A)).

2.6. Fitting of Dynamic Adsorption Models and Adsorption Kinetics Models

The FA adsorption behavior on HPIL-Cl-Xs in a fixed-bed system was assessed using the Boltzmann (eq 4) and Yoon–Nelson dynamic adsorption models (eq 5). To further elucidate the adsorption kinetics of FA, the pseudo-first-order (eq 6), pseudo-second-order (eq 7), Weber–Morris (W–M) kinetic models (eq 8), and Bangham model (eq 9) were employed. The respective kinetic equations are presented in Table .

1. Dynamic Adsorption Model and Adsorption Kinetics Model Formulas.

model	name	equation	no
dynamic adsorption model	Boltzmann	$\frac{C}{C_0}=A_2+\frac{A_1-A_2}{1+{\mathrm{e}}^{(\tau -t)/\mathrm{d}x}}$	eq 4
	Yoon–Nelson	$\ln \left(\frac{C}{C-C_0}\right)=k_1(t-\tau )$	eq 5
kinetic model	pseudo-first-order kinetic mode	$Q_t=Q_e(1-{\mathrm{e}}^{-k_{1}t})$	eq 6
	pseudo-second-order kinetic model	$Q_t=\frac{k_2{Q}_{\mathrm{e}}^{2}t}{1+k_2{Q}_{\mathrm{e}}t}$	eq 7
	Weber–Morris model	$Q_t=k_3{t}^{0.5}+C$	eq 8
	Bangham pore diffusion model	$Q_t=q_{\mathrm{e}}(1-{\mathrm{e}}^{-k_4{t}^z})$	eq 9

In the equations, A ₁, A ₂, τ, and dx are four constants with clear physical meanings. A ₁ and A ₂ are the final and initial values of C/C ₀; τ is the penetration time (C/C ₀ = 50%), min; dx describes the slope of the curve; Q _t is the adsorption capacity of FA at time t, in mg/g; K ₁, K ₂, and K ₃ are the rate constants for the pseudo-first-order kinetic model, pseudo-second-order kinetic model, and Bangham model, respectively, in h^–1; K _p is the internal diffusion rate constant, I n h ^–0.5; z and C are constants.

3. Results and Discussion

3.1. Preparation, Characterization, and Performance Evaluation of Series HPIL-Cl-Xs

A series of HPIL-Cl-Xs were obtained under the following conditions: reaction temperature of 80 °C, reaction time of 24 h, mole ratio of BMZ:FM:DCX as 1:1:2, catalyst as FeCl₃, and solvent as DCE. The FT-IR spectra of HPIL-Cl-Xs are presented in Figure a. The characteristic peak between 3300 and 3500 cm^–1 is attributed to the stretching vibration of the amino group, while the peak between 1400 and 1500 cm^–1 corresponds to the benzene ring skeleton. The peak near 2900 cm^–1 is associated with the stretching vibration of the −CH₂– bridge in the polymer backbone. The peak in the range of 1260–1290 cm^–1 is assigned to the stretching vibration of the C–N bond. For HPIL-Cl-Phe, the peak at 1605 cm^–1 is linked to the stretching vibration of the CO group in the carboxylate anion. The thermal stability of typical HPIL-Cl-Xs was evaluated by using TGA, and the results are shown in Figure S4. All HPIL-Cl-X samples exhibited relatively good thermal stability, with a minor weight loss of approximately 5% at 150 °C, attributed to the volatilization of a small amount of adsorbed water. The polymer skeleton begins to decompose gradually at temperatures above 200 °C. The structure, composition, and adsorption performance of HPIL-Cl-Xs were investigated using N₂ adsorption–desorption experiments, elemental analysis, and dynamic adsorption experiments (T _ad = 30 °C, WHSV = 54,000 h^–1, C _in = 8.6 ppm, RH = 0%). The results are presented in Figure b–d and Table . The adsorption/desorption isotherm curves for HPIL-Cl-Xs display a mixed Type I and IV profile. The significant increase in adsorbed quantity at P/P ₀ < 0.03 indicates the presence of abundant micropores. A higher uptake in the P/P ₀ range of 0.4 to 0.8, followed by a sharp rise at P/P ₀ > 0.8, suggests the presence of mesopores and macropores. Under identical preparation conditions, HPIL-Cl-Xs synthesized with different functional monomers exhibit markedly distinct specific surface areas and pore size distributions. Compared with HPIL-Cl, HPIL-Cl-PA shows an increased specific surface area and a reduced D _BJH of 2.9 nm. Yield analysis indicates that PA is minimally involved in the reaction, as the main processes are the hyper-cross-linked reaction of DCX and BMZ and the self-polymerization of DCX. Additionally, the specific surface areas of the other HPIL-Cl-Xs are lower than those of HPIL-Cl, likely due to the larger molecular volumes of the functional monomers restricting the formation of the pore structure. For instance, in Trp, the amino acid group is bonded to a nitrogen-containing heterocyclic benzene ring, which demonstrates larger steric hindrance than the benzene ring in phenylalanine, thereby causing a reduction in the specific surface area with a S _BET of 353 m²/g. HPIL-Cl-Phe exhibits a micromesoporous structure, with pore sizes primarily concentrated in the microporous region (<2 nm) and the mesoporous region (3.0–4.5 nm) and S _BET of 490 m²/g.

2.

Characterization analysis and FA dynamic adsorption curves of HPIL-Cl-Xs. (a) FT-IR spectra, N₂ adsorption/desorption isotherms (b), and pore width distributions of HPIL-Cl-Xs (c), FA dynamic adsorption curves of HPIL-Cl-Xs (d).

2. Textural Parameters and FA Adsorption Performances of HPIL-Cl-Xs.

adsorbent	S BET (m2/g)	S mic (m2/g)	V total (cm3/g)	V mic (cm3/g)	D BJH (nm)
HPIL-Cl	822	470	0.626	0.249	3.1
HPIL-Cl-DCA	631	407	0.610	0.215	3.9
HPIL-Cl-PA	899	502	0.648	0.270	2.9
HPIL-Cl-Phe	490	297	0.335	0.159	2.5
HPIL-Cl-Trp	353	214	0.248	0.114	2.8

adsorbent	N (wt %)	yield (%)	Q p (mg/g)	Q e (mg/g)	PC (mol/kg/Pa)
HPIL-Cl	2.31	94.56	5.3	8.7	0.34
HPIL-Cl-DCA	1.39	61.15	3.2	4.6	0.18
HPIL-Cl-PA	1.07	46.64	4.7	6.5	0.25
HPIL-Cl-Phe	3.39	76.32	7.9	11.3	0.44
HPIL-Cl-Trp	4.49	84.58	3.4	10.8	0.42

^aBET surface area evaluated from N₂ adsorption isotherms at relative pressures (P/P ₀) from 0.05 to 0.30.

^bMicropore surface area and volume are estimated by the t-plot method.

^cTotal pore volume at P/P ₀ = 0.990.

^dAverage mesopore size calculated by the BJH adsorption equation for P/P ₀ from 0.1 to 0.990.

^eAdsorption capacity calculated through the FA dynamic adsorption curve.

^fYield is calculated based on the mass of reactant and product.

^gPC as partitioning coefficient which was calculated using eq .

^hN element content in HPIL-Cl-Xs detected by EA.

Notable differences in structure, composition, and adsorption performance are observed in the resulting HPIL-Cl-Xs compared to HPIL-Cl. The selected FMs all contain benzene rings, which facilitate hyper-cross-linking, and nitrogen-containing groups that serve as targeted adsorption sites. Due to variations in their physical properties and solubility in DCE, the extent of their participation in the hyper-cross-linking reaction differs significantly. HPIL-Cl-PA and HPIL-Cl-DCA display lower nitrogen content and adsorption capacities compared to HPIL-Cl. The low nitrogen content of HPIL-Cl-DCA and HPIL-Cl-PA as 1.39 and 1.07% indicates the self-polymerization of DCX and hyper-cross-linking of BMZ with minimal involvement of PA and DCA in the reaction. The specific surface areas of HPIL-Cl-Phe and HPIL-Cl-Trp are 490 and 353 m²/g, respectively. Their D _BJH values are 2.5 and 2.8 nm, Q _e values are 11.3 and 10.8 mg/g, Q _p values are 3.4 and 7.9 mg/g, and PC values are 0.44 and 0.42 mol/kg/Pa. The nitrogen contents of HPIL-Cl-Phe and HPIL-Cl-Trp are 3.39 and 4.49%, respectively, both exceeding that of HPIL-Cl. However, HPIL-Cl-Trp exhibits slower mass transfer, resulting in a longer time to reach adsorption equilibrium. SEM images of HPIL-Cl-Phe and HPIL-Cl-Trp are shown in Figure , which reveal similar morphologies to HPIL-Cl. The polymer particles, sized in the range of tens of nanometers, are interconnected to form a cross-linked network. There are larger pores promoting the rapid transportation. This is consistent with the N₂ adsorption/desorption results. HPIL-Cl-PA has the highest specific surface area and the largest pore volume, with its micropore surface area and pore volume far exceeding those of other HPIL-Cl-Xs. However, its low nitrogen content results in fewer functional sites, leading to a lower Q _e value of 6.5 mg/g. HPIL-Cl-Trp, with a higher nitrogen content and more functional sites than HPIL-Cl-PA, has a lower surface area, resulting in a Q _e of 10.8 mg/g and a Q _p of 3.7 mg/g. HPIL-Cl-Phe, with both high nitrogen content and specific surface area, exhibits the highest FA adsorption capacity and PC values.

3.

SEM image of (a–c) HPIL-Cl; (d–f) HPIL-Cl-Phe; (g–i) HPIL-Cl-Trp.

The adsorption capacity of HPIL-Cl-Xs for FA is primarily determined by a combination of the specific surface area, pore size distribution, and type and density of functional groups involved in adsorption. A higher specific surface area provides more available adsorption sites, while a larger micropore surface area and pore volume contribute to enhanced van der Waals interactions. An appropriate proportion of mesopores facilitates the diffusion of FA molecules within the material, thereby improving the mass transfer efficiency and increasing the likelihood of effective adsorption contact. Furthermore, the amine groups introduced via functionalization strengthen the interaction with FA molecules, enabling more efficient adsorption through hydrogen bonding or chemical adsorption mechanisms. Consequently, the overall adsorption behavior of FA on HPIL-Cl-Xs is governed by the synergistic effects of the material structure, surface area, and functional group density. HPIL-Cl-Phe integrates all these favorable characteristics, exhibiting a high specific surface area, a well-balanced micro/mesopore distribution, and an enriched density of functional groups, thereby delivering superior FA adsorption performance.

3.2. Effect of the Mole Ratio of Reactants on the Composition, Structure, and Performance of HPIL-Cl-Phe

During the hyper-cross-linking reaction process, the mole ratio of reactants plays a critical role in determining the composition and structure of HPIL-Cl-Xs. Using HPIL-Cl-Phe as a representative model, the influence of the reactant mole ratio on its structural properties, compositional characteristics, and adsorption performance was systematically evaluated, and the results are presented in Figure and Tables and S2.

4.

N₂ adsorption/desorption isotherms and pore width and FA adsorption curves of HPIL-Cl-Phe prepared with different mole ratios of reactants (a–c).

3. Specific Surface Area, Pore Size Distribution, N Content, and Adsorption Capacities of HPIL-Cl-Phe under Different Reactant Mole Ratios.

mole ratio of BMZ:Phe:DCX	S BET (m2/g)	S micro (m2/g)	D BJH (nm)	V total (cm3/g)	N (wt %)	Q p (mg/g)	Q e (mg/g)
1:0:1	822	470	3.1	0.626	2.24	5.3	8.7
1:1:2	490	297	2.5	0.335	3.39	7.9	11.3
1:2:3	245	138	2.7	0.164	2.20	4.2	6.2
1:3:4	163	133	2.4	0.099	2.14	3.8	5.5
2:1:3	452	347	2.4	0.271	1.55	4.6	6.4
3:1:4	380	236	2.4	0.225	1.42	3.6	5.3
1:1:3	583	416	2.7	0.346	1.55	6.1	8.8
1:1:4	608	399	2.5	0.371	1.20	4.7	6.5

The mole ratio of BMZ:Phe:DCX significantly influences the specific surface area, pore size distribution, and functional group density. Mole ratios of 1:1:2, 1:2:3, 1:3:4, 2:1:3, 3:1:4, 1:1:3, and 1:1:4 were investigated under a T _pre of 80 °C and a reaction time of 24 h. The composition, structure, and adsorption performance results are shown in Figure a–c and Table . When the mole ratio of BMZ and Phe to DCX is fixed at 1:1, increasing the amount of Phe leads to a significant reduction in the specific surface area and pore volume of the prepared HPIL-Cl-Phe, as well as influencing the nitrogen content and FA adsorption performance. Specifically, when the mole ratio of BMZ:Phe:DCX changed from 1:0:1 to 1:3:4, the specific surface area decreases from 822 to 163 m²/g, the pore volume drops from 0.626 to 0.099 cm³/g, and the nitrogen content exhibits a trend of first increasing and then decreasing. When the mole ratio of BMZ and Phe to DCX is fixed at 1:1, increasing the amount of BMZ also reduces the specific surface area and pore volume of the prepared HPIL-Cl-Phe. Adjusting the mole ratio of BMZ:Phe:DCX from 1:1:2 to 3:1:4 causes the specific surface area to decrease from 490 to 380 m²/g and the pore volume to drop from 0.335 to 0.225 cm³/g. By comparing the specific surface area and pore volume of HPIL-Cl-Phe at mole ratios of BMZ:Phe:DCX of 3:1:4 and 1:3:4, it can be observed that increasing the proportion of benzimidazole results in a relatively smaller decrease in specific surface area. This phenomenon can be attributed to the fact that benzimidazole is capable of undergoing both supercrosslinking reactions and nucleophilic substitution reactions with DCX, thereby contributing to the maintenance of pore structure stability. In contrast, the longer side chains of the phenyl group in Phe partially occupy the pore structure, leading to a reduction in pore volume and specific surface area. When the mole ratio of BMZ to Phe is fixed at 1:1, increasing the amount of DCX from 1:1:2 to 1:1:4 results in an increase in the specific surface area of the prepared HPIL-Cl-Phe from 490 to 608 m²/g and the pore volume from 0.335 to 0.371 cm³/g. However, the nitrogen content decreases from 3.39 to 1.20%. In the hyper-cross-linking reaction, increasing DCX promotes its self-polycondensation, resulting in a higher specific surface area and microporous area. However, this also reduces the availability of functional groups for FA adsorption, leading to a decline in adsorption performance. When the mole ratio of BMZ:Phe:DCX is 1:1:2, HPIL-Cl-Phe achieves an S _BET of 490 m²/g, a D _BJH of 2.5 nm, and an N element content of 3.39 wt %. By optimizing the reactant mole ratio and tailoring structure and composition of the polymer, HPIL-Cl-Xs can achieve a higher specific surface area and functional group content, thereby enhancing its adsorption performance. By optimizing the reactant mole ratio and tailoring structure and composition of the polymer, HPIL-Cl-Xs can achieve a higher specific surface area and functional group content, thereby enhancing its adsorption performance.

3.3. Effect of Different Working Conditions on the Adsorption Process of FA Using HPIL-Cl-Phe

To investigate the effect of adsorption temperature on the FA adsorption capacity of HPIL-Cl-Phe, T _ad of 20, 25, 30, 40, 50, and 60 °C were selected, with C _in as 8.6 ppm, WHSV as 54,000 h^–1, and RH as 0%. The experimental results are shown in Figure a and b. With the increase in T _ad, Q _e shows a decreasing trend. Q _p first increases and then decreases. When T _ad is 20 °C, Q _p is 7.9 mg/g, while T _ad as 30 °C, Q _p reaches 8.6 mg/g. The dynamic adsorption curve clearly shows that as T _ad increases, the time required to reach equilibrium after breakthrough decreases. This trend occurs because higher temperatures increase the kinetic energy of molecules, causing them to move and collide more rapidly, which weakens the interactions. As a result, the adsorbent performance driven by physisorption declines with rising T _ad.

5.

FA dynamic adsorption curves and adsorption capacities using HPIL-Cl-Phe under different working conditions; (a and b) different adsorption temperatures; (c and d) different relative humidities; (e and f) different WHSVs.

Different RH values were investigated (0, 15, 30, 45%) under C _in of 8.6 ppm, T _ad of 30 °C, and WHSV of 54,000 h^–1. Experimental results are presented in Figure c and d. As RH increases, the adsorption performance of HPIL-Cl-Phe declines. This is attributed to competitive adsorption between water and FA. Since water molecules have a similar polarity to FA, they occupy the available adsorption sites. Adsorbents with strong FA adsorption capabilities are typically hydrophilic, and their performance is highly sensitive to changes in RH. Different WHSVs were investigated (54,000, 70,000, 85,000, 100,000 h^–1). As the WHSV increased from 54,000 to 100,000 h^–1, Q _e and Q _p decreased by approximately 44.2 and 57.0%, from 11.3 to 6.3 mg/g for Q _e and 7.9 to 3.4 mg/g for Q _p. At low WHSV, FA diffuses slowly across HPILs-Cl-Phe surface, allowing sufficient contact between FA and adsorption sites. High WHSV shortens residence time, preventing FA from effectively interacting with the functional groups. Lots of adsorption sites become underutilized high WHSV which leads to a decrease in adsorption capacity.

3.4. Adsorption Mechinism Exploration for FA Adsorption Using HPIL-Cl-Phe

3.4.1. XPS and FT-IR Analysis of FA Adsorption Using HPIL-Cl-Phe

FT-IR characterization was performed on both fresh and saturated HPIL-Cl-Phe, with the results presented in Figure a. The N–H stretching vibration peak of fresh HPIL-Cl-Phe shifted from 3388 to 3430 cm^–1 in the saturated state, indicating the formation of hydrogen bonds between the −NH group (acting as a hydrogen bond donor) and the aldehyde group. A new absorption peak at 1731 cm^–1 appeared in saturated HPIL-Cl-Phe, attributable to the stretching vibration of the aldehyde group in FA. This suggests that a portion of FA is adsorbed via physical interactions. A double peak at 1672 and 1600 cm^–1 appeared in fresh HPIL-Cl-Phe, corresponding to the stretching vibrations of the primary amine groups. After adsorption, these peaks show a significant decrease, indicating the consumption of primary amine groups during FA adsorption.

6.

FT-IR spectrum (a) and XPS N 1s (b) and O 1s (c) energy spectrum of fresh and saturated HPIL-Cl-Phe.

The N 1s and O 1s XPS spectra and peak fitting results for fresh and saturated HPIL-Cl-Phe are presented in Figure b,c and Table . The N 1s spectrum of fresh HPIL-Cl-Phe shows three distinct peaks at 400.3, 400.0, and 398.7 eV, corresponding to the amino group in Phe, the imine nitrogen, and the amide nitrogen, respectively. In the N 1s spectrum of saturated HPIL-Cl-Phe, a new peak appears at 396.3 eV, corresponding to the nonimidazole −NC– species formed during the adsorption reaction, accounting for 3.53%. The relative intensities of the −NH₂ and −NH– peaks decrease, whereas the content of N– in the imidazole ring remains largely unchanged. This suggests that chemical adsorption occurs during the process, with the consumption of −NH₂ and −NH– groups and the formation of −NC– groups. In the O 1s spectrum, fresh HPIL-Cl-Phe exhibits a single peak at 532.5 eV, corresponding to the carboxyl group in Phe. After adsorption, an additional peak at 534.2 eV appears, attributable to the aldehyde group in FA molecules adsorbed onto the HPIL-Cl-Phe surface, accounting for approximately 11.72%. This supports the presence of physical adsorption, wherein the negatively charged oxygen in the aldehyde group can form hydrogen bonds with the hydrogen atoms in the −NH₂ and −NH– groups, facilitating their separation.

4. Detailed Information of XPS N 1s and O 1s Spectra of Fresh and Saturated HPIL-Cl-Phe.

	O/%		N/%
adsorbent	–COO	–CO	–NH2	–NH–	N–	–NC–
fresh HPIL-Cl-Phe	100	0	34.57	34.34	31.09	0
saturated HPIL-Cl-Phe	88.28	11.72	32.68	32.66	31.13	3.53

3.4.2. Visualization of Noncovalent Interactions in Two Adsorption Systems

The adsorption process of FA on HPIL-Cl-Phe was investigated by using density functional theory (DFT) quantum calculations. The results obtained agreed with the characterization data from XPS and FT-IR, confirming the roles of van der Waals forces, hydrogen bond interactions, and nucleophilic addition reactions in the adsorption process.

The ESP of HPIL-Cl-Phe on different surface areas is visualized and distinguished by different colors in Figure a. , Negative values are depicted in blue, while positive ESP values are depicted in red. The red regions are concentrated at the hydrogen atoms bonded to nitrogen and oxygen in the amino acid groups as well as at the hydrogen atoms of the −NH– groups in the benzimidazole heterocyclic ring. The blue regions are concentrated at the nitrogen and oxygen atoms in the amino acid groups as well as at the nitrogen atoms of the −NH– groups in the benzimidazole heterocyclic ring. The standard RESP atomic charges of active function groups in HPIL-Cl-Phe are indicated in Figure b. Based on the ESP distribution and RESP atomic charges, the interaction strength and reaction possibility can be predicted. The hydrogen bond donor ability of HPIL-Cl-Phe primarily originates from the hydrogen atoms on the −NH–, −NH₂, and −COOH groups. Furthermore, the nitrogen-containing groups serve as potential nucleophilic reagents in nucleophilic addition reactions between FA and HPIL-Cl-Phe, owing to their high electronegativity. Optimized geometric configuration of two adsorption systems can be seen from Figure c,d, and the results are consistent with the ESP distribution and RESP charge calculations. The oxygen atom of the aldehyde group in FA can form a hydrogen bond with the hydrogen atom of the −NH– group in BMZ, with a bond length of 2.019 Å. Additionally, the amino acid groups exhibit >N–H···O hydrogen bond interactions and CO···H–C interactions, where the bond length between −NH···O is 2.240 Å.

7.

ESP isosurface of HPIL-Cl-Phe (red: O, blue: N, green: C, white: H) (a); standard RESP atomic charges for HPIL-Cl-Phe (b); (c, d) optimized geometric configuration of two adsorption systems (blue: nitrogen; red: oxygen; yellow: carbon; white: hydrogen).

The RDG analysis results are shown in Figure , with the 0.6 isosurface displayed. The isosurface is color-coded using a blue–green–red scheme based on the value of sign­(λ₂) ρ (ranging from −0.05 to −0.05), which represents hydrogen bonds, van der Waals interactions, and steric hindrance, respectively. Different types of NCI can be distinguished by color, while the intensity of interactions can be indexed by the color depth. Visualization analysis of noncovalent interactions was achieved through RDG analysis, particularly for gaining a deeper understanding of the −N–H···O (lone pair electrons) hydrogen bonds between the amino group in HPIL-Cl-Phe and the >CO group in FA. The −NH– group in the imidazole heterocyclic cation and the amino acid group are hypothesized to be active sites for FA adsorption. In Figure a, the oxygen (O) atom in the carbonyl group of FA, with its lone pair of electrons, acts as a hydrogen bond acceptor, while the −NH– group of the BMZ cation functions as a hydrogen bond donor. The hydrogen bond interaction is clearly observable through the RDG isosurface plot (highlighted in the red box). The ion pair interaction enhances the hydrogen bond donation ability of the −NH– group, thereby strengthening the hydrogen bond interaction. In Figure b, the oxygen (O) atom in FA forms a hydrogen bond with the −NH₂ group of the amino acid moiety (highlighted in the red box) while also interacting via van der Waals forces.

8.

Color mapped scatter diagrams and the resulting RDG isosurface graphs of FA/HPIL-Cl-Phe: (a) adsorption site at −NH– in the BMZ ring; (b) adsorption site at the amino acid group in Phe.

The CVB index, proposed by Silvi et al., was utilized to quantify the strength of hydrogen bonds, where a larger CVB value signifies a weaker hydrogen bond interaction. In our adsorption systems, the CVB indices for two types of hydrogen bonds were calculated using Multiwfn software. In adsorption system a, the CVB index for the N–H (BMZ cation)···O (FA) interaction is 0.0117, while in adsorption system b, the CVB index for the N–H (amino acid group in Phe)···O (FA) interaction is 0.0331. This indicates that the −N–H···O hydrogen bond interaction in system a is stronger. However, it is important to note that, as shown in Figure b, complete green isosurfaces are observed between the hydrogen atom of FA and the oxygen atom in the carboxyl group and between the oxygen atom in FA and the benzene ring, suggesting a significant van der Waals (vdW) interaction between these fragments.

3.5. Fitting of the FA Adsorption Process Using a Dynamic Adsorption Model and Adsorption Kinetics Model

To further investigate the mechanism of FA adsorption by HPIL-Cl-Xs, the dynamic adsorption data were fitted to several adsorption models, as presented in Figure and Table . The fitting results, as evidenced by the R ² values, indicate that the pseudo-first-order (PFO) model provides a superior fit for the adsorption of FA on HPIL-Cl-Xs. The slight discrepancy in R ² values between the PFO and pseudo-second order (PSO) models suggests that the FA adsorption process may involve both physical and chemical interactions. Moreover, the Q _m values derived from the PFO model exhibit a smaller deviation from Q _e compared to those obtained from the PSO model. This observation further reinforces the predominance of physical adsorption in the overall process. Based on the W–M model (Figure d), the adsorption process can be divided into three distinct stages: external surface diffusion, intraparticle diffusion, and adsorption equilibrium. This multistage adsorption process suggests that the overall adsorption behavior is influenced by both surface adsorption and pore diffusion mechanisms. The Bangham kinetics model achieved the highest correlation coefficient (R ² ≈ 0.999) and exhibited the smallest deviation between Q _e and Q _m, as estimated by the Bangham model. This indicates that the Bangham model effectively describes the FA adsorption behavior of HPIL-Cl-Xs in a fixed bed, which is consistent with the pore diffusion mechanism.

9.

Fitting diagrams of the FA dynamic adsorption process using HPIL-Cl-Xs: (a) pseudo-first order kinetic model, (b) pseudo-second kinetic model, (c) Bangham rate equation, (d) W–M model, (e) Yoon–Nelson model, (f) Boltzmann model.

6. Comparison of FA Adsorption Properties of As-Prepared Samples in This Work with Other Reported FA Adsorbents.

adsorbents	temperature (K)	initial concentration (ppm)	partial pressure (Pa)	adsorption capacity (mg/g)	PC value (mol/kg/Pa)	ref
carbon-based systems for FA adsorption (limited cyclic performance)
PSC	298	0.8	0.08	0.3	0.13
aminosilane-functionalized AC	298	1	0.1	1.2	0.39
NNSC-1	298	3.6	0.36	0.5	0.049 (SA)
BCMFs	298	1870	187	205.0	0.04
AC-BPA, AC-BPAa (after 5th cycle)	298	50	5	82.3, 31.0	0.64, 0.24
commercial AC	303	8.6	0.86	2.8	0.11	this work
non-AC-based FA adsorbents
commercial SBA-15	303	8.6	0.86	0.3	0.01	this work
commercial SPO-30				0.2	0.01
MIL-101	298	150	15	100.0	0.22
ED-MIL-101				165.0	0.37
CBAP-1	298	100	10	105.0	0.35
HPIL-Cl-Phe	298	8.6	0.86	12.0	0.46	this work
HPIL-Cl-Phea (after 6th cycle)	303			14.3	0.55	this work
polycarbazole-based HP	298	160	16	11.0	0.02
static adsorption evaluation process (SA)
MCC/APMDS cellulose-based aerogel	298	5	0.5	9.5	0.63 (SA)
rBTA-2:1	298	200	20	321.4	0.54 (SA)
rBTA-1:1 b				435.2	0.72 (SA)
rBTA-1:2 b				459.8	0.76 (SA)
branched polyamine incorporated aminosilica (BPAS)	298	200	20	171.0	0.29 (SA)

^aDynamic adsorption process.

^bStatic adsorption process (SA).

The rate constants derived from various kinetic models were compared across different HPIL-Cl-X materials. Among them, HPIL-Cl-DCA exhibits the highest rate constant, while HPIL-Cl-Phe and HPIL-Cl-Trp display relatively lower values. This variation in rate constants can be attributed to differences in their compositions and structure. The yield of HPIL-Cl-DCA in the hyper-cross-linking polymerization reaction is relatively low, with fewer functional monomers participating in the reaction, a larger specific surface area, and a D _BJH of 3.9 nm. The pore size distribution of HPIL-Cl-DCA exhibits a greater presence of meso- and macroporous regions compared to other HPIL-Cl-Xs, which accelerate mass transfer and result in a higher rate constant. In comparison, HPIL-Cl-Phe and HPIL-Cl-Trp have higher reaction yields and more functional adsorption sites. Furthermore, their micropore areas account for over 60%, with D _BJH values of 2.5 and 2.8 nm, respectively. The amino acid groups present in FMs are capable of interacting with FA via hydrogen bonding or nucleophilic addition reactions, which can enhance the adsorption capacity for FA. However, these interactions may simultaneously hinder the adsorption kinetics due to increased binding complexity or steric effects.

The Boltzmann and Yoon–Nelson models were used to describe the dynamic adsorption of FA on the HPIL-Cl-Xs. The fitting results are shown in Figure e and f and Table . Both models provided excellent fits for the dynamic adsorption curves of all of the HPIL-Cl-X materials. Notably, the Boltzmann model yielded R ² values greater than 0.99 for HPIL-Cl-Phe and HPIL-Cl-Trp, indicating superior adsorption performance. The predicted half-concentration time (τ) from the Boltzmann model closely matches the experimental data. A steeper slope (1/dx) at the half-concentration point suggests better mass transfer, faster adsorption rates, and a higher utilization efficiency of the adsorbents. The 1/dx fitting results align with the rate constants from the PFO, PSO, and Bangham models. HPIL-Cl-DCA shows the highest value of 1/dx, while HPIL-Cl-Phe and HPIL-Cl-Trp have lower values, indicating faster mass transfer in HPIL-Cl-DCA and slower transfer in HPIL-Cl-Phe and HPIL-Cl-Trp. The half-breakthrough time (τ_m) predicted by the Yoon–Nelson model was closely aligned with the experimental τ_e, demonstrating the predictive capability of the Y–N model. The rate parameter K′ was lower for HPIL-Cl-Phe and HPIL-Cl-Trp, indicating slower adsorption rates within the bed layer, extended breakthrough times, and delayed saturation, which contribute to prolonged adsorption. Overall, the fitting results were consistent with the experimental observations.

5. Fitting Parameters of FA Adsorption Kinetics of HPIL-Cl-Xs .

model	constant	HPIL-Cl	HPIL-Cl-PA	HPIL-Cl-Phe	HPIL-Cl-Trp	HPIL-Cl-DCA
PFO	K 1 (h–1)	0.0577	0.0515	0.0298	0.0389	0.0739
	Q m (mg/g)	11.8	11.5	19.4	14.9	7.9
	R 2	0.9883	0.9917	0.9924	0.9953	0.9939
PSO	K 2 (g/mg/h)	0.0023	0.0015	0.0005	0.0011	0.0033
	Q m (mg/g)	17.8	19.8	33.6	23.6	13.6
	R 2	0.9859	0.9908	0.9919	0.9936	0.9935
Bangham	K 3 (h–z)	0.0121	0.0113	0.0095	0.0205	0.0791
	Q m (mg/g)	8.9	6.5	11.1	11.5	5.1
	z	1.788	1.769	1.732	1.382	1.331
	R 2	0.9996	1.0000	0.9992	0.9993	0.9981
Yoon–Nelson	K′ (min–1)	0.0066	0.0077	0.0047	0.0034	0.0066
	τpre (min)	1088	791	1348	1424	521
	R 2	0.9937	0.9885	0.9893	0.9872	0.9937
Boltzmann	1/dx	0.0060	0.0111	0.0048	0.0024	0.0122
	τpre(min)	1093	787	1347	1525	583
	R 2	0.9956	0.9866	0.9903	0.9973	0.9922
	Q e (mg/g)	8.7	6.5	11.3	10.7	8.7
	τexp (min)	1094	992.5	1577	1661	573

^a Q _m, predicted maximum capacity, mg g^–1; R ² (correlation coefficient) is used to determine the best-fitting model; z is constant. τ is adsorption time at C_t /C ₀ as 0.5.

3.6. Cyclic Regeneration Performance of HPIL-Cl-Phe

Based on the adsorption mechanisms identified through FT-IR, XPS, and DFT calculations, the regeneration of HPIL-Cl-Phe was investigated by using acid washing (0.1 M HCl) combined with N₂ thermal purging (20 min). The adsorption performance of the regenerated HPIL-Cl-Phe was evaluated under the same conditions. The results show that the breakthrough time of the adsorption curve for HPIL-Cl-Phe remains similar, and the time required to reach equilibrium increases compared to the fresh HPIL-Cl-Phe. Over six cycles, Q _p remains stable, while Q _e improves significantly, increasing by approximately 24%. The FT-IR spectrum of the regenerated HPIL-Cl-Phe closely matches the original, with characteristic peaks at 3388, 2925, 1610, 1500, and 1440 cm^–1, confirming that the regeneration process does not alter its physical properties. N₂ adsorption–desorption experiments were conducted to evaluate the specific surface area and pore size distribution of both fresh and regenerated HPIL-Cl-Phe. The results shown in Figure a and b demonstrate that the specific surface area, micropore area, total pore volume, and micropore volume of the regenerated HPIL-Cl-Phe (634 m²/g, 408 m²/g, 0.404 cm³/g, 0.159 cm³/g) have been enhanced to some extent compared to fresh HPIL-Cl-Phe (490 m²/g, 297 m²/g, 0.335 cm³/g, 0.219 cm³/g). The increase in the specific surface area of the regenerated HPIL-Cl-Phe is primarily attributed to the significant enhancement in the micropore area, with its proportion increasing from 60.6 to 64.4%. This can be ascribed to the further intensification of hyper-cross-linking reactions between reactants during the regeneration process, which allows residual functional groups in the polymer that were not fully involved in the initial hyper-cross-linking to engage in additional reactions. Consequently, this results in an effective increase in the micropore area. The increase in porosity and surface area provides more adsorption sites, while the larger pore volume facilitates improved FA transfer efficiency and storage capacity within the adsorbent. These changes explain the higher adsorption capacity of the regenerated HPIL-Cl-Phe compared to its preregeneration state.

10.

Recyclability study of HPIL-Cl-Phe: the N₂ adsorption desorption isotherms (a), pore size distribution (b), and FA adsorption curve (c) during the HPIL-Cl-Phe regeneration process, FT-IR spectra (d) of HPIL-Cl-Phe before and after regeneration, and Q _e (e) and Q _p (f) during the HPIL-Cl-Phe regeneration process.

3.7. Comparative Analysis: FA Sorption Performance of HPIL-Cl-Xs versus Other Sorbents

Adsorption capacity, which is closely related to operational parameters, used as a general performance metric has limited practical applicability. High partial pressure of the adsorbate often yields large adsorption capacity values. So, the partition coefficient (PC), which reflects the degree of partitioning, has proven to be a more reliable measure for assessing the true efficiency of adsorbents. It provides a quantitative estimate of the strength of adsorbate–adsorbent interactions. PC has been widely used as a comprehensive performance metric for evaluating and comparing the performance of different adsorbents. The comparison of FA adsorption performance for HPIL-Cl-Xs with various adsorbents is summarized in Table , and PC values and adsorption capacities were listed.

Significant differences in the PC values and adsorption capacities for FA can be observed. Different adsorbents, different initial concentrations, and different evaluation processes influence PC value and adsorption capacity. For the same adsorbent, the adsorption capacity determined by different evaluation methods shows marked differences. Specifically, the adsorption capacity under static conditions is generally several times greater than that under dynamic conditions. In Table , all adsorption capacities derived from the static adsorption process are clearly identified in italics as ″Static Adsorption (SA)″. The adsorption capacity and PC value of commercial activated carbon can be improved through common modification methods, such as amine and acid treatments. BPA demonstrates superior performance due to its abundant surface functional groups to tackle FA with a PC value of 0.64 mol/kg/Pa, but its cyclic performance needs to be improved with 38% of original adsorption capacity after five cycles. The adsorption performance of noncarbon-based adsorbents can be improved through modification and structural design. MOFs and COPs demonstrate high adsorption capacities and high PC values. The highest adsorption capacity is 165 mg/g at a concentration of 150 ppm, with a corresponding PC value of 0.37 mol/kg/Pa for ED-MIL-101. Both BTA and BPAS show high adsorption capacities, which were achieved under high-concentration static adsorption conditions. HPIL-Cl-Xs in this study exhibits high adsorption capacities and PC values in dynamic adsorption processes under low concentrations and high WHSV. Specifically, the adsorption capacities of HPIL-Cl, HPIL-Cl-Phe, and recycled HPIL-Cl-Phe are 8.7, 11.3, and 14.3 mg/g, with corresponding PC values of 0.34, 0.45, and 0.55 mol/kg/Pa, respectively. More importantly, this kind of adsorbent demonstrated excellent regenerative ability and adsorption capacity, and the PC values even increased after several cycles of regeneration. The superior performance of HPIL-Cl-Xs over other materials can be attributed to their micromesoporous structure, designed function group, and the unique charge environment. These structural and functional characteristics enable HPIL-Cl-Xs to effectively adsorb FA through hydrogen bond interactions and nucleophilic addition reactions.

4. Conclusion

A series of HPIL-Cl-Xs were synthesized via one-step hyper-cross-linked polymerization and quaternization reaction with BMZ, DCE, and amide-functional monomers, featuring a chloride-based ionic environment and amino adsorption sites. They exhibit a prominent microporous structure, tunable functional group density, and unique ionic properties. The effects of polymerization parameters on their composition, structure, and adsorption performance were studied. HPIL-Cl-Phe showed high adsorption capacities (Q _e = 12.0 mg/g and PC value of 0.46 mol/kg/Pa) for FA, outperforming nonfunctional HPIL-Cl and conventional adsorbents. Adsorption capacity is closely related to the specific surface area, pore width distribution, and N content of copolymers. The adsorption capacity of HPIL-Cl-Phe decreases as the temperature increases. Higher space velocities reduce the contact time, leading to under-utilization of adsorption sites. Additionally, water and FA can induce competitive adsorption, and RH significantly affects the adsorption performance. ESP and RESP charges suggest −NH– and −NH₂COOH groups as potential adsorption sites for hydrogen bond interaction and nucleophilic reactions. XPS and FT-IR confirm the existence of a hydrogen bond interaction and nucleophilic reactions. RDG isosurface visually presents the interaction form and intensity between FA and HPIL-Cl-Phe, while the CVB parameters further quantify the strength of hydrogen bonds. Besides hydrogen bond interactions, van der Waals interactions were also observed between FA and HPIL-Cl-Phe. HPIL-Cl-Phe exhibits excellent cycling performance with the adsorption capacity of the sixth cycle higher than fresh. Dynamic fitting analysis confirms both physical and chemical adsorption, with simulation results closely matching experimental data.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/prechem.5c00084.

• Detailed reagents; supplementary data on HPIL-Cl-Phe structural composition under different reactant mole ratios; synthesis and characterization procedures; dynamic adsorption device; FA detection method and standard curve by MBTH; representative fragment of HPIL-Cl-Phe for quantification; and TG curves of HPIL-Cl-Xs (PDF)

T.J.: Funding acquisition, Investigation, Methodology, Project administration, Writing-review and editing. H.F.: Formal analysis, Methodology, Writing-original draft. R.H.: Software, Visualization, Writing–review and editing. S.L.: Data curation, Formal analysis, Investigation, Writing–original draft. P.L.: Conceptualization, Funding acquisition, Methodology, Writing–review and editing. Y.Z.: Funding acquisition, Methodology, Writing–review and editing. W.Z.: Funding acquisition, Investigation, Methodology, Writing–review and editing. X.L.: Data curation, Methodology, Supervision. H.Z.: Formal analysis, Methodology, Writing–review and editing. X.L.: Methodology, Writing–review and editing. S.F.Y.L.: Methodology, Supervision, Writing–review and editing.

The authors declare no competing financial interest.