Cr(VI) anion-imprinted polymer synthesized on mesoporous silicon via synergistic action of bifunctional monomers for precise identification and separation of Cr(VI) from aqueous solution by fixed-bed adsorption

In the process of industrial production such as electroplating, metallurgy, leather making, metal processing and wood preservative, large quantities of wastewater containing Cr(VI) are generated and discharged into the ecological environment, becoming one of the most common pollutants in soil, groundwater and surface water (Kong et al. 2021). Chromium in water mainly exists in the forms of Cr(VI) and Cr(III). The toxicity of chromium is related to the valence state, and compounds with different valence states can be converted into each other (Li et al. 2023b). Cr(III) is an important micronutrient in the human body, with stable properties and easy-to-produce hydroxide precipitation, which is 100 times less toxic than Cr(VI) (Zhou et al. 2020). Different from Cr(III), Cr(VI) has strong mobility and toxicity (Fang et al. 2021), which can lead to cell cancerization and even to the death of living organisms (Zhang et al. 2019). Therefore, the content of Cr(VI) is an important indicator of water pollution control. The maximum levels of Cr(VI) in drinking water and industrial effluent are regulated by the US Environmental Protection Agency (EPA) within 50 and 200 μg/L, respectively (Jasim et al. 2023). As a result, precise detection of Cr(VI) presence in ambient water is critical.

The most commonly used determination methods of chromium are Raman scattering spectroscopy (Wang et al. 2022), X-ray fluorescence spectrometry (Ferreira et al. 2023), inductively coupled plasma mass spectrometry (ICP-MS) (Pluhacek et al. 2023) and atomic absorption spectrometry (AAS) (Rajaram et al. 2023). However, these techniques can only measure the total content of Cr, which means the amount of Cr(VI) cannot be detected accurately in water. The interference of Cr(III) should be avoided in order to identify the amount of Cr(VI). Therefore, it is essential to preconcentrate and separate Cr(VI). Undoubtedly, one of the most popular and efficient separation techniques is solid phase extraction (SPE) (Yaida et al. 2023). However, general SPE sorbents, such as nano-MnO₂ (Li et al. 2023a), ceria decorated titanate nanotubes (Biswas et al. 2023), polyethyleneimine (Bagdat et al. 2023) and Fe₃O₄@SiO₂@DPC (Assi et al. 2019) have poor selectivity. In addition, SPE sorbents for Cr(VI) such as Mil-100 (Fe) (Li et al. 2022), Fe-Zn@BC (Zheng et al. 2022), water-based magnetic fluid-spores (WMFSs) (Ren et al. 2022) and Cit-nZVI@BC (Zhou et al. 2022) have the problem of low adsorption capacity. Therefore, it is necessary to develop a preconcentration and separation method for Cr(VI) with high selectivity and adsorption capacity.

Ion-imprinted polymer (IIP) is a synthetic polymer with the function of target ion recognition (Islam & Rais 2023). Due to its unique predetermined recognition performance and applicability, IIP has been widely used in the field of SPE (Giove et al. 2023). Some research teams have developed different kinds of IIPs to separate metal ions with different valence states, such as Mo(IV)/Mo(VI) (Fallah & Taghizadeh 2020), As(III)/As(V) (Chen et al. 2020), Sb(V)/Sb(III) (Wang et al. 2020), and so on. However, these IIPs only use single functional monomers during the pre-assembled phase. Imprinted cavities formed by coordination between a single functional monomer and template ions cannot recognize target ions accurately, and the imprinted cavities have a low affinity for target ions, resulting in unsatisfactory selectivity of IIP (Bao et al. 2023). Especially for the separation of different valence states of the same metal elements, due to their close mass, size, coordination number and spatial configuration (Hassanzadeh et al. 2018), the imprinted cavity formed by a single functional monomer cannot identify the target ions accurately. In addition, different valence states of the same metal element exist in different forms in solution. For example, Cr(III) generally exists in solution as a cation, while Cr(VI) generally exists as an anion. Most of the traditional IIPs focused on the selective separation of metal cations, but there are few reports dedicated to the preparation of metal anionic imprinted polymers. IIPs for specific anions will achieve more precise identification of the target ions in valence state, spatial configuration and size. At present, the separation of Cr(VI) anion by a single functional monomer is usually poor in selectivity. Therefore, it is important to develop new forms of interaction between functional monomers and template ions to prepare novel metal anionic imprinted polymers, so as to improve and enhance the selectivity of traditional IIPs.

In recent years, some researchers have developed a novel interaction relationship between functional monomers and template molecules to further improve the identification accuracy of IIP, which was constructed by the synergistic action of bifunctional monomers. A batch of new IIPs was synthesized using 4-vinylpyridine (4-VP)/ethylene glycol dimethacrylate (EGDMA) (Nchoe et al. 2020) and glycidyl methacrylate (GMA)/acrylamide (AM) (Luo et al. 2020), etc., as bifunctional monomers for the separation of Cr(VI) anion. Selectivity studies showed that in the presence of competitive ions, IIP synthesized by bifunctional monomers had a higher selective adsorption capacity for Cr(VI) anion, which was improved significantly compared with traditional IIP prepared by a single functional monomer. However, these IIPs prepared by traditional emulsion, suspension or bulk polymerization will still face problems such as elution difficulty, the poor affinity of the binding site and large mass transfer resistance due to template ion embedding. Therefore, introducing surface imprinting technology based on SBA-15 is an effective method to solve these problems (Xu et al. 2021a).

SBA-15 consists of parallel cylindrical pores with axes arranged in a hexagonal unit cell. The regular structure prevents changes in the shape of imprinted sites in the organic polymer (Dobrzyńska et al. 2021). The ordered mesoporous structure endows it with a large specific surface area and pore volume. At the same time, it also has the characteristics of a controllable aperture, thick hole wall and good hydrothermal stability (Zhang et al. 2023). The large aperture of SBA-15 is conducive to the distribution of imprinted sites in the channel and Cr(VI) transport (Hashami et al. 2022). The IIP prepared by a surface imprinting technique based on the unique structural features of SBA-15 not only boosts the adsorption capacity of Cr(VI) but also achieves an excellent dynamic adsorption effect.

In this study, a new type of Cr(VI)-IIP was synthesized on the surface of mesoporous silicon-based material SBA-15 by using surface imprinting technology and a bifunctional monomer pre-assembly system to enhance the adsorption capacity and selective separation capability of Cr(VI) anion. In which, methacrylic acid (MAA) and 4-VP were used as functional monomers and was selected as the template ion. The separation of Cr(VI) and Cr(III) in an aqueous solution can be better achieved due to the specific recognition and binding affinity of the imprinted sites for ⁠. The effects of a series of experimental conditions, such as pH, initial Cr(VI) concentration, temperature and adsorption time on adsorption were investigated in static adsorption experiments, and the mechanism in the adsorption process was studied by adsorption isotherm model and adsorption kinetic model. Since the static adsorption experiments focus on theoretical research and the dynamic adsorption experiments were more focused on practical applications, the dynamic adsorption capacity and dynamic regeneration capacity of Cr(VI)-IIP were investigated in fixed-bed experiments and excellent practical applications were confirmed.

Mesoporous silicon (SBA-15) was synthesized according to the procedure described by Zhao et al. (1998), methacrylic acid (MAA), 4-vinylpyridine (4-VP), EGDMA and 2,2′-azobis-isobutyronitrile (AIBN) were purchased from Aladdin Reagent Co. Ltd (China). Potassium dichromate (K₂Cr₂O₇), nickel nitrate hexahydrate (Ni(NO₃)₂·6H₂O), copper nitrate trihydrate (Cu(NO₃)₂·3H₂O), cadmium nitrate (Cd(NO₃)₂·4H₂O), chromium trichloride (CrCl₃), sodium hydroxide (NaOH), hydrochloric acid (HCl), methanol (CH₃OH), hexamethylenetetramine (HMA), 1,5-diphenylcarbonyl dihydrazine and 3-methylacryloxypropyl trimethoxysilane (MPS) were purchased from Sinopharm Holding Co. Ltd (China). All reagents used in the experiment were analytical grade and doubly distilled water (DDW) used throughout the experiments was obtained by the laboratory purification system.

TAS-986 flame atomic adsorption spectrometer (FAAS) (Purkinje General, China) and inductively coupled plasma atomic emission spectrometer (ICP-AES) (Varian, USA) were combined to measure the concentrations of Cr(VI) and other mental ions such as Cr(III), Cu(II), Cd(II) and Ni(II). Fourier transmission infrared spectra (FT-IR, 4,000–400 cm⁻¹) in KBr were recorded on a NICOLET NEXUS 4700 FT-IR apparatus (Nicolet, USA). A NOVA2000-specific surface and porosity analyzer (Quntachrome, USA) was used to measure the surface area and pore size distribution. Low-angle X-ray powder diffractometer (XRD) measurements were carried out with a Bruker D8 diffractometer (Bruker, Germany) using an angle of 0.5–8.0° with 1°/min. The morphologies of SBA-15 and Cr(VI)-IIP were observed by a JSM-6480 scanning electron microscope (SEM) (JEOL, Japan) and a JEM-2010 transmission electron microscope (TEM) (JEOL, Japan). Thermogravimetric analysis (TGA) was carried out using a differential scanning calorimeter/differential analysis (DSC/DTA-TG) (STA 449C Jupiter Nexe, Germany). A pHS-3C precision pH meter (Lida, China) was used for the pH value adjustments.

First, SBA-15 was dispersed into 3 mol/L HCl for reflux for 24 h. After activation, the activated SBA-15 was washed with DDW until to neutral and dried. Next, the 0.1 g dried SBA-15 was dispersed into 40 mL ethanol solution with ultrasonic dispersion. Then, the mixture of 2.5 mL MPS and 10 mL ethanol was added slowly into the solution under the condition of 50 °C water bath heating and magnetic stirring and reacted for 12 h. The product was centrifuged and cleaned with ethanol to remove unreacted coupling agents. Finally, MPS-modified SBA-15 was dried under vacuum at 50 °C for 12 h.

In a 100 mL three-necked flask, 73.5 mg K₂Cr₂O₇ (0.5 mmol Cr(VI)), 86 μL MAA (1 mmol) and 108 μL 4-VP (1 mmol) were added into 40 mL mixed solutions (CH₃OH/DDW, v/v 3:1) and reacted for 2 h with magnetic stirring. Then, the pre-assembly solution was purged with N₂ and then 10 mg AIBN, 10 mL EGDMA and 0.1 g MPS-modified SBA-15 were dispersed into this suspension. Afterwards, under N₂ protection, the mixture was refluxed at 65 °C and stirred for 6 h. After polymerization, the polymer was cooled with an ice-water bath and collected by suction filtration with a Buchner funnel. To remove the residual monomers, the polymer was washed with CH₃OH/DDW (1:1, v/v) for three times. Then, the chelated Cr(VI) in the polymer was eluted with 2 mol/L HCl under magnetic stirring. ICP-AES was applied to determine the Cr content (Kachbi et al. 2022). The eluate was detected by ICP-AES until no template ion left in the eluate to ensure that Cr(VI) was eluted completely in the polymer. Finally, the polymer was washed with DDW until neutral and dried under vacuum at 60 °C for 24 h. As a comparison, the NIP was synthesized under the same conditions without K₂Cr₂O₇.

The static binding behaviors of Cr(VI)-IIP and NIP for Cr(VI) under different experimental conditions were investigated in batch adsorption experiments. The accurate amount of Cr(VI) in the solution was detected by ICP-AES. The detailed experimental procedure and calculation equations were shown in Supplementary Data.

The selective recognition ability of Cr(VI) by Cr(VI)-IIP and NIP was evaluated by adsorption tests in binary solutions containing Cr(VI) and interfering ions M such as Cr(III), Cu(II), Cd(II) and Ni(II). The selective adsorption experiments were studied with 10 mg/L of each metal ion and 10 mg Cr(VI)-IIP or NIP in 25 mL colorimetric tube under room temperature. In particular, diphenylcarbazide spectrophotometric method combined with ICP-AES detection technology was used to detect Cr(VI) and Cr(III) (Alahmad et al. 2019). First, the concentration of total Cr was measured by ICP-AES, and then diphenylcarbazide and Cr(VI) formed a stable purple-red chrome compound, and its absorbance was measured at the maximum absorption wavelength of 550 nm with a spectrophotometer, thereby obtaining Cr(VI) concentration (Mohana Rangan et al. 2021). Finally, the concentration of Cr(III) was thus obtained by subtraction of Cr(VI) concentration from the total concentration of Cr (Liu et al. 2013). Other ion concentrations were detected by ICP-AES. The relevant calculation formulas were given in Supplementary Data.

0.1 g of adsorbent (Cr(VI)-IIP/NIP) was added into a self-made dynamic column with a diameter of 1.0 cm and a length of 15 cm. The concentration of Cr(VI) in the inflow solution was 10 mg/L, and the flow rate was controlled at 1 mL/min. The outflow solution was collected every 5 min, and the content of Cr(VI) in the outflow solution was detected via ICP-AES. When the fixed bed reached the adsorption equilibrium, the saturated Cr(VI)-IIP was eluted continuously with 2 mol/L HCl at a flow rate of 1 mL/min until Cr(VI) was not detectable in the eluate. Then, the Cr(VI)-IIP was washed to neutral with copious amounts of DDW and continued for the next fixed-bed cycle. The dynamic adsorption capacity q_e, the total adsorption capacity q_total, the volume of solution flowing through the fixed bed at adsorption equilibrium V_e (L), empty bed contact time (EBCT) and the separation efficiency (Y%) were given in Supplementary Data.

To verify the accuracy and applicability of the experimental method, river water samples from the Yudai River (Zhenjiang, China) and tap water from our laboratory were selected for analysis. All samples were filtered off suspended particles through a 0.45 μm Millipore membrane, then acidified with nitric acid (pH of water samples adjusted to 2) and stored at 4 °C in darkness. After that, the concentrations of Cr(VI) in water samples and the recovery rate of Cr(VI)-IIP to Cr(VI) were analyzed by the standard addition method.

The influence of C₀ (initial concentration of Cr(VI)) on adsorption performance of adsorbents was investigated at room temperature. As presented in Figure 7(b), the adsorption capacities of the two adsorbents increased sharply at lower C₀, and then the adsorption capacities reached saturation and remained constant after reaching a certain concentration. For Cr(VI)-IIP, when C₀ reached 200 mg/L, the adsorption amount of Cr(VI)-IIP reached saturation, and its saturation adsorption amount was 96.32 mg/g. For NIP, when C₀ exceeded 100 mg/L, the adsorption amount of NIP reached saturation, and its saturation adsorption amount was 35.68 mg/g. The saturation adsorption amount of Cr(VI)-IIP was about 2.7 times more than that of NIP. This was mainly because the functional monomers 4-VP and MAA constituted a large number of Cr(VI) imprinted sites after coordination with template ions, which had strong specific recognition ability for Cr(VI). By comparison, there were no imprinted sites on NIP, and the NIP only had a simple electrostatic interaction with Cr(VI), so the adsorption amount was limited.

The temperature had a significant impact on the adsorption of Cr(VI). The adsorption capacities of the two adsorbents at different temperatures were investigated. As presented in Figure 7(c), both absorbents showed the best adsorption performance for Cr(VI) at 25 °C, and the adsorption capacities decreased gradually as temperature increased. This phenomenon might be due to the adsorption processes of adsorbents being exothermic, and the low temperature was conducive to adsorption. Therefore, the temperature of 25 °C was selected as the condition for a series of subsequent experiments.

Since the adsorption rate was a critical parameter to describe the adsorption kinetics, the adsorption capacities of the adsorbents at different times were investigated. Figure 7(d) showed that the adsorption amounts of both adsorbents increased gradually with the extension of adsorption time, and finally tended to equilibrium state. The adsorption equilibrium time of Cr(VI)-IIP and NIP were 70 and 100 min, respectively. Cr(VI)-IIP reached adsorption equilibrium significantly earlier than NIP, which was not only because of the imprinted sites formed by bifunctional monomers and template ions had strong coordination ability with Cr(VI), but also because of the easy contact of imprinted sites on the mesoporous materials SBA-15 with low mass transfer resistance. Moreover, due to the lack of Cr(VI) imprinted sites on NIP (Li et al. 2011), resulted in a lack of binding affinity for Cr(VI), which made the adsorption rate relatively slow.

The specific recognition and binding ability of Cr(VI)-IIP and NIP for Cr(VI) were investigated in binary mixed solutions of Cr(VI)/Cr(III), Cr(VI)/Cu(II), Cr(VI)/Cd(II) and Cr(VI)/Ni(II), respectively. These four metal ions were selected as competing ions for Cr(VI) based on the principle of the same element or the same charge. The experimental results were given in Table 2. The K_d value of Cr(VI) (K_d = 4,351 mL/g) was much larger than that of Cr(III) (K_d = 290 mL/g), illustrating that the prepared Cr(VI)-IIP could realize the separation of different valence states of Cr element. Meanwhile, the K_d of NIP for competing ions was close, leading to the k_NIP around 1. In contrast, the k_IIP for competing ions was far greater than 1. These data proved that Cr(VI)-IIP had a strong selective adsorption ability for Cr(VI) under the interference of other metal ions. Moreover, according to the values of k′, Cr(VI)-IIP had a higher selective adsorption capacity for Cr(VI) than NIP. These were because during the imprinted process, functional monomers 4-VP and MAA were coordinated to Cr(VI) to form abundant Cr(VI) imprinted sites, which had a strong specific recognition ability for Cr(VI) (Fei et al. 2021). However, functional monomers in NIP were polymerized randomly, resulting in a random arrangement of functional groups on the polymer surface (Guo et al. 2013). These functional groups do not have specific recognition ability for Cr(VI), so the NIP had no selective adsorption ability for Cr(VI).

The regeneration performance of the adsorbent was very important in fixed-bed experiment. Therefore, the adsorption–desorption experiments of Cr(VI)-IIP were also studied. Under the same adsorption and elution experimental conditions, the adsorbed Cr(VI) was eluted with 2 mol/L HCl, and the adsorption performance of the same batch of adsorbents filled into the dynamic column was studied. The results were shown in Table 3, In the first cycle: during the adsorption process, the t_b and t_e of the Cr(VI)-IIP in the fixed bed were 80 and 140 min, respectively, and the equilibrium adsorption amount was 10.02 mg/g. In the elution process, only 15 mL of 2 mol/L HCl was used to elute the adsorbent, and the elution efficiency was up to 98.2%. With the increase in the number of experimental cycles, it could be found that the penetration time and equilibrium time of the adsorbent in the fixed bed shorten slightly, but after five cycles, the equilibrium adsorption capacity was still up to 9.03 mg/g. In addition, after 20 mL 2 mol/L HCl was eluted, the elution efficiency was still up to 90.0%. The high elution efficiency might be determined by the mesoporous structure of the adsorbent. Large specific surface area and fast mass transfer rate enable the eluent to reach the imprinted sites quickly. Moreover, the eluent of the mobile phase in the fixed bed was more conducive to the elution of the target ion. Overall consideration, the Cr(VI)-IIP had strong regeneration ability and great potential for metal ion separation in fixed-bed experiments.

The dynamic column filled with Cr(VI)-IIP adsorbent was used to treat the actual water samples from tap water and Yudai River (Zhenjiang, China) successfully. From the experimental results in Table 4, the recoveries of Cr(VI) in the two samples ranged from 97.9 to 104.8%, and the relative standard deviations ranged from 0.1 to 2.3%, indicating that the potential application of the prepared novel Cr(VI)-IIP combined with ICP-AES for the separation and detection of trace metal ions in water samples.

The adsorption performance of Cr(VI)-IIP prepared in this experiment was compared with that of other Cr(VI) adsorbents, such as equilibrium time, adsorption capacity and selectivity. As shown in Table 5, first, the Cr(VI)-IIP had a great advantage in selectivity compared with other common Cr(VI) adsorbents (Faleschini et al. 2023; Li et al. 2023c, 2023d; Mao et al. 2023), this was attributed to the strong specific recognition ability of Cr(VI) by the imprinted cavity left after the functional monomers chelated with the target ion during the preparation of Cr(VI)-IIP, illustrated that the effectiveness of ion imprinting technique. Afterwards, compared with Cr(VI)-imprinted polymers prepared from single functional monomer such as 4-VP (Nchoe et al. 2020), 3-aminopropyltriethoxysilane (APTES) (Zhang et al. 2022) and 1-vinylimidazole(1-VI) (Hassanzadeh et al. 2018), Cr(VI)-IIP showed a large improvement in adsorption capacity and selectivity. Finally, compared to Cr(VI)-imprinted polymers prepared with bifunctional monomers such as 4-VP/methyl methacrylate (MMA) (Neolaka et al. 2018) and 4-VP/2-hydroxyethyl methacrylate (HEMA) (Taghizadeh & Hassanpour 2017), the advantages of Cr(VI)-IIP were mainly reflected in the higher adsorption amount and superior selectivity, though its adsorption equilibrium time was not the shortest. Clearly, the advantage here was the establishment of the synergy between the pre-assembly system of bifunctional monomers and surface imprinting technology based on mesoporous silicon material SBA-15.

A novel Cr(VI)-IIP was synthesized on the surface of SBA-15 using MAA and 4-VP as bifunctional monomers and Cr(VI) anion as template ion by surface thermal initiation imprinting technique. A series of static adsorption experiments showed that Cr(VI)-IIP exhibited the best adsorption performance when the solution pH was 2 and the temperature was room temperature, and the saturation adsorption capacity was obviously higher than that of NIP under the same conditions. The adsorption behavior of Cr(VI)-IIP was consistent with the Langmuir isotherm model and the PSO model, which indicated that the adsorption process for Cr(VI) tended to monolayer adsorption and was a chemical adsorption process. In addition, in the selective adsorption experiments, Cr(VI)-IIP showed a strong selective adsorption ability for Cr(VI) in the presence of competing ions such as Cr(III), Cu(II), Cd(II) and Ni(II), which demonstrated that Cr(VI)-IIP could be applied in complex aqueous environments to achieve efficient separation of Cr(VI), while enabling the separation of different valence states of elemental Cr. Finally, in the fixed-bed experiment, Cr(VI)-IIP showed ideal dynamic adsorption performance and high regeneration ability for Cr(VI), indicating that Cr(VI)-IIP possessed a fast adsorption rate and good stability. This study provided an efficient and novel method to precise identification and separation of Cr(VI), and the prepared Cr(VI)-IIP had enormous potential application in the preconcentration and separation of Cr(VI) from aqueous solution.

This work was financially supported by National Natural Science Foundation of China (No. 22178151) and Postgraduate Research Innovation Program of Jiangsu Province (No. SJCX22-1857).

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.