Preparation of grafting copolymer of acrylic acid onto loess surface and its adsorption behavior

Among all the heavy metals, lead ions (Pb²⁺) are one of the most common and abundantly available pollutants in the environment (Arfin & Tarannum 2019). Pb²⁺ is nonbiodegradable and easily accumulated in living organisms for a long time (Zhang et al. 2011; Hashemzadeh et al. 2019), causing threatening diseases. In general, most of the current heavy metal removal technologies, including chemical precipitation (Fu et al. 2012), ion-exchange (Lasheen et al. 2016), reverse osmosis, and electrochemical and membrane process (Song et al. 2011) are costly and inefficient for pollutants (Liu et al. 2018). Hence, the effective and economic removal of heavy metals from wastewater is exceedingly important. In the past decades, adsorption technology was considered as the most efficient and inexpensive method compared to the other methods described above.

Various kinds of competent adsorbents have been reported, such as biosorbents (Shen et al. 2009) and activated carbon (Gong et al. 2007), maize husk adsorbent (Indah et al. 2016) and clay minerals like montmorillonite (Kumar et al. 2012; Yang et al. 2020), bentonite (Ecer et al. 2018), and apatite (Li et al. 2019) etc. Loess soil is widely distributed throughout Asia, Russia, the Middle East, and North America, and is especially abundant in China, covering an area of approximate 640,000 km² (Wang et al. 2009). As a wastewater treatment material, loess is an inexpensive and easily obtained eco-friendly silicate mineral that has been modified by many existing methods to improve its adsorption performance. However, in a previous study, the polymers were physically composited with loess (He et al. 2012; Shen et al. 2019), and therefore, the disadvantages were that the proportion of loess in composite is too low and its cost is high.

In this study, a novel surface grafted functional copolymer (LC-PAVP) was synthesized by grafting copolymerization, which produced to a remarkable adsorbent that was used to remove Pb²⁺ from aqueous solution. During preparation, the proportion of clay in the composite material was increased and the amount of polymer was reduced, and thus the production cost of polymer adsorbent was significantly decreased. As a new absorbent material, LC-PAVP was used in wastewater treatment. The optimal conditions to remove Pb²⁺ and adsorption mechanism of LC-PAVP were also investigated.

Loess clay (LC) was collected in Lanzhou, China. After being ground, it was sieved with a 100 mesh, and then stored in a desiccator for further use. Hydrochloric acid (AR, HCl) was supplied by Beijing Chemical Plant. γ-Methacryloxypropyl trimethoxysilane (AR, KH-570) was supplied by Shanghai Jingchun Reagent Co. Ltd. Acrylic acid (AR, AA) was provided by Tianjin BASF Chemical Co. Ltd. N-vinyl pyrrolidone (AR, NVP) was provided by Xuzhou Zhuoyuan Chemical Co. Ltd. N,N-methylenebisacrylamide (AR, MBA) was purchased from Shanghai Zhongqin Chemical Reagent Co. Ltd. Acetic acid (AR, CH₃COOH) was supplied by Shanghai Boer Chemical Reagent Co. Ltd. Ethanol (AR, EtOH) was supplied by Tianjin Fuyu Fine Chemical Co. Ltd. Sodium hydroxide (AR, NaOH), ammonia hydroxide (AR, NH₃·H₂O) and potassium persulfate (AR, KPS) were purchased from Yantai Shuangshang Chemical Industry Co. Ltd. Distilled water was used in all the experiments.

First, LC was pretreated with HCl. The dried LC was ground and sieved with a 100 mesh, 10.00 g LC was added to 100 mL HCl solution (4 mol·L⁻¹) at 80 °C and stirring for 2 h. Then, it was cooled to room temperature. The pretreated LC was filtered and washed with distilled water until the eluant became neutral. The acidifying LC (HLC) was dried at 60 °C for 5 h with 76.5% of yield.

Second, the surface of the loess was modified with KH-570. In a three-necked flask, HLC (5.00 g), H₂O (10 mL) and ethanol (EtOH) (30 mL) were mixed with stirring for 1 h. The pH value adjusted to 3.0–4.0 by acetic acid (HOAc) solution. Adding KH-570 (1.00 g), the mixture was stirred for 30 min at room temperature. Then, the pH value of the mixture was adjusted to 9–10 with ammonia hydroxide (NH₃·H₂O), heated to 80 °C and stirring for 3 h. The product was filtered and washed with EtOH three times. After vacuum drying at 50 °C for 5 h, silane coupling agent modified loess (KH-LC) (4.87 g) was obtained with 81.20% of yield.

Finally, the LC was modified by surface grafting copolymerization. KH-LC (12.50 g), distilled water (15 mL) and NaOH (0.56 g) was added to a three-necked flask with stirring rod, condenser and nitrogen pipe and stirred. Then, AA (1.45 g) was added drop by drop and stirred mechanically at room temperature for 30 min. NVP (1.1 g) and MBA (0.13 g) were dispersed in distilled water (4 mL) and added into above solution and mechanically stirred for 30 min at 45 °C under nitrogen atmosphere. Then, KPS (0.15 g) was added and the reaction was continued at 80 °C by stirring for 30 min. The obtained product was washed with distilled water three times, and dried at 50 °C, which provided the loess-based grafting copolymer (LC-PAVP) with 96.30% of yield.

LC-PAVP was characterized and analyzed using Fourier-transform infrared spectroscopy (FT-IR), thermal gravimetric analysis (TG), scanning electron microscopy (SEM). Before measuring FT-IR, the samples were ground and sieved with 100–200 mesh. Then, it was mixed with potassium bromide (ratio being 1:100) and compressed to a transparent sheet.

AA and NVP were selected as comonomers because AA is a vinyl monomer with a lower cost and is less polluting to the environment. Due to the carboxyl group (–COOH) in AA, the pH value of its polymers had an effect on the degree of ionization, which further affected the polymerization rate. The homopolymerization of AA is actually the copolymerization of AA and carboxylate (Scott & Peppas 1997), which can be used to prepare microspheres (Yu et al. 2014), gels, and so on (Ryu et al. 2016). AA can therefore be used for two, three, and multicomponent copolymerization with many monomers (Poggi et al. 2015; Yan et al. 2015). Acrylic copolymers have been widely used in many fields, such as dyes, water absorbent resins, scale inhibitors, coatings etc. (Cilurzo et al. 2010). NVP, a type of lactam compound containing vinyl group, is easily hydrolyzed and polymerized (Parambil et al. 2012). The NVP homopolymer and copolymer show good biocompatibility and low toxicity (Reyes et al. 2014), which was widely used in food, health care, cosmetics, and other novel materials (Guinaudeau et al. 2012; Narayana Reddy et al. 2012; Rajeswari et al. 2012). A number of copolymers of AA and NVP were prepared (Narayana Reddy et al. 2012; Jin et al. 2013; Ding et al. 2014).

The process for preparation of LC-PAVP is shown in Figure 1. First, the acidifing pretreated loess (LC) was modified by KH-570, which introduced vinyl groups (–CH=CH₂) onto the surface of the loess particles. Then, using AA and NVP as functional comonomers and MBA as a cross-linking agent, loess particles were modified by surface grafting copolymerization, which produced loess-based grafting copolymer (LC-PAVP).

We found that the copolymer could be grafted onto loess particles (KH-LC) using different monomers, and the results are shown in Table 1. However, the yields largely decreased when other monomers (such as 2-hydroxyethyl methacrylate (HEMA) or maleic anhydride (MA)) were used as monomers. It was also found that the ratio of LC:comonomers should be more than 5:1.

The obtained loess-based grafting copolymer (LC-PAVP) was then characterized by SEM, TG, and FT-IR, and compared with intermediates (KH-LC) and materials (LC).

The surface morphology of LC-PAVP was observed by SEM and compared with its materials. The results are shown in Figure 2. It was found that the surface of loess particles (Figure 2(a)) was uneven, porous, very rough and had many small surface cracks. After being modified by KH-570 (Figure 2(b)), some particles were stacked on its surface, and there was not much difference. After modification by grafting copolymers of AA and NVP (Figure 2(c)), the surface of the loess particles became smoother, and a copolymer film was formed on the loess surface that wrapped around the interparticle gaps. The results showed that AA and NVP were grafted by copolymerization onto the surface of loess particles.

The FT-IR spectra of LC-PAVP and its materials (LC, KH-LC) are shown in Figure 3. In LC, the absorption peak of the Si–O–Si stretching vibration was near 1,090 cm⁻¹, and the flexural vibration absorption peak of Si–O–Si was near 559 cm⁻¹ and 497 cm⁻¹. The characteristic peaks at 3,626 cm⁻¹ and 3,450 cm⁻¹ can be assigned to the stretching vibration of surface hydroxyl groups (–OH). Flexural vibrations of –OH was near 1,635 cm⁻¹ (Tang et al. 2009). In KH-LC and LC-PAVP, the main silicate characteristic peaks of the loess (LC) were unchanged. In KH-LC, the characteristic absorption peak of methylene in KH-570 was near 2,950 cm⁻¹. The peak at 1,716 cm⁻¹ was the characteristic absorption peak of the ester carbonyl group (Liu et al. 2015). In LC-PAVP, the characteristic peaks near 1,716 cm⁻¹ and 1,570 cm⁻¹ were attributed to symmetric and antisymmetric stretching vibration absorption peak of C=O in COOH (Jin et al. 2013; Ma et al. 2013), and the peaks around 1,550 cm⁻¹ was attributed to N–H in NVP (Roy et al. 2013). The bending vibration peak of C–H was near 1,410 cm⁻¹ (Rafiee et al. 2016). The results showed that AA and NVP were successfully grafted by copolymerization onto the surface of loess.

In order to study the thermal stability of LC-PAVP, TG analysis was carried out and compared with LC and KH-LC. The results are shown in Figure 4. In LC, a weight loss zone (2%) appeared near 100 °C, which was the water loss of silicate. Water loss was not evident in KH-LC and there was a 3.2% in total weight loss. That meant LC was successfully modified by KH-570. In LC-PAVP, the copolymers of AA and NVP were decomposed between 200 °C and 500 °C, and approximately 10.4% of the weight loss was due to the copolymer chain. Homopolymers of AA and NVP were therefore washed out during preparation, indicating that the copolymer was successfully grafted onto the surface of loess.

The obtained LC-PAVP was used to remove Pb²⁺ because it has many hydroxyl groups (–OH) and carboxylates (–COO⁻). Some reported inorganic adsorbing materials, such as zeolite, kaolinite clay, hydroxyapatite, and pine cone activated carbon, were compared with LC-PAVP. The adsorption capacities of different adsorbent materials for removing Pb²⁺ are shown in Table 2. Compared with other inorganic adsorbing materials, the obtained LC-PAVP is important for the removal of Pb²⁺ from aqueous solution.

Other parameters related to the effect on the adsorption process, such as initial concentration of Pb²⁺, adsorbent dosage, time, pH value, and temperature, were also investigated.

The effect of initial Pb²⁺ concentrations was investigated at 25 °C using 0.05 g of LC-PAVP as the adsorbent. Adsorption capacity of LC- PAVP with various initial concentration of Pb²⁺ is shown in Figure 5. If the concentration of Pb²⁺ was lower than 200 mg·L⁻¹, the removal rate was above 98%, and decreased in a small range. The removal rate of LC-PAVP was 98.20% when the concentration of Pb²⁺ was 200 mg·L⁻¹. Subsequently, the removal rate decreased rapidly with the increase in concentration. This may be because the LC-PAVP had enough active sites to bind with Pb²⁺ when Pb²⁺ concentration was low. With increasing Pb²⁺ concentration, the active site of LC-PAVP decreased (Sun et al. 2010).

The effect of the amount of adsorbent (LC-PAVP) on the Pb²⁺ adsorption from 50 mL of Pb²⁺ solution (200 mg·L⁻¹) was investigated at 25 °C, and the results are shown in Figure 6. With an increase in the sorbent dosage from 0.01 to 0.08 g, the removal rate of Pb²⁺ also increased noticeably. When LC-PAVP dosage increased from 0.08 to 0.20 g, the removal rate increased slightly. The reason is that all active sites on the adsorbent surface are occupied with increasing coverage, the fraction of the bare surface rapidly diminishes, and Pb²⁺ ions have to compete for the adsorption site. An increase in adsorbent dosage will therefore not provide higher uptake of Pb²⁺ (Aroua et al. 2008). When the dosage of LC-PAVP increased to 0.08 g, the removal rate of Pb²⁺ reached 99.49%. In relation to cost, the optimal dosage of LC-PAVP was 0.08 g when used in the treatment of Pb²⁺ solutions of 200 mg·L⁻¹ concentration.

The effect of adsorption time on the removal rate of Pb²⁺ was studied for 50 mL of Pb²⁺ solution (200 mg·L⁻¹) at 25 °C using 0.08 g of LC-PAVP as the adsorbent, and results are shown in Figure 7. Initially, the adsorption rate of Pb²⁺ on LC-PAVP was relatively fast, reaching up to 98.64% within 10 min. The reason was that Pb²⁺ in the solution formed an initial complex with functional groups in the hydroxyl on the surface of LC. Afterwards the adsorption rate increased slowly and gradually. After 60 min, the removal rate reached 99.47% and remained flat. These results may be due to the initiation of LC-PAVP adsorption on Pb²⁺ mainly occurring on the outer surface of the composite, which can be completed in a short time. The repulsive force of Pb²⁺ increased in line with an increase in adsorption capacity. The resistance of free Pb²⁺ to LC-PAVP micropores was also enhanced, and it took longer to reach adsorption saturation (Doğan et al. 2009). As a consequence, the optimal adsorption time was 60 min.

The influence of temperatures on adsorption of Pb²⁺ solution (200 mg·L⁻¹) was studied using 0.08 g of LC-PAVP as the adsorbent, and the results are shown in Figure 8. The removal rate of Pb²⁺ was above 99.0% at both high and low temperatures, and the highest removal rate of Pb²⁺ was 99.70% at 15 °C. The influence of temperature on the whole adsorption process is not significant. It indicated that the adsorption material was very stable at different temperatures. Adsorption of LC-PAVP on Pb²⁺ can be applied to a wide scope of temperatures.

The pH value of the solution was an important factor that may affect adsorbate intake and the degree of ionization that caused the variation in reaction kinetics and equilibrium of the adsorption process (Jin et al. 2008). The effect of pH value on the adsorption of Pb²⁺ was measured at 25 °C for 60 min using 0.08 g of LC-PAVP as the adsorbent, and the results are shown in Figure 9. With the increase in pH value from 2.0 to 5.0, the removal rate of Pb²⁺ increased from 95.91% to 99.32%. It was demonstrated that H⁺ and Pb²⁺ compete for adsorption sites on the LC-PAVP surface at lower pH values. When the pH value reached 5.44, the removal rate was 99.46%, which was the highest in this experiment. When the pH value was more than 6.0, Pb²⁺ began to form lead hydroxide (Pb(OH)₂)), and the precipitates on the surface of LC-PAVP prevented Pb²⁺ from contacting with the active site, which decreased the removal rate (Farooq et al. 2010). LC-PAVP can therefore be used in acid environments.

Using Freundlich and Langmuir isothermal models, the nonlinear fitting is made on the relationship between concentration and adsorption capacity. Results are shown in Figure 10 and Table 3. It was closer to the Freundlich equation (Figure 10(b)), and the correlation coefficient got to 0.9858. The maximum adsorption capacity was 68.75 mg·g⁻¹, which fits the Langmuir model in the experiment.

Nonlinear equations in the measurement of model parameters were more effective than linear equations (Lin & Wang 2009). The results are shown in Figure 11 and Table 4. It was obvious that the pseudo-second-order equation (Figure 11(b)) was better in describing the adsorption kinetics, as shown by the higher determination coefficient (R² = 0.9292) value, indicating that a chemical reaction is the rate-controlling step (Ho & McKay 1999). The results showed that the chemical adsorption played a primary role in the adsorption of LC-PAVP on Pb²⁺.

To sum up, a novel loess-based polymer absorbent (LC-PAVP) using AA and NVP as functional monomers was successfully synthesized through grafting copolymerization, which significantly reduced the proportion of synthetic polymers. The effects of adsorbent dosage, contact time, system temperature, and initial pH value on the adsorption performance of Pb²⁺ were systematically investigated. The results revealed that LC-PAVP had remarkable capacity of adsorbing Pb²⁺ and had a removal efficiency up to 99.49%. The adsorption process can be explained with the pseudo-second-order type kinetic model and the thermodynamics were best fitted with the Freundlich model. It was concluded that the obtained adsorbent would be a promising, high-efficient and economic material of removing Pb²⁺ in wastewater treatment processes.

The work was supported by the National Natural Science Foundation of China (Grant No. 21865030, 21364012).

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