Acrylic acid (AA), acrylamide (AM), ammonium persulfate (APS), N,N-methylene bisacrylamide (MBA) were provided by Kelong Chemical Reagent Plant (Chengdu, China). Hydrochloric acid (HCl), sodium hydroxide (NaOH), copper sulfate (CuSO₄), were. purchased from Sinopharm Chemical Reagent CO., Ltd. And all reagents that used were analytical grades.

Firstly, 5 mL AA was neutralized with 28.3 mL of 10 wt.% NaOH solution to reach 100% neutralization degree. Then, 2 g of AM and 0.3 g of MBA were added under the magnetic stirring at 70°C. Thereafter, the dissolved APS solution (0.2 g of APS to 5 mL distilled water) was added to the mixed solution along with heating at 70°C. Once the mixed solution was turned into a solid gel, the magnetic stirrer was turned off, and the reaction was continued for 3 h at 70°C. Finally, the gel was cut into granular particles, washed with absolute alcohol for three times to remove unreacted monomers, and dried at 70°C for 24 h. Through smashing and sieving, P(AANa/AM) gel resin had a particle size within 0.45 mm to 1 mm.

Three indexes, including adsorption capacity, released sodium ion amounts by resin, and water absorbency, were measured together [17]. Firstly, CuSO₄ was dissolved in distilled water to prepare 1000 mg/L reserve solution and diluted to different concentrations according to experimental requirements. Adsorption experiments were carried out by adding different samples to 100 mL CuSO₄ solution with the initial concentration of 250 mg/L. Samples were mechanically agitated at 180 rpm at 25°C. After reaching the equilibrium state, P(AANa/AM) were separated by stainless steel mesh. The concentrations of copper and sodium ions in residual solution were detected by a flame adsorption atomic spectrophotometer (AAS, AA-7050, Beijing East & West Analytical Instruments CO., Ltd., China). The filtered adsorbents were weighted by a analytical balance. The adsorption capacity, removal rate to copper, released sodium ion amounts by resin, and water absorbency of P(AANa/AM) were calculated by the following Eq. (1), Eq. (2), Eq. (3), and Eq. (4). To check the reproducibility, each test specimen was prepared in triplicate.

where Q_ad (mg/g) and η mean the adsorption capacity and removal rate to Cu²⁺, respectively. Q_RNa (mg/g) means the released sodium ion amounts by resin. Q_ab (mg/g) means the water absorbency for resin in aqueous solution. C₀ (mg/L) and C_e (mg/L) are the initial and equilibrium concentrations of Cu²⁺, respectively. V (L) is the volume of CuSO₄ solution. m₀ (g) and m₁ (g) are the mass of the dried adsorbent and the filtered adsorbent [18, 19].

The relationship between adsorption capacity, released sodium ion amounts by resin, and water absorbency was evaluated through continuous adsorption. P(AANa/AM) of 0.1 g was added into 100 mL of 50 mg/L CuSO₄ solution, and stirred for 1 h. The adsorbents were filtered and reused for a fresh adsorption in another 50 mg/L CuSO₄ solution. This process was repeated until the concentration of Cu²⁺ in residual solution did not decrease. The adsorption capacity, released sodium ion amounts by resin, and water absorbency values in each process were calculated and evaluated of correlation analysis by SPSS.

The swelling characteristic of P(AANa/AM) in regeneration experiments was observed. A fixed amount of 0.1 g of P(AANa/AM) was immersed sequentially in 100 mL of 250 mg/L CuSO₄ solution, 0.1 mol/L HCl solution, and 0.1 mol/L NaOH solution. The bottle was placed in a shaker for 1 h at 180 rpm. And the end of the experiment, the solid adsorbent was filtered and then immersed in distilled water. Water absorbency of adsorbent in each steps were measured.

The surface metal ion contents were scanned by Bruker EDX detector and coated with Au vapor. The accelerating voltage was set to 15 kV. The working distance was set to 12.5 mm. The macro appearance of adsorbents in each batch experiments was observed by an industrial camera with light source and 30 times magnification (GP-300C, Kunshan Gaopin Precision Instrument CO., LTD., China).

Fig. 1(a) illustrated the influence of initial copper ion concentration on heavy metal adsorption capacity, released amount of sodium ions, and water absorbency. Heavy metal adsorption capacity increased dramatically with increasing the initial copper ion concentration. The maximum adsorption capacity was 201.63 mg/g (converting into 3.17 mmol Cu²⁺/g), in which the saturation threshold of initial copper ion concentration was 250 mg/L (converting into 7.41 mmol Na⁺/g). The adsorption saturation state for P(AANa/AM) gel resin was reached when initial copper ion concentration was higher than that of 250 mg/L. Otherwise, the functional adsorption sites in adsorbent were not all occupied by heavy metals. The curve trend of the released sodium amounts was same as adsorption capacity. And the maximum value of the released sodium amounts was 170.48 mg/g, which was derived from the diffusion of −COONa in adsorbent. Significantly, water absorbency of gel adsorbent was synchronously measured in adsorption experiment. When initial Cu²⁺ concentrations were 10, 250, and 500 mg/L, water absorbency values of adsorbent were 39.04, 3.56, and 3.67 g/g, respectively. Fig. 1(b) shows the relativity of adsorption capacity, released sodium ion amounts, and water absorbency for P(AANa/AM) gel adsorbent under different initial Cu²⁺ concentrations. Strong positive correlation between adsorption capacity and the released sodium ion amounts indicated that copper ion in solution was exchanged with sodium ion in adsorbent. Furthermore, strong negative correlation between adsorption capacity and water absorbency was found, which indicated water absorbency could be served as an indicator to determine the endpoint of adsorption saturation for gel adsorbent [20–22].

Fig. S1 showed the morphologies of P(AANa/AM) adsorbent after immersing in CuSO₄ solutions with different initial concentrations. The original morphology of P(AANa/AM) gel resin was dried particle with the size of 0.45 to 1 mm. The swelling behavior was observed upon immersion in a 10 mg/L CuSO₄ solution. According to our earlier research, the −COONa functional group was classified as a negative group that exhibited the capacity to adsorb cationic heavy metal ions [23]. Herein, the hydrophilicity of −COONa for gel adsorbent should be taken into account simultaneously in heavy metal adsorption experiments. The swollen particle was the P(AANa/AM) adsorbent’s current state when adsorption saturation was not attained. Water molecules were now dispersed in the inner area of resin. Once the initial copper ion concentration exceeded 250 mg/L, the massive shrinking phenomenon was noticed. These findings supported the theory that gel resin would likewise exhibit swelling behavior during the heavy metal adsorption process. [24–26].

Fig. 2 illustrated the influence of P(AANa/AM) gel resin adding dose on removal rate, adsorption capacity, released sodium ion amounts, and water absorbency. Removal rates to copper ion was increased dramatically with increasing the adding dose of adsorbent, and the maximum removal rate was 94.08%. This observation can be illustrated by that the increase of adding dose of adsorbents may multiply the amount of the adsorption active sites, hence leading to the increase of removal rates. When the adding dose of adsorbent was lower than 0.10 g, adsorption capacity, released sodium ion amounts, and water absorbency values remained roughly constant of 200.83 mg/g, 177.93 mg/g, and 3.35 g/g, respectively. In these moments, adsorption saturation states were all reached for P(AANa/AM). With further rising in adding dose, water absorbency was increased dramatically, which was mainly related to the binding degree of −COONa to heavy metals. The remaining vacant sites in adsorbent were not all occupied by copper ions, but to combine with water molecules [27, 28]. It was related that the shrank phenomena in Fig. S2 were observed when the adding dose of P(AANa/AM) was lower than 0.12 g. The gel adsorbent swelled with further increasing adding dose. Thereafter, the hydrophilicity property of −COONa should be attached to study the heavy metal adsorption process.

Fig. 3 illustrated the influence of reaction time on adsorption capacity, released sodium amounts, and water absorbency in CuSO₄ solutions under different initial concentrations of 20, 100, 250, and 500 mg/L. It can be seen that adsorption capacity and released Na⁺ amounts increased rapidly during the first 15 min, then plateaued by 40 min. The number of the adsorbed copper ion by adsorbent was increased, which was accompanied with the releasing process of Na⁺ from adsorbent into the aqueous solution. To understand water molecules releasing mechanisms during heavy metal adsorption process, water absorbency was measured simultaneously. Swelling behavior of P(AANa/AM) gel resin was also occurred in heavy metal solution. P(AANa/AM) swelled quickly after immersing the dried adsorbent particles into CuSO₄ solution. At 7.5 min, water absorbency values were 35.95, 32.66, 27.29, 18.51, 12.42 g/g in distilled water, CuSO₄ solutions with initial concentration of 20, 100, 250, and 500 mg/L, respectively. There was a gradual decrease of water absorbency during the subsequent soaking process. At 60 min, water absorbency values were 39.82, 38.67, 23.48, 3.71, and 3.69 g/g. Such behavior is due to a synergistic effect of −COONa to combine heavy metal and water molecules. As a polyelectrolyte, water molecules could easily release into the inner area of gel resin and combine with −COONa groups through hydrogen bond. The stretching network polymer chains provide a sufficient space to accommodate water molecules. Further immersing adsorbents in CuSO₄ solution, water absorbency was decreased with the increase of adsorption capacity. This could be attributed to the fact that the copper ions and water molecules competed for −COO⁻ sites for resin in heavy metal adsorption process.

In order to further confirm the relationship of adsorption capacity, released Na⁺ amounts, and water absorbency in heavy metal adsorption process, the continuous adsorption experiment in initial 50 mg/L CuSO₄ solution was conducted, and the results were showed in Fig. 4 and Fig. S3. In first immersion in 50 mg/L CuSO₄ solution, values of adsorption capacity, released Na⁺ amounts, and water absorbency were 33.85 mg/g, 36.64 mg/g, and 34.85 g/g, respectively. After immersing in fresh 50 mg/L CuSO₄ solutions for five times, the saturation state was reached. And the values of adsorption capacity, released sodium ion amounts, and water absorbency were 199.01 mg/g, 170.80 mg/g, and 3.72 g/g. The reduction in water absorbency accompanied with the increase of adsorption capacity and released sodium ion amounts. A significant positive correlation was found between adsorption capacity and released sodium ion amounts. This result revealed that Na⁺ was released from adsorbent particles into the outer solution. Meanwhile, adsorption capacity had a significant negative correlation with water absorbency, which indicated that water molecules were also lost during the heavy metal adsorption process [29, 30]. Such phenomena are due to the hydrophilic property of −COONa groups in the crosslinked net polymer chains. The quickly swelling behavior would be occurred once soaking the dried gel adsorbent into aqueous solution. Synchronous releases of sodium ion and water molecule were accompanied with the adsorption of copper ions. Hence, on the basis of the exchanging process between heavy metals and sodium ions, water molecule acts as a co-transported species in the ion exchange process [31–33]. The dynamic swelling and shrinking changes should be added to study adsorption mechanism of gel adsorbent in heavy metal solution.

Prior research on the regeneration experiment mostly assessed the sustainability of adsorption capacity or removal rate for adsorbent [34–36]. It has not been possible to observe the morphological changes that occurs over the whole soaking cycle of an adsorbent in acid or alkali solutions. Fig. 5 and Fig. 6 illustrate water absorbency and morphologies of P(AANa/AM) gel resin during the whole regeneration process. Soaking the dried adsorbent into 250 mg/L CuSO₄ solution, water absorbency of adsorbent was 3.68 g/g. Later immersing in 0.1 mol/L HCl solution and distilled water, water absorbency were 0.97 and 1.26 g/g. All of the adsorbed Cu²⁺ in adsorbent was rinsed out in this phase. The resin with −COOH groups exhibited the extremely low water absorption capacity. Finally, recovery of −COONa functional groups was achieved following immersion in 0.1 mol/L NaOH solution and distilled water, which can be confirmed by the listed EDS analysis results in Fig. 5. The final state of P(AANa/AM) gel resin was the swelled particle with 39.51 g/g water absorbency. Immersing in 250 mg/L CuSO₄ solution once again, water loss phenomenon was observed for resin. Morphology observations in Fig. 6 were compatible with the water absorbency results. The swelling and shrinking behavior for gel adsorbent should be recommended in column filter application. The loading height or quality of gel adsorbent must be correlated by water absorbency.

Fig. 7 illustrates the heavy metal adsorption process of gel resin containing −COONa groups. Crosslinked polymer backbones provide the insoluble but swelling property for gel resin. The −COONa groups are linked in the polymer chains, in which the mobility is limited. Attributing to the hydrophilic property of −COO⁻, gel resin can absorb water molecules through hydrogen bond. Thus, Immersing dried gel resin into heavy metal solution, the preferential behavior of absorbing water is firstly observed. The polymer chains in the swollen particles generate the elastic stretch synchronously, reflecting in the increase of the particles size. Meanwhile, the ionization of −COONa is accompanied with the quick expanding behavior. Part of Na⁺ was released into the outer solution.

The −COONa group in the polymer network chains could also adsorb heavy metals. In case −COO⁻ group is not combined with heavy metal, −COO⁻ group in the polymer chains combines with water molecule through hydrogen bond. The maximum water absorbency would be obtained in the early stage. As the adsorption reaction proceeding, the adsorption active sites in −COO⁻ groups would gradually be occupied by copper ions. sodium ion in resin was exchanged and released into the outer solution. In fact, the electrostatic attraction theory explains the adsorption force between anionic −COO⁻ group and cationic copper ion. Ion exchange theory emphasizes the mass migration process to maintain zero charge in adsorbent and aqueous solution. The orbital hybridization theory perfectly interprets the combining form for unequal charge number between −COO⁻ and copper. To achieve a stable conformation, divalent copper ion needs to bind with four numbers of −COO⁻ groups to develop a tetrahedral structure through dsp² orbital hybridization. Owing to this special ionic crosslinkage, the space area for accommodating water molecules is decreased. Thus, water absorbency is decreased along with the adsorption process. When adsorption saturation is reached, all water molecules and sodium ions are released. The final state of the adsorbent is the shrinkage particle. Thus, the absorption and release of water by gel resin represent the participation of the water molecules during heavy metal adsorption process.